Monetised Benefits and Costs Manual

f
Copyright information
This publication is copyright © Waka Kotahi NZ Transport Agency. Material in it may be reproduced
for personal or in-house use without formal permission or charge, provided suitable acknowledgement
is made to this publication and Waka Kotahi NZ Transport Agency as the source. Requests and
enquiries about the reproduction of material in this publication for any other purpose should be made
to the Manager, Corporate Information and Records Management, Waka Kotahi NZ Transport
Agency, Private Bag 6995, Wellington 6141.
The permission to reproduce material in this publication does not extend to any material for which the
copyright is identified as being held by a third party. Authorisation to reproduce material belonging to
a third party must be obtained from the copyright holder(s) concerned.
Disclaimer
Waka Kotahi NZ Transport Agency has endeavoured to ensure material in this document is
technically accurate and reflects legal requirements. However, the document does not override
governing legislation. Waka Kotahi does not accept liability for any consequences arising from the use
of this document. If the user of this document is unsure whether the material is correct, they should
refer directly to the relevant legislation and contact Waka Kotahi.
More information
Waka Kotahi
First published August 2020
Version 1.6.1, June 2023 (Version 1.6 originally published April 2023)
ISBN 978-1-98-856173-8 (online)

20-151
If you have further queries, call our contact centre on 0800 699 000 or write to us:
Waka Kotahi NZ Transport Agency
Private Bag 6995
Wellington 6141
This document is available on the Waka Kotahi website at
https://www.nzta.govt.nz/resources/monetised-benefits-and-costs-manual
Monetised benefits
and costs manual
June 2023 │ Version 1.6.1

Back to contents page >>
Contents
List of tables ......................................................................................................................................6
List of figures ..................................................................................................................................12
Introduction .....................................................................................................................................13
Purpose of this manual ..................................................................................................................13
Who this manual is for ...................................................................................................................13
1. Concepts ...................................................................................................................................14
1.1 Social cost–benefit analysis ............................................................................................14
1.2 Equity or distributional effects of land transport initiatives ..............................................14
1.3 Steps in BCR calculation .................................................................................................15
1.4 Counterfactuals and the do-minimum .............................................................................16
1.5 Alternatives and options ..................................................................................................17
1.6 Period of analysis ............................................................................................................17
1.7 Benefits ...........................................................................................................................18
1.8 Costs ...............................................................................................................................20
1.9 Discounting......................................................................................................................22
1.10 Benefit–cost ratios and other appraisal tools ..................................................................23
1.11 Sensitivity analysis ..........................................................................................................24
2. Demand estimation and mode share .....................................................................................27
2.1 Demand estimates and importance to economics ..........................................................27
2.2 Key concepts ...................................................................................................................28
2.3 Key industry guidance and references ............................................................................31
2.4 Methods for demand estimation ......................................................................................31
2.5 Definitions of transport model types ................................................................................32
2.6 Future guidance development areas ..............................................................................33
2.7 Drivers of trip-making and mode share ...........................................................................34
2.8 Land use projections and regional planning structures ..................................................35
2.9 Mathematical methodologies and elasticities ..................................................................36
2.10 Demand estimation and transport modelling in New Zealand ........................................36
2.11 Optimism bias .................................................................................................................37
2.12 Factors and considerations influencing demand estimation ...........................................38
2.13 Fixed trip matrix and variable trip matrix assessments ...................................................42
2.14 Demand estimation uncertainty ......................................................................................44
2.15 Checks, reporting and reviewing .....................................................................................46
3. Benefits .....................................................................................................................................49
3.1 Impact on social cost of deaths and serious injuries ......................................................51
3.2 Impact of mode on physical and mental health...............................................................54
3.3 Impact of air emissions on health ...................................................................................55
3.4 Impact on greenhouse gas emissions ............................................................................57
3.5 Impact of noise and vibration on health ..........................................................................59
3.6 Impact on network productivity and utilisation ................................................................63
Behavioural values of travel time for demand modelling ..............................................64
Equalised values of travel time for benefit calculations ................................................64
Monetised benefits and costs manual │ version 1.6.1, June 2023 // 4

3.7 Impact on system reliability .............................................................................................79

3.8 Impact on user experience of the transport system ........................................................81
3.9 Dynamic WEBs and land use benefits and costs ...........................................................88
3.10 Wider economic benefit (productivity) .............................................................................90
3.11 Wider economic benefit (employment impacts) ..............................................................95
3.12 Wider economic benefit (imperfect competition) .............................................................97
3.13 Wider economic benefit (regional economic development) ............................................98
3.14 Composite value for abatement of marginal congestion costs .....................................102
3.15 Other monetised impacts ..............................................................................................103
3.16 Impacts during implementation/construction.................................................................105
3.17 Behavioural change composite benefits .......................................................................106
4. Evaluation procedures ..........................................................................................................108
4.1 Introduction to procedures ............................................................................................108
4.2 Evaluation of walking and cycling activities ..................................................................111
4.3 Evaluation of road renewal and improvement activities ................................................119
4.4 Evaluation of public transport service activities ............................................................145
4.5 Evaluation of travel demand management activities ....................................................174
4.6 Evaluation of education, promotion and marketing activities ........................................180
4.7 Evaluation of freight activities........................................................................................184
4.8 Evaluation of private sector financing and road tolling .................................................202
5. Discounting ............................................................................................................................211
6. Benefit–cost ratios.................................................................................................................214
6.1 National benefit–cost ratio.............................................................................................214
6.2 Government benefit–cost ratio ......................................................................................215
6.3 Incremental cost–benefit analysis .................................................................................216
6.4 First year rate of return ..................................................................................................217
7. Sensitivity and risk analysis .................................................................................................218
7.1 Sensitivity analysis overview .........................................................................................218
7.2 Sensitivity tests .............................................................................................................219
7.3 Demand estimation sensitivity tests ..............................................................................219
7.4 Risk analysis overview ..................................................................................................225
7.5 Risk analysis procedures ..............................................................................................229
7.6 Adaptive decision-making .............................................................................................240
List of commonly used acronyms ...............................................................................................241
References .....................................................................................................................................243
Appendices ....................................................................................................................................248
Appendix 1: Demand estimation methods and guidance ............................................................248
Appendix 2: Crash analysis .........................................................................................................275
Appendix 3: Traffic data and travel time estimation ....................................................................303
Appendix 4: Vehicle operating cost tables ..................................................................................340
Appendix 5: Passing lanes ..........................................................................................................377
Appendix 6: Discount factors .......................................................................................................403
Appendix 7: Risk analysis worksheets ........................................................................................407
Appendix 8: Worked examples ....................................................................................................413

List of tables
List of tables
Table 1: Guidance on potential suitability of sources of, and approaches to, demand estimation for
different geographic contexts and transport environments ......................................................... 32
Table 2: Guidance on factors affecting demand – project model and calculation focus ...................... 39
Table 3: Guidance on factors affecting demand – transport modelling focus ....................................... 41
Table 4: High-level guidance on potential for VTM or FTM approaches .............................................. 43
Table 5: Relationship between benefits included in the Waka Kotahi Land Transport Benefits
Framework and EEM’s benefits .................................................................................................. 50
Table 6: New pedestrian facility benefits ($/pedestrian km – 2021) ..................................................... 54
Table 7: New cycle facility benefits ($/cyclist km – 2021) ..................................................................... 55
Table 8: Calculating traffic-related emission loads ............................................................................... 56
Table 9: Emissions damage costs ($/tonne – 2021) ............................................................................. 57
Table 10: Calculating traffic-related emission loads ............................................................................. 58
Table 11: Shadow price of carbon (NZ$2022 per tonne of CO2 equivalent) ........................................ 59
Table 12: Average noise design levels (Leq(24 hour)) ................................................................................ 61
Table 13: Behavioural values of travel time for vehicle and PT occupants, pedestrians and cyclists for
all road categories and all time periods ($/h/person July 2021) ................................................. 64
Table 14: Equalised values of travel time for all road categories, all time periods, all users*
($/hr/person July 2021) ............................................................................................................... 64
Table 15: Values of travel time for vehicles used for work travel purposes ($/h/vehicle July 2021) .... 65
Table 16: Composite values of travel time, plus maximum increments for congestion, for different road
categories and different time periods ($/h/vehicle July 2021) .................................................... 65
Table 17: Volume to capacity (VC) ratios for level terrain, overtaking sight distance and percentage of
time delayed (PTD) following slow vehicles................................................................................ 67
Table 18: Volume to capacity (VC) ratios for rolling terrain, overtaking sight distance and percentage
of time delayed (PTD) following slow vehicles ........................................................................... 67
Table 19: Volume to capacity (VC) ratios for mountainous terrain, overtaking sight distances and PTD
following slow vehicles ................................................................................................................ 68
Table 20: Increased service headway .................................................................................................. 70
Table 21: Breakdown of base VOC by component (%) ........................................................................ 72
Table 22: Running cost by speed and gradient regression coefficients (cents/km – July 2015) .......... 73
Table 23: Additional VOC due to congestion regression coefficient by vehicle class (cents/km –July
2015) ........................................................................................................................................... 74
Table 24: Additional VOC due to congestion regression coefficients by road category (cents/km – July
2015) ........................................................................................................................................... 74
Table 25: Additional VOC due to roughness – regression coefficients (cents/km – July 2015) ........... 76
Table 26: Increase in VOC per vehicle – kilometre per 1mm increase in Benkelman beam deflection
(July 2015) .................................................................................................................................. 77
Table 27: Additional VOC due to bottleneck delay by vehicle class (cents/minute – July 2015) ......... 77
Table 28: Additional VOC due to bottleneck delay by road category (cents/minute – July 2015) ........ 77
Table 29: Valuation of public transport (PT) user benefits/disbenefits due to a price change ............. 78
Table 30: Equivalent time to a minute-late ratios .................................................................................. 79
Table 31: Summary of user experience benefits .................................................................................. 81
Table 32: Vehicle feature values for public transport services – rail..................................................... 84
Table 33: Vehicle feature values for public transport services – bus.................................................... 85
Table 34: Infrastructure features value for public transport – bus ........................................................ 86
Table 35: Relative benefit for different types of cycle facilities ............................................................. 87
Table 36: Data requirements ................................................................................................................ 91
Table 37: Weighted average agglomeration elasticities for New Zealand by industry ......................... 92
Table 38: Imperfect competition parameters ........................................................................................ 97
Table 39: Categories of visitors ............................................................................................................ 99
Table 40: International visitor activity and estimated economic activity, by region ($2018) ............... 100
Table 41: Diversion rates and composite benefit values ($2021) for abatement of marginal congestion
costs for major urban corridors (refer also to simplified procedure SP9 worksheet SP9.5) ..... 103
Table 42: Composite benefit values ($2021) for abatement of marginal congestion costs for major
urban corridors by PT modes ($/additional passenger boarding) (refer also to simplified
procedure SP10 worksheet SP10.4) ........................................................................................ 103

List of tables
Table 43: Workplace travel plan benefit ($/employee/year – 2008) ................................................... 106
Table 44: School travel plan benefit ($/student/year – 2008) ............................................................. 106
Table 45: Household community-based activity benefits ($/capita/year – 2008) ............................... 107
Table 46: Simplified procedures in relation to full procedures ............................................................ 109
Table 47: Simplified procedure summaries ......................................................................................... 110
Table 48: SP11 Walking and cycling facilities procedure template .................................................... 113
Table 49: Steps in the SP11 evaluation of walking and cycling activities ........................................... 113
Table 50: Benefit factors for different types of cycle facilities ............................................................. 113
Table 51: Full procedure for walking and cycling activities ................................................................. 114
Table 52: Cycle demand ..................................................................................................................... 116
Table 53: Cycling commute share – has been deleted ....................................................................... 117
Table 54: SP1 Road renewals procedure template ............................................................................ 121
Table 55: Steps in the SP1 evaluation of road renewal activities ....................................................... 121
Table 56: SP2 Structural bridge renewals procedure template .......................................................... 123
Table 57: Steps in the SP2 evaluation of structural bridge renewal activities .................................... 123
Table 58: Cost factors for different type of heavy trucks .................................................................... 123
Table 59: SP3 Road improvement activities procedure template ....................................................... 124
Table 60: Steps in the SP3 evaluation of road improvement activities ............................................... 124
Table 61: SP4 Seal extensions procedure template ........................................................................... 125
Table 62: Steps in the SP4 evaluation of seal extension activities ..................................................... 125
Table 63: SP5 Isolated intersection improvements procedure template ............................................ 126
Table 64: Steps in the SP5 evaluation of isolated intersection activities ............................................ 126
Table 65: Multiplication factors for items with an estimated life of less than 40 years ....................... 126
Table 66: Full procedures for evaluation of road improvement activities ........................................... 127
Table 67: Breakdown of vehicle operating costs (VOC) by component ............................................. 136
Table 68: Recommendations on diversion rates to/from public transport from changes in car travel
costs .......................................................................................................................................... 136
Table 69: Coefficients to calculate standard deviation of travel time .................................................. 137
Table 70: Travel time variability – rural two-lane road, level terrain ................................................... 138
Table 71: Travel time reliability – rural two-lane road, rolling terrain .................................................. 139
Table 72: Travel time variability – rural two-lane road, mountainous terrain ...................................... 139
Table 73: Adjustment factors to apply to variability calculations table ................................................ 139
Table 74: Contingency allowances ..................................................................................................... 144
Table 75: SP9 New public transport services procedure template ..................................................... 146
Table 76: Steps in the SP9 evaluation of new public transport service activities ............................... 147
Table 77: SP10 procedure template ................................................................................................... 147
Table 78: Steps in the SP10 evaluation of new public transport service activities ............................. 148
Table 79: Stages of analysis for the evaluation of public transport services ...................................... 148
Table 80: Issues in elasticity estimation and application .................................................................... 152
Table 81: Overall (short run) direct elasticity estimates (at 12 months after service etc change) ...... 154
Table 82: Summary of evidence on component elasticities for key variables .................................... 155
Table 83: Typical fare and service level (short-run) relative elasticities by time period...................... 156
Table 84: Patronage ramp-up profile data by category of initiative .................................................... 158
Table 85: Prior modes of new public transport passengers resulting from urban public transport
initiatives ................................................................................................................................... 159
Table 86: Recommendations on diversion rates to/from public transport from changes in car travel
costs .......................................................................................................................................... 161
Table 87: Bus operating cost variables ............................................................................................... 168
Table 88: Unit cost categories and allocation ..................................................................................... 169
Table 89: Unit cost rates, 2009/10 prices (standard diesel bus)......................................................... 170
Table 90: Sensitivity tests ................................................................................................................... 173
Table 91: Stages of analysis for TDM ................................................................................................. 175
Table 92: Stages of analysis for evaluation of education, promotion and marketing activities .......... 181
Table 93: SP6 HPMV route improvements procedure template ......................................................... 186
Table 94: Steps in the SP6 evaluation of HPMV route improvement activities .................................. 186
Table 95: SP8 Freight services procedure template ........................................................................... 187
Table 96: Steps in the SP8 evaluation of freight service activities ..................................................... 187
Table 97: Stages of analysis for freight activities ................................................................................ 188
Table 98: Elasticities for freight commodities ...................................................................................... 190
Table 99: Calculate reduction of road user charges (RUC) revenue .................................................. 198

List of tables
Table 100: Sensitivity tests ................................................................................................................. 201

Table 101: Stages of analysis for private sector financing, and road tolling activities ........................ 204
Table 102: Annual single payment present worth factors ................................................................... 212
Table 103: Guidance on importance of sensitivity tests – project model and calculation focus ......... 221
Table 104: Guidance on importance of sensitivity tests – regional transport modelling focus ........... 222
Table 105: Likelihood rating (Z/44) ..................................................................................................... 227
Table 106: Consequence ratings ........................................................................................................ 227
Table 107: Risk consequence – qualitative rating criteria (Z/44) ........................................................ 227
Table 108: Risk matrix ........................................................................................................................ 229
Table 109: Benefit risks from base year models/source data used to develop demand estimations . 232
Table 110: Benefit risks from future year forecast demand estimations ............................................. 234
Table 111: Benefit realisation risks ..................................................................................................... 236
Table 112: Cost risks .......................................................................................................................... 237
Table 113: Risk treatment options example ........................................................................................ 239
Table A1: Forecasting demand for improvements to services or facilities ......................................... 249
Table A2: Fixed or variable trip matrices from a regional transport model ......................................... 254
Table A3: Travel behaviour change target population ........................................................................ 256
Table A4: Composite evaluation of TBhC packages .......................................................................... 257
Table A5: Scoring system for workplace travel plan diversion rates .................................................. 258
Table A6: Workplace diversion rates .................................................................................................. 258
Table A7: School diversion rates ........................................................................................................ 258
Table A8: Household and community diversion rates ........................................................................ 259
Table A9: Steps to apply growth constraint to the do-minimum/activity trip matrix ............................ 260
Table A10: Steps to apply matrix scaling ............................................................................................ 262
Table A11: Steps to apply incremental matrix capping ....................................................................... 262
Table A12: Steps to apply the shadow network technique ................................................................. 263
Table A13: Steps to apply elasticity methods ..................................................................................... 264
Table A14: Long-run generalised cost elasticities .............................................................................. 265
Table A15: Steps to apply variable matrix methods ........................................................................... 267
Table A16: Steps to apply variable matrix method B .......................................................................... 268
Table A17: Guidelines for estimating user time and vehicle operating costs ..................................... 270
Table A18: Required cost and trip matrices ........................................................................................ 271
Table A19: Suggested checks for fixed and variable trip matrix calculations/adjustments ................ 273
Table A20: Suggested checks for capacity ......................................................................................... 274
Table A21: Crash trend adjustments factors ....................................................................................... 286
Table A22: Growth adjustment factors ................................................................................................ 286
Table A23: Ratio of fatal to serious crash severities by movement for 50km/h speed limit areas ..... 288
Table A24: Ratio of fatal to serious crash severities by movement for 70km/h speed limit areas ..... 288
Table A25: Ratio of fatal to serious crash severities by movement for 100km/h speed limit areas ... 288
Table A26: Factors for converting from reported injury crashes to total injury crash ......................... 289
Table A27: Factor for converting from reported non-injury crashes to total non-injury crashes ......... 289
Table A28: Cost per crash by movement and vehicle involvement for fatal injury crashes in 50km/h
speed limit areas ....................................................................................................................... 289
Table A29: Cost per crash by movement and vehicle involvement for serious injury crashes in 50km/h
Table A30: Cost per crash by movement and vehicle involvement for minor injury crashes in 50km/h
Table A31: Cost per crash by movement and vehicle involvement for non-injury crashes in 50km/h
Table A32: Cost per crash by movement and vehicle involvement for fatal injury crashes in 100km/h
Table A33: Cost per crash by movement and vehicle involvement for serious injury crashes in
100km/h speed limit areas ........................................................................................................ 292
Table A34: Cost per crash by movement and vehicle involvement for minor injury crashes in 100km/h
Table A36: Cost per reported injury crash .......................................................................................... 293
Table A37: Cost per reported injury crash for specific sites ............................................................... 294
Table A38: Cost per reported injury crash by mode ........................................................................... 294

List of tables
Table A39: Ratio of fatal and serious/all injury factors by crash type, ONRC grouped and by alignment
type, 80–100km/h ..................................................................................................................... 295
type, 60–70km/h ....................................................................................................................... 296
type 40–50km ........................................................................................................................... 297
Table A42: Ratio of fatal and serious/all injury as a function of crash type by ONRC for 80–100km/h
.................................................................................................................................................. 298
Table A43: Ratio of fatal and serious/all injury as a function of crash type by ONRC for 60–70km/h 299
Table A44: Ratio of fatal and serious/all injury as a function of crash type by ONRC for 40–50km/h 300
Table A45: Vehicle classes ................................................................................................................. 303
Table A46: Road categories................................................................................................................ 304
Table A47: Traffic composition (%) ..................................................................................................... 304
Table A48: Steps to separate an activity into its component sections ................................................ 305
Table A49: Steps to divide the year into time periods ........................................................................ 306
Table A50: Vehicle occupancy and travel purpose ............................................................................. 307
Table A51: Axle pair adjustment factors ............................................................................................. 309
Table A52: Steps to determine traffic volumes ................................................................................... 312
Table A53: Steps to calculate the free speed travel time ................................................................... 313
Table A54: Steps to determine the free speed of a multi-lane road ................................................... 314
Table A55: Steps to determine the free speed of a two-lane rural road ............................................. 315
Table A56: Steps to determine the free speed of an 'other urban road' ............................................. 316
Table A57: Steps to select the appropriate procedure for determining the capacity of road sections 318
Table A58: Steps to determine the capacity of a motorway section with separate motorway
components in each direction of travel ..................................................................................... 318
Table A59: Steps to determine the capacity of a multi-lane road ....................................................... 319
Table A60: Steps to determine the capacity of a two-lane rural road ................................................. 320
Table A61: Steps to determine whether the effects of vehicle interactions are significant ................. 321
Table A62: Steps to determine the peak interval ................................................................................ 324
Table A63: Steps to calculate the average peak interval traffic intensity ........................................... 325
Table A64: Steps to determine the volume to capacity (VC) ratio ...................................................... 326
Table A65: Steps to calculate additional travel time ........................................................................... 326
Table A66: Steps to calculate bottleneck delay .................................................................................. 329
Table A67: Steps to determine whether to consider peak spreading ................................................. 331
Table A68: Steps to determine the additional travel time of speed change cycles from substandard
curves........................................................................................................................................ 332
Table A69: Steps to calculate the time period total average travel time per vehicle .......................... 333
Table A70: Lane width factors ............................................................................................................ 334
Table A71: Approach grade factors .................................................................................................... 334
Table A72: Parking factors .................................................................................................................. 335
Table A73: Locality factors .................................................................................................................. 335
Table A74: Arrival type ........................................................................................................................ 335
Table A75: Delay adjustment factor .................................................................................................... 336
Table A76: Platoon dispersal distances (m) ....................................................................................... 336
Table A77: Critical gap (Tg) ................................................................................................................. 338
Table A78: Average peak interval delay ............................................................................................. 338
Table A79: Passenger car VOC by speed and gradient (cents/km – July 2015) ............................... 340
Table A80: LCV VOC by speed and gradient (cents/km – July 2015) ................................................ 341
Table A81: MCV VOC by speed and gradient (cents/km – July 2015) ............................................... 342
Table A82: HCVI VOC by speed and gradient (cents/km – July 2015) .............................................. 343
Table A83: HCVII VOC by speed and gradient (cents/km – July 2015) ............................................. 344
Table A84: Bus VOC by speed and gradient (cents/km – July 2015) ................................................ 345
Table A85: Urban arterial VOC by speed and gradient (cents/km – July 2015) ................................. 346
Table A86: Urban other VOC by speed and gradient (cents/km – July 2015) .................................... 347
Table A87: Rural strategic VOC by speed and gradient (cents/km – July 2015) ............................... 348
Table A88: Rural other VOC by speed and gradient (cents/km – July 2015) ..................................... 349
Table A89: Urban additional VOC due to roughness by vehicle class (cents/km – July 2015) .......... 350
Table A90: Rural additional VOC due to roughness by vehicle class (cents/km – July 2015) ........... 351
Table A91: Additional VOC due to roughness by road category (cents/km – July 2015) ................... 352

List of tables
Table A92: Urban arterial and urban other – additional VOC due to congestion by vehicle class
(cents/km – July 2015) .............................................................................................................. 353
Table A93: Rural strategic and rural other – additional VOC due to congestion by vehicle class
(cents/km – July 2015) .............................................................................................................. 354
Table A94: Motorway – additional VOC due to congestion by vehicle class (cents/km – July 2015) 355
Table A95: Additional VOC due to congestion by road category (cents/km – July 2015) .................. 356
Table A96: Passenger car additional travel time due to speed change cycles (seconds/speed cycle)
.................................................................................................................................................. 357
Table A97: Passenger car additional VOC due to speed change cycles (cents/speed cycle – July
2015) ......................................................................................................................................... 358
Table A98: LCV additional travel time due to speed change cycles (seconds/speed cycle) .............. 359
Table A99: LCV additional VOC due to speed change cycles (cents/speed cycle – July 2015) ........ 360
Table A100: MCV additional travel time due to speed change cycles (seconds/speed cycle) ........... 361
Table A101: MCV additional VOC due to speed change cycles (cents/speed cycle – July 2015) ..... 362
Table A102: HCVI additional travel time due to speed change cycles (seconds/speed cycle) .......... 363
Table A103: HCVI additional VOC due to speed change cycles (cents/speed cycle – July 2015) .... 364
Table A104: HCVII additional travel time due to speed change cycles (seconds/speed cycle) ......... 365
Table A105: HCVII additional VOC due to speed change cycles (cents/speed cycle – July 2015) ... 366
Table A106: Bus additional travel time due to speed change cycles (seconds/speed cycle) ............ 367
Table A107: Bus additional VOC due to speed change cycles (cents/speed cycle – July 2015) ....... 368
Table A108: Urban arterial additional travel time due to speed change cycles (seconds/speed cycle)
.................................................................................................................................................. 369
Table A109: Urban arterial additional VOC due to speed change cycles (cents/speed cycle – July
2015) ......................................................................................................................................... 370
Table A110: Urban other additional travel time due to speed change cycles (seconds/speed cycle) 371
Table A111: Urban other additional VOC due to speed change cycles (cents/speed cycle – July 2015)
.................................................................................................................................................. 372
Table A112: Rural strategic additional travel time due to speed change cycles (seconds/speed cycle)
.................................................................................................................................................. 373
Table A113: Rural strategic additional VOC due to speed change cycles (cents/speed cycle – July
2015) ......................................................................................................................................... 374
Table A114: Rural other additional travel time due to speed change cycles (seconds/speed cycle) . 375
Table A115: Rural other additional VOC due to speed change cycles (cents/speed cycle – July 2008)
.................................................................................................................................................. 376
Table A116: Crash rates for rural mid-block locations (/108 veh/km) ................................................. 380
Table A117: Traffic flow profiles .......................................................................................................... 380
Table A118: Classification of passing lane costs ................................................................................ 381
Table A119: Passing lane average costs ($ 2005) ............................................................................. 382
Table A120: Steps to determine passing lane spacing strategy ......................................................... 383
Table A121: Combined terrain classification....................................................................................... 383
Table A122: Terrain relationship to passing sight distance ................................................................ 384
Table A123: Traffic growth correction factors for BCR graphs ........................................................... 384
Table A124: Steps to refine passing lane spacing strategy ................................................................ 384
Table A125: Limiting lengths m for consideration of climbing lanes ................................................... 385
Table A126: Steps for assessment of individual passing lanes .......................................................... 390
Table A127: Traffic growth correction factors for travel time and VOC graphs .................................. 391
Table A128: Traffic growth correction factors for driver frustration graphs ......................................... 391
Table A129: Passing lane length factors for travel time delays and vehicle operating cost savings .. 392
Table A130: Passing lane length factors for frustration cost savings ................................................. 392
Table A131: Traffic growth correction factors for crash savings graphs ............................................. 393
Table A132: Quarterly single payment present worth factors ............................................................. 403
Table A133: Annual uniform series present worth factors .................................................................. 403
Table A134: Annual arithmetic growth present worth factors ............................................................. 405
Table A135: Summary of benefit risks ................................................................................................ 407
Table A136: Summary of cost risks .................................................................................................... 408
Table A137: Summary of other risks ................................................................................................... 409
Table A138: Identified high or critical risks ......................................................................................... 410
Table A139: Risk adjusted BCR ......................................................................................................... 412
Table A140: Do-minimum VC ratios ................................................................................................... 417
Table A141: Option VC ratios ............................................................................................................. 418

List of tables
Table A142: Do-minimum flow matrices ............................................................................................. 418

Table A143: Option flow matrices ....................................................................................................... 418
Table A144: Standard deviations of travel time (minutes) .................................................................. 418
Table A145: Standard deviations of intersection travel times ............................................................. 419
Table A146: Do-minimum matrices of standard deviations of travel times ......................................... 419
Table A147: Option matrices of standard deviations of travel times................................................... 419
Table A148: Do-minimum matrices of flow × standard deviation of travel time .................................. 420
Table A149: Option matrices of flow × standard deviation of travel time ............................................ 420
Table A150: Example expected value calculations ............................................................................ 422
Table A151: Example funding gap calculation .................................................................................... 423
Table A152: Example costs for the 2nd half of year 1 ........................................................................ 423
Table A153: Example costs for the 1st half of year 2 ......................................................................... 424
Table A154: Example costs for the 2nd half of year 2 ........................................................................ 424
Table A155: Mutually exclusive options ranked by cost ..................................................................... 427
Table A156: Calculating the incremental BCR of mutually exclusive options .................................... 427
Table A157: Example data for bottleneck delay calculation ............................................................... 427
Table A158: Example traffic signal data ............................................................................................. 428
Table A159: Example data for relative risk indicator calculation ........................................................ 429

List of figures
List of figures
Figure 1: Decision process for selecting crash analysis methods ........................................................ 53
Figure 2: Percentage of time delayed (PTD) two-lane rural roads level terrain .................................... 68
Figure 3: PTD for two lane rural roads, rolling terrain ........................................................................... 69
Figure 4: PTD for two lane rural roads, mountainous terrain ................................................................ 69
Figure 5: Procedure for estimating changes in VOC due to congestion ............................................... 75
Figure 6: Graphs of driver frustration benefits for all terrain ................................................................. 83
Figure 7: Step by step guidance for agglomeration benefits ................................................................ 94
Figure 8: Estimating imperfect competition benefits ............................................................................. 98
Figure 10: Decision chart for bridge replacement ............................................................................... 122
Figure 11: Stages for estimating travel time ....................................................................................... 134
Figure 12: Risk analysis process ........................................................................................................ 142
Figure 13: Risk analysis steps ............................................................................................................ 142
Figure 14: Typical public transport patronage ramp-up profiles from service changes ...................... 157
Figure 15: Flow chart for estimating road section travel time ............................................................. 164
Figure 18: Flow chart for estimating road section travel time ............................................................. 193
Figure 21: Risk management process ................................................................................................ 230
Figure A1: Decision process for selecting crash analysis methods .................................................... 276
Figure A2: Method A flow chart ........................................................................................................... 279
Figure A3: Method B flow chart ........................................................................................................... 281
Figure A4: Method C flow chart .......................................................................................................... 283
Figure A5: Stages for estimating road section travel time .................................................................. 311
Figure A6: Vehicle interaction delay and bottleneck delay ................................................................. 322
Figure A7: Average peak interval traffic intensity ............................................................................... 323
Figure A8: Benefit length of installing passing lanes .......................................................................... 377
Figure A9: Selection of passing lane analysis procedure ................................................................... 378
Figure A10: Graphs of strategy BCR for flat terrain ............................................................................ 386
Figure A11: Graphs of strategy BCR for rolling terrain ....................................................................... 387
Figure A12: Graphs of strategy BCR for hilly terrain .......................................................................... 388
Figure A13: Graphs of strategy BCR for mountainous terrain ............................................................ 389
Figure A14: Graphs of vehicle operating cost and delay savings for all terrain .................................. 394
Figure A15: Graphs of driver frustration benefits for all terrain ........................................................... 395
Figure A16: Graphs of crash savings for flat terrain ........................................................................... 396
Figure A17: Graphs of crash savings for rolling terrain ...................................................................... 397
Figure A18: Graphs of crash savings for hilly terrain .......................................................................... 398
Figure A19: Graphs of crash savings for mountainous terrain ........................................................... 399
Figure A20: The relative risk indicator for project costs ...................................................................... 411
Figure A21: Township bypass overview ............................................................................................. 417
Figure A22: Example cashflows .......................................................................................................... 424
Figure A23: Incremental BCR between two options ........................................................................... 426

Introduction > Purpose of this manual
Introduction
Purpose of this manual
The Monetised benefits and costs manual (MBCM) is Waka Kotahi NZ Transport Agency’s
standardised guidance for assessing the monetised benefits and costs of proposed investments in
land transport (activities). The primary purpose of this manual is to establish consistency,
transparency and comparability between activities to aid the evaluation of their economic efficiency.
This manual is a substantial refresh of the earlier Economic evaluation manual (EEM). With the
introduction of appraisal summary tables (AST) for capturing a wider range of impacts in social cost–
benefit analysis, including non-monetisable and monetisable impacts, this manual has been
refocused to only cover costs and monetised benefits (including disbenefits). This manual is one of
the primary tools for assessing economic efficiency.
The MBCM includes guidance for assessing the 12 monetised benefits of the 25 benefits within the
benefits framework. It is designed to be read in conjunction with the Land Transport Benefits
Framework measures manual and the Benefits management guidance on our website, as together
these will provide all the information necessary for preparing an AST.
Who this manual is for

The information contained within this manual is designed to assist three primary types of user.
Proposal submitters
The cost–benefit analysis and appraisal concepts at the start of this manual should be used to guide
the development of investment proposals and options. They are references that will help ensure that
proposals have been correctly scoped.
Additionally, the summaries of monetised benefits will encourage the systematic identification of a
proposal’s benefits and disbenefits.
Transport analysts
The procedures, values and worksheets in this manual have been designed to assist analysts in
developing fit-for-purpose evaluations for transport activities. The material in this manual enables
activities to be assessed using standardised approaches, however, professional judgement and
supporting evidence can be utilised for bespoke analyses.
Decision makers
Decision makers use benefit–cost ratios, an output of the monetised components of cost–benefit
analysis as a decision support tool. By standardising the methods of evaluation, this manual is able to
assist decision makers when proposals are compared or put forward for funding decisions.

1. Concepts > 1.1 Social cost–benefit analysis
1. Concepts
1.1 Social cost–benefit analysis
Social cost–benefit analysis, generally referred to as economic cost–benefit analysis or CBA, differs
from financial analysis by incorporating social, economic, environmental and cultural impacts. A CBA
measures costs and benefits at a national level and is a systematic method of organising information
about the costs and benefits of a proposed activity.
CBA is primarily a decision support tool and so the level of effort put into measuring impacts should
reflect the scale, scope and complexity of the decision that needs to be made. This manual sets out
standardised guidance for measuring impacts and monetising them, but a CBA can include non-
monetised costs and benefits.
Benefit–cost ratios, or BCRs, are often confused with CBA. BCRs are an indicator of economic
efficiency in the CBA framework, but they focus solely on monetised benefits and costs. Waka Kotahi
uses BCRs as one measure of efficiency, but decisions are further supported by the impacts captured
in the appraisal summary table (AST), including non-monetised benefits.
The appraisal summary table is a structured way to show the monetised and non-monetised
benefits and costs of short-listed options and the preferred solution. This tool plays a key role in
demonstrating how a preferred solution contributes to outcomes and also enables Waka Kotahi to
track benefits.
BCRs indicate whether activities will generate more benefits than they cost, make it possible to
compare activities, and enable prioritisation of activities under funding constraints – all within a well-
defined framework. For this reason, they remain the primary measure of economic efficiency used by
Waka Kotahi when assessing an activity from a purely monetised point of view.
Refer to the Waka Kotahi Planning and Investment Knowledge Base for information on when a BCR
is required to support a funding application.
The process for preparing an analysis of an activity is detailed step by step throughout this section of
the manual and specific information about BCRs, and other appraisal tools, is discussed in section
1.10.
For a more in-depth discussion of social cost–benefit analysis in a New Zealand context, please
refer to Treasury’s Guide to social cost benefit analysis
1.2 Equity or distributional effects of land transport initiatives

The term equity is normally used to refer to the ethical desirability of distributional effects among
groups of individuals. Equity impacts of transport service activities should be quantified wherever
possible and reported as part of the evaluation (separately from the economic efficiency calculation).
The potential benefits related to distributional issues and implications have been also included in the
description of benefits in the Land Transport Benefits Framework.
While an analysis of the distribution of benefits and costs among different groups of people is not
required for economic efficiency analysis, evaluations of an activity should report the distribution of
benefits and costs, particularly where they relate to the needs of transport disadvantaged
populations. This reporting forms a part of the funding allocation process.
When it is required, distributional effects should be reported separately from, but alongside, the
CBA results.

1. Concepts > 1.3 Steps in BCR calculation
1.3 Steps in BCR calculation
• Do-minimum
• Improvement alternatives and options
• Consider whether the improvement(s) should be part of a package and/or
Options programme of activities
• Collect data
Demand • Use of transport models and calibration
forecast
• Measure and monetise the impacts (benefits and disbenefits) for the do-minimum
and options
Benefits
estimates
• Investigation and design

• Property
• Construction, including preconstruction and supervision
• Maintenance, renewal and operation
Costs • Risk management and mitigation of external impacts
estimates • Residual values
• Undertake risk analysis when there are significant unpredictable events that may
affect or be affected by the improvement activity
Risk
analysis
• Discount the monetised benefits and costs over the analysis period to obtain present
values
Discounting
• Determine the benefit–cost ratios (BCR) of the preferred option

BCR
calculation
• Use incremental cost–benefit analysis to select the preferred option for mutually
exclusive options
Incremental
analysis
• Perform sensitivity tests on the preferred option to determine how robust the
calculations are and whether a small change in one of the input parameters has a
Sensitivity large change on the evaluation outcome
analysis

1. Concepts > 1.4 Counterfactuals and the do-minimum
1.4 Counterfactuals and the do-minimum

A counterfactual is a future in which a proposed activity does not occur. Typically, a CBA will analyse
counterfactuals known as the ‘do-nothing’ and the ‘do-minimum’.
There should be careful consideration of what the counterfactual is, as this is what the activity will be
measured against. Overstating or understating the counterfactual can have an adverse effect on the
CBA. Effort should therefore be applied early in the development of the analysis to define the future
state if an activity did not proceed in order to establish a realistic baseline that options can be
assessed against.
The analysis of the counterfactual should match the analysis period applied to the CBA (see section
1.6). The do-nothing and do-minimum will usually involve a multi-year forecast, as opposed to a point-
in-time estimate. Such forecasts should estimate the costs and benefits likely to occur in the absence
of further intervention. A common example is increased congestion in response to population growth.
In the case of greenhouse gas (GHG) emissions, the counterfactual should reflect baseline emissions
forecasts for the relevant network or region, using data and tools described in section 3.4.
In some cases it is possible that through a comprehensive CBA it is determined that the
counterfactual is the preferred option. This could occur if the expected costs of alternative options are
very high relative to their expected benefits.
Do-nothing
Most forms of activity evaluation involve choices between different options or courses of action. In
theory, every option should be compared with the option of doing nothing at all, ie the do-nothing.
However, this is often not practical as usually there is a minimum level of expenditure required to
keep a facility or service functioning.
Do-minimum
For many transport activities, it is often not practical to do nothing. A certain minimum level of
expenditure or activity may be required to maintain a minimum level of service. This minimum level of
expenditure or activity and the resultant performance is known as the do-minimum, and should be
used as the basis for evaluation, rather than the do-nothing. It is important not to overstate the scope
of the do-minimum.
The do-minimum may include maintaining the status quo and should account for committed and
funded transport activities. For the purposes of this manual, the do-minimum is defined as the least
cost option that provides a minimum level of service.
Particular caution is required if the cost of the do-minimum represents a significant proportion of, or
exceeds, the cost of the options being considered. In such cases, the do-minimum should be re-
examined to see if it is being overstated.
If an activity’s option results in cost savings compared with the developed do-minimum, then the
option becomes the new do-minimum that all other options should assessed against.
Do-minimum for safety activities

For safety activities where reducing the speed limit is a potential option, the do-nothing scenario is the
existing baseline conditions of the network, based on the existing speed limit, operating speed,
infrastructure and services.
Where a road controlling authority decides to introduce one or more interventions to address
unacceptable levels of collective and/or personal risk, to re-set the speed limit, and/or to manage
speeds on a particular piece of road, the do minimum can include benefits and costs of implementing
a new safe and appropriate operating speed.
In such situations the do-minimum should be compared to both the do-nothing and the other activity
options in order to determine whether the do-minimum is the preferred option (ie the optimal solution),
or whether additional improvements are justified over and above the do-minimum, and if these
additional improvements are therefore the preferred option.

1. Concepts > 1.5 Alternatives and options
When undertaking safety interventions addressing speed the following information should be
referenced:
• Land Transport Rule: Setting of Speed Limits 2017
• the Waka Kotahi Speed management guide, and
• the Waka Kotahi MegaMaps tool that is used with the Speed management guide.
1.5 Alternatives and options

Rigorous consideration of alternatives and options is a key component of the Waka Kotahi investment
process and a requirement of the Land Transport Management Act 2003 (LTMA).
Alternatives
Alternatives are different means of achieving the same objective as a proposed activity. Alternatives
should, amongst other things, try to consider whether non-transport solutions, such as changes to
existing policy, are suitable responses to an identified problem and can achieve the outcomes sought.
An alternative can be a response to a problem or opportunity by applying a whole-of-system approach
(this can include corridor or network planning). For example, exploring the potential for different land-
use arrangements or encouraging greater use of other modes to address projected growth in network
demand. Alternatives may have been identified as part of development strategies and spatial plans
but may also be developed as part of the business case development process. In developing
alternatives, it is important to consider the intervention hierarchy, which addresses:
• demand – for example, ways in which the need for travel can be reduced
• productivity – for example, by making sure the current system is optimised
• supply – for example, provision of new services or infrastructure.
Options
Options are variants of a proposed activity. Activity options may differ in scale, scope, or even
alignment, and all realistic options for addressing the problem must be evaluated.
It is a common mistake for evaluations to concentrate on a single preferred option. Typically, this is
caused by a failure to understand the problem that needs to be addressed, by overstating the do-
minimum, or from narrowing the scope of analysis too early.
Some options may be classed as being mutually exclusive. Mutually exclusive options occur when
proceeding with a specific option would preclude another option from being progressed. For example,
when choosing between two different alignments for a road, the choice of one alignment precludes
the choice of the other alignment. The two alignment options are therefore mutually exclusive.
The concept of mutually exclusive options is important for incremental assessment.
Multi-modal options
When considering the possible options to solve a problem, the solutions should not be constrained to
a specific mode. Solutions to problems can come from different modes and can even be combinations
of interventions targeting multiple modes. These options should be considered in the analysis.
1.6 Period of analysis

The procedures outlined in this manual are time dependent. It is important to set appropriate critical
times and analysis periods. These periods should be applied consistently to all options, including the
do-minimum or do-nothing. There are three critical times to be set up for the analysis process which
are described in more detail below.
Time zero
Time zero is the date that all future cost and benefit streams are discounted to. Time zero for all
proposed activities is standardised to 1 July of the financial year in which the analysis is submitted.

1. Concepts > 1.7 Benefits
Time zero is independent of the construction date of a proposed activity and therefore all options
being assessed must use the same time zero.
Base date
The base date is used to standardise the valuation of all monetised impacts to a common year. The
base date for all proposed activities is standardised to 1 July of the financial year in which the analysis
is prepared.
The base date does not need to coincide with time zero. It is common for the base date to be one
year earlier than time zero.
This manual contains factors for converting the value of monetised impacts from earlier base years to
the current financial year.
Analysis period
The analysis period, starting from time zero, is the period for which all costs and benefits are included
in the BCR calculations.
The time period used in economic evaluations must be sufficient to cover all costs and benefits that
are significant in present value terms.
The standard analysis period is 40 years for a 4% discount rate.
An increase of the analysis period to 60 years is permitted where appropriate for the activity under
consideration and where the benefits and costs can reliably be forecast for up to 60 years, in order to
ensure that the whole-of-life costs and benefits of long-lived infrastructure activities are captured. An
extension of the analysis period increases the importance of demand forecasting. Emphasis should
be placed on developing a range of options and scenarios, and on reporting uncertainty in the
business case and economic evaluation, when the analysis period is extended. Examples of where it
can be appropriate to use a 60-year analysis period are for major infrastructure projects that have an
expected life of least 60 years and potentially much longer, such as an additional Waitematā Harbour
crossing or a new railway line to a major urban area,
The appropriate period of analysis may also be less than the standard 40 years. It is important to
consider the useful lifespan of an activity and adjust the analysis period accordingly. For activities with
short-lived assets, or activities where benefits dissipate quickly, it may only be necessary to assess
the activity over a 5- to 10-year period. In these circumstances changes to the analysis period should
be used as a sensitivity test.
1.7 Benefits
Benefits are any positive or negative impacts that are attributable to an activity. The benefits within
this manual have been named impacts to explicitly account for the generation of both positive benefits
and negative benefits (disbenefits).
The Waka Kotahi policy is that any expenditure on delivery, maintenance operations and renewal
is treated as a cost while all the negative impacts are treated as disbenefits.
The impacts of transport activities may affect individuals outside of the transport system.
Externalities – the impacts that affect individuals outside the transport system – must be considered in
the BCR calculation, as the analysis is conducted from a national viewpoint.
As a rule, only changes to real resources should be considered an impact in the analysis. Where
there is a transfer of resources between parties, such as a road user paying a toll, this should be left
out of the BCR. Similarly, any change in resources that is not attributable to the activity should be left
out of the BCR.
If there is no change in resources between the do-minimum and the activity scenario then there can
be no impact.

1. Concepts > 1.7 Benefits
This manual provides standardised methodologies for monetising a range of impacts generated by
transport activities.
Transfers
Care must be taken to ensure that a change in real resources for one set of individuals is not offset by
a change of real resources in the opposite direction for another set of individuals. Where this transfer
occurs, there is no net effect on national resources and therefore these transfer payments must be
excluded from the BCR calculation.
Specifically, tolls, which are a cost to transport system users, simultaneously benefit transport
operators and are excluded from the BCR. It is important, however, that tolls are considered during
demand estimation.
Similarly, any changes to business or retail profitability as a result of a transport activity are also
considered transfers and must be excluded from the BCR unless there are economy-wide efficiencies
from increased competition. In this instance the wider economic impact procedures for calculating
reductions in imperfect competition, contained within section 3.12 of this manual, must be followed.
Double counting
The benefits listed in this manual generally constitute the total economic impact of improved levels of
service, accessibility or safety.
Certain external impacts of activities, such as increased land values, may arise because of an
improved level of service and accessibility, but these impacts must be excluded from the BCR
calculation. The capitalisation of reduced travel costs leads to an increase in the underlying land
value, but including this in the BCR would be double counting as direct travel-cost benefits and wider
economic benefits should already have been calculated using the benefits in this manual.
Wider economic impacts

In some instances, second order impacts may be generated by transport activities of significant scale
and scope. These are termed wider economic impacts as they can alter the distribution of economic
activity generated by firms, households and workers.
Wider economic impacts are additional to conventional transport system benefits, but care must be
taken to ensure that they are not already captured within the analysis to avoid double counting.
Level of data collection and analysis

The primary impacts of transport activities and the monetised parameter values associated with these
impacts need to be included in the BCR calculations. This manual contains defined parameter values
and assessment procedures to estimate the economic costs and benefits of transport activities.
The level of work associated with collecting information on parameters (and their associated values),
and the data needed for economic analysis, should be commensurate with the likely impact of those
parameters on the analysis together with the decision-making needs of both the activity proponents
and Waka Kotahi.
This manual also contains both full procedures (extensive) and simplified procedures for the BCR
calculations. The simplified procedures contain standardised default assumptions, which can be used
to assess the majority of transport activities, and they match the level of effort to the appropriate level
of complexity of the activity. In many cases the full procedures analysis will be proceeded by a
simplified analysis to indicate whether the BCR will be sufficiently viable to continue with a fuller and
more expensive analysis. Activities should be considered on a case-by-case basis to determine the
appropriate level of data collection and analysis to apply.
Monetisation
After benefits have been quantified they should be monetised, where possible. Chapter 3 of this
manual contains all impacts ascribed standardised monetary values for transport appraisal in New
Zealand. Section 3.15 contains advice on approaches for monetising impacts not contained in this
manual, while the Land Transport Benefits Framework measures manual contains information on

1. Concepts > 1.8 Costs
non-monetised measures, namely quantitative and qualitative measures, for all of the benefits
included in the benefits framework.
Types of monetised benefits

There are two broad types of monetised benefits.
The first type of benefit, market benefits, have values that can be directly derived from real world
goods and services. For example, vehicle operating cost benefits are comprised of changes in the
costs of fuel, tyres, repairs and maintenance, oil, and depreciation, all of which have market values,
and therefore vehicle operating cost benefits are a market value.
Non-market benefits form the second type of benefit. While non-market benefits have standardised
monetised values, the benefits are not based on traded goods and services, and therefore they do not
have directly observable values. The valuation must instead be established through research,
typically via surveys or indirect valuation methodologies.
While market and non-market benefits can be compared, in many instances the market prices for
goods and services do not equal their resource cost due to taxes or market imperfections. It is
necessary in those situations to substitute the market price with a shadow price that is adjusted to
equate the market benefit with its true resource cost. All the benefits in this manual take into account
any differences between market prices and resource costs, and therefore they do not require any
further adjustment.
Rule of half
The rule of half is a simplifying assumption used to calculate the benefits that accrue to transport
system users who change their travel behaviour, such as by switching their mode of travel, as a result
of changes to the cost or quality of travel.
In the do-minimum, users experience benefits from their existing travel behaviour. If they choose to
change their travel behaviour in response to a new or improved activity, then it must be the case that
they experience a higher level of benefits as a result of the activity. However, upon changing their
travel behaviour, the users must also forgo the benefits of their previous travel behaviour in the do-
minimum, which offsets the increase in benefits after the change. Therefore, the transport system
users who change their travel behaviour receive only an incremental increase in benefits between the
do-minimum and activity scenarios.
The rule of half assumes that, on average, transport system users will receive half of the incremental
benefits after changing their travel behaviour.
In the case of transport system users who change their mode of travel, some new users may have
been almost indifferent between the two modes in the do-minimum. After changing their behaviour in
response to an improvement in the activity scenario, they receive the full value of the incremental
benefits. Other new users may only be marginally better off in the activity scenario compared with the
do-minimum, and they receive almost zero benefit from the improvement. If it is assumed that new
users are evenly distributed along the demand curve, then the average new user gains one half of the
maximum incremental benefits. The sum of new user benefits can then be approximated by
multiplying half of the maximum incremental benefit by the number of new users. This is also known
as a consumer surplus calculation.
Without this assumption extensive surveys of potential travel behaviour change would be required to
establish the willingness to pay of any improvements, which is not realistically feasible for the majority
of activities.
A worked example of consumer surplus and the rule of half is provided in Appendix 8: Worked
examples.
1.8 Costs
The costs taken into account in a BCR calculation include all costs necessary for the planning and
investigation, delivery, maintenance, operation and renewal of a transport activity. These whole-of-life

1. Concepts > 1.8 Costs
costs cover all resource costs incurred at any time during the analysis period including indirect and
administration costs incurred by the approved organisation or by Waka Kotahi.
Indirect costs
Indirect costs are costs that are not directly attributable to a specific activity. They are indirectly
applied, typically by way of a separate allocation, for example overheads. Indirect costs may be either
fixed or variable, and include administration and personnel costs.
Administration costs
Administration costs are an overhead cost incurred in the delivery of activities. They are not integral to
the delivery of an activity but must be provided to support the delivery of the activity.
Expected cost
The costs included in the BCR calculation should be expected costs, which are the 50th percentile, or
p50, costs.
The expected cost is based on probability and risk theory. That is, if the activity was theoretically
delivered 100 times, in 50 instances the total cost of delivery would be below the expected cost and
the total cost of delivery for the remaining 50 instances would be above the expected cost. Full
information on estimating expected costs is contained within the Waka Kotahi Cost estimation manual
(SM014).
The expected cost must include any contingency that has been allowed for in the cost estimate.
In some instances, it may be appropriate to use 95th percentile, or p95, costs as a sensitivity test of
the cost risks of an activity.
Sunk costs
Sunk costs are costs that have been irrevocably committed and which have no realisable value
through resale or salvage.
Sunk costs that have already been incurred, such as prior investigation or design costs, must not be
included in the BCR calculation. If a pre-committed cost has a market value that could be realised in
the future, such as land, then this must continue to be included in the BCR calculation.
Avoided costs and cost savings

An activity may prevent costs in the do-minimum from being incurred. These avoided costs must be
included in the BCR calculation as a cost saving.
Typically avoided costs relate to reductions in ongoing maintenance or operation costs, but can
include cost savings from deferring future physical infrastructure. For technology projects there may
be cost savings as a result of corporate efficiencies or reduced staffing costs. These avoided costs
must be considered where they are applicable.
Interest costs
Interest payments are generally excluded from BCR calculations irrespective of any arrangements to
finance an activity by way of loans. Interest is excluded as it forms part of the cost of capital and is
accounted for in the discount rate. In the analysis, capital costs should be included as cash flows in
future years according to when the costs are incurred.
An exception to the interest payment rule applies only when Waka Kotahi borrows to fund its share of
an activity, whether this is through a traditional loan or alternative funding arrangements such as
public private partnerships (PPPs). In this instance the costs are treated as cash flows from the
National Land Transport Fund (NLTF) in the future years that actual payment is predicted to occur.

1. Concepts > 1.9 Discounting
Escalation
Costs must be measured in real terms and reported in constant present-day dollars using the base
date year. In practical terms, this means that escalation is not applied to future costs and that all costs
must be calculated according to the prices of inputs in the financial year that the analysis is prepared.
Inflation and escalation are often confused. Inflation is defined as an increase in general prices
throughout the full economy. Escalation refers to an increase in the cost of inputs relevant to an
activity. The rate of escalation can be different to the inflation rate, and the rate of escalation may
even differ between inputs. Full information on cost estimation and escalation is available in the Cost
estimation manual (SM014).
No adjustments to the discount rate should be made to account for future inflation or escalation, as
the discount rates in this manual are real discount rates.
Funding gap
All BCR calculation procedures in this manual use economic costs based on changes to real
resources. This differs from financial analysis, which does include the effects of transfer payments
between parties.
The simplified and full procedures for public transport services require an additional step that includes
financial analysis of the expected funding gap between future revenue and cost of operating a
service.
The funding gap is the deficit in cash flow that needs to be funded by local and central government if
the activity is to be financially viable from the public transport service provider’s point of view, based
on the best estimate of service provider revenue and the service provider’s desired rate of return.
The service provider costs can be compared with the predicted revenue or increase in revenue if
there is a pre-existing service, using a net present value methodology to determine whether or not the
activity is viable in a financial sense.
Full information on conducting funding gap analysis for public transport services is contained within
section 4.4 of this manual.
1.9 Discounting
There is a trade-off between consuming resources now and in the future. In most instances people
demonstrate a preference in favour of immediate consumption rather than delaying consumption to
future years. This preference is the time preference or, alternatively, the time value of money.
Over a 40-year analysis period activities will have a profile of costs and benefits that are generated
over time. Furthermore, different activities will generate, often quite substantially, different profiles of
costs and benefits. However, due to the time value of money, costs and benefits in one period cannot
be treated with the same weighting as costs and benefits in another period. To ensure that costs and
benefits that occur in the future are given less weight than those incurred today, and to ensure that
costs and benefits from different years and different activities can be compared in a common unit,
they must be discounted.
Discount rate
The discount rate serves two purposes. Firstly, the discount rate represents the rate at which society
is willing to trade off present benefits and costs against future benefits and costs, thus capturing the
time value of money. In this instance, a high discount rate indicates a high degree of impatience in the
time value of money or more simply a greater preference towards immediate consumption of
resources.
Secondly, as resources committed to one activity preclude those resources being committed to
another purpose, the discount rate reflects the opportunity costs of resource expenditure. Here a high
discount rate indicates that the committed resources may have a higher return if put to an alternative
use.
In either case a high discount rate means costs and benefits incurred in future years are given much
less weight than those occurring immediately or in the near term. Therefore, particular care must be

1. Concepts > 1.10 Benefit–cost ratios and other appraisal tools
taken when allocating cost and benefit flows to the first five years of the analysis period as this can
have a large impact on an activity’s economic efficiency compared with costs and benefits occurring
in future years.
Waka Kotahi has revised the discount rate from 6% to 4%. This is based on the social opportunity
cost of resources methodology.
Further information on the methodology used to calculate and revise the discount rate can be found in
Heerdegen (2013) and Waka Kotahi NZ Transport Agency (2019a).
Present value
The discount rate is used to calculate discount factors for future years according to the formula:
1
(1 + i)n
where: n is time in years after time zero; and

i is the discount rate expressed as a decimal, ie for 4% i = 0.04.
The discount factor for each year is then applied to the costs and benefits that occur in that year,
yielding a present value of costs and benefits.
The present value of a future benefit or cost is therefore its value discounted back to the present day
or, more commonly for transport activities, to the base date.
As an activity will have a series of benefits occurring over the analysis period, the present value of net
benefits is found by summing the discounted benefits from all years in the analysis period. Similarly,
the present value of net costs is found by summing the discounted costs. It is these present values of
net costs and benefits that enable activities to be compared with the do-minimum and other activities,
despite their different cost and benefit profiles.
Chapter 5 of this manual contains the formulas used to calculate discount factors, while Appendix 6:
Discount factors contains tables of discount factors for 3%, 4% and 6% discount rates to assist in
calculating the present value of costs and benefits.
1.10 Benefit–cost ratios and other appraisal tools

The primary purpose of conducting a CBA is to support decision making. The CBA indicates to
decision makers whether the benefits of proceeding with an activity outweigh its costs, indicates the
optimal timing for delivering an activity, or, when comparing activities, indicates activity is the most
economically efficient.
BCRs are the primary measure of economic efficiency with the CBA framework used by Waka Kotahi
when assessing an activity from a purely monetised point of view. A BCR must be developed for any
activity submitted for funding from the NLTF, and it is expected that BCRs will be reported for all
shortlisted options assessed. Additionally, further information on an activity’s net present value and
first year rate of return may be required.
Benefit–cost ratio
The benefit–cost ratio (BCR) of an activity is calculated by dividing the present value of net benefits
by the present value of net costs.
A BCR greater than 1.0 indicates that an activity generates benefits in excess of its costs, while a
BCR less than 1.0 indicates that the costs are greater than its benefits.
Additionally, it is possible for an activity to have a negative BCR. This can occur when an activity
generates net disbenefits when evaluated against the do-minimum. If an option is of lower cost than
the do-minimum, then the option is treated as the new do-minimum.

1. Concepts > 1.11 Sensitivity analysis
BCRs are used as an aid for prioritising project and programme options against other options. An
option with a higher BCR than another option indicates that the option with the higher BCR delivers
greater benefits per dollar of cost (to the extent that all costs and benefits can be monetised).
There are two types of BCR in this manual, the national benefit–cost ratio (BCRN) and the government
benefit–cost ratio (BCRG). The procedures for estimating these BCRs are included in Chapter 6. The
generic term BCR is used in this guidance for BCRN, which is a measure of economic efficiency from
a national perspective. A national benefit–cost ratio must be calculated for all the shortlisted options.
The BCRG is not an alternative to the BCRN. Rather the BCRG is additional information that is helpful
when considering the business case and the financing for an activity or a programme of activities.
Net present value

The net present value (NPV) of an activity is simply the present value of net benefits less the present
value of net costs.
A positive NPV indicates that an activity generates more benefits than it costs, while a negative NPV
indicates that the costs of the activity outweigh its benefits.
The conceptual simplicity of an NPV is useful for communicating to decision makers whether an
activity is economically efficient, however, unlike BCRs, NPV should not be used to rank activities
when funding is constrained, as is the case of activities funded from the NLTF.
First year rate of return

First year rate of return (FYRR) helps indicate the optimal start date of an activity.
FYRR, expressed as a percentage, is calculated by dividing the present value of benefits in the first
full year following completion of construction by the activity’s full present value of net costs. The
formula for FYRR is given by:
present value of the activity benefits in first full year following completion × 100
FYRR =
present value of the activity costs over the analysis period
The FYRR is useful for sequencing activities when funding is constrained, but it should not be used to
evaluate whether an activity is economically efficient. The FYRR indicates the extent to which the
benefits of an activity arise immediately, or are dependent on future growth, but the overall economic
efficiency cannot be evaluated on the basis of the activity’s benefits in the first year of operation.
It is a requirement that the FYRR is reported for the preferred option of any activity submitted for
funding from the NLTF. Ideally, the FYRR should be calculated and reported for a range of possible
implementation start dates. This allows changes in the FYRR over time to inform the optimal timing of
investment.
Back to 1.1 Social cost–benefit analysis >>
1.11 Sensitivity analysis

Conducting a CBA requires making assumptions and predictions about the future. Moreover, it
requires not only predicting what the future looks like if an activity went ahead, but also what the
future would look like in a situation where the activity doesn’t proceed. As the future is unknowable, it
is entirely possible, indeed even likely, that these futures do not come to pass.
Due to the inherent uncertainty involved in predicting the future it is important to test the sensitivity of
the assumptions and predictions that underlie the analysis.
Chapter 7 of this manual details the methodologies that should be followed to undertake sensitivity
and risk analysis. It is important, however, to record the assumptions and predictions that have been
made during the development of a proposal, and keep in mind at all times the risks and uncertainties
that could have a material impact on the analysis.

Risk and uncertainty

The terms risk and uncertainty are often confused with each other or used interchangeably.
References to risk and uncertainty in this manual are established upon formal definitions of risk and
uncertainty based on probability theory.
A risk has known objective probabilities of outcomes occurring. The simplest example of a risk is
calling a coin toss. There is a 50% risk of the coin landing on heads when tails has been called.
Uncertainties arise when it is impossible to define all possible outcomes or when the objective
probabilities of outcomes occurring are unknown. Future population growth is classed as an
uncertainty because it is both impossible to define all possible outcomes, and it is impossible to define
the probability of those outcomes occurring.
Care needs to be taken when assessing risks and uncertainties to ensure that uncertainties are not
misclassified as risks by relying on subjective probabilities. A subjective probability is a best guess
estimate of the probability of outcomes occurring that combines probability data with personal
judgements about the probability distribution.
Scenario testing
One of the most powerful sensitivity analysis tool available is scenario testing. Scenarios are plausible
states of the future and are developed by changing key assumptions such as population and
employment growth rates, future land use patterns and future travel behaviour.
Before defining the do-minimum or developing alternatives and options, consideration should be given
to how sensitive the problem being addressed is to changes in the assumptions being made about the
future. If the nature or scale of the problem is likely to change substantially based on changes in the
assumptions, then multiple do-minimums and scenarios should be developed. Doing so early will
ensure that an appropriate range of alternatives and options are developed that are adaptable to
forecasting uncertainties.
Sensitivity testing
Sensitivity testing is a simple method of checking the sensitivity of a BCR to changes in assumptions
and uncertain input variables. The most basic method of sensitivity testing involves manipulating a
single variable, such as an activity’s cost, for a range of values to produce a BCR range. A more
robust method, which can highlight interactions between assumptions, is to manipulate multiple
variables at the same time. This should be used as a precursor to full risk analysis.
Sensitivity testing is useful for quickly testing the veracity of the analysis and demonstrating to
decision makers the robustness of the BCR to often extreme changes in key assumptions. It should
be noted, however, that sensitivity testing is unable to provide information on the probability of
outcomes occurring, and the choice of variables tested can greatly impact on the credibility of the
analysis.
When conducting sensitivity testing focus should be given to variables that have the highest impact
and are uncertain.
Risk analysis
Risk analysis is a more detailed type of sensitivity testing that involves describing the probability
distributions of the input variables and those of the resulting estimates of benefits and costs. For a risk
analysis to be possible, both the benefits and costs arising from each of the possible outcomes and
their probability of occurrence have to be estimated.
Risk analysis can support the development of methods for minimising, mitigating and managing
uncertainties.
Adaptive decision-making
Adaptive decision-making involves considering all possible outcomes when selecting options for
further investigation where there is deep uncertainty. It provides the analyst with additional analysis

options and methods during the assessment process. For further information on adaptive decision-
making refer to section 7.6.

2. Demand estimation and mode share > 2.1 Demand estimates and importance to economics
2. Demand estimation and mode share

This chapter brings together the different elements of travel demand estimation, which play a more
significant role in economic assessments. It provides information and some guidance to assist with a
systematic and balanced approach to demand estimation carried out as part of an economic
assessment.
2.1 Demand estimates and importance to economics

A demand estimate is a prediction of the use of a transport facility, service or travel mode. The term
‘demand estimate’ can be applied to the production of any representation of trip-making in an area or
location, including both base-year (existing/current) and future-year trip-making levels.
The term ‘forecasting’ is generally used in reference to the production of future-year trip-making
levels.
How demand estimates will be developed for the assessment of a transport activity will vary
depending on such factors as the type, significance, and location of the activity. For example, different
approaches are likely to be deployed for a rural intersection upgrade in Southland compared to a
large-scale public transport project in Auckland.
Demand estimates are important in most economic assessments because the amount of predicted
use of the facility, service, or mode is often a key, if not critical, driver of the potential benefits.
Demand estimation covers a wide range of information, factors and aspects related to transport trip-
making. For example, mode choice is just one of many components that make up a potential demand
estimate – this starts from whether an individual person has a desire/need to make a trip, where they
start from, the purpose of the trip, where they wish to go, the modes available to get there (or to other
destinations related to the same purpose), down to details such as the departure time and route
chosen.
This chapter broadly covers the following areas:
• high-level principles of demand estimation and key concepts related to it

• clarification of important definitions and linkages noted with key industry guidelines
• key considerations around inputs that are likely to affect demand estimates and forecasts
• broad description of key elements in the New Zealand context, for example, the availability of
transport models
• the influence demand estimates and particularly forecasts can play on economic
assessments, and key considerations such as optimism bias
• broad considerations related to the application of different methods for estimating demands,
including transport modelling aspects
• considerations relating to fixed and variable trip matrix approaches.
• aspects of uncertainty in demand estimation and forecasts.
Sensitivity and risk analysis of demand estimation is included in Chapter 7: Sensitivity and risk
analysis. Further considerations and high-level guidance can be found in these sections on carrying
out sensitivity and risk assessment relating to demand estimates.
Appendix 1: Demand estimation methods and guidance contains more detailed information, some
specific technical information and guidance on applying certain methodologies, and key factors such
as elasticity parameters.

2. Demand estimation and mode share > 2.2 Key concepts
2.2 Key concepts

This section provides a definition of key terms and concepts related to demand estimation.
The following definitions are specifically and directly related to demand estimation.
Travel demand
The representation of movements (vehicles, persons, etc) across the area spatially and over time.
Commonly in the form of origin–destination (OD) trip matrices and the profile of demand or
proportions through the time period modelled. (Transport model development guidelines (NZ
Transport Agency 2019)).
Trips
A trip is a journey made between a start and end point for a specific reason or purpose. A trip can
involve multiple modes, for example, a trip to work could involve walking from home to a bus stop,
catching the bus, getting off the bus at the bus stop and then scootering to the office – all of which is
one home-to-work trip. A trip chain is a series of linked trips made from point A to B, and then from
point B to point C, etc. Traditional regional transport models model trips rather than trip chains, which
are from a start point to an end point for an associated travel purpose. For example, a home-to-
recreation trip.
Demand estimation
The estimation of travel, including trip numbers, destination choices, mode choices, time choices,
route choices, etc. Demand estimation encompasses multiple and different types of analyses. One
example is developing vehicle trip origin–destination matrices representing base or future years.
Another example is estimating future-year cycle link volumes or pedestrian crossing volumes at a
particular location.
Mode share
The percentage split of demand using a particular travel mode, typically related to person travel
demand. Typically assessed travel modes are: private vehicle/car (sometimes differentiated by driver
and passenger), public transport (bus, rail, ferry, etc), walking, and cycling. Freight (for example,
commercial vehicle trips) is usually considered separately and is not considered a travel mode.
Elasticities
Elasticity is a general term for the relative rate of change of demand when compared to a causal
variable. Economists refer to elasticities as the percentage change in consumption/use of a ‘good’
caused by a 1% change in its price or other characteristic. In transport demand estimation, elasticities
typically describe a percentage change associated with a demand response linked or related to a
change in the transport system or environment. For example, public transport elasticities are defined
as the percentage change in patronage resulting from a 1% change in the relevant transport service
attribute, for example, fare level or service frequency.
Elasticities are one of many methods that may be used to develop future demand and mode choice
estimations.
Rule of half
The rule of half is applied in situations where there is a difference in the travel demand (number of
trips between any origin and destination) in the study area between the do-minimum scenario and the
activity scenario for the same evaluation year. When estimating demands and carrying out economic
assessments, it is important to distinguish situations where the rule of half is and isn’t applied; in other
words what constitutes a ‘real’ difference in trip numbers between the do-minimum and what does

not. This is important for how economic benefits are calculated from the demands used in these
assessments. (Refer to ‘Rule of half’ in Chapter 1 for more information.)
The following elements of potential changes relating to travel may produce variable trip matrices in
the activity scenario and may have the rule of half applied. The four elements listed below may be
collectively described as ‘demand responses’, particularly with reference to changes in travel between
the do-minimum and activity scenarios.
Pure induced demand/supressed demand

Pure induced travel demand relates to entirely new trips that would not have been made without the
activity or supply. For example, if an activity (or collection of activities) improves access to a shopping
location, a person who in the do-minimum scenario would make an average of one trip to the shop
per week may make an average of two trips to the shop.
Supressed demand is effectively the opposite of induced demand. It is when people would like to
undertake trips but the travel impedance is too great for the trip to occur. It is also when people who
previously made a trip decide to no longer undertake that trip because travel impedance increased
(for example, congestion increases).
Re-distributed trips
A trip where the destination is changed due to the transport or land use activity causing the
impedance on travel to move around different areas of the network, making another destination more
attractive. For example, a home-to-shop trip, where the shopping destination alters as a particular
retail location has become easier to travel to or provides greater opportunities with the activity in
place.
Mode-shifted trips
A trip which switches from one travel mode to an alternative mode due to changes in the transport
system and/or land use bought about by the activity. For example, when a home-to-work trip
previously made by car changes to being made on public transport due to transport system changes
associated with the activity.
Macro-time shifted trips (from one discrete time period to another)

A trip which shifts from one discrete time period to be made in another period. For example, if the
morning commuter period is assessed as 7am to 9am, a trip which no longer occurs within this time
period and instead is made in the inter-peak (after 9am, and before the start of the afternoon or
evening period).
The following elements of potential changes relating to travel do not typically have the rule of half
applied.
Micro-time shifted trips (within discrete time periods)

A trip that has its departure time altered within a discrete time period. For example, a vehicle trip
which leaves 10 minutes later in the morning peak due to an activity reducing the travel time to reach
a destination. This also may be called ‘peak spreading’, in reference to increasing travel delays and
people leaving earlier or later to account for longer trip times.
Re-assigned (re-routed) trips

A trip which continues to travel from A to B in the same time period and by the same mode, but takes
a different route to get there. For example, a home-to-work car trip which travels on Road X in the do-
minimum, but which switches to Road Y with the activity in place due to changed conditions on the
road network.

The study area for an economic assessment should be defined such that any potentially significant
reassignment effects associated with changes due to the activity are captured within the study area.
Where origin–destination demand matrices are used in transport economic analyses and assessment,
the OD trip volumes in the do-minimum and activity scenarios should not differ due to reassignment
(re-routeing).
The following general terms and concepts have important links with demand estimation.
Land use
Land use is used in transport assessments to describe the data and characteristics relating to the use
of land parcels (often zones in a transport model, developed from census area units, statistical areas
or meshblocks) which is used as input to estimating travel demand. Typically includes information
such as the population, households, education roll numbers, employment numbers by type, and
further information relating to family units (for example, numbers of vehicles available, income, age
band, number of school-aged children, number of working aged adults, etc).
Generalised cost and generalised time

Broadly, generalised cost/time is the sum of monetary and non-monetary components of a trip across
all modes. For generalised cost, the non-monetary elements (for example, travel time or wait time) are
converted to monetary units using value of time. Generalised time is the same concept but expressed
in units of time (for example, hours or minutes) and converts the monetary elements (parking charges,
fuel costs, etc) to time units using vehicle operating costs. Generalised time is sometimes also
referred to as ‘generalised cost’.
Travel/transport capacity
The maximum number of travellers or vehicles, typically by travel mode, that can pass through a
system or a specific point in a set period of time. For example, the maximum number of vehicles that
can pass through a traffic lane on a road each hour, or the maximum number of passengers on a bus,
or the maximum number of passengers that can be carried along a rail corridor in one direction in the
morning commuter period.
Travel/transport supply
The total available capacity in the transport system by travel mode and over a time period.
Supply influences people’s choices in travel patterns, such as the destination, mode or route chosen.
As such, the transport system supply influences demand estimation.
Tolling
A fee charged for the use of part of a transport system, typically a road or waterway.
Road pricing
Fee charged for the use of roads. As well as tolling, the following are examples of road pricing:
• distance or time-based charges

• congestion charges
• vehicle-type charges
• fuel-type charges.

2. Demand estimation and mode share > 2.3 Key industry guidance and references
2.3 Key industry guidance and references

In New Zealand there have been two main sources of guidance related to transport demand
estimation.
Historical Economic evaluation manual (EEM) and Monetised benefits and costs manual
(MBCM) guidance
Older versions of the EEM and MBCM have contained demand estimation guidance. The majority of
this content has been updated or redeveloped within this chapter and Appendix 1: Demand estimation
methods and guidance.
Waka Kotahi Transport model development guidelines

The key purpose of the Transport model development guidelines (TMDG) is to provide guidance for
the comparisons carried out between observed and modelled outputs, commonly during the base
model development phase of a project.
Given the focus of this chapter on estimating demand and particularly predicting future demand, there
is little-to-no crossover with the TMDG. This MBCM guidance naturally follows on from guidance in
the TMDG focused on developing ‘base’ models that typically represent current or existing transport
environments and conditions.
2.4 Methods for demand estimation

There are a number of potential approaches to estimate demand, which very broadly include:
• First principle estimates: includes factoring, daily traffic volume estimates and broad simple
estimates of predicted facility use based on comparable examples in other locations.
• Simple mathematical models: includes growth trend equations/calculations, trip generation
rate calculations, mathematical relationship models and elasticity techniques.
• Project transport models: models which do not have the capability to provide travel demand
estimates from land use. May be fed by relatively simple trip generation (and potentially
distribution) calculations (or similar) to approximate future-year demand. May already exist in
a study area or be specifically developed for a project. May include the application of elasticity
techniques and/or other mathematical models.
• Regional transport models: models with the capability to provide travel-demand estimates,
notably for future years, from land use inputs. May or may not have mode share estimation
capabilities. As described further in section 2.10, in New Zealand access to regional models,
which can provide future-year demand estimates, is relatively prevalent in urban areas. These
are typically maintained as transport planning assets and not developed for specific projects.
Some regional models have more comprehensive demand estimation capabilities (for
example, consideration of parking cost and supply or changes in real value-of-time over time),
particularly in the major urban centres. and some have more limited forecasting capabilities
(for example, vehicle trips only).
Given the wide availability of transport models in New Zealand with demand estimation capability,
models are likely to play some role in demand estimation in many economic assessments. This is
particularly notable in assessments involving estimating future use of a transport facility, service or
travel mode in an urban area. Because of this, some of the content of this chapter is focused on
transport modelling. However, the core principles and approaches to demand estimation apply,
irrespective of whether a more sophisticated transport model is applied or another method used (for
example, trend analysis, trip rate analysis, etc).
For transport activities within urban centres, the economic assessment of many transport activities will
involve transport modelling and often applying (and potentially developing) models in order to predict

2. Demand estimation and mode share > 2.5 Definitions of transport model types
future demand estimates. This is particularly relevant in New Zealand’s six main urban centres with
populations greater than 100,000, and is likely to apply to urban areas and towns with populations
down to around 30,000, depending on the activity being assessed and the local transport
environment.
For these reasons, guidance in this chapter is intrinsically linked with the application of transport
models and estimating future-year demand. In effect, guidance and information is targeted towards
analysts utilising transport models and model outputs when carrying out economic assessments.
Some guidance is also provided in other areas not involving transport modelling, for example traffic
volume growth estimates. This information is briefer and more succinct because of the generally
greater complexity involved with applying transport models to estimate demand.
Table 1 provides some guidance on the potential suitability and availability of key sources of demand
estimation information in New Zealand, based around geographic context and transport environment.
Table 1: Guidance on potential suitability of sources of, and approaches to, demand
estimation for different geographic contexts and transport environments
Geographic Potential source/approach to demand estimation
context/transport Regional Regional Project Simple First
environment model with model with models mathematical principle
comprehensive more (may be models (eg estimates
forecasting limited fed by growth (eg
capability forecasting regional trends, trip engineering
capability model generation estimate of
and/or rate etc) facility use)
simple
math
models)
Major urban centre S P P U U
Moderate urban centre S S P P U
Small urban centre P S S P P
Township U U S S S
Rural corridor U U S S S
S = Generally suitable and likely to be available
P = Potentially useful and possibly available
U = Generally unsuitable and/or unlikely to be available
For some locations, it will be preferable to use combinations of models and techniques. For example,
in major urban centres an analyst might apply a regional transport model that takes account of
changing land use patterns, changing public transport services and the impact of new infrastructure
on destination choice, in combination with a project model for more detailed and robust analysis of
route choice and/or estimation in changes in travel time for a specific mode (such as vehicle trips).
2.5 Definitions of transport model types

Table 1 above covers the broad range of methodologies that may be used to develop demand
estimates for an economic assessment: first principle estimates, mathematical models, project and
regional transport models. Project and regional transport models relate to more sophisticated forms of
transport models that are likely to be developed and applied using purpose-built transport modelling
software. The main transport models of this type can be broadly grouped and defined as:
• Project/assignment model: a model that assigns demand (vehicles or persons) to a network

and has no direct incorporation of land use, demographics, mode choice etc. Examples
include traffic assignment models, where vehicles constitute the demand, and public transport

2. Demand estimation and mode share > 2.6 Future guidance development areas
(PT) assignment models, where people using PT make up the demand. This form of model
requires the demands to be specified as an input. They may be developed with a specific task
or transport project in mind. These models range from single intersections to entire inter-
urban areas, and are built using a wide range of types of software. Traffic assignment models
will generate travel times between zones, and may produce vehicle emissions, queues, etc.
PT assignment models will determine the number of people using each of the modes and
services incorporated, as well as calculating travel times for people and generating fare
revenues.
• Regional transport model: a transport model that consists of an assignment model with a
demand model that responds to changes in land use and transport supply. These models are
concerned with the movement of people, vehicles, or goods. Some will provide information on
mode share, induced travel, heavy vehicle volumes, etc. Transport models are built on
relationships between factors such as land-use activity, demographics and transport supply,
and commonly cover the movement of transport demand across larger geographic areas
(although not necessarily regions). The forms of these models can be 3-stage (or step), 4-
stage, 5-stage, tour-based, or activity-based. May also be referred to as strategic,
macro(scopic), or demand models.
• Land-use model: a model that estimates use of land parcels, including residential, industrial,
business, etc. It is often responding to accessibility provided by the supply. When
complemented with a transport model, it forms a land-use transport interaction (LUTI) model,
which balances the distribution of new land use with the accessibility of the transport system.
Project/assignment models and regional transport models are relevant to demand estimation and are
often directly applied to economic assessments. Currently land use models are not prevalent and less
directly used in transport activity economic assessments.
2.6 Future guidance development areas

The MBCM is a living document. Currently this chapter does not cover the following areas, but there
is the potential for guidance to be developed in these areas in the future:
• coastal shipping
• air transport
• land use transport interaction (LUTI) modelling and how this may/can be deployed in
developing demand estimates used for economic assessments
• significant technology disrupters such as autonomous vehicles (AV)
• dramatic shifts in current vehicle ownership models; for example, those brought about by car
sharing, AVs or other disruptions.
• significant changes to mode share brought about by technology and/or significant policies
and/or behaviour changes; for example, micro mobility technologies, electric bicycle uptake,
large-scale climate change behaviour response, or significant climate change transport policy
changes.
The last two points above are currently impacting travel behaviour and are therefore of interest in
relation to transport demand estimation. Some aspects, for example micro mobility options (such as
e-scooter share services) and electric bicycle uptake, may already be having some effect on
transport. Section 7.3 Demand estimation sensitivity tests provides some further background to these
issues and some suggestions around sensitivity tests that may partially reflect these changes.

2. Demand estimation and mode share > 2.7 Drivers of trip-making and mode share
Demand estimation principles and considerations

This section describes at a high level:
• the drivers of trip-making and mode share

• considerations relating to land-use projections and regional planning structures
• transport modelling in New Zealand and how this relates to demand estimation
• how demand estimates can influence economic assessments, and key factors and
elements related to this (such as optimism bias in demand forecasts), and some more
specific considerations relating to fixed trip or variable trip matrix assessments
• considerations relating to uncertainties in demand estimation.
2.7 Drivers of trip-making and mode share

This section describes important, high-level, background concepts that relate to transport demand
and mode share estimation.
Trip-making considerations
The demand for transport is derived, it is not an end in itself. People travel in order to satisfy a need
(for example, related to work, leisure or health), undertaking an activity at particular locations (Ortúzar
and Willumsen 2011). Transport demand is the distribution of these activities over space and time.
Transport demand and supply has a strong dynamic element, for example, in urban areas a large
proportion of this activity takes place during weekday morning and evening peak periods.
Transport demand (trip-making) is influenced by various economic and socio-economic factors such
as composition of household or family unit, household size, age (school or working age), employment
status, disposable income, accessibility to activities and services, and social and environmental
beliefs. Historically, trip-making was also considered to be influenced by vehicle availability and/or
ownership, although this is likely to be evolving, particularly in larger urban areas, with access to car-
sharing companies and a shift to more social and environmental responsibility.
Equilibrium of transport demand and supply

Transport demand and supply should be balanced or aligned. This has been equated with the
concept that the total demand in a network across a time period should not significantly exceed the
total network capacity
More generally, it is anticipated that transport demand and transport supply should be broadly
correlated, particularly during typical peak periods. For example, when carrying out a study in a small
town where through traffic is provided for by a two-lane road, it would be anticipated that demand
passing through the town wouldn’t significantly exceed (for example, by a factor of two or greater) the
capacity of a two-lane road for long periods of time (more than one to two hours) during typical peak
periods.
In New Zealand’s major urban centres there are clearly cases where demand exceeds supply for
periods of time, and care is required in these circumstances. Some degree of ‘reasonableness’ in the
balance between supply and demand still applies in these locations; for example, it is unlikely that
people would choose to live in a location that would result in spending excessively long times (more
than several hours) in stationary queues during regular day-to-day travel.
Model developers consider and account for existing and potential future congestion where feasible.
For example, the current trend is to develop models (all forms) that cover increasingly larger
geographic areas and longer time periods. These two factors are likely to improve the application of
models to testing do-minimum and activity scenarios, and estimating economics values.
Analysts should identify in reporting scenarios where demand is exceeding supply for long periods of
time, detail any adjustments or treatments that have been applied to account for or improve the

2. Demand estimation and mode share > 2.8 Land use projections and regional planning structures
robustness of the assessment due to excess demand, and note how these aspects may effect
outcomes. If demand is significantly greater than supply, this will lead to undesirable situations where,
for example, the model may gridlock, there may be instability in the model results, and the models
may fail to reach convergence.
Mode choice considerations

Mode choice is important as it affects the general efficiency with which we can travel in urban areas,
the amount of urban space devoted to transport functions, and whether a range of choices is available
to travellers (Ortúzar and Willumsen 2011). The characteristics that influence mode choice include:
• the characteristics of the trip maker: vehicle availability and/or ownership, driver licence,
household structure, income, cost sensitivity, residential density, and social or environmental
responsibility
• the characteristics of the journey: trip purpose (for example, journey to work may be easier by
public transport), time of day, whether alone or with others, weather, and the origin and
destination and how well these align with access by the various available modes
• the characteristics of the transport facility:
o quantitative aspects: travel time (in-vehicle, waiting, modal transfers, and walking),
monetary costs (fares, tolls, fuel, etc), availability and cost of parking, reliability of
journey travel time, and regularity of service
o qualitative aspects: comfort, convenience, safety, perception of safety, opportunity to
complete other tasks while travelling
o relativity of these aspects with other available modes and facilities.
Where there are several methods and/or routes available, an individual’s choice of mode can be
broadly reflected in the modelling system as minimising their generalised cost while considering
qualitative aspects that cannot be valued. In a transport model, the public transport generalised cost
is likely to be made up of some, or many, of the above quantitative aspects of the transport system
(as is the choice to travel by public transport). Some characteristics influencing mode choice cannot
be explicitly represented in a transport model and simplifications and/or factors are incorporated. For
example, it is difficult to explicitly model qualitative aspects of transport facilities, and the social and
environmental beliefs of individuals.
Estimating demands and mode share for the purpose of an economic assessment, particularly in
future years, effectively includes considering, and often accounting for, quantitative aspects (for
example, travel time) and qualitative aspects (for example comfort and perceived safety).
2.8 Land use projections and regional planning structures

As described in section 2.1: Key concepts, land use data is an input to regional transport models.
Future-year travel demand estimates are developed from the projected future land use, an agreed
description of the future transport supply (infrastructure, services, etc), and, for some models,
assumptions on policies and/or charges.
Population is a component of land use data. Importantly, Statistics NZ produces population estimates
and projections. In this context, estimates describe the size and characteristics of the population at a
past date. Projections describe trends to show what the population may be like in the future. These
population projections are a key driver of future travel demand and are likely to be an important input
to travel demand estimation.
Land use data is often provided by local authority planning departments. It is often their role to
develop and check this data, and transport planners may inherit this information for developing travel
demand estimates.

2. Demand estimation and mode share > 2.9 Mathematical methodologies and elasticities
In addition to population, other land use data that may be input to regional transport models includes
the number of households, household composition, work or labour force, vehicles, employment by
type, and school roll.
In some regions, the road controlling authorities have systems and structures in place that may
develop a ‘given’ set of future land use projections, which are sometimes linked to the Statistics NZ
population projections. In addition, a ‘given’ set of transport system supply and service assumptions
may exist. These in turn will feed an associated set of future-year travel demand estimates. In other
regions these structures may be less formalised.
This data and these regional systems and structures are important aspects to consider when
receiving and/or developing future demand estimates. Where regional assumptions around land use
projections are less structured, more effort may be required by the analyst to check the underlying
assumptions feeding demand estimates and to be particularly mindful of the potential for optimism
bias in these assumptions. Optimism bias relating to projections and demand estimates is described
further in the section 2.11.
An Excel workbook is available on the MBCM web page to record checks on model specification, as
well as coarse and detailed checks of model outputs. These are designed to help an analyst or peer
reviewer to carry out and record some of the required sense checks, and provide Waka Kotahi with
information for assurance purposes.
Where these regional structures do exist, a level of discussion and collaboration is anticipated
between the analyst and the organisations involved in the development of land use and/or transport
supply information, so that the underlying assumptions in the demand estimates used in an economic
assessment are well established and understood. For example, existing and future traffic demands for
a subset area of a regional transport model may be provided for a project model. In this case, it is
important to understand the assumptions in the regional transport model for the future year. What PT
services are anticipated? What is the assumption of PT fares or parking costs or parking availability?
All of these assumptions in the regional transport model will be inherent in the project model and the
analyst should have, and communicate, a clear understanding on the basis of the demand estimate.
2.9 Mathematical methodologies and elasticities

A range of mathematical methods, approaches, and models may be used in estimating the demand
for a transport activity. These range from a straightforward estimate of the number of people using a
particular facility, vehicle volume trend analysis or trip generation/distribution assessments, through to
more complex mathematical relationships.
Elasticity calculations and applications are one mathematical approach to estimating changes in
demand. As described in section 2.1: Key concepts, elasticities are defined as the relative change of
demand when compared to a causal variable. Elasticities may be used to estimate a range of demand
responses, more commonly the estimated change in public transport patronage due to a change in
the relevant transport service attribute (such as fares), estimated change in parking rates due to a
change in service attribute (for example, parking supply or parking costs), and estimated change in
freight by mode based on the relative price between road and rail transit. More information on
elasticities and cross-elasticities, including public transport and freight service values, are provided in
Appendix 1.
A mix of approaches and methods may be applied. For example, within a project/assignment
transport model a mathematical calculation of trip generation rates and assumed trip distribution may
be applied to estimate demands for a development area. Within the same model, an elasticity may be
used to predict some further aspect of demand response.
2.10 Demand estimation and transport modelling in New Zealand

As described in section 2.4, transport modelling is likely to play an important role in producing
demand estimates used in economic assessments in urban areas, depending on the transport activity

2. Demand estimation and mode share > 2.11 Optimism bias
being assessed. Broadly there are two types of models that are currently most relevant to the
estimation of demands and application in economic assessments:
• regional models, which have some capability to estimate demand responses to changes in
land use and transport supply
• project models, which have no direct incorporation of land use, demographics etc. where
demands are specified as an input to the model.
Many of New Zealand’s urban centres with populations greater than around 30,000 are represented
by a regional model. All urban areas with populations above 100,000 are represented by a regional
model.
The larger the population the more complex the transport environment, and generally the more
components and aspects to the transport model system that are available. The transport models that
cover the main urban regions (Auckland, Tauranga, Waikato, Wellington, Christchurch, Queenstown
Lakes and Dunedin) have, or are likely to have in the future, some aspect of mode choice estimation
capability. Some smaller urban area models have an aspect of mode choice capability, but most of
these types do not.
A list of transport models is provided in the first table in Appendix A in Urban transport modelling in
New Zealand – data and practice and resourcing (Smith 2019). This largely focuses on regional
models – that is, models with capability to produce future-year demand estimates. As it is only a
snapshot taken at a certain time, it cannot be considered a comprehensive list of available transport
models. However, it does provide a good indication of the coverage of this form of transport model
across New Zealand’s urban centres.
2.11 Optimism bias

Demand estimates are often an important, if not key, driver of estimated benefits. If the use of a
transport activity is projected to be low, it is more probable that the economic returns will be low and,
conversely, if projected use is estimated to be high there is a higher probability that the economic
returns will be greater. The following examples illustrate this point.
• If one aspect of upgrading a priority intersection to an alternative intersection form is related

to reducing travel delays, the projected future-year volumes at the intersection are likely to be
an important driver of the estimated value of travel time benefits in the economic outcomes.
• If assessing a significant public transport project in an urban centre, the estimated change in
mode share is likely to drive a number of important economic elements, such as the benefits
to PT users, potential reduction in other travel modes (for example, reducing the volume of
private car trips) and the associated travel time saving benefits for these modes.
Optimism bias is an important consideration due to the likely correlation between demand estimates
and economic outcomes. Optimism bias (or appraisal optimism) is the demonstrated systematic
tendency for people to be overly optimistic about the outcome of planned actions. This includes
overestimating the likelihood of positive events and underestimating the likelihood of negative events
(Charles 2011).
In economic assessments, optimism bias has been described and recorded as resulting in costs
tending to be underestimated and/or demand (and hence benefits) overestimated (Australian
Transport Assessment and Planning 2018c).
Overestimation of demand and, notably, the projected use of a transport activity in economic
assessments is an important concern and the analyst should be mindful of responding appropriately
to this when developing and using demand estimates.
Sensitivity tests are one potential response to optimism bias in demand estimation. The following
examples demonstrate this point.

2. Demand estimation and mode share > 2.12 Factors and considerations influencing demand estimation
• Using lower levels of land use uptake to produce alternative future-year demand estimation
scenarios with fewer vehicle trips.
• Adjusting a model input, parameter or assumption that is sensitive and known to reduce PT
mode share to produce an alternative future-year demand estimation with a higher volume of
vehicle trips and lower number of PT trips (or vice versa). For example, reducing the
frequency of planned PT services, increasing modal transfer penalties reflecting people’s
general dislike of transferring.
See section 7.3: Demand estimation sensitivity tests for further discussion around sensitivity tests
relating to demand estimates.
Other reasonable responses to optimism bias include:
• analysts and clients being aware of the potential for it to occur and simply taking steps to
avoid or reduce the influence of optimism bias (for example, by not basing economic
assessments on a single demand estimation scenario with potentially high population levels
and land use development)
• clearly identifying in reporting any areas of the assessment/analysis where optimism bias may
be present and noting the effect it may have on outcomes.
2.12 Factors and considerations influencing demand estimation

As noted, when developing demand estimates there are a wide range of elements and factors that
influence trip-making and mode choice. In some locations, environments and scenarios these factors
will have little to no affect, because they are unlikely to change or influence travel responses in the
future and therefore will have no real effect on the assessment of transport activities. For example,
the cost of parking has no impact in a rural town where parking is predominantly free, abundant, and
likely to remain so in the future. In other situations and locations, certain elements and factors will play
an important role in demand estimation; for example, PT mode choice in a major urban centre with a
congested road network and a significant supply of reliable PT services.
Table 2 and Table 3 provide overview guidance on when certain elements may influence demand
estimation and may need to be considered when developing or using demand estimates in an
economic assessment. While these should not be considered a comprehensive list of all the potential
elements influencing demand estimation, they provide context of the type and form of certain aspects
and where and/or when these may be important.
Table 2 is focused on project models that do not have demand estimates fed by regional models,
calculations, spreadsheet methodologies, trend analyses and similar. Table 3 is focused on regional
transport modelling.

Table 2: Guidance on factors affecting demand – project model and calculation focus
Factors affecting demand estimates Project model/calculation approaches
Network Short Spreadsheet Straightforward
project corridor/ or similar calculations
model intersection equations/
(not linked model (not models/
to/fed by linked calculations
regional to/fed by
model) regional
model)
Suitability and appropriateness of I P P U
elasticity methods, relationships and
values
Knowledge and certainty of local land I I P P
use changes
Knowledge and certainty of local I I P P
transport system and supply changes
Source and suitability of trip rates I P P U
Source, method and appropriateness I P P U

of distribution analysis
Suitability of factors/trends selected in I I P U
factoring methods
Robustness and sample size of P P I I
historical data used in trend analysis
Suitability/appropriateness of U U P I
engineering estimate methods of
predicted facility use
I = Generally an important element
P = Potentially or partially important
U = Generally unimportant and non-critical
Expanding on Table 2, some key high-level considerations relating to each of the elements, methods
and approaches that may be used to develop demand and mode share estimates are noted below.
• Elasticity methods, relationships and values: elasticities are based on historical and
international studies into the relationships and responses in transport demand and
characteristics of the supply. An elasticity value, such as the public transport demand change
response to a fare change, may be based on analysis from a particular city over a particular
time period. The relevance and suitability of any elasticities used need to be carefully
considered when applying them to a particular activity in a certain location. Sensitivity tests –
that is, varying the elasticity value or values used – are one approach to examine the
response and suitability in relation to the specific activity in the local context.
• Local land use changes: understanding local land use changes, and the degree of certainty
related to them, are an important consideration when accounting for them in demand
estimations. Examples include new residential or retail developments nearby but outside the
study area, the impact of which will affect the demand estimate. Optimism bias may be an
issue; for example, assuming all potential plan changes occur, and are fully developed, within
a short-term horizon could overestimate future demands and the benefits associated with a
particular activity.

• Local transport system and supply changes: as with land use, the degree of certainty
around local potential transport system and supply changes and the timing of these may be
an important consideration.
• Trip rates: the suitability of trip rates used in the development of existing and future demand
estimations should be checked and understood. For example, a retirement village trip rate
based on a small number of surveys in urban Sydney may not be relevant to a rural town in
New Zealand.
• Distribution analysis: the method and data used to develop first principle trip distributions
associated with a certain area and/or specific land use activity should be carefully considered.
Depending on the location and scale of the activity, if a first principle distribution approach is
used to estimate origin–destination demands it may be appropriate to check the estimated
distribution against observed data (for example, a vehicle number plate survey determining a
key distribution aspect, such as the number of trips passing directly through a town centre).
• Factoring methods: the relevance and suitability of any factors used to estimate demand
and mode share should be considered when applying this approach to a particular activity in a
certain location. For example, annual freight growth rates in rural Southland may not be
appropriate in urban locations with low volumes of freight passing through the study area.
• Trend analysis: trend analysis, which is used in the development of existing and future
demand estimations, should be carried out in a careful and considered manner. Seemingly
small inaccuracies in a trend analysis may have a significant effect on future projections; for
example, using a small sample of historical counts to estimate an annual growth rate and
applying this growth rate to predict 20, 30 or 40 years into the future. As an example of
checks and considerations, if historic traffic counts are used to estimate growth these should
be based on a sufficient sample (the number of robust data points over time) and be checked
for robustness (such as consistent vehicle classification data, seasonality effects, local
issues/events, longer term patterns in the wider economy, etc.). Sensitivity tests may be one
approach to account for broader issues such as wider economic patterns.
• Engineering estimates of predicted facility use: similar to elasticities and trip rates, the
relevance and suitability of an engineering estimate approach should be considered when
applying a method to a particular activity in a certain location. Again, using approaches that
are based on examples in major urban areas and applying these to more rural locations is an
example of relevancy or suitability check, and sensitivity testing may be one approach to
account for this form of issue.
Table 3 provides some high-level guidance on elements that may influence demand estimation,
focused on regional transport modelling.

Table 3: Guidance on factors affecting demand – transport modelling focus

Factors affecting demand estimates Geographic context/transport environment
Major urban Moderate Small urban Township,
centre urban centre rural
(population centre (population corridor/
roughly (population ~30,000– area
greater than roughly 100,000) (population
500,000) between roughly less
100,000– than
500,000) 30,000)
Population structure/make-up I I I I
(particularly age)
Household/family structure (retired, I I I I
school-age children, in workforce, etc)
Vehicle availability/access to a vehicle I I I I
Access to alternatives modes and I I P U

infrastructure (public transport, cycling,
etc)
Public transport – service coverage, I I P U
service frequency, charges
Residential density – accessibility to I I P U
activities
Parking – charge and availability of I I P U
supply
Road congestion/delay I I I I
Road pricing/tolling I P U U
Route choice I I P P
Policies and practices (work hours, I P U U

working from home, travel plans, etc)
Technology influencing behaviour I P U U
(online shopping, work/school travel
plans)
I = Generally an important element
P = Potentially or partially important
U = Generally unimportant and non-critical
Expanding on Table 3, key high-level considerations relating to each of the elements, methods and
approaches that may be used to develop demand and mode share estimates are noted below.
• Population structure/make-up: a person’s age (for example, school age, working age,
retirement age) and characteristics (for example, whether they are in the workforce or not) are
a key driver for transport demand. This is particularly important considering localised demand
around specific attractors such as schools.
• Household/family structure: the household or family unit strongly drives transport demand
in terms of the number of trips for certain purposes.
• Vehicle availability/access to a vehicle: access to a vehicle influences mode choice and
was historically a measure of overall trip-making, with households with more cars generally
undertaking more trips.

2. Demand estimation and mode share > 2.13 Fixed trip matrix and variable trip matrix assessments
• Access to alternative modes and infrastructure: examples include proximity at both ends
of a trip to a bus stop/service, or whether there is a quality cycleway available. These
elements will influence both transport demand (including the destination a person chooses)
and mode choice.
• Public transport coverage/frequency/charges: access to a bus stop, a public transport
service with good connectivity between a person’s origin and destination, the frequency of the
service, and the monetary fares will all influence demand by mode and also the destination
selected (for example, for shopping, recreation, etc).
• Residential density: how close a person is to activities (for example, shops or school) is a
key driver of demand (destination chosen and the resulting trip length) and mode choice.
• Parking: parking charges, availability of supply, and location of supply relative to the final
destination influence mode choice and demand (destination selected, particularly for
discretionary trips such as shopping or recreation).
• Road congestion/delay: travel times by road are a key driver of choice of mode, for
example, rail verses car. Congestion and delays are a significant factor in which route a driver
chooses, and a primary component of a demand estimate, for example, the demand for a
proposed new bypass.
• Road pricing/tolling: monetary charges are a key driver of the mode a person selects,
heavily influenced by the reason why that person is travelling. For example, a business
traveller is more likely to pay higher out-of-pocket costs to minimise the time they are not
productive. Human responses also need to be considered, as theoretically ‘cheaper’ choices
are not necessarily preferred by all people.
• Route choice: the route people choose, whether by road, cycleway, or public transport
service, influences the demand for each element of the transport system (for example,
specific roads).
• Policy and practices: these include government policies and commercial and/or employment
practices. Examples include whether employers offer more flexible working hours or working
from home, and the influence of work or school travel plans.
• Technology influences: examples of technologies that influence demand and/or mode
choice are online shopping, food delivery services, car sharing companies, and electric bikes
and electric scooters. These technological advances influence travel behaviour in terms of
how people choose to travel and where they decide to go, particularly for discretionary trips.
2.13 Fixed trip matrix and variable trip matrix assessments

As described in section 2.1: Key concepts, demand estimates assigned to the transport network are
commonly represented in the form of origin–destination (OD) trip matrices. When considering the OD
demand matrix in a study area for an economic assessment, there are broadly two approaches that
may be carried out:
• fixed trip matrix (FTM): over time, the OD demand for a particular travel mode is the same in
the do-minimum and activity scenarios
• variable trip matrix (VTM): over time, the OD demand for a particular travel mode is different
between the do-minimum and the activity scenario due to the influence on demand response
in the study area from the activity.
How significantly the demand for a particular travel mode in the study area is influenced by the activity
and, associated with this, whether an FTM or VTM approach may be required, is an important
consideration.

2. Demand estimation and mode share > 2.13 Fixed trip matrix and variable trip matrix assessments
The ‘rule-of-half’ is a simplifying assumption that is used to the calculate the benefits to transport
system users where there is a difference in the do-minimum and activity demand estimates. That is,
the rule-of-half is applied in VTM analyses.
Demand responses that would typically result in a VTM assessment and have the rule-of-half applied
include:
• pure induced demand: entirely new trips that would not have occurred without the activity
• re-distributed trips: trip destination is changed due to the activity
• mode shifted trips: trip changes from one travel mode to an alternative due to the activity
• macro-time shifted trips: trip shifts from one discrete time period to another.
The following elements of potential changes relating to travel do not typically have the rule-of-half
applied. Or, in other words, these responses do not result in a difference in OD demand matrices
used in the do- minimum and activity scenarios;
• micro-time shifted trips: trip departure time is changed within a discrete period (more
information is provided in Applying peak spreading in Appendix 1)
• re-assigned trips: a trip travelling from A to B in the same period and by the same mode but
takes a different route to get there.
The above demand responses are described further in section 2: Key concepts.
Table 4 provides some high-level guidance on when VTM and FTM approaches may need to be
considered.
Table 4: High-level guidance on potential for VTM or FTM approaches

Factors influencing variable or fixed Geographic context/transport environment
trip matrix approaches Major urban Moderate Small Township,
centre urban centre urban rural
(population (population centre corridor/
roughly roughly (population area
greater than between ~ 30,000– (population
500,000) 100,000– 100,000) roughly
500,000) less than
30,000)
Large-scale PT/active mode activity VTM VTM VTM VTM
Moderate-scale PT/active mode VTM VTM VTM P
activity
Smaller-scale PT/active mode activity P P P P
Large-scale roading-based activity VTM VTM P P
Moderate-scale roading-based activity VTM P P FTM
Smaller-scale roading-based activity P P FTM FTM
VTM = Probable that activity will influence modal demand significantly, VTM likely
P = Potential for VTM or FTM approach
FTM = Unlikely activity will influence modal demand significantly, FTM likely
Table 4 provides general guidance, rather than specifying a prescribed approach. For example, there
are cases where an FTM approach could be applied to assess a large-scale roading activity in a
major urban centre. This could involve assigning the do-minimum demands to the activity network
scenario (or vice versa, assigning the activity demands to the do-minimum) directly in the regional
model; the FTM and/or VTM approaches may be considered as a sensitivity test.
Methods for applying FTM and VTM techniques are described further in Appendix 1 (Fixed trip
matrices and Variable trip matrices).

2. Demand estimation and mode share > 2.14 Demand estimation uncertainty
Applying regional transport models

As indicated in Table 4, a more significant activity focused on public transport and active travel
modes, and which has a key or supporting objective of driving mode change and may achieve a level
of demand response, is likely to involve a VTM approach for vehicle, person, and active mode
demand estimations. Similarly, a large-scale roading activity in a major urban centre would also be
likely to involve a VTM approach.
Where a demand response is anticipated, in New Zealand’s main urban centres demand estimation
and, to some degree, assessing economics outcomes of the activity is likely to involve some form of
application of existing regional transport models. Review of the model is required to determine if it is
suitable for the assessment. For example, the assessment of a major new cycleway in an urban area
would need to consider how cyclists are represented in the modelling tools available. See section
2.10 for more information on the availability of regional models in New Zealand.
2.14 Demand estimation uncertainty

Demand estimates, and particularly estimates of future-year forecasts, all have some degree of
uncertainty. This should not be considered a weakness, or necessarily a problem, but acknowledged
as a reality. In many cases this can present an opportunity to test and understand the range of
potential outcomes associated with an activity; for example, by considering key uncertainties in the
demand estimates a range of forecasts could be developed and used as sensitivity tests on economic
outcomes.
Section 7.2: Demand estimation sensitivity tests contains information and guidance relating to
sensitivity tests and risk assessment with particular reference to demand forecasts.
In transport demand forecasting, there is necessarily some reliance on past and current trip-making,
travel behaviour and trends. The current awareness around local and global environmental issues
highlights a need to be flexible moving forward. In developing future transport estimates, analysts and
the methodologies applied should consider the ability to adapt, adjust and examine elements that
could have a large impact on travel demand. Important considerations into the future can and may
include:
• climate change: particularly government policy responses and personal behavioural changes
that affect transport and travel
• social equity in transport: analysts may need to consider the structure and specification of
models and their resulting ability to assess and interrogate transport equity changes and
outcomes
• technology disruption: particularly those leading to changes in vehicle ownership levels; for
example, autonomous vehicles, micro-mobility, e-mobility, etc. Other potential future changes
will bring another dimension to consider.
Forecast horizons and uncertainty

The standard analysis period is 40 years (see section 1.6). Although there are exceptions (for
example, short-lived operations activities such as signal optimisation), carrying out an economic
assessment will generally involve development of one or a number of future-year demand estimates.
Economic inputs are then typically projected (for example, with a linear trend) between the different
assessment years (for example, base year and future year(s)).
Generally, the further the forecast horizon (that is, future-year demand estimate) is from the current or
base year, the greater the associated uncertainty will be. One way to think of this is that the longer the
forecast horizon, the wider the range of possible futures.
The analyst will need to determine an approach to estimating the inputs to the economic assessment
across the analysis period. This will include weighing up the increasing uncertainty in future-year
forecasts, along with potential demand/capacity challenges in longer-term forecasts, against the

2. Demand estimation and mode share > 2.14 Demand estimation uncertainty
ability to appropriately or adequately project economic outcomes. The discounting rate may also need
to be considered; for example, there could be little value in forecasting long-term horizons where the
discounted benefits in long-term future years become small.
Forecasting travel demand into the future may very broadly be thought of in terms of the following
horizons:
• short-term horizon, forecasting forward 10 years or less

• medium-term horizon, forecasting forward 10–20 years.
• long-term horizon, forecasting forward more than 20 years.
In congested urban centres, regional models can and do produce longer-term demand forecasts.
However, there is a noted challenge and often concern with use of longer-term forecasts in project
models being applied for economic assessment, particularly the types of project models where
vehicles are physically represented on the network, for example, microsimulation traffic models.
Techniques for scenarios where demand exceeds network capacity are described further in Appendix
1 (Fixed trip matrices and Variable trip matrices).
The issue of congested urban networks and project model application complicates considerations
around forecasting horizons for economic assessments.
Some broad considerations around forecast horizons are noted below:
• Rural area, smaller town and similar (largely uncongested): for smaller-scale activities,
although short- to medium-term forecasts may be sufficient for an uncomplicated projection
and economic assessment, in contrast it may be straightforward to produce a small number of
long-term forecasts without the sensible analysis of scenarios suffering from issues
associated with higher levels of congestion.
For larger-scale activities, it is likely to be relevant to develop a small number of longer-term
forecasts. That is, develop several sensitivity test scenarios for demand estimates 20 years or
more into the future.
• Small- to moderate-sized town, inter-urban area, urban periphery and similar (low to
medium levels of congestion): for smaller-scale activities, a small number of short- and
medium-term forecasts are likely to be sufficient for economic assessments.
For larger-scale activities, several medium- to longer-term forecasts may be appropriate, and
there is the potential that applying longer-term forecasts in project models may generate the
need to apply techniques to manage scenarios (for example, the do-minimum) where the
demand exceeds network capacity or breaches analytical limits.
• Moderate to major urban centre (medium to high levels of congestion): for smaller-scale
activities, several short- to medium-scale forecasts and demand sensitivities may be
appropriate.
Large-scale activities in congested networks are the most complex scenario and require more
careful consideration around the development and treatment of forecasts. As noted above,
regional models can and do produce longer-term forecasts; however, there are issues with
these. such as the level of uncertainty in the estimates and the levels of congestion that these
forecasts may produce in project models.
A consideration in carrying out economic assessments in larger urban areas, where higher
levels of congestion are experienced, is to test economic outcomes in two models: a regional
model, utilising longer-term forecasts where demand may not be constrained to supply when
passing through the network; and a project model, utilising short- to medium- term forecasts
where the demand and capacity representation is more overt. The project model application is
likely to include techniques to manage demand exceeding capacity. Sensitivity tests of
demand forecasts would be likely in both models.

2. Demand estimation and mode share > 2.15 Checks, reporting and reviewing
2.15 Checks, reporting and reviewing

There are several pieces of information that are expected to be derived and reported, regardless of
the methodology or methodologies that are used to develop demand forecasts and/or predict the use
of a transport facility, service or travel mode. Some are expected in order to complete certain
evaluation procedures described in this manual, and/or as generally anticipated information to be
provided in assessment documentation and through review processes.
This section describes reporting that is expected in relation to MBCM evaluation procedures, the
sections that follow describe more general demand estimation and forecast checks and potential
reporting elements.
Reporting demand estimates

Methodologies and assumptions
It is anticipated that the methodology or methodologies that are used to develop demand estimates in
the do-minimum and/or activity scenarios are described in reporting. This includes any assumptions
made, particularly regarding future transport numbers and growth rates.
Some further information and guidance on reporting information around demand estimates and
forecasts is provided in Sense checking forecasts.
Demand estimates and forecasted facility use
The following information is generally anticipated to be reported:
• For an existing specific facility, the current or ‘base’ number of users and the change
anticipated by the activity. For example, improvements to an existing facility, such as
widening, safety improvement to an existing active mode facility, or increasing the PT service
rate. This also includes any anticipated change in the future trend in user numbers; for
example, an anticipated higher rate of growth from year X to year Y compared to the do-
minimum.
• For a new specific facility, the estimated numbers predicted to use the facility and the source
of these new users. For example, a new roading connection, active mode connection, or new
public transport service. This includes any anticipated trends in user numbers; for example,
rate of growth from year X to year Y. Identifying new users may include mode transfer, newly
generated trips, or change from a parallel existing facility to the new facility.
• Total existing and future-year study area demands, by time period and by mode, and
associated growth (absolute difference, percentage difference, and it is often useful to
express it as the percentage growth per annum).
• In situations where a variable trip matrix approach is being applied or is being considered, the
differences between the do-minimum and the activity demands, by time period, and by mode.
• Where there may be a potential change in mode share, the range of information provided is
likely to increase. For example, modelled parameters and assumptions may require more
detailed documentation and explanation.
• The time period demand estimates and any interpolation applied to estimate outcomes. For
example, information may be only be available for, or focused on, weekday morning,
interpeak, and evening periods, and estimates may be made of overnight, daily, and/or
weekend/holiday periods. Reporting should describe any time period factors, interpolation,
and/or methodologies.
• For activities that feature user charges (for example, public transport facilities and services),
the assumptions around the user charges applied in the assessment will need to be
documented. Including any base existing average user charge, proposed user charges and
potentially the maximum charge users may be willing to pay. In this manual the maximum

user charge is defined to be the price above which no one would use the service under
consideration.
Sense checking forecasts

This section describes some more general sense checking and reporting that may be carried out in
association with demand estimates, with a particular focus on transport modelling.
Regional model transport demand forecast checks and reporting
While not an exhaustive list, the following are key elements that are likely to be checked and reported
when producing a set of forecasts from a regional transport model. This may apply to general regional
transport modelling tasks; for example, updating the transport demand forecasts for a region, and
may be relevant to directly applying a regional transport model to estimate demand forecasts for a
transport activity.
• Growth in model inputs compared with estimated transport demand growth. For example,
input land use growth (population, households, and vehicle availability being the most critical)
compared with total trips growth.
• Total trips by travel mode year to year, and the mode split year to year.
• Estimated base and future trip rates (for example, trips per person or trips per household).
• Sector-to-sector travel growth trends. Sense checking the geographic travel pattern growth
trends between key areas of the network (for example, central business district to/from north,
south, east, west) against the areas of anticipated future land use development.
• Sense checking the trip length distribution. For example, if an urban area is ‘spreading’, it may
be anticipated that future years have longer trips.
• Plots/figures of changes are likely to be useful. For example, network flow difference plots
(base to future) and/or figures showing transport trip growth by geographic location.
The above information may be reported for any sensitivity tests and/or scenarios that have been
developed. Reporting would include noting what parameters have been adjusted as a sensitivity and
the logic behind this.
Project/assignment transport demand forecast checks and reporting
For a project/assignment model that is not fed by a regional transport model (for example a project
model which uses first principles trip generation/distribution demand inputs), the above sense checks
and reporting are likely to apply.
For a project model that is fed by a regional transport model, the following key elements are likely to
be checked and reported, with sense checks focused to some degree on comparison between the
regional model input and the project model incorporation of these inputs.
• Total OD matrix growth by time period, and where appropriate, vehicle classification and
travel mode (absolute difference, percentage difference, and it is often useful to express it as
a percentage growth per annum).
• Sector growth by time period and, where appropriate, vehicle classification and travel mode.
Particularly relevant may be the percentage sector growth in the regional model inputs (for
example, cordon matrix), compared with the percentage sector growth in the estimated
project model demands. Similar geographic patterns would be anticipated; for example,
growth to/from areas where higher rates of future land use development are anticipated.
• Trip length distribution by time period and, where appropriate, vehicle classification and travel
mode, comparing the regional model input (for example, cordon matrix) with the calculated
project model growth.
• Flow difference plots and geographic location growth figures are also likely to be useful.
Reporting would be expected to describe the method and calculations used to develop forecasts from
a regional model into a project/assignment model, any factors applied (for example, factors where the

project model represents different time periods than the regional model), and any adjustments made
to the ‘raw’ growth estimated from the regional model (for example, any growth constraint techniques
applied, such as those described in Appendix 1 (Fixed trip matrices procedures).
Any sensitivity tests developed are likely to be reported and include sense checks, such as those
described above. Reporting would include describing the logic behind any sensitivity test scenarios
used.
Mathematical methodologies and elasticities
Any growth rates (for example, trend rates, trip generation rates and similar) and any elasticity values
used in demand estimation should be directly reported. Key reported information relating to
forecasted or estimated demand is likely to include:
• per annum demand growth rate by time period and, where appropriate, vehicle classification
and travel mode
• total study area demand growth/change by time period and, where appropriate, vehicle
classification and travel mode
• change in travel mode by time period (total trip change and percentage mode split change)
• change in study area demand with and without an activity.
Any sensitivity tests developed are likely to be reported and include the key information above.

3. Benefits
3. Benefits
Waka Kotahi has developed a common Land Transport Benefits Framework for use across the entire
benefit management process. These benefits are mode neutral and are aligned to the Ministry of
Transport’s enduring Transport Outcome Framework (MoT TOF). High level benefit clusters have
been developed to demonstrate meaningful alignment between the new mode neutral benefits and
the MoT TOF.
In addition, the assessment process has evolved towards a more comprehensive approach through
the introduction of appraisal summary tables (AST). AST systemise the inclusion of non-monetised
impacts in the appraisal process alongside monetised benefits and costs. Thus, the new benefits
framework includes both monetised and non-monetised benefits.
In summary, the new benefits framework:
• is aligned with the enduring MoT TOF
• is used in all stages of benefit management, from business cases to economic evaluation,
and through to post-implementation benefits realisation
• includes three groups of benefits: monetised, quantitatively described and qualitatively
described
• captures the actual benefits to people, society and the environment, rather than functioning as
benefit indicators
• is mode neutral
• is direction neutral, by using the term ‘impacts’ to cover benefits, disbenefits and costs.
This manual only provides guidance on the monetisation of the benefits with provided standard
monetary value. Qualitative and quantitative measures, associated to all of the benefits included in
the benefits framework except wider economic benefits, are provided in the Land Transport
Benefits Framework measures manual.
the benefits provided in this manual are not sufficient or relevant for the improvements under
consideration, then the benefit and cost parameters can be adjusted, subject to agreement from
Waka Kotahi in writing. Table 5 shows the whole framework in relation to the benefits specified in the
now-superseded EEM. The benefits included in this manual are highlighted in turquoise.
The monetary values presented in this manual have been measured and monetised at different
points in time and therefore have varying base dates. To update these values to the current year
Waka Kotahi provides update factors annually. The update factors also account for any escalation
in construction costs. The update factors for benefits and constructions costs are available on the
Waka Kotahi website.

3. Benefits
Table 5: Relationship between benefits included in the Waka Kotahi Land Transport Benefits Framework and EEM’s benefits
Benefit cluster Benefit In EEM indicated by

MoT TOF
and
Monetised Monetised
1. Changes in user safety Impact on social cost of deaths and serious injuries
Crash cost savings (social cost of crash)
Impact on a safe system n/a

Healthy and safe people
2. Changes in perceptions of Impact on perceptions of safety and security -

safety
3. Changes in human health Impact of mode on physical and mental health Walking and cycling health benefits
Impact of air emissions on health Vehicle emission reduction benefits (air

pollutants)
Impact of noise and vibration on health Other external benefits (noise)
Other external benefits (vibration)
4. Changes in impact of Impact of system vulnerabilities and redundancies Risk reduction benefits
and security
Resilience
unplanned disruptive events on (natural/environmental risks, eg water flows)

access to social and economic
opportunities
Risk reduction benefits (human-made risks)
5. Changes in transport costs Impact on system reliability Journey time reliability benefits
Impact on network productivity and utilisation Travel time saving
Vehicle operating cost savings
Parking user cost savings

Economic prosperity
PT users cost savings due to change in the

user charge
Walking and cycling cost savings
6. Wider economic impact Wider economic benefit (productivity) (WEB) Productivity
Wider economic benefit

(WEB) Labour supply
(employment impact)

(WEB) Imperfect competition
(imperfect competition)

-
(regional economic development)
Environmental sustainability
7. Changes in natural Impact on water External benefits (water quality and flows)
environment
Impact on land and biodiversity Other external benefits (ecological impact)
8. Changes in climate Impact on greenhouse gas emissions Vehicle emission reduction benefits
(greenhouse gas emissions)
9. Changes in resource Impact on resource consumption -

consumption
10. Changes in access to social Impact on user experience of the transport system Driver frustration reduction benefits
and economic opportunities
Seal extension benefits
User benefits from new or improved facilities:

public transport and cycling
Impact on mode choice -
Inclusive access
Impact on access to opportunities -
Impact on community cohesion Other external benefits (community

severance)
Other external benefits (isolation)
11. Changes in liveability of Impact on heritage and cultural values

Other external benefits (special area)
places
Impact on landscape Other external benefits (visual impacts)
Impact on townscape Other external benefits (overshadowing)
12. Changes to te ao Māori Impact on te ao Māori* Other external benefits (eg iwi, Māori values)
* It is not part of neither the MoT TOF nor the Government Policy Statement on Land Transport priorities, but a requirement of the Waka Kotahi Māori
strategy, Te ara kotahi.
Back to 1.7 Benefits: Monetisation >>

3. Benefits > 3.1 Impact on social cost of deaths and serious injuries
3.1 Impact on social cost of deaths and serious injuries

This section (and its related Appendix 2: Crash analysis) provides guidance on calculating crash costs for
the do-minimum and option scenarios for a route or site.
For the purposes of this manual, a crash is an event involving one or more transport system users that
results in personal physical injury, including to pedestrians and cyclists, and any damage to property.
Crash rates, crash prediction models and crash reduction factors can be found in the Crash estimation
compendium.
Changes in crash costs are a function of predicted numbers of crashes and unit crash costs. Unit crash
costs vary by crash type and severity, and vehicle speed, while predicted crash numbers need to take
account of the road environment, under-reporting and the exposure to the risk of having a crash.
Deciding whether to conduct a crash analysis

Not all project evaluations require a crash analysis. Several factors affect the decision as to whether or
not to undertake a crash analysis; these include:
• the nature of the site (eg average annual daily traffic (AADT) and length)
• the availability of reliable crash history and crash rates or crash prediction models
• whether crash savings are claimed as part of a project evaluation
• the size and type of a project.
Crash cost analysis for project evaluations can generally be separated into four groups.
Safety improvement projects
These are projects in which most of the benefits are the result of a reduction in crashes. Safety
improvement projects are undertaken where a route or site (eg curve, railway crossing, bridge etc) has a
high occurrence of crashes, or when the risk of crashes is considered high. For guidance on identifying
high-risk sites and routes refer to the High-risk intersections guide and High-risk rural roads guide. Given
that the majority of benefits for such schemes arise from a reduction in crashes, it is important that a
robust assessment is undertaken. Analysts should avoid basing their assessment on a small number of
historical crashes or using unsuitable crash rates, crash prediction models or crash reduction factors. It is
also important to undertake sensitivity testing to understand how sensitive the benefit–cost ratio is to the
crash history, crash prediction and crash reduction factors.
Other roading improvement projects
These are projects where reducing crashes is not the primary outcome sought, but there are safety
impacts that need to be included in the economics (eg installation of a bus transit lane, extension of turn
lanes, traffic capacity focused schemes). For these projects, benefits arise primarily from travel time and
vehicle operating costs savings. Crash benefits can still be a key contributor to the benefit–cost ratio, but
they are unlikely to be the key determinate in whether a project is funded.
While the majority of roading improvement projects will reduce crashes, some may increase them,
especially if safety is not given adequate consideration in their design. Examples of projects that may
result in an increase in crashes include four- and six-lane roads with bus or high-occupancy vehicle lanes,
and seal extensions in curvy and tortuous terrain. The safety disbenefit in these cases can be minimised
or eliminated through high-quality design. For example, on seal extensions through curvy realignment,
providing improved delineation and safety barriers may mitigate the safety disbenefits that would
otherwise arise.
Large projects
Large projects, for example a new motorway link, will have a network-wide impact on crashes. However,
in most cases the crash benefits of such a project are only a small proportion of the overall benefits.
When this is the case it may be appropriate to use a more basic level of analysis than that specified in
Appendix 2: Crash analysis, although some of the crash rates and costs provided can still be used. When
considering projects of an area-wide nature, such as the evaluation of an urban traffic network (eg for
transport planning or a traffic management study), it is insufficient to calculate crash costs from changes
in global totals of vehicles per kilometre of travel. Where a new road link is being added to a network, or a

network change will result in major redistributions of traffic, analysis is required of the incidence of
crashes on the links to which the traffic is being diverted and on the links for which traffic volumes reduce.
For a new link, using crash rates appropriate to its intended design, speed limit and intersections along it
can be applied. In some situations, the use of the site (or route) specific crash rates may be more
appropriate than using the crash rates provided in the Crash estimation compendium.
The full analysis process for calculating crash costs for area-wide changes in traffic networks is not
outlined in this manual, and practitioners should refer to an expert in the field of crash analysis during
this process.
Walking and cycling projects

Some new or improved walking and cycling facilities effectively eliminate hazards along an established
route used by pedestrians or cyclists, such as through the provision of over bridges, underpasses, bridge
widening and intersection improvements. In these cases a more detailed analysis of the changes in crash
types, numbers and costs should be completed using the procedure in Appendix 2: Crash analysis.
Reduction or elimination of hazards on a walking or cycling route is likely to be a factor in attracting new
users or additional use of the facility. The evaluation should quantify, preferably through surveys or
research, the extent to which the existing hazards are an impediment to new users, or additional use, and
provide supporting information on pedestrian or cycle numbers.
There is also evidence that the crash rate per cyclist or per pedestrian reduces significantly as the
number of cyclists or pedestrians increases, and that the overall number of crashes (motor vehicles,
cyclists and pedestrians) does not change substantially when private vehicle trips are diverted to cycling
or walking. This means that, in most cases, there are no significant negative crash costs associated with
diverting private vehicle trips to walking and cycling trips.
Crash analysis steps

The general process for a crash analysis is as follows:
1. Select the appropriate analysis procedure(s) using Appendix 2: Crash analysis and, depending
on the method(s) selected:
i. determine the historic crash performance via analysis of crash records, typically over the
last five years
ii. select crash rates or crash prediction models for the do-minimum and option cases from
the Crash estimation compendium.
2. Apply crash reduction factors or crash modifying factors (CMFs) as required.
3. Calculate the annual crash performance and corresponding crash costs for the do-minimum and
the option scenarios. Adjust for general trends in crash occurrence.
Calculate the annual crash cost savings. These are the future annual crash costs of the do-
minimum less the future annual crash costs of the options.
Crash analysis methods

There are three types of crash analysis methods available:
• method A: crash-by-crash analysis
• method B: crash rate analysis
• method C: weighted crash procedure.
Follow the flowchart steps in Figure 1 to determine the need for a crash analysis and the selection of
crash analysis method(s). Details on each step in the process are discussed Appendix 2: Crash analysis.

Figure 1: Decision process for selecting crash analysis methods
Crash migration
When undertaking a crash analysis, it is important that the potential crash migration is considered. It is
possible that when a site (or route) is upgraded, crashes will be reduced at the site but may migrate to
a different site downstream. In this case the benefits that have occurred at the site will be offset by the
disbenefits that occur downstream. Ideally the potential for crash migration should be identified and
addressed.

3. Benefits > 3.2 Impact of mode on physical and mental health
Crash migration downstream of the treated site is normally not an issue in the urban road environment
(under 70km/h). Crash migration is more prevalent on rural roads and in close proximity to the site being
treated. The migration of crashes from the improved site down to the next curve or substandard road
element (eg a narrow bridge) is more likely than migration to a similar element 20km downstream.
To assess the possibility of crash migration, 1–2 kilometres either side of the study area should be
considered. If road elements, such as low-design speed curves (75km/h or less), narrow bridges and
railway crossings occur within this 1–2 kilometres, the analysis should assess whether an increase in
travel speeds through the project area will increase crashes at the adjoining road elements. If there is an
expected increase in the crash occurrence, then:
• the negative impacts (disbenefits) need to be included in the economic evaluation
• improvements need to be made to downstream road elements to eliminate or reduce the crash
migration, or
• a reduced estimate of crash savings should be used in the analysis.
A similar exercise should be undertaken for a longer length, up to 5 kilometres either side of the study
area, if the speed change from the site improvements is expected to influence speeds and driver
perception over a wider area. This may be the case for major realignments, which significantly impact on
the speed environment.
Valuation of social cost of crashes

Unit values of crash costs are provided in Appendix 2: Crash analysis for each crash type by movement
category, speed limit, severity and vehicle involvement.
The values in the tables in Appendix 2 are based on Denne T (2023). The values per injury (as at July
2021) are:
• $12.5 million per fatality

• $660,100 per serious injury
• $68,000 per minor injury.
A worked example of the crash cost procedure is provided in Appendix 8: Worked examples.
3.2 Impact of mode on physical and mental health

The impact of mode on physical and mental health relates to people who change modes. This could be
people who switch from private vehicle use to walking or cycling, and therefore switch from being inactive
to being active.
Physical health benefits are included in benefit values for assessing pedestrian and cycling facilities. At
this time there is no standard monetised value for mental health impacts.
Walking
The health benefits value for new pedestrian users is shown in Table 6.
Table 6: New pedestrian facility benefits ($/pedestrian km – 2021)
Benefit Health benefits for new user ($/km) Maximum annual benefit per new user
Pedestrian benefit $9.90 $3,100
Where a new facility eliminates or improves a site (eg an underpass) that is an impediment to safe
walking, a health benefit shown in Table 6 may be ascribed to new pedestrians using the facility. The
benefit is irrespective of the length of the improvement. It is calculated using the average pedestrian trip
length of 1 km times the value given above. The benefit is ascribed to users not users’ trips.
The annual benefit calculated for each new individual pedestrian cannot exceed the maximum annual
benefit shown in Table 6, which is the total estimated economic health benefit for converting an inactive
person to an active person. The maximum annual benefit is irrespective of the pedestrian trip length
because of diminishing returns, from a national perspective, an already active person walking additional
kilometres won’t deliver additional health benefit.

3. Benefits > 3.3 Impact of air emissions on health
Cycling
Values for new cyclists using conventional and electric-assisted cycling are shown in Table 7.
Table 7: New cycle facility benefits ($/cyclist km – 2021)

Benefit Health benefits for new user ($/km) Maximum annual benefit per new
user
Conventional cycling benefit $4.90 $6,200
Electric-assisted cycling benefit $2.50 $4,600
Where a new facility eliminates or improves a site that is an impediment to safe cycling, a benefit of
$14.70 may be ascribed to cyclists using the facility. The benefit is irrespective of the length of the
improvement. It uses the average cycle trip length of 3 km times the composite benefit given above. The
benefit is ascribed to users not users’ trips.
The annual benefit calculated for each new individual cyclist cannot exceed the maximum annual benefit
in Table 7, which is the total estimated economic health benefit for converting an inactive person to an
active person either using conventional cycling or electric-assisted cycling. The maximum annual benefit
is irrespective of the cyclist trip length because of diminishing returns, from a national perspective, an
already active person cycling additional kilometres won’t deliver additional health benefit.
A worked example of cycling benefits is provided in Appendix 8: Worked examples.
3.3 Impact of air emissions on health

Vehicle emissions
Vehicle emissions are a complex mixture of gases and particles, with pollutants typically split into:
• harmful air pollutants, which cause adverse health effects and have local impacts, such as
particulate matter (PM10 and PM2.5), nitrogen dioxide (NO2), carbon monoxide (CO), sulphur
dioxide (SO2) and volatile organic compounds, and
• greenhouse gases or climate pollutants, which cause global warming and have global impacts,
for example carbon dioxide (CO2), black carbon (BC) and methane (CH4).
The impact and assessment of air pollution is covered within this section. The impact of transport services
on greenhouse gases is described in more detail in section 3.4.
Impact of air pollution

For transport-related sources, the harmful pollutants of most concern are the following:
• Particulate matter (PM2.5 – matter that is 2.5 microns (μm) or smaller), which predominantly
impacts on respiratory and cardiovascular systems. Effects can range from reduced lung
function, to increased medication use and increased hospital admissions, right through to
reduced life expectancy and death.
• Nitrogen dioxide (NO2), a gas that causes increased susceptibility to infections and asthma. It
reduces lung development in children and has been associated with increasingly more serious
health effects, including reduced life expectancy (Kuschel et al 2022).
• Carbon monoxide (CO), a gas that is readily absorbed from the lungs into the bloodstream. It
attaches more readily to haemoglobin in the blood than oxygen, causing headaches, dizziness
and weakness. It can also aggravate heart conditions.
• Volatile organic compounds, hydrocarbons that include a wide range of chemicals, some of which
are carcinogenic to humans. Volatile organic compounds can also react with NOX (nitrogen oxide
and NO2) in the presence of sunlight to form ozone (O3), which is a lung irritant.
Greenhouse gas emissions from road users contribute to global climate change. Changes in climate may
also impact state highways through sea level rise, heavy rainfall and more frequent flooding.

3. Benefits > 3.3 Impact of air emissions on health
Assessment of air pollution

Waka Kotahi has a comprehensive process for assessing the potential air quality effects of roading
projects. Information on the assessment process, as well as procedures and resources for assessment,
are available on the Waka Kotahi website.
Calculating traffic-related emission loads

This procedure has been developed from the Vehicle Emissions Prediction Model (VEPM). Emission
rates are to be extracted from the latest version of the VEPM, which is freely available. To obtain a
password to access to the spreadsheet, please email [email protected].
Emission rates are available for various speeds, gradients and traffic compositions, or other variables
such as vehicle load.
Table 8: Calculating traffic-related emission loads
Step Action
1 Determine the:
• traffic composition
• time period’s total average travel time per vehicle (min).
2 Convert the traffic composition vehicle classes into light and heavy emission classes according
to % of vehicle kilometres travelled (VKT):
Emission class Vehicle classes (Table A45)
Light Passenger cars
(vehicles less than 3.5 tonnes) Light commercial vehicles
Heavy Medium commercial vehicle (MCV)
(vehicles greater than 3.5 tonnes) Heavy commercial vehicle I (HCVI)
Heavy commercial vehicle II (HCVII)
Buses
3 Calculate average speed on the link road, either using a model, or according to the formula:
Speed (km/h) = 60 × length / TT
where: length = road link length (km)
TT = time period total average travel time per vehicle (min).
4 Calculate the emission rates for light and heavy vehicle types using average speed on the link
road from step 3, and emission factors from the latest VEPM version.
5 Weight the calculated emission rates by vehicle flow composition (g/vkt):
= % light vehicles × light vehicle fleet average emission factor + % heavy vehicles ×
heavy fleet average emission factor
6 Multiply the weighted emission rates by the time period’s total vehicle volume and the road’s
length (ie VKT) to give the emission load (g).
While emission rates are provided for all vehicle classes over the speed and gradient ranges, certain
combinations of vehicle class, speed and gradient do not occur in practice, such as the sustained
operation of laden heavy vehicles at high speed on steep gradients. Emission rates at these extremes are
not available, so the closest available rate is used.
Valuation of emissions
The external impacts of air emissions are costed using the damage cost approach. This assigns a cost to
each tonne of pollutant emitted to reflect the damage done to the surrounding environment, including
people and ecosystems. Emissions are calculated for each assessment scenario and then multiplied by
the costs per tonne so that the likely impacts can be compared. The damage cost approach is simpler
than undertaking exposure modelling, which requires detailed understanding of the sources, receptors,
terrain and meteorology to arrive at predicted concentrations to which exposure response functions are
then applied. However, it utilises factors which apply to the project as a whole, rather than at a local
scale.

3. Benefits > 3.4 Impact on greenhouse gas emissions
To calculate the damage cost of each option, use the total calculated emissions for each option and the
damage cost per tonne from Table 9.
Damage costs in Table 9 have been calculated based on the 2021 value of life years (VoLY) derived from
the 2021 value of statistical life (VoSL) of $12,500,000 (NZ$). For project evaluations in future years,
damage costs for all pollutants should be updated based on the update factor published for injury costs.
Table 9: Emissions damage costs ($/tonne – 2021)

Pollutant Urban costs in Rural costs in National costs in
NZ$/tonne NZ$/tonne NZ$/tonne
PM2.5 $853,824.00 $49,075.00 $530,676.00
NOX $865,797.00 $24,040.00 $325,312.00
CO $4.87 $0.19 $2.99
Volatile organic compounds $1,545.00 $61.00 $949.00
SO2 $39,334.00 $1,546.00 $24,160.00
Note: These damage costs for New Zealand have been based on Kuschel et al (2022).
The predicted value of any change in emissions should be calculated and included in the BCR. Projects
which include both urban and rural areas should model and monetise urban and rural emissions
separately, before summing them together. National damage costs should only be used when it is not
practicable to separately model urban and rural emissions. Emissions should be quantified in tonnes and
reported in the appraisal summary table.
A worked example of the Vehicle emissions procedure is provided in Appendix 8: Worked examples.
3.4 Impact on greenhouse gas emissions

The greenhouse effect is caused by greenhouse gases, particularly CO2 and water vapour, trapping heat
in the lower atmosphere. These gases let energy from the sun travel down to the earth relatively freely,
but then trap some of the heat radiated by the earth.
Greenhouse gas emissions from road users contribute to global climate change. Changes in climate may
also impact transport system through sea level rise, heavy rainfall and more frequent flooding.
Note: Several harmful pollutants (especially black carbon) are direct climate pollutants, in that they have
a direct warming effect on the atmosphere. However, many of the remaining harmful pollutants (eg SO 2
and CO) are indirect climate pollutants. This means they do not warm the atmosphere themselves, but
react with other gases to increase greenhouse gas concentrations. Therefore, initiatives that address
harmful air pollutants typically yield both health and climate change benefits.
Impact of CO2
While CO2 occurs naturally, in the last 200 years the concentration of CO 2 in the earth’s atmosphere has
increased by 25%. As these extra amounts of CO2 are added to the atmosphere they trap more heat,
causing the earth to warm. This extra warming is called the enhanced greenhouse effect and is predicted
to significantly alter the earth’s climate.
CO2 makes up around half of the extra greenhouse gases and a significant proportion of this extra CO 2 is
emitted by motor vehicles.
Assessment of CO2 emissions

The following procedure is to assess changes in the level of greenhouse gas (GHG) emissions as the
result of transport activities.
This procedure has been developed from the Vehicle Emissions Prediction Model (VEPM) to calculate
traffic-related emission loads. GHG emission rates are to be extracted from the latest version of the
VEPM, which is freely available. To obtain a password to access to the spreadsheet, please email
[email protected].

3. Benefits > 3.4 Impact on greenhouse gas emissions
Emission rates are available for various speeds, gradients and traffic compositions, or other variables
such as vehicle load.
Table 10: Calculating traffic-related emission loads
Step Action
1 Determine the:
• traffic composition
• time period’s total average travel time per vehicle (min).
2 Convert the traffic composition vehicle classes into emission classes:
Emission class Vehicle classes (Table A45)
Light Passenger cars
(vehicles less than 3.5 tonnes) Light commercial vehicles
Heavy Medium commercial vehicle (MCV)
(vehicles greater than 3.5 tonnes) Heavy commercial vehicle I (HCVI)
Heavy commercial vehicle II (HCVII)
Buses
3 Calculate average speed on the link road, either using a model, or according to the formula:
Speed (km/h) = 60 × length / TT
where: length = road link length (km)
TT = time period total average travel time per vehicle
4 Calculate the emission rates (g/km) for light and heavy vehicle types using average speed on
the link road from step 3, and emission factors from latest VEPM version.
5 Weight the calculated emission rates by vehicle flow composition (g/vkt):
= % light vehicles × light emission rate + % heavy vehicles × heavy emission rate
6 Multiply the weighted emission rates by the time period’s total vehicle volume and the road’s
length to give the emission load (g).
While emission rates are provided for all vehicle classes over the speed and gradient ranges, certain
combinations of vehicle class, speed and gradient do not occur in practice, such as the sustained
operation of laden heavy vehicles at high speed on steep gradients. Emission rate at these extremes are
not available, so the closest available rate is used.
Valuation of CO2 emissions

The whole-of-government agreed shadow price1 of carbon ($ per tonne of CO2 equivalent) emissions, in
Table 11, is to be used for calculating the economic impact of carbon for transport activities.
This means applying the central price path as the default analysis in the economic evaluation of transport
proposals and accompanying this with sensitivity analysis based on the low and high price paths.
Note that these carbon values will be updated when new information on the value of C02 equivalent
becomes available.
Should your analysis require shadow prices for years prior to 2023, email [email protected].
1For more information see the technical paper Economic valuation of greenhouse gas emissions (Waka Kotahi NZ
Transport Agency 2021) on the Waka Kotahi website.

3. Benefits > 3.5 Impact of noise and vibration on health
Table 11: Shadow price of carbon (NZ$2022 per tonne of CO 2 equivalent)

Year 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034
Low $59 $65 $72 $78 $85 $91 $98 $104 $108 $112 $116 $120
Middle $87 $97 $107 $116 $126 $136 $146 $155 $161 $167 $174 $180
High $171 $182 $193 $203 $214 $219 $224 $230 $235 $241 $247 $253
Year 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046
Low $124 $129 $133 $137 $141 $145 $149 $153 $157 $161 $165 $169
Middle $186 $192 $198 $204 $210 $216 $222 $228 $235 $241 $247 $253
High $259 $265 $271 $278 $284 $291 $298 $305 $313 $320 $328 $336
Year 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058
Low $173 $178 $182 $186 $188 $189 $191 $193 $195 $197 $199 $201
Middle $259 $265 $271 $277 $286 $294 $303 $312 $321 $331 $341 $351
High $344 $352 $361 $369 $387 $407 $427 $448 $471 $494 $519 $545
Year 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070
Low $203 $205 $207 $209 $211 $214 $216 $218 $220 $222 $224 $227
Middle $362 $373 $384 $395 $407 $419 $432 $445 $458 $472 $486 $501
High $572 $601 $631 $662 $695 $730 $767 $805 $845 $887 $932 $978
A shadow price places a value on future greenhouse gas emissions emitted or reduced, usually
concerning international and/or national emissions goals.
Shadow prices are different from market traded prices in the Emissions Trading Scheme (ETS), which
do not currently reflect the full marginal cost of achieving New Zealand’s emission targets. An ETS is
typically only one of the many policies that governments implement to meet their climate targets.
3.5 Impact of noise and vibration on health

Noise is a disturbing or otherwise unwelcome sound, which is transmitted as a longitudinal pressure wave
through the air or other medium as the result of the physical vibration of a source. Noise propagation is
affected by wind and intervening absorbent and reflective surfaces, and is reduced with distance.
Road traffic noise sources include:
• engine and transmission vibration

• exhaust systems
• bodywork and load rattle
• air brake and friction brakes
• tyre/road surface contact
• horns, doors slamming, car audio systems
• aerodynamic noise.
Two types of vibration are evident alongside traffic routes: ground-borne vibrations and low-frequency
sound, which can result in building vibrations.
The primary cause of ground-borne vibrations is the variation in contact forces between vehicle wheels
and the road surface. The interaction between vehicle tyres and road surface irregularity can result in the
release of significant energy. Therefore, roads with surface irregularities generate more vibrations than
new, smooth roads. Once produced, ground conditions markedly affect the way in which ground-borne
pressure waves are spread. Also, distances between the road and dwelling locations will determine how
much vibration energy actually reaches nearby properties.

Air-borne low-frequency sound below 100Hz can also induce building vibration. The primary cause of
these vibrations is low-frequency vehicle-produced sound, which enters the building and can excite the
building structure and/or the contents. This excitation at the natural frequency of the structure being
excited is highly dependent upon the type of building structure and its proximity to the road. In general,
air-borne vibration is taken into account in the assessment of noise effects, ie locations likely to
experience significant air-borne traffic-induced vibrations are likely to have been assessed as high noise
areas and the impact determined.
Traffic-induced vibrations are evident in many parts of New Zealand and variations occur because of sub-
soil geological factors such as high water tables, light volcanic sub-soil, or peaty soils. Generally the
levels of vibration perceived will be a function of vehicle size, speed, proximity to the road, subsoil
geology, building characteristics and sensitivity at the receiver location.
Impact of road traffic noise

Road traffic noise is generally continuous and long-term exposure can have significant adverse effects.
These can be categorised as disruptive impacts, such as sleep disturbance and speech interference, and
psychological impacts such as annoyance reaction and other behavioural impacts. While there is no
evidence of permanent hearing loss from road traffic noise, there is a great deal of evidence to show that
noise can cause adverse health effects in people due mainly to stress-related factors.
While the untrained ear will generally only detect noise level differences of three decibels (dB) or more,
smaller increases will still affect people’s wellbeing. To increase the noise level by 3dB requires traffic
volumes to double.
Design guidelines for road traffic noise

Design guidelines for the management of road traffic noise on state highways are given in the Guide to
assessing road-traffic noise, using NZS 6806 for state highway asset improvement projects. These
guidelines apply to noise-sensitive facilities adjacent to either new state highway alignments or to any
other state highway improvements, which require a new designation.
The assessment point at which the design criteria apply is 1m in front of the most exposed point on the
façades of existing residential buildings or educational facilities. An exception is in the case of noise
buffer strips, where the assessment point is the outer limit of the buffer strip.
The two criteria in the guidelines, both of which apply, are:
1. Average noise design criteria

• The average noise design levels for residential buildings and educational facilities at the
assessment point are set out in Table 12.
• If it is not practicable or cost effective to meet the average design noise criterion at the
assessment point given in Table 12, then the guidelines specify internal noise design criteria.
These criteria apply to all living rooms (including kitchens) and bedrooms in residential
buildings, or teaching areas in educational facilities, with windows closed on the exposed
walls.
• The internal noise level criterion for residential buildings is either the level given in Table 12
minus 20dB(A), or 40dB(A) Leq(24 hour), and for educational facilities the internal noise level
criterion is either the level given in Table 12 minus 20dB(A), or 42dB(A) Leq(24 hour), in each
case whichever is the higher.

Table 12: Average noise design levels (Leq(24 hour))

Noise area Ambient noise level Average noise design level
(dB(A)) (dB(A))
Low Less than 43 55
Areas with ambient noise levels of less than 43–50 Ambient + 12
50dB(A) Leq(24 hour)
Medium 50–59 62
Areas with ambient noise levels of 50 to 59–67 Ambient + 3
High 67–70 70
Areas with ambient noise levels of more than more than 70 Ambient
2. Single noise event design criterion

a. A single noise event is the maximum noise level emitted by a single vehicle passing the
assessment point.
b. Where the assessment point for residential buildings and educational facilities is less than
12m from the nearside edge of the traffic lane, the Guide to assessing road-traffic noise using
NZS 6806 for state highway asset improvement projects require noise reduction measures to
reduce noise by at least 3dB(A). This is designed to provide a level of protection to properties
from the noise effects of single vehicles.
Mitigation of road traffic noise impacts

There are various options for reducing the effects of road traffic noise. These include realignment to
increase the distance between the roadway and the assessment points, noise buffer strips, barriers,
alternative road surfaces and building insulation.
Where project optimisation requires noise mitigation measures, the cost of such measures will be
identified and included in the project cost as discussed in section 1.8.
Measurement and prediction of road traffic noise impacts

Traffic volumes used for noise predictions shall be based on forecasts of traffic flow 10 years after the
completion of the project.
Equipment and methods for the measurement of noise shall comply with NZS 6801: 2008 Acoustics –
measurement of environmental sound. Prediction of road traffic noise shall be carried out using the
United Kingdom Calculation of road traffic noise (UK Department of Transport 1988) method, calibrated to
New Zealand conditions (refer to Traffic noise from uninterrupted traffic flow) and converted to the
appropriate Leq index.
The conversion formulae to calculate Leq values from the L10 values derived from the UK Calculation of
road traffic noise (1988) method are:
Leq(24 hour) = L10 (18 hour) - 3dB(A)
Leq(1 hour) = L10 (1 hour) - 3 dB(A)
Valuation of road traffic noise impacts

There have been no specific studies carried out in New Zealand to determine the cost of road traffic
noise; however, there is evidence to suggest that road traffic noise levels of 53 to 62dBA do encourage
people to move out of an area more quickly (Dravitzki et al 2001).
A British survey (Tinch 1995) of international (predominantly hedonic price) valuations suggests that the
costs of noise are approximately 0.7% of affected property values per dB. A Canadian survey (Bein 1996)
found that hedonic pricing revealed typical costs of 0.6% of affected property prices per dB, and the
OECD recommends noise valuation based on 0.5% per dB. Bein argues that the total costs of noise are
much higher than the change in property values because:

• consumers may not consider the full effects at time of purchase (supported by a German study
which showed increased willingness to pay with increased understanding of noise)
• effects on other travellers and on occupants of commercial or institutional buildings are not
captured
• hedonic studies typically consider values of homes which experience noise above and below
certain levels (a German study shows increasing willingness to pay as base noise rises).
A reasonable figure for New Zealand is suggested as being 1.2% of value of properties affected per dB of
noise increase (0.6% multiplied by a factor of 2 to take into account the factors mentioned by Bein). Using
average values for urban property of $640,000 according to residential property data (REINZ) February
2020, and occupancy of 2.8 persons according to Census 2018 data, this suggests a present value cost
of $7680 per dB per property and $2740 per dB per resident affected ($495 per household or $177 per
person per year). This figure should be applied in all areas, since there is no reason to suppose that noise
is less annoying to those in areas with low house prices. It is arguable as to what range of noise increase
the cost should be applied to, but a conservative approach would be to apply it to any increase above
existing ambient noise. This reflects a belief that most people dislike noise increases, even if the resulting
noise is less than 50dB.
Costs of road noise shall be valued at:
$495(2020) per year × dB change × number of households affected
Where noise affects schools, hospitals, high concentrations of pedestrians and other sensitive situations
an analysis may be required to determine the cost of noise that is site specific. The methodology for
undertaking a valuation of noise at sensitive sites should be appropriate to the site (ie willingness-to-pay
surveys may be appropriate for sites with high concentrations of pedestrians and inappropriate for
hospital sites).
Reporting of road traffic noise impacts

The number of residential dwellings and the educational facilities affected by a change in road traffic
noise exposure shall be reported in terms of:
• the predicted change from the ambient noise level
• the difference between the predicted noise level and average noise design levels given in
Table 12.
Guidance on predicting and managing noise impacts can be found in Waka Kotahi guidance on noise and
vibration.
Where noise is a significant issue, plans shall be prepared distinguishing each type of land use. These
plans shall show:
• contours of noise exposure in the do-minimum and for each project option, and changes in
noise exposure in bands of 3dB(A), ie 0 to 3dB(A), >3 to 6dB(A), >6 to 9dB(A)
• the number of residents in each band
• where the predicted noise level is above the average noise design levels given in Table 12 or
where the single event criterion should apply.
Where projects incorporate measures to mitigate noise, the incremental costs and benefits of these
measures shall be reported. If appropriate, these costs and benefits shall be reported for various levels of
noise mitigation.

3. Benefits > 3.6 Impact on network productivity and utilisation
3.6 Impact on network productivity and utilisation

The impact of transport initiatives on network productivity and utilisation is measured using the following
indicators:
1. changes in travel time (all modes)
2. changes to the financial cost of transport use (including public transport fares, vehicle operating
costs, and cycling costs).
Changes in travel time

Travel time savings are a function of travel times and traffic volumes, and vary by travel purpose and
mode, vehicle occupancy, traffic composition and congestion. The value of travel time savings is based
upon the theory that time spent travelling is an opportunity cost to both individuals and businesses, and
that therefore any reduction in travel time can be represented as a cost saving. The unit value of travel
time savings can be interchangeably referred to as the value of time or VoT.
This section contains values of time for vehicle occupants, public transport users, pedestrians, cyclists,
and freight vehicles. The road user values are used to produce composite travel time values for the
different road categories for uncongested and congested traffic conditions. Values are also provided to
calculate the values for changes in road user journey time reliability. Unit values are provided for
vehicle occupant, vehicle and freight-time costs, along with values for travel in congested conditions.
Unit travel time values are given for standard traffic compositions on urban arterial, urban other, rural
strategic and rural other roads by time period.
New trips generated or induced as a result of travel time savings for existing traffic (see Appendix 1:
Demand estimation methods and guidance) shall be assessed at half the benefits from travel time
saving per vehicle for existing traffic. This assumes that the benefits to new trips will be uniformly
distributed between zero and the maximum following the rule of half.
Where a proposed walking or cycling facility improvement reduces the existing walk or cycle travel
time, for example, by adding a pedestrian or cyclist priority phase at a signalised crossing, there will be
travel time cost savings to existing pedestrians or cyclists other than those covered by the procedures
in section 4.2. The standard values of time given in Table 14 may be used to calculate these benefits.
These benefits may, however, be offset by increased delays to motor vehicles, which should also be
taken into account depending on the road and community context.
Travel time estimation procedures

Travel times shall be estimated from a suitable transport model or according to the procedures in
Appendix 3: Traffic data and travel time estimation. Where a specific procedure is not given, the travel
time shall be determined according to a recognised procedure compatible with the procedures in the
appendix.
Travel time values

This section contains travel time values for all modes. The road user values are used to produce
composite travel time values for the different road categories for uncongested and congested traffic
conditions. Values are also provided to calculate the values for changes in road user journey time
reliability.
The travel time benefits for a project option shall be calculated as the difference between the do-minimum
and option travel time costs as follows:
Total travel time savings = base travel time benefits for improved flow or shorter trips
+ travel time benefits for reduced traffic congestion (if applicable)
+ travel time benefits for improved trip reliability (if applicable).

Travel time values are presented in this section under the following headings:
• Behavioural values of travel time (for demand modelling)
• Equalised values of travel time (for benefit calculations to determine the BCR)
o Work travel time values (for vehicles and for vehicles plus occupants)
o Composite values of travel time and congestion.
Behavioural values of travel time for demand modelling

Table 13 provides behavioural values of travel time that are to be used for transport demand modelling
purposes (refer to Denne et al). The values are for all mode users by trip purpose in $/hour/person.
Table 13: Behavioural values of travel time for vehicle and PT occupants, pedestrians and cyclists
for all road categories and all time periods ($/h/person July 2021)
Vehicle occupants, pedestrians, cyclists Work travel Commuting Other non-work
purpose to/from work travel purposes
Behavioural values of time for uncongested traffic ($/h/person)

Drivers and passengers in cars, LCVs, 37.92 30.90 31.21
MCVs, HCVs, motorcycles
Seated bus and train passengers 37.92 8.16 6.61
Standing bus and train passengers 37.92 11.88 10.33
Pedestrians and cyclists 37.92 11.88 10.33
Maximum increment for congestion ($/h/person)

Drivers and passengers in cars, LCVs, 26.34 26.34 24.47
MCVs, HCVs, motorcycles
Seated bus and train passengers 26.34 6.96 5.18
Standing bus and train passengers 26.34 10.13 8.10
Equalised values of travel time for benefit calculations

Table 14 to Table 16 provide travel time values to be used for the calculation of the economic benefits in
the BCR calculations. These tables contain travel time values that have been equalised across modes.
Table 14 provides equalised values of travel time by trip purpose for all modes combined. These values
are used for calculating travel time benefits in the economic calculations of the BCR. The values are in
$/hour for all trip purposes for uncongested traffic conditions and a maximum increment for congested
conditions. Note that the maximum values apply only when there is extreme congestion. Refer to the
procedures for traffic congestion below
Table 14: Equalised values of travel time for all road categories, all time periods, all users*
($/hr/person July 2021)
Vehicle occupants, pedestrians, cyclists Work travel Commuting Other non-work
purpose to/from work travel purposes
Values of time for uncongested traffic 37.92 19.53 18.91
($/h/person)
Maximum increments for congestion 26.34 16.65 14.83
($/h/person)
* All users means: vehicle drivers and passengers, sitting and standing PT users, pedestrians and cyclists

Work travel time values (for vehicles and for vehicles plus
occupants)
Table 15 provides values of travel time for vehicles and freight where these vehicles are used for work
purposes. The values are for vehicle and freight time, and vehicle and freight time plus occupants in
$/hour. These tables are used for calculating travel times values for non-standard traffic compositions.
For standard traffic compositions use the values in Table 16.
Table 15: Values of travel time for vehicles used for work travel purposes ($/h/vehicle July 2021)
Vehicle type Vehicle and freight Vehicle and freight Maximum increments
time, excluding time, including for congestion for
occupants occupants work travel purposes
($/h/vehicle) ($/h/vehicle) (CRV $/h/vehicle)
Passenger car 2.16 59.04 41.67
Light commercial vehicle 4.61 61.49 44.12
Medium commercial vehicle 11.93 61.22 46.17
Heavy commercial vehicle I 30.54 79.84 64.79
Heavy commercial vehicle II 49.39 98.68 83.63
Bus 30.54 79.84 64.79
Composite values of travel time and congestion

The composite travel time values in Table 16 combine vehicle occupant time, vehicle time and freight
time, based on the vehicle types for standard traffic compositions defined in Table A47. These composite
values are calculated using the default traffic volumes for the four road categories defined in Table A46.
Because the composite values in Table 16 include vehicle occupant time, vehicle time and freight time,
the values from Table 15 (which are provided for non-standard traffic compositions) must not be added
to the travel time values in Table 16.
Table 16 provides the maximum additional values for traffic congestion (CRV), to be applied as
described in this manual.
Table 16: Composite values of travel time, plus maximum increments for congestion, for different
road categories and different time periods ($/h/vehicle July 2021)
Road category and time period Composite value of time Maximum increments for
($/h/vehicle) congestion (CRV $/h/vehicle)
Urban arterial
Morning commuter peak 35.52 24.79
Daytime inter-peak 37.78 25.12
Afternoon commuter peak 35.35 24.43
Evening/night-time 35.62 23.79
Weekday all periods 37.55 25.48
Weekend/holiday 38.32 27.00
All periods 38.24 26.17
Urban other
Weekday 37.79 25.76
Rural strategic
Weekday 50.70 32.33

Rural other
Weekday 49.91 32.24
Traffic congestion values (CRV)

Road users value relief from congested traffic conditions over and above their value of travel time saving.
The maximum increments for congestion values apply to vehicle occupants or road category and time
periods as indicated in Table 13, Table 14, Table 15, and Table 16. The actual additional value for
congestion used in the evaluation is adjusted according to the requirements for each category of delay
set out in below.
Worked examples of selected traffic congestion procedures are provided in Appendix 8: Worked
examples.
Treatment of passing lanes
An exception to the calculation below is made in the case of driver frustration benefits as the results of
passing lane projects evaluated using the procedures in section 3.8. When a separate value for changes
in driver frustration is calculated using the valuation procedure in section 3.8, no additional allowance
shall be made for congestion or improvements in trip reliability. Similarly, if passing lanes are evaluated
using the values for congestion and/or reliability outlined in this section, then no allowance can be
included for driver frustration.
Congested traffic conditions – rural two-lane highways
To allow for congestion, the following addition should be made on sections of rural two-lane highways.
Section lengths for this analysis should normally be greater than 2 kilometres.
Peak traffic intensity and volume to capacity (VC) ratio are first calculated in the normal manner (see
Appendix 3: Traffic data and travel time estimation). Using the VC ratio, terrain type and percentage no-
passing for the road section, the percentage of time delayed (PTD) following slower vehicles is selected
from Figure 2, Figure 3 or Figure 4, or Table 17, Table 18 or Table 19. Alternatively, the formulas shown
in at the bottom of Figure 2, Figure 3 or Figure 4 can be used to calculate PTD, within a limiting range of
PTD greater than or equal to 30%. For lower values of PTD the curves are linear.
Incremental value for congestion = CRV × PTD/90 ($/h)
Percentage of time delayed has a maximum limit of 90%, and so for situations where PTD is ≥90%, the
maximum increment for congestion (CRV) should be added to the base value of travel time.
Congested traffic conditions – urban roads, multi-lane rural highways and motorways
To allow for congestion, the following addition should be made to road section travel time values where
the time period VC ratio exceeds 70%.
Incremental value for = CRV × (road section traffic volume - 70% of road section capacity volume)
congestion ($/h) 30% of road section capacity volume
Bottleneck delay
For all bottleneck delay, the maximum increment for congestion should be added to the base value of
travel time.

Table 17: Volume to capacity (VC) ratios for level terrain, overtaking sight distance and
percentage of time delayed (PTD) following slow vehicles
PTD % Level terrain – percentage of overtaking sight distance

100 90 80 70 60 50 40 30 20 10 0
0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
7.5 0.04 0.03 0.03 0.02 0.02 0.02 0.01 0.01 0.01 0.00 0.00
15.0 0.07 0.07 0.06 0.05 0.04 0.03 0.03 0.02 0.01 0.01 0.00
22.5 0.11 0.10 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.02 0.02
30.0 0.15 0.13 0.11 0.10 0.09 0.07 0.06 0.06 0.05 0.04 0.04
37.5 0.18 0.16 0.15 0.14 0.12 0.11 0.10 0.09 0.09 0.08 0.07
45.0 0.23 0.22 0.20 0.19 0.18 0.17 0.16 0.15 0.14 0.13 0.12
52.5 0.30 0.29 0.28 0.26 0.25 0.24 0.23 0.22 0.21 0.21 0.20
60.0 0.39 0.38 0.37 0.36 0.35 0.34 0.33 0.32 0.31 0.30 0.30
67.5 0.50 0.49 0.49 0.48 0.47 0.46 0.45 0.44 0.44 0.43 0.42
75.0 0.64 0.63 0.63 0.62 0.61 0.61 0.60 0.60 0.59 0.58 0.58
82.5 0.80 0.80 0.80 0.79 0.79 0.79 0.78 0.78 0.78 0.77 0.77
90.0 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
Table 18: Volume to capacity (VC) ratios for rolling terrain, overtaking sight distance and
percentage of time delayed (PTD) following slow vehicles
PTD % Rolling terrain – percentage of overtaking sight distance
100 90 80 70 60 50 40 30 20 10 0
0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
7.5 0.04 0.03 0.02 0.02 0.02 0.01 0.01 0.00 0.00 0.00 0.00
15.0 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.01 0.00 0.00 0.00
22.5 0.11 0.09 0.07 0.06 0.05 0.03 0.02 0.02 0.01 0.01 0.01
30.0 0.15 0.12 0.10 0.08 0.06 0.05 0.04 0.03 0.03 0.02 0.02
37.5 0.18 0.15 0.13 0.11 0.10 0.09 0.07 0.06 0.06 0.05 0.04
45.0 0.23 0.20 0.18 0.16 0.15 0.13 0.12 0.11 0.10 0.09 0.08
52.5 0.30 0.27 0.25 0.23 0.21 0.20 0.18 0.17 0.16 0.15 0.13
60.0 0.38 0.36 0.33 0.32 0.30 0.28 0.27 0.25 0.24 0.23 0.21
67.5 0.49 0.47 0.44 0.42 0.41 0.39 0.38 0.36 0.35 0.34 0.32
75.0 0.62 0.60 0.58 0.56 0.54 0.53 0.52 0.51 0.49 0.48 0.47
82.5 0.78 0.76 0.74 0.73 0.71 0.70 0.69 0.68 0.67 0.67 0.66
90.0 0.97 0.96 0.94 0.93 0.92 0.92 0.91 0.91 0.90 0.90 0.89

Table 19: Volume to capacity (VC) ratios for mountainous terrain, overtaking sight distances and
PTD following slow vehicles
PTD % Mountainous terrain – percentage of overtaking sight distance
100 90 80 70 60 50 40 30 20 10 0
0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
7.5 0.03 0.03 0.02 0.01 0.01 0.01 0.00 0.00 0.00 0.00 0.00
15.0 0.07 0.05 0.04 0.03 0.02 0.01 0.00 0.00 0.00 0.00 0.00
22.5 0.10 0.08 0.06 0.04 0.03 0.02 0.01 0.01 0.00 0.00 0.00
30.0 0.14 0.10 0.07 0.06 0.04 0.03 0.02 0.02 0.01 0.01 0.01
37.5 0.17 0.13 0.11 0.08 0.07 0.05 0.04 0.03 0.03 0.02 0.02
45.0 0.22 0.18 0.15 0.13 0.11 0.09 0.08 0.06 0.05 0.05 0.04
52.5 0.28 0.24 0.21 0.18 0.16 0.14 0.13 0.11 0.10 0.09 0.08
60.0 0.36 0.32 0.29 0.26 0.24 0.22 0.20 0.18 0.16 0.15 0.13
67.5 0.46 0.42 0.39 0.36 0.34 0.31 0.29 0.27 0.26 0.24 0.22
75.0 0.58 0.55 0.52 0.49 0.47 0.45 0.43 0.41 0.39 0.37 0.35
82.5 0.73 0.70 0.68 0.65 0.63 0.62 0.60 0.58 0.57 0.55 0.53
90.0 0.91 0.89 0.87 0.86 0.84 0.83 0.82 0.81 0.80 0.79 0.78
Figure 2: Percentage of time delayed (PTD) two-lane rural roads level terrain

Figure 3: PTD for two lane rural roads, rolling terrain
Figure 4: PTD for two lane rural roads, mountainous terrain

Changes in service frequency for public transport

Changes in service frequency may also be described as changes in waiting time, headway, or queuing
time. The impact of the headway changes depends on the existing frequency of the service.
Calculate the service frequency transport service user benefits using the information in Chapter 2 to give
the projected new patronage level, as follows:
Frequency benefit per transport service user:
FB = WTf × 2 × VOT
Change in net total benefits for existing transport service users:
Bfexisting = FB × Q1
Apply the rule of half for the total benefits for new transport service users:
Bfnew = FB × (Q2− Q1) × ½
Total service frequency benefits:
Bftotal = Bfexisting + Bfnew
where: Q1 is the existing number of passengers (patronage)
Q2 is the projected new total number of passengers
WTf is the wait time benefit (in minutes) from Table 20
VOT is the value of vehicle occupant time ($/minute) for by trip purpose from
Table 14
For improved frequency of services waiting time is valued at two times the value of VOT.
Using the existing headway/service frequency (minutes) and the appropriate trip purpose from Table 20,
identify the changes in wait time in minutes from improving service frequency. If the proposed new
headway/service frequency is significantly less than the existing (ie 20 minutes compared with 40
minutes) an average of the wait time benefit for the two frequencies should be used.
Table 20: Increased service headway

Existing headway Wait time (minutes)
(minutes) Commute Other Combined
5.0 2.4 3.2 2.5
10.0 3.3 4.0 3.3
15.0 4.1 4.8 4.2
20.0 5.0 5.6 5.1
30.0 6.6 7.2 6.8
45.0 9.8 10.6 10.1
60.0 11.7 12.3 11.9
Interchange reduction for public transport

In addition to the wait and/or walk time to transfer time that applies to service frequency benefits, there is
a five-minute IVT ‘interchange penalty’.
Calculate the interchange reduction transport service user benefits for public transport using the
information in Chapter 2 to give the projected new patronage level, as follows:
Interchange reduction benefit per public transport service user:
IB = (WTi × 2 + 5) × VOT
Change in net total benefits for existing public transport service users:

Biexisting = IB × Q1
Apply the rule of half for the total benefits for new public transport service users:
Binew = IB × (Q2 - Q1) × ½
Total interchange reduction benefits:
Bitotal = Biexisting + Binew
where: Q1 is the existing number of passengers.
Q2 is the projected new total number of passengers.
WTi is the existing wait and/or walking time to transfer between public transport
services (minutes).
VOT is the value of vehicle occupant time ($/minute) for by trip purpose from Table 14.
For improved frequency of services waiting time is valued at 2 times the value of VOT.
Changes in financial costs of transport use

This section provides values for cost of all mode use. The costs of transport use in this section are
provided as resource costs, ie exclusive of duties and indirect taxation, such as excise and other taxes
on fuel, import duties, and GST on all cost inputs.
Vehicle operating costs (VOC)

This section provides values for vehicle operating costs (VOC), categorised into running costs, road
surface related costs, speed change cycle costs, congestion costs and costs while at a stop. Values are
provided by vehicle classes and for standard traffic compositions on four different road categories.
The VOC value for each vehicle class is based on the weighted average costs of the vehicles of different
types within each class. The vehicle classes are defined in Table A45 within Appendix 3: Traffic data and
travel time estimation.
VOC values are provided for the standard traffic compositions using four road categories: urban arterial,
urban other, rural strategic and rural other (for more information see their definitions in Table A46 within
Appendix 3: Traffic data and travel time estimation). The road category costs contained in the tables in
this section are for the ‘all time periods’ traffic mix.
The section includes:
• base VOC and VOC by speed and gradient
• VOC changes due to congestion
• VOC changes due to road surface conditions
• VOC changes due to bottleneck delay
• VOC changes due to speed change cycles.
To assist analysts, regression equations are provided which can be used to predict the VOC when using
spreadsheets or other applications.
The regression coefficients vary between vehicle classes and road categories.
The regression equations were used to generate the corresponding VOC tables so the results will be
consistent, irrespective of which approach is used.
Minor differences will arise when calculating road category costs from individual vehicle class costs due
to the regression equations being developed from the road category data. Where high precision is
required, the vehicle class equations should be summed and used in preference to the road category
equations.

The total VOC are calculated by adding the following components:

VOC= base running costs by speed and gradient
+ road roughness costs (if appropriate)
+ road surface texture costs (if appropriate)
+ pavement elastic deflection costs (if appropriate)
+ congestion costs (if appropriate)
+ bottleneck costs (if appropriate)
+ speed change cycle costs (if appropriate).
All components except the base running costs are marginal costs that reflect the additional cost due to
that component.
Base VOC and VOC by speed and gradient
The base VOC value is comprised of costs for fuel, tyres, repairs and maintenance, oil, and the proportion
of depreciation related to vehicle use. Standing charges, ie those incurred irrespective of use, are
excluded from these costs. Such charges are included in the travel time costs for vehicle types (Table 15)
and the composite travel time values (Table 16).
The breakdown of the base VOC by component is given in Table 21.
Table 21: Breakdown of base VOC by component (%)

Vehicle class Percentage of total base VOC by component
Fuel and oil Tyres Maintenance and Depreciation
repairs
PC 35.7 6.2 37.2 21.0
LCV 39.6 7.2 29.4 23.8
MCV 38.6 4.2 44.2 13.0
HCVI 42.0 8.3 42.1 7.6
HCVII 37.3 10.4 43.0 9.3
BUS 46.1 6.0 36.9 11.0
Road category Fuel and oil Tyres Maintenance and Depreciation
repairs
Urban arterial 36.8 7.3 38.1 17.8
Urban other 36.9 6.9 37.9 18.3
Rural strategic 37.5 7.9 39.4 15.2
Rural other 37.4 7.7 39.1 15.8
Information for VOC by speed (between 10km/h and 120km/h) and gradients (between 0% and 12%) is
provided in Table A79 through to Table A88 in Appendix 4: Vehicle operating cost tables. Each table is
accompanied by a graph, and the tables and graphs are generated based on the regression coefficients
and equation in Table 22. The tables give calculated values for each 5km/h and percentage gradient.
The values are the average of the uphill and downhill gradient costs. While VOC values are provided for
all vehicle classes over the speed and gradient ranges, certain combinations of vehicle class, speed and
gradient do not occur in practice, for example sustained operation of laden heavy vehicles at high speed
on steep gradients. VOC estimates at these extremes are less reliable than those in the range of normal
operation.
Intermediate values should be interpolated or predicted using the regression equation. To use the graphs,
the line of average traffic speed on the X-axis shall be read upwards to where it intersects with the
appropriate gradient curve and then the running costs read off the Y-axis.
For all vehicle classes and road categories, the graph curves slope steeply upwards at low speeds. This
is because as vehicle speeds decrease the fuel consumption is governed by the minimum fuel
consumption of the vehicle. As vehicle speeds increase above 60–70km/h the graph curves start to rise
due to the effects of increasing aerodynamic drag.

Table A79 through to Table A84 provides VOC values for individual vehicle classes for use when an
evaluation requires costs for a particular vehicle class or road category, and where the traffic composition
does not fall into one of the four standard road categories. One set of tables is provided for each vehicle
class and these combine the VOCs for both urban and rural road categories.
Where a non-standard traffic composition is considered, the combined VOCs are estimated from the
costs of the individual vehicle classes, and the mean speed of each vehicle class shall be used rather
than the mean speed of the traffic stream as a whole.
Table A85 through to Table A88 provides the VOC values for standard traffic compositions in the four
road categories.
Buses are not included in these standard traffic compositions. If buses form a significant component of
the traffic stream, they shall be included in proportion to their representation.
The regression coefficients for running costs by speed and gradient are provided in Table 22.
Table 22: Running cost by speed and gradient regression coefficients (cents/km – July 2015)
Regression Vehicle class Road category
coefficient PC LCV MCV HCVI HCVII Bus Urban Urban Rural Rural
arterial other strategic other
a 22.63 18.46 28.17 0.878 -171.45 -64.59 18.22 20.08 13.80 15.73
b (× 10-2) -24.11 -68.84 78.13 29.81 2,370.30 229.00 21.90 0.39 73.45 48.93
c 26.89 47.75 63.91 153.83 352.54 167.92 37.50 35.47 48.56 44.72
d (× 10-4) -274.79 -112.69 2,489.40 6,744.40 9,495.20 2,018.00 62.39 59.27 523.50 396.59
e -13.25 -21.98 -28.20 -60.64 -116.34 -62.44 -16.96 -16.38 -20.74 -19.48
f (×10-2) 21.93 42.74 -21.64 -51.71 -1,237.60 -29.16 -2.80 8.23 -31.06 -18.24
g (× 10-4) 12.11 9.64 -175.46 -389.28 -856.38 -231.53 -13.28 -10.43 -46.44 -35.85
h 1.62 2.65 3.46 6.99 12.24 7.21 2.02 1.97 2.43 2.30
i (× 10-3) -34.05 -60.44 -41.35 8.86 1,605.70 -134.22 -3.59 -19.13 30.33 14.28
j (×10-3) 8.16 12.40 61.02 97.13 350.59 121.09 17.38 15.29 27.96 23.96
VOCB = a + b × 10–2 × GR + c × ln(S) + d × 10–4 × GR2 + e × [ln(S)]2 + f × 10–2 × GR × ln(S) + g × 10–4 × GR3 +
h × [ln(S)]3 + i × 10–3 × GR × [ln(S)]2+ j × 10–3 × GR2 × ln(S)
Where: VOCB = base vehicle operating costs in cents/km
GR = absolute value of average gradient (ie >0) over range of 0–12%
S = speed in km/h over range of 10–120km/h
In = natural logarithm
VOC changes due to congestion
The congestion costs are the additional VOCs due to vehicle accelerations and decelerations arising from
traffic congestion. At low volume-to-capacity ratios (VC ratio) there are few accelerations or decelerations,
so the congestion values are relatively low; but they increase with increasing VC ratio, eventually
becoming asymptotic as traffic flows approach capacity (VC ratio = 1.0).
The congestion costs by vehicle class are supplied in Table A92 through to Table A95 for three different
types of operating conditions:
• urban arterial and urban other roads.

• rural strategic and rural other roads.
• motorways.
Motorway costs are based on the rural strategic traffic composition.

Road category costs (all vehicle classes combined) are also provided in Appendix 4: Vehicle operating
cost tables, Table A95, while Table 23 provides regression coefficients for predicting the congestion costs
by vehicle class and Table 24 by road category.
Table 23: Additional VOC due to congestion regression coefficient by vehicle class (cents/km –
July 2015)
Road type Parameter Regression coefficient by vehicle class
PC LCV MCV HCVI HCVII Bus
Urban a 4.41 7.12 9.03 26.03 69.54 16.75
b -3.81 -2.71 -1.50 -0.30 0.86 -1.03
c 5.41 4.85 3.96 3.80 3.65 4.08
Two-lane a 3.86 6.02 7.49 24.22 70.57 13.07
highway b -9.17 -12.14 -0.63 0.57 2.08 -1.25
c 10.48 13.93 2.74 2.66 2.39 3.78
Motorway a 3.30 5.00 7.20 23.10 70.02 12.07
b -23.38 -27.76 -4.98 -3.90 -1.75 -6.64
c 24.51 29.32 7.13 7.13 6.21 9.14
VOCCONG = min {a, exp(b + c x VC) - exp(b)}

Where: VOCCONG = Additional vehicle operating costs due to congestion in cents/km.
VC = volume to capacity ratio.
Table 24: Additional VOC due to congestion regression coefficients by road category (cents/km –
July 2015)
Parameter Urban Rural two-lane Motorway
Strategic Other
a 9.21 7.70 6.98 7.08
b -1.90 -1.24 -1.56 -5.93
c 4.33 3.21 3.41 7.87
When considering congestion costs, the analyst must take into account the amount of time over the year
when traffic is at different levels of congestion (ie different VC ratio). A minimum of five different one-
hourly flow periods should be adopted, reflecting low to high flows, and the number of hours per year the
traffic is at each flow level calculated (summing to 8,760 h/year).

Figure 5: Procedure for estimating changes in VOC due to congestion
• Determine the capacity of the road (see Appendix 3: Determining the capacity of road
Step 1 sections)
• Determine the traffic flow in pcu/hr and the corresponding VC ratio, for each of the
Step 2 hourly flow periods, (see Appendix 3: Calculating the volume to capacity ratio).
• Determine the speed for each of the hourly flow periods (see Appendix 3: Travel times
Step 3 and speed).
•Determine the unadjusted VOC (including roughness, texture and deflection) for each of
Step 4 the hourly flow period speeds, using the VOC tables,.
• Determine the congestion cost corresponding to the VC ratio for each of the hourly flow
Step 5 periods, from Table A84 through to Table A87.
• Determine the total VOC for each flow period as the sum of the unadjusted VOC and
Step 6 the congestion costs.
• Determine the total annual VOC by weighting the costs for each flow period by the
Step 7 percentage of the year that flow is experienced.
VOC changes due to road surface conditions

Road roughness
For some projects road surface roughness is an important contributor to VOC. Projects for which
roughness measurements are necessary include shape correction, seal extension and any other work in
which the riding characteristic of the road surface is changed by the project. The base VOC and VOC by
speed and gradient outlined in the previous part of this section are calculated assuming zero road
roughness (as measured on the International Roughness Index or IRI m/km scale) and shall be
supplemented for the additional costs caused by road roughness when relevant to the project evaluation.
Roughness costs are made up of two components: vehicle costs, and values for vehicle occupants’
willingness to pay to avoid rough road conditions. The willingness-to-pay values reflect the preference of
road users for driving on smooth roads and are based on New Zealand research. The willingness-to-pay
values indicate that road users on rural roads have a higher willingness-to-pay value for a given
roughness than urban users because of their higher average speeds. However, at very high roughness
levels the willingness-to-pay values are the same for both urban and rural road users. These two
components are combined in tables in Appendix 4: Vehicle operating cost tables. Table A89 and Table
A90 provide the additional costs due to road roughness for individual vehicle classes for urban and rural
conditions. Table A91 provides the costs for the standard traffic composition on the four road categories.
To use the VOC tables for road roughness requires the measurement of road roughness. Previously,
NAASRA counts/km was the primary measure, but with the increased use of profilometers the
International Roughness Index (IRI) has been adopted as the primary measure. The NAASRA roughness
can be estimated from the IRI using the conversion 1 NAASRA counts/km = 26.49 × IRI m/km - 1.27.
If the current average roughness is less than 100 NAASRA then there is no actual benefit. Benefits
calculated for pavements with initial roughness less than 100 NAASRA (3.8 IMI) must not be used in
any BCR calculation.

Table 25 provides the regression coefficients for predicting the roughness costs.
Table 25: Additional VOC due to roughness – regression coefficients (cents/km – July 2015)
Road category Vehicle Regression coefficients – July 2015
class a b c d e f g h
Urban PC -18.73 60.26 -70.37 33.96 -5.01 0.00 1.56 6.17
LCV -40.74 125.22 -138.55 63.63 -9.39 0.00 1.64 11.00
MCV -5.38 32.10 -56.08 34.69 -5.17 0.00 4.04 10.60
HCVI -11.80 55.79 -87.72 51.60 -7.63 0.00 5.23 17.28
HCVII -11.85 57.47 -93.84 56.90 -8.27 0.00 7.67 11.04
Bus 8.22 -5.31 -20.76 21.20 -3.18 0.00 4.81 8.80
Rural PC -218.08 820.79 -1,196.82 841.35 -284.85 37.74 1.59 5.74
LCV -354.11 1,315.51 -1,894.73 1,318.30 -443.12 58.36 1.70 10.22
MCV -385.18 1,489.60 -2,226.16 1,597.09 -548.76 73.55 4.07 10.18
HCVI -615.52 2,362.79 -3,510.17 2,508.61 -860.67 115.23 5.23 17.39
HCVII -548.46 2,126.70 -3,187.29 2,291.28 -787.49 105.62 7.76 9.91
Bus -354.29 1,392.35 -2,110.17 1,530.53 -529.69 71.39 4.84 8.44
Urban All -20.16 65.27 -76.73 37.25 -5.49 0.00 1.77 6.96
Rural strategic All -267.66 1,011.64 -1,481.01 1,044.76 -354.66 47.09 2.09 7.00
Rural other All -261.54 987.73 -1,444.98 1,018.75 -345.70 45.89 2.00 6.87
2 3 4 5
VOCRI = min ( {a + b x ln(RI) + c x [ln(RI)] + d x [ln(RI)] + e x [ln(RI)] + f x [ln(RI)] }, {g x RI + h} )
Where: VOCRI = additional vehicle operating costs due to roughness in cents/km
RI = max (2.5, roughness in IRI m/km)
In = natural logarithm.
Road surface texture
A vehicle’s rolling resistance is influenced by the macrotexture of the road surface and impacts on fuel
and tyre consumption. The base VOC and VOC by speed and gradient provided in Appendix 4: Vehicle
operating cost tables are calculated on the basis of 0 texture.
The effect of surface texture on VOC is as follows:
1mm increase in surface macro texture = 0.20 cents/km/vehicle (all vehicle classes combined)
Macrotexture is expressed in millimetres either as a mean profile depth (MPD) or a sand circle (SS). The
conversion between the two measures are:
SS = 0.2 + 0.8 MPD
The additional VOC due to road surface texture is added to the VOC in Table A80 through to Table A88
and is applied to the total traffic volume using the road.
Pavement elastic deflection
Most road pavements in New Zealand are of a bituminous flexible construction. Pavement elastic
deformation under heavy wheel loads depends on the type and strength of the pavement layers and sub-
grade. It influences rolling resistance and therefore fuel and tyre consumption.
The pavement elastic deformation costs from Table 26 are added to the VOC in Table A81 through to
Table A88 for MCV, HCVI, HCVII and buses and the four road categories.
Use of these costs should be accompanied by an adequate statistical sample of Benkelman beam test
results for existing pavements, or Benkelman beam equivalent values from another recognised non-
destructive test method.

Table 26: Increase in VOC per vehicle – kilometre per 1mm increase in Benkelman beam
deflection (July 2015)
Vehicle class Cents/veh/km
MCV 2.50
HCVI 3.90
HCVII 5.30
Bus 3.90
Road category
Urban arterial 0.20
Urban other 0.21
Rural strategic 0.46
Rural other 0.39
VOC changes due to bottleneck delay

Table 27 and Table 28 show the additional VOC by vehicle class and road category for a vehicle while
experiencing bottleneck delay (ie VC ratio ≥1.0). They are calculated from the fuel consumption while
idling and are in cents/minute.
Table 27: Additional VOC due to bottleneck delay by vehicle class (cents/minute – July 2015)
PC LCV MCV HCVI HCVII Bus
1.82 2.56 3.06 4.41 4.41 3.25
Table 28: Additional VOC due to bottleneck delay by road category (cents/minute – July 2015)
Rural other Rural strategic Urban arterial Urban other
2.10 2.15 2.00 1.99
VOC changes due to speed change cycles

When a vehicle travelling at its cruise speed has this speed interrupted due to road geometry or other
road features (eg one-lane bridges or intersections), it decelerates to a minimum speed (which may be a
complete stop) before accelerating back to its original cruise speed. The speed change cycle values are
the difference in travel time and VOC for travelling the distance of the speed cycle at the original cruise
speed versus through the speed cycle.
Additional VOC due to speed change cycles are only to be used for specific situations where traffic
follows a speed cycle comprised of a single deceleration from an initial cruise speed to a minimum speed
before returning to the original cruise speed. These situations typically consist of:
• curves
• traffic signals
• one-lane bridges
• intersections
• work zones.
Table A96 through to Table A115 in Appendix 4: Vehicle operating cost tables provide additional travel
time (in seconds per speed cycle) and additional VOC (in cents per speed cycle) due to a speed change
cycle for:
1. the individual vehicle classes, and
2. the standard traffic compositions in the four road categories.
Since the speed change cycle costs are additional VOC, care must be taken to ensure that there is no
double counting of travel time benefits. For example, when considering traffic signals, the average speed
excluding delays at traffic signals would be used to calculate the travel time and VOC. For those vehicles
delayed by traffic signals, the additional time and additional VOC associated with the speed change
would then be added. In the case of one-lane bridges, the average speed excluding the delay at the
bridge would be used to calculate the travel time and VOC. The additional time and additional VOC due
to the bridge would then be added.

Changes in public transport fares

Transport service activities will provide benefits to new and existing transport service users. These may be affected by user charge levels, travel time, quality
of service, and additional user costs, and can be positive or negative. The purpose of this section is to monetise the net public transport (PT) user benefits and
disbenefits of a PT proposal where there is a change in the user charge. For the purpose of this analysis, PT users are people being moved.
Table 29: Valuation of public transport (PT) user benefits/disbenefits due to a price change
Proposal Users Net benefits calculations description Net benefit calculation equation and data sources
For new PT Those who have It is based on the difference between the proposed Net user benefits = (Pmax - Pnew) × Qnew × ½
services transferred from other and the maximum user charge (at which no one Where: Pnew is the proposed user charge
modes would use the service). The result is then divided in Pmax is the maximum user charge
Those who are half, based on the rule of half.
Qnew is the projected number of
completely new users new service users (see Chapter 2)
(generated trips)
For change to The calculation of PT user benefits for a price Bptotal = Bpexisting + Bpnew
existing PT change on an existing service is based on the
services difference between the existing average user charge
and the proposed average user charge.
Existing users Existing users receive the full benefit of the Bpexisting = (P1 – Pnew) × Q1
improvement.
New users (including both New users are considered to receive on average Bpnew = (P1 - Pnew) × (Q2 - Q1) × ½ (rule of half)
transferred users and one half of the existing user’s benefit based on the
generated trips) rule of half. Where: Pnew is the proposed user charge
P1 is the existing average user
charge
Q1 is the existing number of
passengers (patronage)
Q2 is the projected total number of
service users (see Chapter 2)

3. Benefits > 3.7 Impact on system reliability
Changes in walking and cycling user costs

Cycle operating costs and walking costs are assumed to be included in the perceived costs of changing
to and using these modes.
The term perceived and behavioural costs are referring to the costs that road users perceive and
therefore base their mode choice decision on (Australian Transport Council 2006). ‘Transport users may
ignore some costs when making decisions. Car drivers may see fuel and other vehicle operating costs as
a fixed cost they pay periodically, rather than a variable cost that changes with distance and speed.
“Perceived cost” is derived by deducting from generalised cost the costs that users are assumed not to
perceive.’ (Australian Transport Assessment and Planning 2018b)
The difference between perceived and resource costs is most readily apparent for the value of travel time
savings as presented in Table 13 and Table 14. For the purposes of calculating a BCR, resource costs
are represented by an equity value of time, differentiated only by trip purpose, which applies to all modes
(Table 14). However, the perceived value of time used for modelling differs by the type of transport user
as well as trip purpose (Table 13), which allows for variation in mode choice as generalised costs change.
3.7 Impact on system reliability

Reliability relates to the uncertainty in the time taken to travel from the start to the end of a person’s
journey.
This section outlines how likely variations in journey time can be assessed and the benefits from
improvements to trip time reliability incorporated into project evaluation. Trip time reliability is measured
by the unpredictable variations in journey times, which are experienced for a journey undertaken at
broadly the same time every day. The impact is related to the day-to-day variations in traffic congestion,
typically as a result of day-to-day variations in flow. This is distinct from the variations in individual journey
times, which occur within a particular period. It contains methods for valuing the impact of public transport
and road projects on journey time reliability for users.
Public transport’s journey reliability

For a public transport journey, reliability can affect users in two ways:
• as a delay when picking up the passenger, and
• as a delay when the passenger is on the service.
Unreliable services cause adjustments in an individual’s desired trip-making behaviour, for example by
catching earlier services to get to their destination on time. Therefore, an improvement in reliability
generates a benefit to users in time savings and may also create demand for the service.
The number of passengers affected for the calculation of departure impacts is the number of passengers
boarding, and the number of passengers affected when calculating in-vehicle travel impacts is the
number of passengers already on the service. Generally, just the number of passengers boarding can be
used.
Table 30 below contains the minute-late ratios for each minute the service is late.
Table 30: Equivalent time to a minute-late ratios

Segment Departure In-vehicle travel Combined
All 5.0 2.8 3.9
Train 3.9 2.4 3.1
Bus 6.4 3.2 4.8
Work 5.5 2.8 4.1
Education 3.0 3.8 3.4
Other 5.4 2.0 3.7
The total reliability benefit cannot exceed any travel time saving.
Services running more than 10 minutes late should be treated as 10 minutes late.

3. Benefits > 3.7 Impact on system reliability
The combined value assumes a 50:50 split between departure and in-vehicle time delay en route.
Calculate the user reliability benefits using the formula below:
Reliability benefit = EL × (VTT($/h)/60) × AML × NPT
where: EL is the equivalent time to a minute late ratio from Table 30
VTT is the vehicle travel time ($/h) from Table 15
AML is the reduction in average minutes late (minutes)
NPT is the number of passengers affected.
Road journey reliability

Journey times tend to vary throughout the day, particularly between peak and off-peak periods, and
between weekdays and weekends. This type of variation is well-known to regular drivers and is taken into
account in calculating the travel time values (including congestion values).
Trip reliability is a different type of variability, which is much less predictable to the driver. (For example,
drivers who make a particular journey at the same time every day find some days it takes as little as 20
minutes, and on other days as much as 40 minutes.) Hence, when drivers plan their trips, they have to
consider not just the expected travel time but also its variability. Where an activity improves trip reliability,
the benefits apply to both work and non-work trips.
The following steps in the process for evaluating reliability benefits are discussed in detail in section 4.3
Evaluation of road renewal and improvement activities:
1. Calculate the standard deviation of travel time on each link between intersections and for each
intersection movement or approach.
2. Square the standard deviations to produce variances.
3. Sum variances along each origin-destination path to obtain the total variance for journeys
between each origin and destination.
4. Take the square root of the aggregated variance for a journey to give the standard deviation
of the journey time.
5. Multiply the total trips for each origin–destination pair by the standard deviation of travel
time and sum over the matrix to give the network-wide estimate of the variability.
6. Calculate the difference in variability between the project and do-minimum networks.
7. Assess the percentage of variance occurring outside of the selected study area and select the
adjustment factor.
8. Calculate the impact of changes in trip reliability using following formula:
0.9 × travel time value ($/h) (Table 14, Table 15 or Table 16)
× (reduction in the network variability (in min)/60)
× traffic volume for time period (veh/h)
× correction factor (Table 73)
Where: The reduction in network variability is the difference between the sums of the variability
for all journeys in the modelled area for the do-minimum and project option. The 0.9
factor is the value of reliability based on a typical urban traffic mix.
For projects with a significantly different vehicle mix, evaluators should use 0.8 for cars
and 1.2 for commercial vehicles.

3. Benefits > 3.8 Impact on user experience of the transport system
3.8 Impact on user experience of the transport system

This benefit refers to the changes in user experience as the result of new or improved transport
infrastructure and/or services. This benefit is additional to ‘Impact on network productivity and utilisation’
(ie changes to travel time and cost).
Table 31 summarises benefit indicators and their value proxies for impact of changes in user experience
related to each mode.
Table 31: Summary of user experience benefits

User type Intervention Indicator Value proxy
2
Road users Passing lane Driver frustration Driver frustration benefits are derived
from the ‘time spent following’
information with a willingness-to-pay
value for the provision of passing lanes of
3.5 cents per vehicle per kilometre of
constructed passing lane
Sealing an unsealed Comfort and A value of 10 cents per vehicle per
road productivity gain kilometre can be used for road user
comfort, which takes account of the other
benefits associated with avoiding
unsealed roads.
Public Infrastructure and In-vehicle time in In-vehicle time value based on stated
transport vehicle features relation to preference study
users infrastructure and bus
and train attributes
Demand change Probability of being Vehicle occupant time value
left and proportion of
standing passengers
Pedestrians Cycle facilities quality Four types of cycling Relative value for different type of
and cyclists improvement facility’s quality cycling facilities
improvements
Footpath and walking Improvement of Additional time someone would be
environment quality different aspects of the willing to spend walking to obtain the
improvement (interim pedestrian realm improvement (min)
methodology)
Driver frustration related to passing lanes

Driver frustration benefits are derived from the ‘time spent following’ information (given in the TRARR
OUT file). Research by Koorey et al (1999) established a willingness-to-pay value for the provision of
passing lanes of 3.5 cents per vehicle per kilometre of constructed passing lane (this is in addition to
other benefits such as travel time savings). This benefit is applied to all vehicles that are freed from a
platoon (see definition in Appendix 5: Passing lanes) at the passing lane over the length they remain free
from a platoon. The value of 3.5 cents/veh/km shall only apply to vehicles travelling in the direction of the
passing site. The vehicle-km that should be applied to the willingness-to-pay factor should be determined
by multiplying the traffic volume by the road length and the change in time spent following.
The following travel time and driver frustration benefits are generated when passing lanes reduce the
amount of time drivers spend travelling in platoons. The demand for passing and consequently the
benefits, are a function of a number of parameters including:
2 The procedures in Appendix 5: Passing lanes include a separate value for the reduction in driver frustration and the
effect of reducing travel time variability. When evaluating passing lanes using the procedures in Appendix 5: Passing
lanes, no additional allowance shall be made for congestion or improvements in trip reliability. Similarly, if passing
lanes are evaluated using the values for congestion and/or reliability outlined in this appendix, then no allowance can
be included for driver frustration.

Traffic variables Road variables
• traffic volume • terrain/alignment

• percentage of HCVs • grades
• initial platooning • available passing lanes (sight distance)
• directional split of traffic • passing lane lengths and frequency
• vehicle speed distributions
Calculate the driver frustration savings, using graphs in Figure 6. If necessary, multiply by the traffic
growth correction factor in Table A128 and the driver frustration update factor from the most recent
update factors, available on the Waka Kotahi website.

Figure 6: Graphs of driver frustration benefits for all terrain

Road users comfort due to sealing unsealed roads

Road user comfort benefits and productivity gains from sealing an unsealed road should also be taken
into account. Simplified procedure SP4 – seal extensions provides information on productivity gains. A
value of 10 cents per vehicle per kilometre can be used for road user comfort, which takes account of the
other benefits associated with avoiding unsealed roads.
Public transport user experience

The value of public transport service user benefits (other than fare change benefits, increased service
frequency benefits and interchange reduction benefits), eg improved comfort, is usually based on a
willingness-to-pay value derived from a stated preference (SP) survey or on values derived for similar
service improvements in other areas.
Public transport infrastructure and vehicle features
Public transport users value infrastructure and vehicle features. Typical user valuations expressed in
terms of in-vehicle time (IVT) are provided in Table 32, Table 33 and Table 34. These may be converted
to generalised costs by multiplying by the value of time given in Table 14. All values represent the
difference between the do-minimum and an improvement. These values have been drawn from stated
preference surveys and are the perceived benefits of an individual feature.
Table 32: Vehicle feature values for public transport services – rail
Attribute Sub-attribute Valuation Comment
(IVT minutes)
Driver/staff Train attendant 1.6
Ride 1.2 Quiet and smooth
Facilities CCTV 2.0
On-board toilets 0.6
Information Interior 1.1 Frequent and audible train
announcements
Seating Comfortable 1.5
Layout 0.7 Facing travel direction
Maintained 1.5 Clean and well maintained
Comfort Ventilation 1.5 Air conditioning

Table 33: Vehicle feature values for public transport services – bus
(IVT
minutes)
Boarding No steps 0.1 Difference between two steps up and no steps
No show pass 0.1 Two stream boarding, no show pass relative to single file
past driver
Driver Attitude 0.4 Very polite, helpful, cheerful, well presented compared
with business like and not very helpful
Ride 0.6 Very smooth ride (no jerkiness) compared with jerky ride
causing anxiety and irritation
Cleanliness Litter 0.4 No litter compared with lots of litter
Windows 0.3 Clean windows with no etchings compared with dirty
windows and etchings
Graffiti 0.2 No graffiti compared with lots of graffiti
Exterior 0.3 Very clean everywhere compared with some very dirty
areas
Interior 0.3 Very clean everywhere compared with some very dirty
areas
Facilities Clock 0.1 Clearly visible digital clock showing correct time
compared with no clock.
CCTV 0.7 CCTV, recorded, visible to driver, and driver panic alarm
compared with no CCTV
Information External 0.2 Large route number and destination front/side/rear, plus
line diagram on side relative to small route number on
front/side/rear
Interior 0.2 Easy to read route number and diagram display compared
with no information inside bus
Info of next stop 0.2 Electronic sign and announcements of next stop and
interchange compared with no information next stop
Seating Type/layout 0.1 Individual-shaped seats with headrests, all seats facing
forward compared with basic, double-bench seats with
some facing backwards
Tip-up 0.1 Tip-up seats in standing/wheelchair area compared with all
standing area in central aisle
Comfort Legroom 0.2 Space for small luggage compared with restricted legroom
and no space for small luggage
Ventilation 0.1 Push-opening windows giving more ventilation compared
with slide opening windows giving less ventilation
1.0 Air conditioning

Table 34: Infrastructure features value for public transport – bus

(IVT
minutes)
Stop/ Condition 0.1 Excellent condition, looks like new compared with basic working
shelter order but parts worn and tatty
Size 0.1 Double-sized shelters compared with single-sized
Seating 0.1 Seats plus shelter versus no shelter and seats
Cleanliness 0.1 Spotlessly clean compared with some dirty patches
Litter 0.2 No litter compared with lots of litter
Graffiti 0.1 No graffiti compared with lots of/offensive graffiti
Type 0.2 Glass cubicle giving good all-round protection compared with no
shelter
Ticketing Roadside 0.1 Pay by cash (change given), credit/debit card compared with pay
machines by coins (no change given)
Availability of 0.2 At busiest stops compared with none
machines
Sale of one-day 0.1 Sale on bus, same price as elsewhere compared with no sale of
pass one-day pass
Cash fares 0.3 Cash fares on the bus, driver giving change compared with no
cash fares on bus
Two ticket 2 × 1 ticket –
transfer transfer
Security Security point 0.3 Two-way communication with staff compared with no security
point
CCTV 0.3 Recorded and monitored by staff if alarm raised compared with
no CCTV
Lighting 0.1 Very brightly lit compared with reasonably well lit
Information Terminals 0.1 Screen with real-time information for all buses from that stop
compared with current timetable and map for route
Maps 0.2 Small map showing local streets and key locations versus no
small map
Countdown 0.8 Up to the minute arrival times/disruptions, plus audio compared
signs/real-time with no countdown sign
information
Clock 0.1 Digital clock telling correct time compared with no clock
Contact number 0.1 Free-phone number shown at stop compared with no number
Location of 0.1 One payphone attached to shelter compared with no payphone
payphones
Simple timetable 0.4 Simpler more user-friendly
Stations Up to 3.0 Includes bright lighting, CCTV, cleaned frequently, customer
service staff walking around at info desk, central electronic sign
giving departure times, snack bar, cash-point, newsagent,
landscaping, block paving and photo-booths
Experience from other SP surveys indicates that the perceived benefits of multiple features are
less than the sum of individual components. When multiple features are combined, the values should
be divided by two to adjust for any overestimation.

Demand change impact on existing public transport users

If there is a significant detrimental effect of the new level of demand on existing transport service users
then the disbenefits to existing users should be subtracted from the total user benefits.
Possible negative effects of demand change on existing transport service users include:
• the proportion of standing passengers is increased

• the probability of being left behind is increased.
Service demand disbenefit = (increased waiting time × VOT) × probability of being left behind × total
number of existing users.
Where: VOT is the monetary value of vehicle occupant time from Table 14.
Different types of pedestrian and cycling facilities

Where a quality improvement (amenity, comfort or security) is proposed to existing walking and cycling
facilities, or where new walking and cycling facilities are proposed, the value of different levels of quality
must be assessed. The valuation should be based on a stated preference (SP) survey or information from
similar improvements to facilities in other areas.
The Pedestrian planning and design guide describes a SP methodology and study to identify preferences
for different types of cycling facilities. The study determined the additional time that cyclists would spend
travelling on each type of facility (the incremental attractiveness of that type of facility) compared with a
base case of 20 minutes of travel in traffic with road-side parking. The study gave the values in Table 35.
The relative benefit values should be used in an incremental analysis.
Table 35: Relative benefit for different types of cycle facilities

Type of cycle facility Relative benefit
On-street with parking (no marked cycle lane) 1.0
On-street with parking (marked cycle lane) 1.8
On-street without parking (marked cycle lane) 1.9
Off-street cycle path 2.0
Waka Kotahi NZ Transport Agency (2020b) provides interim methodology and parameter values for
pedestrian facilities improvements. The interim methodology may be used but is not incorporated until
New Zealand specific values are estimated.
Walking distances
Activities that involve mode change need to be careful not to claim unrealistic walking distances. Statistics
on walking used in this manual are based on the New Zealand Household Travel Survey. The average
pedestrian trip length is estimated at 1km.
Cycling distances
Statistics on cycling provided are based on the New Zealand Household Travel Survey. The current
average cycle trip length is estimated at 3km. This applies equally to new and existing users.

3. Benefits > 3.9 Dynamic WEBs and land use benefits and costs
Wider economic benefits

Wider economic benefits (WEBs) are impacts that can result from transport investment. These have
been used internationally to improve transport cost–benefit analysis and can be thought of as impacts
that are additional to the conventional benefits to transport users. Great care is required to ensure that
the estimates for wider economic benefits are truly additional to conventional benefits to avoid double
counting.
Additionally, only the most significant infrastructure improvements are likely to generate WEBs.
Generally, these would need to change the distribution or density of households and firms within a
major metro area, or deliver significant improvements in accessibility between regions, in order for
wider effects to arise.
WEBs can be both static and dynamic.
Static WEBs are more commonly assessed. They are influenced by changes in factors such as travel
times and travel costs (due to the transport investment option) based on static (unchanged) land use.
The MBCM includes guidelines for assessing these effects in sections 3.10, 3.11, 3.12 and 3.13.
Dynamic WEBs estimate additional productivity benefits from a change in location or level of
jobs/workers as a result of changing land use (dynamic). Given the reliance on understanding land use
changes caused by a project, they are less commonly estimated. Since static WEBs leave productivity
changes associated with land use change unaccounted for, capturing dynamic WEBs in addition to the
traditional static WEBs can enable a more complete estimate of wider economic benefits. For transport
projects with sizeable land use change assumptions or objectives, this may be material and impact
upon the viability of the project.
3.9 Dynamic WEBs and land use benefits and costs

Where transport policies affect land use, this will lead to changes in people’s choice of destinations,
modes and routes. Allowing for dynamic land use change provides more accurate estimates of
conventional transport user benefits, which comprise the bulk of benefits for most transport policies.
Changes in conventional transport user outcomes are also the main channel through which wider
economic benefits arise.
Part of the challenge in quantifying the benefits of land use changes is disentangling the degree to which
private investment is truly additional. It may be that private investment in developing land would have
occurred anyway, though the timing may be accelerated, or the location of private investment may have
changed in response to the transport investment. While the magnitude of these impacts will vary, they are
nonetheless likely to result in additional benefits, specifically:
• The traditional benefits of increased network capacity are larger than they otherwise would have
been once the project is delivered.
• Land use and transport changes (eg higher-use redevelopment, alleviated congestion) may be
realised earlier in the project lifecycle, increasing the present value of benefits.
• Agglomeration economies may form around new transport nodes, leading businesses and
residents to cluster in those areas. Where people have moved from areas of lower effective
density or less favourable industrial mix, the net agglomeration productivity impact arising from
this dynamic ‘sorting’ will be positive.
• Land use outcomes may be better and more coordinated, driving higher productivity and more
efficient land use.
In general, relaxing the fixed land use assumption and explicitly modelling land use change presents an
opportunity to gain a more complete picture of the effects of a transport project.

3. Benefits > 3.9 Dynamic WEBs and land use benefits and costs
Productivity gains from land use

The productivity gains from transport investments arise through several mechanisms that affect urban
processes, with land use changes affecting scale, density and sector mix for business activity, as well as
residential patterns. Effects include, but are not limited to, agglomeration benefits (such as industry
specialisation effects). Broadly, transport investment can impact through two types of clustering:
• Static clustering: Transport investment alters the effective density by allowing individuals and
firms to move around the location more easily, lowering transaction costs, thereby facilitating
interactions. This does not change land-use patterns.
• Dynamic clustering: Transport investment can lead to land use change, affecting the scale,
sector mix and the density of a location by inducing a change in the level and/or location of
economic activity (land use change), thereby facilitating new/different interactions. Changes in
the level and location of economic activity are related to labour supply and demand interactions.
At the higher level, land use outcomes affect urban form and therefore affect efficiencies across
urban spatial economies. This can impact effective density because interaction with other
locations is improved, so that efficiency is improved and/or transaction costs lowered.
Land use change estimation

Different models and approaches have varying strengths and limitations, and there is no single best
approach or model for measuring land use impacts. In practice, the adequate selection of the approach to
measuring and quantifying land use impacts will depend on a number of factors, including the type of
infrastructure project, the purpose behind the project, policy scenarios that need to be tested, data
requirements and availability, and modelling efforts. A greater level of sophistication is to be expected in
larger projects and/or those with economic appraisals that rely heavily on land use impacts. It is also
dependent on where the project is in the business case lifecycle, ie high-level estimations may be
appropriate for a strategic case or programme business case, while more detailed analysis would be
expected in a detailed or single-stage business case. It is not expected that all projects will generate land
use changes, and even projects that do may not be expected to deliver significant dynamic benefits/costs.
Therefore, all transport projects should undertake an initial qualitative analysis to estimate the scale of the
land use change (and therefore the benefits associated with it) at an early stage of a project to
understand whether it warrants a further investigation. If the project warrants further investigation on its
potential for inducing land use change, the following quantitative analysis approaches are recommended:
• Land capacity analysis: This supply-side approach estimates land use changes by focusing on
the ability of land in a particular location to support densification – taking into account local and
district level opportunities and constraints. Geographic information system (GIS) tools are used to
support this analysis.
• Land use and transport interaction model (LUTI) or land use attractiveness model: It is
broadly accepted that these methods are able to establish the most confidence as they are
typically based on historical land use relationships. The advantage of these types of statistical
models is that challenges around reverse causality may be addressed.
• Detailed corridor capacity analysis: This approach combines supply-side and demand-side
analysis and uses real-time measures to inform congestion and capacity as well as land use
constraints and potential in a corridor, from a bottom-up approach. This approach provides quality
results for a smaller area, ie corridor for a project, compared to the modelling approaches.
Estimating land use benefits

There are a number of qualitative and quantitative approaches that can be undertaken for estimating land
use benefits and its various categories, including land value changes, public infrastructure cost changes,
second round transport externalities, second round user benefits and costs, and public health cost
changes. Generally, it is recommended these benefits are quantified where there is sufficient information
and parameters. Tailored parameters (including elasticities) for New Zealand will be used, where

3. Benefits > 3.10 Wider economic benefit (productivity)
available. A qualitative approach may be undertaken if the project faces time or cost constraints or the
scale of the project dictates.
Types of dynamic WEBs

The primary dynamic WEBs that Waka Kotahi will consider for inclusion in an evaluation are:
• Dynamic agglomeration: if the relocation of workers or firms results in an increase in net

density, existing firms and workers will become more productive. These productivity gains
(agglomeration benefits) are net additional to the cost–benefit appraisal. It is recommended that a
full calculation of dynamic agglomeration is performed as this will capture both static and dynamic
effects. Parameters for this benefit calculation already exist in the MBCM and should be utilised.
• Move to more productive jobs (M2MPJ): by improving accessibility for commuters, an
infrastructure project may induce workers to change their location of work. If the project induces
the worker to take up a more productive job, there is an additional benefit to society. This benefit
is the average tax take on the marginal increase in wages that the worker earns. It is
recommended that a full calculation of M2MPJ is performed. Where parameters for this benefit
already exist in New Zealand they should be utilised. Where these parameters have either yet to
be established, or only exist at a different spatial level, proxy parameters may be utilised. In some
cases, it may not be appropriate to assume that transport investment would cause land use
change. Therefore, it is appropriate to outline the cause of dynamic land use which makes a
dynamic assumption realistic when assessing project appraisal. Provided that specific guidelines
are followed, there would not be a risk of any double counting of benefits or costs between static
and dynamic WEBs, nor between either and conventional user benefits.
Estimating dynamic WEBS

Waka Kotahi has released a technical paper, Transformative transport projects (dynamic WEBS and land
use benefits and costs) (Waka Kotahi NZ Transport Agency 2020c), which documents emerging guidance
on dynamic WEBs. This guidance will be updated as new research or methodologies become available.
3.10 Wider economic benefit (productivity)

It is widely recognised that economic density – the clustering of activity in towns and cities – has a
positive impact on productivity, and that such clustering is dependent on effective transport systems.
Some of the productivity effects, commonly known as agglomeration, come from interactions between
different economic agents that are not fully internalised, creating market failure and wider economic
benefits, as recognised in this manual.
‘Agglomeration effects are characterised as the productive impact of employment in surrounding areas on
a firm’s production technology. Agglomeration is acknowledged to have three underlying sources,
sharing, matching, and knowledge spillovers.’ (Duranton and Puga 2003).
The required spatial concentration of economic activity for realising agglomeration benefits is only
likely to occur in the major industrial and urban centres of New Zealand. It is only the large and
complex urban transport activities that will provide the relevant conditions that justify an analysis of
agglomeration benefits.
This section sets out a step-by-step process for estimating agglomeration benefits of transport
investment.
The method requires transport modelling data for the urban area of influence. Generally this will be
extracted to a spreadsheet from a regional or sub-regional strategic transport network model, using the
model zoning system or an aggregation of zones appropriate to the activity (more detailed zoning in the
urban centre and around the locality of the activity, and coarser zoning for peripheral areas). The selected
zones should give a reasonable compromise between detail and practicality.

Step A: Define spatial zoning system

Capturing this requires a spatial zoning system to be defined for the purpose of assessing agglomeration
economies. The main criteria for a spatial zoning system are:
• full coverage of the study area and as large a geographic area as possible
• a reasonable level of detail (for instance by area units)
• ability to be tied to a set of boundaries for which one can extract detailed statistical information on
employment and output.
Since much of the data needed for the assessment will come from one or more transport models, the
model zoning system(s) should be the starting point. Transport models tend to have a high degree of
geographical detail in the study area and much less detail for external zones. It is usually not possible or
desirable to disaggregate model zones in a sensible way, so in practice a zoning system needs to use the
transport model zones as building blocks.
Step B: Gather economic data
Step B sets out in detail the economic data that is required for the analysis.
B1: Employment data
Zonal employment data (full-time equivalent employees) is required for the year or years for which the
assessment is being made. Ideally, separate employment projections for the do-minimum and option
scenarios would be used, but it is most likely that only fixed land use and employment projections will be
available and will be acceptable.
B2: Economic output data
An estimate of gross domestic product (GDP) per zone is obtained by distributing the regional GDP for
the assessment year in proportion to zonal employment. Regional GDP estimates can be derived from
Statistics New Zealand data. Sector disaggregation by Australian and New Zealand Standard Industrial
Classification (ANZSIC) should be used for the analysis and be undertaken individually for each industrial
sector.
Table 36: Data requirements

Data Variable Disaggregation Source
Demand D Origin–destination pair, do-minimum, Transport model
option, mode, purpose, year
Generalised cost GC Origin–destination pair, do-minimum, Transport model
option mode, purpose, year
Base year employment – Zone, full-time equivalents ANZSIC Statistics New Zealand
Future year employment E Zone (option) Transport model/other
Agglomeration elasticities ε ANZSIC Table 37
Output GDP Zone/ANZSIC Statistics New Zealand

B3: Agglomeration elasticities by zone

Current estimates for the relationship between density and productivity are shown in Table 37.
Table 37: Weighted average agglomeration elasticities for New Zealand by industry
ANZSIC 2006 Industry Agglomeration
elasticity (ε)
A Agriculture, forestry and fishing 0.032
B Mining 0.035
D Electricity, gas, water and waste services 0.035
C Manufacturing 0.061
E Construction 0.056
F Wholesale trade 0.086
G Retail trade 0.086
H Accommodation and food services 0.056
I Transport, postal and warehousing 0.057
J Information media and telecommunications 0.068
K Finance and insurance services 0.087
M Professional, scientific and technical services 0.087
N Administrative and support services 0.087
O Public administration and safety 0.087
L Rental, hiring and real estate services 0.079
P Education and training 0.076
Q Health care and social assistance 0.083
R Arts and recreation services 0.053
All industries 0.065
An intermediate step is to calculate the agglomeration elasticities for each study zone using evidence of
each zone’s sector composition of employment. This is done by calculating the weighted average of the
elasticities using employment proportion of each sector for each zone as weights.
∑𝑆 (𝜀𝑖𝑆 × 𝐸𝑖𝑆 ) where:
𝜀𝑖 = ∑𝑆 𝐸𝑖𝑆
ε = agglomeration elasticity
E = employment
This operation requires data on base year workplace-based employment by study zone for each of the
sectors for which agglomeration elasticities are provided, as well as total employment (for the remainder
of the economic sectors a zero elasticity is assumed). Employment growth forecasts and output forecasts
are required by sector for each assessment year.
B4: Transport model outputs
The transport model data required is origin–destination matrices of demand and generalised cost for:
• each modelled transport mode

• the following journey purposes/user segments:
o work travel purpose (including freight)
o commuting to and from work
o non-work travel purposes
• the do-minimum or option scenarios
• one or more future assessment years.
A typical scenario could include two variables for public transport and car modes, three purposes, two
scenarios and one future year, which produces 24 origin–destination matrices. When gathering and
preparing the transport data, there are a number of things to bear in mind:

• Coverage of all major modes: although the transport activity under consideration may only
affect one mode, all travel modes need to be included in the analysis, as it is the relative change
in travel costs that drives agglomeration benefits. If the transport model only represents a single
mode, it will be necessary to make assumptions on journey costs for other modes and the
proportion of demand by mode.
• Separately identified user groups: if the demand and cost data is not available separately for
the required journey purposes and/or user segments, they will need to be estimated. This is
feasible by adjusting the time-cost element of generalised cost for differences in values of time
between user groups.
• Complete cost matrices: for the analysis the cost matrices need to contain cost information for
all origin–destination movements where there is travel demand. This is to avoid weight being
given to origin–destination pairs where the costs are set arbitrarily high or low (transport model
matrices frequently contain zeros or very high cost on pairs where there is no cost information).
This includes intra-zonal movements. There should be no zeros or empty cells.
Where the available data does not cover all modes or there are missing cells, the matrices should be
complemented with evidence from other sources. Possible sources include:
• time, cost or demand data from other transport models
• distance and/or journey time data from GIS or journey planning tools
• assumptions on average time/cost per kilometre
• census travel to work data
• travel surveys.
Step C: Calculate weighted average costs for in-work and travel to work across all modes
The relevant measure of journey costs for the purpose of assessing agglomeration impacts is the
weighted average generalised cost for work travel purposes (including freight where relevant) and
commuting to and from work, across all modes.
Demand should be used as weights. So, for a given origin–destination pair, the relevant generalised cost
for the do-minimum or options:
where:
AGC is the average generalised cost
∗,𝑚,𝑝 𝑆,𝑚,𝑝 D is the demand
∑𝑚,𝑝 𝐷𝑖,𝑗 𝐺𝐶𝑖,𝑗
𝐴𝐺𝐶𝑖𝑗𝑆 = ∗,𝑚,𝑝
GC is the generalised cost
∑𝑚,𝑝 𝐷𝑖,𝑗 S is the do-minimum or option
m is the mode
p is purpose
I is origin
J is destination
Note: the superscript * on demand reflects that these weights need to be identical for both the do-
minimum and option, eg the sum of the do-minimum and option demands.
Step D: Calculate effective density by zones for each scenario
The effective density of employment is calculated for each scenario and assessment year using the AGC
from step C and the total employment by zone gathered in step B, using the following relationship:
𝐸𝑗𝑆 Where:
𝐸𝐷𝑖𝑆 = ∑ ED is effective density
𝐴𝐺𝐶𝑖𝑗𝑆 E is employment
𝑗
Step E: Calculate productivity gains by zone

The productivity gains from agglomeration are calculated for each zone by applying the agglomeration
elasticities to the change in density in each zone:

where:
𝜀 δPR is relative increase in productivity
𝐸𝐷𝑖𝑂𝑃𝑇 𝑖 OPT is option
𝛿𝑃𝑅𝑖 = ( 𝐷𝑀 ) − 1 DM is do-minimum
𝐸𝐷𝑖
ε is agglomeration elasticity
i is zone
The absolute increase in productivity by zone is then obtained by multiplying the percentage increase with
the output by zones:
where:
𝑑𝑃𝑅𝑖 = 𝛿𝑃𝑅𝑖 × 𝐺𝐷𝑃𝑖 dPRi is absolute increase in productivity in dollars
GDPi is total output for each zone
If the agglomeration analysis is undertaken by industrial sectors, this step will have to be repeated for
each of the sectors where there is agglomeration evidence (in other words there will be another subscript
for all variables in the two equations, except for the effective densities, since these are always calculated
based on total employment by sector).
Step F: Sum output increases across all zones in the study area
The final step in estimating the impact of the intervention on productivity is to sum the agglomeration
gains across the study zones:
where:
𝐴𝑔𝑔𝑙 = ∑𝑖 𝑑𝑃𝑅𝑖 Aggl is total agglomeration benefits from the
interventions.
Figure 7: Step by step guidance for agglomeration benefits
Step G: Profiling and calculation of net present values

Standard guidance on profiling impacts over the analysis period is to interpolate between the base year
and the analysis years, and to extrapolate from the last year of the analysis period. While the interpolation
can be done by linear annual increments, the extrapolation is done by assuming all variables remain
constant from the last analysis year, ie demand and employment, but allowing productivity to grow
annually. Benefits must be based on constant dollars.

3. Benefits > 3.11 Wider economic benefit (employment impacts)
The extrapolation of agglomeration gains is straightforward. The benefits for the last modelled year are
assumed to grow by the rate of productivity growth until the last year of the analysis period. The full
stream of agglomeration benefits is then discounted to the base year and summed to derive the net
present value.
Step H: Interpretation and presentation
The main output of the assessment is total productivity gains from agglomeration as the total net present
value of benefits. The results can also be presented in several other ways:
• as a proportion of conventionally measured evaluation benefits

• productivity gains per worker, or
• productivity gains for a future year.
It can also be useful to demonstrate how the agglomeration benefits are distributed across the study area.
This is an indication only, as it will only ever represent the location of the first round of impacts and not
their final incidence. There is therefore a clear trade-off between the level of spatial disaggregation and
robustness. For New Zealand, an appropriate balance between the two may be to present findings at the
level of territorial units.
Finally, if the analysis has been undertaken at an industry sector level, the impact on different parts of the
economy could be illustrated.
3.11 Wider economic benefit (employment impacts)

Job creation is often held up as a major impact of transport investment, with two distinct mechanisms
being suggested:
• impact on the supply side: better transport may make it easier for people to get to work, and
may replace discouraged worker effects
• impact on the demand side: induced investment creating new employment opportunities when
there is unemployment, ie displacement is not 100%.
The methodologies for estimating each of the employment impacts are discussed in this section.
Labour supply: participation and tax wedges

Individuals’ labour force participation decisions are based on comparing the costs of working (including
commuting costs), against the wages earned from a job. By reducing the cost (in time and money) of
getting to work, a transport investment is likely to increase the returns to working; some people, for whom
the net returns to entering the labour market were initially not worthwhile, may decide to enter. Such an
increase in labour supply and employment raises gross value added (GVA) but, in the simplest
circumstances, does not increase welfare. Initially, the individual was not working because the utility from
leisure exceeded that from working, net of commuting costs. If a transport improvement triggers work, the
benefit to the individual cannot be greater than the user-benefit received (if it were, the individual would
have chosen to work in the first place). However, this conclusion changes if there is an income tax wedge
(or loss of state benefits). The individual does not receive the full value of work undertaken because a
fraction of it accrues to government. The full gain from entering employment is then the user-benefit plus
tax revenue paid (or benefits not received).
The impact of the transport investments on labour supply should be estimated using the following steps.
Step 1: Calculate commuting costs
The first step requires an estimate of the change in commuting costs for workers living in zone i and
working in zone j. Calculate for the do-minimum and option the total annual commuting costs for each
origin–destination pair (ie home to work, from i to j, and work to home, from j to i), averaged across all
modes.

3. Benefits > 3.11 Wider economic benefit (employment impacts)
where:
GijOPT,c,f is the average generalised cost across mode
(m), commuting purpose (c), forecast year (f), and
option (OPT), between origin zone (i) and destination
zone (j)
𝑂𝑃𝑇,𝑚,𝑐,𝑓 𝑂𝑃𝑇,𝑚,𝑐,𝑓 𝐷𝑀,𝑚,𝑐,𝑓 gijOPT,c,f is generalised cost for mode (m), purpose (p),
𝑂𝑃𝑇,𝑐,𝑓
∑𝑚(𝑔𝑖𝑗 + 𝑔𝑗𝑖 ) × 𝑇𝑖𝑗
𝐺𝑖𝑗 = forecast year (f), and option (OPT), between origin zone
∑𝑚 𝑇𝑖𝑗𝐷𝑀,𝑚,𝑐,𝑓 (i) and destination zone (j)
TijDM,m,c,f is the total number of annual home to work
trips for mode (m), commuting purpose (c), forecast
year (f), between home zone (i) and work zone (j) in the
do minimum scenario. (Also known as the ‘home to
work’ matrix in do-minimum (DM)
The total annual commuting cost savings for workers living in zone i is calculated by multiplying the
change in commuting cost for each destination by the number of commuters and summing.
where:
dGif is total annual commuting cost savings for workers
living in zone (i) and forecast year (f)
WijOPT,f is the number of workers commuting from zone i
to zone j in option (OPT) and forecast year (f)
𝑓 𝑂𝑃𝑇,𝑓 𝑂𝑃𝑇,𝑐,𝑓 𝐷𝑀,𝑐,𝑓 GijOPT,c,f is the average generalised cost across
𝑑𝐺𝑖 = ∑ 𝑊𝑖𝑗 (𝐺𝑖𝑗 − 𝐺𝑖𝑗 )
𝑗
commuting purpose (c), forecast year (f), and option
(OPT), between origin zone (i) and destination zone (j)
GijDM,c,f is the average generalised cost across
commuting purpose (c), forecast year (f), and do-
minimum (DM), between home zone (i) and work zone
(j)
Step 2: Labour supply response

The labour supply response can now be calculated by assessing the magnitude of the commuting cost
changes in relation to workers’ net wage for each area and multiplying by the labour supply elasticity.
where:
dEi is total labour supply in zone (i)
1
𝑑𝐸𝑖 = 𝜀 𝑙𝑠 𝑑𝐺
𝑓 ls is the elasticity of labour supply with respect to
𝑦𝑖 − 𝜏𝑙 ) 𝑖
(1 effective (real) wages
yi is gross mean residence based earnings in zone (i)
l is factor to convert gross to net earnings
dGif is total annual commuting cost savings for workers
living in zone (i) and forecast year (f)
Step 3: Gross labour supply impact
The increased output from the increased labour supply impact is estimated using the product of the
increased labour supply and the net productivity per worker for new entrants.
where:
𝐿𝑎𝑏𝑜𝑢𝑟 𝑠𝑢𝑝𝑝𝑙𝑦 𝑖𝑚𝑝𝑎𝑐𝑡 = ∑ 𝑑𝐸𝑖 𝜂𝑚𝑖
dEi is total labour supply in zone (i)
𝑖
 is the productivity of marginal labour market entrants
relative to the average
mi is gross mean residence based GSP per worker in
zone (i)
Step 4: Net labour supply impact
The wider economic impact from increased labour supply is the proportion of the additional output taken
in taxation.
𝑊𝑖𝑑𝑒𝑟 𝑒𝑐𝑜𝑛𝑜𝑚𝑖𝑐 𝑖𝑚𝑝𝑎𝑐𝑡 𝑓𝑟𝑜𝑚 𝑖𝑛𝑐𝑟𝑒𝑎𝑠𝑒𝑑 𝑙𝑎𝑏𝑜𝑢𝑟 𝑠𝑢𝑝𝑝𝑙𝑦 = 𝐿𝑎𝑏𝑜𝑢𝑟 𝑠𝑢𝑝𝑝𝑙𝑦 𝑖𝑚𝑝𝑎𝑐𝑡 × 𝜏 𝑙𝑠
where: ls is tax take on increased labour supply
This can add up to an additional 10% of wider economic benefits over conventional benefits.

3. Benefits > 3.12 Wider economic benefit (imperfect competition)
Labour demand and unemployment

In demand side, if new jobs are created in one place, then the value of output produced by each new job
is the wage, and this is set against the value of what workers would have done, absent the new jobs.
For workers drawn out of involuntary unemployment the alternative is of low value, so the net benefit is
large. This may be an important effect in developing economies, in regions with significant structural
unemployment (or underemployment) or in special economic conditions, such as after a pandemic.
However, for long-run transport projects in reasonably well-functioning market economies it seems likely
that the labour market will adjust to some ‘natural rate’ of unemployment which is independent of
transport investment. If this is the case then an increase in labour demand is met either by increased
labour force participation or by drawing workers out of other employment. If demand is met by increased
labour force participation then its value is, as above, the tax wedge on income. If it is met by withdrawing
labour from other activities, then the value is the alternative wage. There is no net benefit if wages are the
same in both jobs. Displacement is 100%, so demand induced employment effects should, from the
national perspective, be ignored.
A qualification to this argument is conceptually important, although perhaps not quantitatively large for
any single transport project. To draw labour from other activities there may have been an increase in
wage rates in the area affected or more broadly. Given the level of productivity, an increase in wages
must be financed either by a reduction in profits (or more generally, payments to other inputs), or by an
increase in prices. The increase in wages is therefore just a transfer, of no value to aggregate income,
unless the people paying for it (consumers and recipients of profits) are, for some reason, people that we
do not value. A standard approach would be to suggest that benefit arises to the extent that the increase
in price is paid by foreigners, ie represents a terms of trade improvement, so the country is able to sell its
exports at higher price. This is an additional source of benefit, although one that is unlikely to be
quantitatively significant for any single transport project (Venables 2016).
Labour demand is more likely to impact through the national labour market and, as suggested above, is
likely to displace workers from other jobs.
3.12 Wider economic benefit (imperfect competition)

Conventional transport economics assumes all transport-using sectors operate in perfect competition,
where price equals marginal costs. The value of the additional production is identical to the gross
marginal labour cost of the additional hour worked. Conventional economics therefore measures the
value of the travel time saving as a saving in gross labour cost.
However, if price-cost margins exist, they cause a wedge between gross labour costs and the market
value of what is produced. Hence, where there are price-cost margins, a transport-induced increase in
output will cause a wider economic impact identical to the size of this wedge.
Imperfect competition parameters

The average price-cost margin in the New Zealand economy is 20%. Together, with evidence on how the
economy responds to a reduction in transport costs at an aggregate demand elasticity of -0.6, this gives
an estimated wider economic benefit from increased competition of 10.7% of business user benefits.
Table 38: Imperfect competition parameters

Parameter Description Value
εad Aggregate demand elasticity -0.6
pcm Price-cost margin 20.0%
τ Imperfect competition uplift to business user benefits 0.107
𝐼𝑚𝑝𝑒𝑟𝑓𝑒𝑐𝑡 𝑐𝑜𝑚𝑝𝑒𝑡𝑖𝑡𝑖𝑜𝑛 𝑖𝑚𝑝𝑎𝑐𝑡 = ∑ 𝜏 × 𝐵𝑢𝑠𝑖𝑛𝑒𝑠𝑠 𝑢𝑠𝑒𝑟 𝑏𝑒𝑛𝑒𝑓𝑖𝑡𝑠 𝑓

𝑓

3. Benefits > 3.13 Wider economic benefit (regional economic development)
where: Business user benefits is total conventional business user benefits from travel time and vehicle
operating cost savings
f is forecast year
τ is imperfect competition uplift factor.
Figure 8: Estimating imperfect competition benefits
Total business user Imperfect

benefits by forecast   Business user benefits f competition
year f impact
This can typically add up to an additional 5% of wider economic benefits over conventional benefits.
Back to 1.7 Benefits: Transfers >>
3.13 Wider economic benefit (regional economic development)

The economic benefits that may arise from increased tourism activity are established on a separate
theoretical underpinning to the traditional wider economic benefits of agglomeration, increased labour
supply and imperfect competition reduction. The three traditional WEBs relate to supply-side
improvements that lift economic output, for example by increasing the size of the labour force or
increasing the productivity of existing firms and workers. Here, increased activity within the economy is
driven by more efficient utilisation of existing capital and labour resources.
Tourism benefits, by contrast, arise due to changes in the demand for goods and services produced in
New Zealand. Hence tourism benefits may be thought of as a demand-side response to increased activity
and therefore there may need to be a reallocation or expansion of capital and labour resources to meet
the new level of demand.
When estimating tourism benefits, particular care must be taken to ensure that there is a real increase in
the level of output, and that there is not simply a displacement of activity from one sector of the economy
to another. For this reason, domestic tourists must be excluded from the analysis, reflecting the fact that
domestic tourists are likely to have otherwise spent their money on other local goods and services.
Domestic tourists may generate user benefits and these should be captured through traditional transport
benefit appraisal.
Tourism demand estimation

Predicting the number of tourists that will make use of a new facility is typically the most challenging
element of any evaluation of tourism benefits.
Where possible, demand estimates should:
• draw upon project- or sector-specific surveys of visitor intentions and/or discussions with tourism
operators
• seek out case studies of similar facilities
• check the reasonability of predictions against the scale of the overall market.
After estimating total visitor activity, it is necessary to distinguish between domestic tourists and
international tourists. The calculation of economic benefits shall relate to international tourists only, as
domestic tourists are to be excluded.

It is also necessary to consider the additionality of international tourist activity. Economic benefits shall
only be calculated for international visitors who would not have come to the country if the facility was not
available, or who would have come to the country but would have spent less time or money. Where
existing international visitors would be diverted from other regions, or where they are dissuaded from
visiting due to the investment, displacement effects must be calculated.
Table 39 summarises five different categories of visitor and outlines the approach for valuing the
economic benefits associated with each category.
Table 39: Categories of visitors

User origin Alternative without Added economic Calculate or exclude
facility benefits/disbenefits?
New Zealand Spend time or money in No Exclude from benefit calculations
NZ
Travel overseas Potentially Exclude unless there is specific evidence
to identify the number of people who will
do this
International Spend same amount of No Exclude from benefit calculations
time or money in NZ
Spend less time or Yes Calculate changes in itinerary or spending
money in NZ and convert to value added
Spend more time or Yes, disbenefit. Calculate changes in itinerary or spending
money in NZ and convert to value added
Do not visit NZ Yes Calculate changes in itinerary or spending
and convert to value added
Measuring economic benefits from increased tourism

After estimating the net increase in international visitor activity in New Zealand, it is necessary to convert
an increase in visitor numbers into increases in economic activity in New Zealand.
An increase in international tourist activity in New Zealand can be thought of as an increase in New
Zealand’s exports. The economic benefits of this increase in demand relate to the resulting increase in
New Zealand’s total economic output, ie its gross domestic product (GDP) or gross national income.
Value added, or impact on national GDP, is the measure used to calculate the economic benefits of
increased international visitor activity. It represents the additional work undertaken in the impacted
sectors, excluding intermediate inputs that are imported or purchased from other sectors.
Value added can be calculated as follows:
𝑽𝒂𝒍𝒖𝒆 𝒂𝒅𝒅𝒆𝒅 = 𝐸𝑚𝑝𝑙𝑜𝑦𝑒𝑒 𝑐𝑜𝑚𝑝𝑒𝑛𝑠𝑎𝑡𝑖𝑜𝑛 + 𝑝𝑟𝑜𝑓𝑖𝑡𝑠 𝑎𝑛𝑑 𝑚𝑖𝑥𝑒𝑑 𝑖𝑛𝑐𝑜𝑚𝑒 + 𝑛𝑒𝑡 𝑡𝑎𝑥𝑒𝑠 𝑜𝑛 𝑝𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑜𝑛
𝒐𝒓 = 𝐺𝑟𝑜𝑠𝑠 𝑟𝑒𝑣𝑒𝑛𝑢𝑒 − 𝑖𝑛𝑡𝑒𝑟𝑚𝑒𝑑𝑖𝑎𝑡𝑒 𝑖𝑛𝑝𝑢𝑡𝑠
Table 40 estimates the average value added per international visitor night on a regional basis. The full
methodology used to estimate these figures is available in Waka Kotahi NZ Transport Agency, 2019b.

Table 40: International visitor activity and estimated economic activity, by region ($2018)
Region International guest Estimated spending Estimated value
nights (000s, per guest night added per guest night
unadjusted) (adjusted to match (adjusted to match
IVS) IVS)
Northland 702 $112 $41
Auckland 3,503 $367 $143
Waikato 1,114 $178 $69
Bay of Plenty 1,521 $114 $46
Hawke’s Bay, Gisborne 388 $148 $56
Taranaki, Manawatū, Wanganui 454 $165 $60
Wellington 1,038 $228 $91
Nelson, Marlborough, Tasman 902 $122 $45
Canterbury 2,603 $164 $62
West Coast 912 $98 $35
Otago 3,758 $159 $63
Southland 700 $108 $39
Total 17,595 $194 $75
Sources: Statistics NZ International Visitor Survey, Accommodation Survey, Tourism Satellite Account,
and Annual Enterprise data; MBIE Monthly Regional Tourism Estimates.
The data indicates that there are regional variations in spending patterns and economic impacts, and
therefore displacement of activity between regions has real effects on national benefit levels. In addition,
it implies that, on average, each additional dollar of tourist spending results in around $0.39 in value
added.
The estimated value added data in Table 40 should be relied upon for calculating the economic benefits
of increased international tourist activity. The use of non-standard values will require submission of
sufficient supporting evidence and the agreement of Waka Kotahi in writing.
For help, or to discuss project requirements, please contact the MBCM team through
[email protected].
Reporting tourism benefits

Tourism benefits must not be included in any BCR calculation, even as a sensitivity test, or added to the
quantum of traditional transport WEBs. Tourism benefits should instead be reported as a separate
monetised benefit in the AST.
The current domestic and international evidence base concerning the effects of transport infrastructure on
international tourism activity is limited and covers a variety of approaches. As there has been no
consistent quantification methodology for benefits to date, it is unclear whether benefits estimated using
the value added approach are wholly additive to those included in a traditional BCR calculation. Similarly,
due to a lack of monitoring, it is unclear whether tourism benefits are sustained over the medium- to long-
term horizons captured by transport appraisal.
While there is not sufficient confidence to include monetised tourism benefits in BCR calculations at this
time, Waka Kotahi may revise this policy in the future once an evidence base is developed, and their
inclusion in the AST ensures they are captured in the decision-making process.
Any analysis of tourism benefits must be accompanied by the following measures:
• net national change in international visitor guest nights

• net national change in international visitor spending, and
• net national change in value added (or GDP equivalent).
The economic benefit that may be reported in the appraisal summary table is the net national change in
value added, or the equivalent impact on national GDP.
Visitor spending must not be reported as a measure of economic benefits as this does not account for the
cost to serve visitor spending. Similarly, regional economic benefits must not be reported, as Waka Kotahi
takes a national view of economic benefits and costs and any displacement effects must be accounted
for.
Sensitivity to displacement effects

Additional tourism activity may displace, or crowd out, other economic activity. Where this occurs, the
economic benefits from international visitors will be smaller than their total expenditure, ie the net GDP
benefits from international visitors will be less than their gross contribution to GDP.
There are three primary ways crowding out can occur:
• capacity constraints at tourism facilities may mean that it is not possible to accommodate
additional visitors
• increased tourism activity may draw resources away from other industries, and subsequently
these industries may reduce in size, or
• increased international visitor activity may cause the exchange rate to appreciate, crowding out
other exports.
Analysis of tourism benefits should consider whether capacity constraints at tourism facilities are likely to
apply. For instance, it may be necessary to assess trends in local hotel occupancy rates to identify
capacity constraints that prevent international tourists from increasing the length of their stay or adding
the destination to their itinerary.
The second and third mechanisms are general macroeconomic effects and should be assumed to apply
to most projects.
Sensitivity to multiplier effects

Theory and evidence suggest that displacement effects dominate multiplier effects. The Treasury’s Guide
to social cost benefit analysis (2015) states that ‘unless there is significant unemployment of people with
the requisite skills, it is likely that multiplier effects do not exist.’ For the purpose of analysis, multiplier
effects should be ignored unless there is clear evidence of underutilised resources.
Summary principles for estimating tourism benefits
1. Analysis of economic benefits should be based on the expected net change in international
tourism. Domestic tourism should be excluded.
2. Demand estimation must assess whether tourists would have come to New Zealand without the
project and, if so, discount their added spending accordingly.
3. The analysis must identify any short- and long-term capacity constraints that would prevent the
sector from serving additional visitors.
4. Estimates of economic benefits must be based on net impact on value added, which will always
be smaller than the total added expenditure.
5. Value added estimates must account for crowding out effects and exclude multiplier effects
unless there is evidence of underutilised resources.
Input–output and CGE modelling

Input–output analysis and CGE (computable general equilibrium) modelling remains recommended on a
case-by-case basis for projects that are likely to substantially alter the distribution of economic activity
within New Zealand. In these instances, it is recommended that advice is sought from Waka Kotahi on the
most appropriate form of analysis to be utilised.
3. Benefits > 3.14 Composite value for abatement of marginal congestion costs
3.14 Composite value for abatement of marginal congestion costs

In the presence of traffic congestion, ie where the road corridor has at least one point that operates at
greater than 80% capacity during the peak period, the removal of some traffic will generally provide
positive benefits to remaining road users. Some activities, however, may achieve their improved transport
service level by reducing the available road capacity for other road users. The level of traffic congestion to
remaining users may then be increased, creating a negative impact. Also, traffic congestion may be
increased where a proposed transport service increases the number of public transport vehicles on roads
shared with other traffic.
The effect of increased transport output on overall traffic congestion will depend on:
• the change in the number of public transport vehicles per hour per period
• their size and performance characteristics
• the reduction in the number of trips
• the do-minimum composition of road traffic flow.
A composite benefit is defined for valuation of benefits related to changes to road traffic that usually
include impact on network productivity and utilisation (travel time cost and vehicle operating cost), and
impact on social cost of deaths and serious injuries.
Impact on greenhouse gas emissions, which was included as part of the composite values in the previous
version of the MBCM, is no longer included in the values as it needs to be calculated using the new
procedure (see section 3.4) and reported separately in the AST.
Road traffic reduction benefits critical to the outcome of the evaluation may include:
• traffic volumes, particularly model results, growth rates and the assessment of diverted and
generated traffic and transport service users
• travel speeds
• crash reduction.
For each significant factor the following shall be listed:
• the assumptions and estimates on which the evaluation has been based
• an upper and lower bound of the range of the estimate
• the resultant BCR at the upper and lower bound of each estimate.
Extra caution for double counting is required when the composite value for abatement of marginal
congestion costs is used to calculate activities benefits.
With respect to transport services, road traffic reduction benefits shall generally be limited to peak
periods. The evaluator shall specify, and justify, the peak period times.
In some cases, for instance with most freight transport services, it may be appropriate to also consider
off-peak period road traffic reduction benefits.
For a new walking or cycling facility, in addition to the walking and cycling benefits a composite benefit for
the abatement of marginal congestion costs of $0.22 ($2021) per km of trip length may be applied for
each new pedestrian or cyclist using the new facility.
The composite values in Table 41 are used in the simplified procedures, SP8 Freight services, and SP9
New public transport services.
3. Benefits > 3.15 Other monetised impacts
Table 41: Diversion rates and composite benefit values ($2021) for abatement of marginal
congestion costs for major urban corridors (refer also to simplified procedure SP9 worksheet
SP9.5)
Urban area Diversion rate (vehicle/km removed Road traffic reduction benefit
from road per new public transport ($/vehicle/km per year removed from
passenger km) road – $2021)
Auckland 0.725 (72.5%) $2.99
Wellington 0.777 (77.7%) $1.83
Christchurch/other 0.675 (67.5%) $0.40
The composite values in Table 42 are used in simplified procedure SP10 Existing public transport
services.
Often, changes to existing public transport services are limited to additional peak period services that
remove commuters from private vehicles. In such cases the cost of the service should only include the
capital costs and the maintenance and operating costs of providing the additional peak period public
transport services where there are road traffic reduction benefits.
Table 42: Composite benefit values ($2021) for abatement of marginal congestion costs for major
urban corridors by PT modes ($/additional passenger boarding) (refer also to simplified procedure
SP10 worksheet SP10.4)
Urban area Mode Average trip Road traffic reduction benefits
length (km) ($/additional passenger boarding –
$2021)
Peak Off peak
Auckland All 7.70 19.77 1.88
Rail 16.50 27.50 3.42
Bus 6.60 18.27 1.69
Wellington All 12.14 20.40 2.57
Rail 22.76 27.23 3.65
Bus 6.97 18.73 2.05
Christchurch All 8.05 4.24 3.20
Other All 7.86 2.63 2.32
Where the trip length for a public transport activity is significantly different to the average trip length
indicated in Table 42, calculate the traffic reduction benefits using the diversion rates and road traffic
reduction benefits ($/vehicle/km per year) from Table 41.
3.15 Other monetised impacts

The benefits framework provides a comprehensive list of transport benefits, however, not all of these are
assigned a standardised value in this manual. These benefits may be monetised subject to data
availability and agreement with Waka Kotahi in writing. Some common valuation methodologies and data
sources are described below.
Valuation methods
There are two types of consumer preference surveys – revealed preference (RP) surveys and stated
preference (SP) surveys:
RP surveys observe actual behaviour under varying conditions, for example the modes of travel used by
household members relative to the level of service of public transport. This information is then analysed to
identify and quantify the factors that influence travel decisions.
3. Benefits > 3.15 Other monetised impacts
SP methods ask individuals how they would respond to various situations. Two techniques used in SP
analyses are contingent valuation and conjoint analysis. Contingent valuation (attitudinal) surveys ask
respondents directly how they would respond to various situations, or ask them to rate or rank their
preferences for various levels of service, facility or situation. This often gives values several times higher
than what they would be in reality, because people often do not do what they say they would do. This
type of survey tends to be better suited to evaluating relative preferences and for estimating the maximum
possible response to an action, than to predicting actual changes in travel.
Conjoint analysis (hypothetical choice) surveys require respondents to make choices between
hypothetical alternatives with varying attributes. It is necessary to have forced trade-offs so that a better
environment might be coupled with higher costs or a higher travel time. This forces the respondent to
relate the value of each component of preference.
SP surveys need to be stratified by audience: current users versus potential users. Current users should
be asked to respond to questions about factors that would provide for a more comfortable or attractive
journey through different types of environments, facilities or levels of service.
For potential users, it is important to create scenarios based on constructed markets. For example,
questions could include what mode they would choose for work and non-work trips based on the quality
of the transport environment, including travel by private vehicle, public transport, walking and cycling. It
would query residents about the degree to which they perceive different levels of service or facilities
would improve the conditions of their commute, recreational activities and so forth. By measuring how
demand might change, one can ascertain the preferences of current non-users, some of whom would
become users if certain improvements were made.
Analysts may wish to consult other sources for guidance as to the design and implementation of SP
surveys to derive willingness-to-pay values. Waka Kotahi may be able to provide some assistance in this
regard.
Benefit transfer is also one of the common methods is used for economic evaluation and specifically for
environmental benefits:
Benefit transfer, also known as value transfer, is simply using results of previous studies of analogous
situations (source values) to provide information about values of the case under consideration (study
values). Benefit transfer can be inexpensive and rapid if suitable source studies are available. Source
values can be transferred to the study project as point estimates, value functions, or as meta-analyses.
Meta-analyses, which draw information from a large number of previous studies, provide useful
information on source study valuation contexts, and identify adjustments required to transfer source
estimates to the study case. Meta-analysis also provides an indication of the variability of value estimates
and so is recommended rather than point and value function transfers from individual sources or a small
number of studies.
Great care is required to match source and study scenarios. Non-market values are highly sensitive to
context and can vary because of differences in the nature of the resource, the availability and prices of
substitutes and complements, underlying preferences, cultural context, environmental value orientations,
socio-economic characteristics, demographics, population density, transport availability, and other
matters.
Even the most careful and comprehensive benefit transfer studies can be extremely inaccurate. Hence,
benefit transfer is recommended primarily as a useful aid in determining whether non-market values are
likely to be significant for the project under evaluation, and whether a primary valuation study is
warranted.
Data and information sources

As Treasury (2015a) says: ‘The international Environmental Valuation Reference Inventory (EVRI
http://www.evri.ca/) provides source study information for benefit transfer from many thousands of non-
market studies undertaken worldwide, including studies from New Zealand. However, international benefit
3. Benefits > 3.16 Impacts during implementation/construction
transfer adds additional complications because of currency and cultural differences. New Zealand studies
can be identified at www2.lincoln.ac.nz/nonmarketvaluation/.’
Waka Kotahi has commissioned research to develop a database of values and valuation methods for
monetisation of other potentially monetisable benefits. It will also provide guidance on how to use the
database. A link to the research results and database will be provided when it is ready for publication later
in late 2021.
Impact on system vulnerability and redundancy

Impact on system vulnerabilities and redundancies is the transport system’s ability to enable communities
to withstand and absorb impacts of unplanned disruptive events, perform effectively during disruptions,
and respond and recover functionality quickly. It requires minimising and managing the likelihood and
consequences of small-scale and large-scale, frequent and infrequent, sudden and slow-onset disruptive
events, caused by natural or man-made hazards.
Impact on system vulnerabilities and redundancies can be measured by reduction in expected costs of
disruption. To account for risk, the costs of disruptions should generally be valued as expected values;
that is, the average costs of disruption, weighted by their likelihood of occurrence. This approach, which
treats expected values as being equivalent to certain amounts, is common practice in conducting a social
cost–benefit analysis and is consistent with other sections of this manual.
𝐸𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑎𝑛𝑛𝑢𝑎𝑙 𝑑𝑖𝑠𝑟𝑢𝑝𝑡𝑖𝑜𝑛 𝑐𝑜𝑠𝑡𝑠 = Σ𝑖 𝐴𝑛𝑛𝑢𝑎𝑙 𝑝𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑑𝑖𝑠𝑟𝑢𝑝𝑡𝑖𝑜𝑛𝑖 × 𝑐𝑜𝑠𝑡 𝑜𝑓 𝑑𝑖𝑠𝑟𝑢𝑝𝑡𝑖𝑜𝑛𝑖
The methodology for estimating probability and cost of disruption may vary depending on the nature of
the disruption, availability of data, interdependency and criticality of infrastructures. There are some
methodologies developed from the relevant research (see McWha 2020 and Hughes 2020) that can be
referred to, but the analysis is more on a case-by-case basis as there has not been one methodology that
fits all situations.
Back to 1.7 Benefits: Monetisation >>
3.16 Impacts during implementation/construction

Disbenefits considered in the economic evaluation during implementation should in most cases be
restricted to travel-time delays only, but this applies to all modes. It does not need to include vehicle
operating costs, crash cost, noise, dust, etc.
Where the activity/option results in minimal disruption (eg a tie-in that does not require reduction in
capacity during construction) there is no need to incorporate the disbenefits in the economic evaluation.
Where the impact of disruption is material, then the disbenefits of the activity/option shall be included in
the evaluation.
The impact should be determined through sensitivity analysis, eg a preliminary estimate of the disbenefits
to adjust the BCR. If the adjusted BCR remains within its funding efficiency profile level (ie low, medium,
or high), then there is no need to undertake a detailed evaluation of the disbenefits, provided the
difference between the BCRs is less than 10%. However, if the adjusted BCR falls to a lower profile level,
which could impact the activity's priority or funding source, then a detailed evaluation of the disbenefits
must be undertaken. If the adjusted BCR falls more than 10% then a detailed evaluation should be
undertaken.
Where disbenefits are included, they should cover the disbenefits to existing users and existing transport
services (public transport operators) as well as the costs of dislocation and disruption to all modes.
3. Benefits > 3.17 Behavioural change composite benefits
Generally, these costs/disbenefits could include:
• increased travel time

• increased travel discomfort
• loss of service
• change in demand for public transport
• loss of revenue
• any significant costs to the wider community during construction.
3.17 Behavioural change composite benefits

This section provides guidance on the monetisation of changes in workplace, school and
household/community-based travel plans. Analysis is based on the reduction in private car travel resulting
from travel plans being initiated.
Composite benefit values have been derived for a range of travel behaviour change (TBhC) activity types
and situations. The composite benefit values include benefits to the people changing their travel
behaviour as well as benefits to remaining road users and the general community, such as reduced
health costs and accident cost savings, vehicle operating cost (VOC) savings and environmental benefits.
Composite benefit values are the average annual benefit for all people in the workforce, school or
community targeted by the TBhC activity (and take account of the proportion that do not participate or
change their travel behaviour).
The composite benefits also incorporate the default diversion rate assumptions for each TBhC activity
type as well as the average trip length for each mode affected by the proposal. If evaluators consider they
have strong reasons why a different diversion rate is more appropriate for the situation they can
interpolate a composite benefit value (based on the values given below and the particular situation
compared with the default diversion rates) for workplace travel plans, or use a computer spreadsheet
programme (available from Waka Kotahi) to forecast a diversion rate and calculate a composite benefit
value for any TBhC proposal.
Table 43: Workplace travel plan benefit ($/employee/year – 2008)

Location Workplace CBD Non-CBD
Diversion Low Medium High Low Medium High
Auckland Standard 0.00 188.51 0.00 165.51
Alternative 0.00 214.47 616.23 0.00 191.47 556.89
Wellington Standard 0.00 170.88 0.00 147.88
Alternative 0.00 191.97 554.77 0.00 168.97 495.43
Christchurch/ Standard 0.00 61.97 0.00 61.97
other Alternative 0.00 58.21 196.51 0.00 58.21 196.51
Based on 100% of changed trips being in peak period.
Standard = without public transport improvements or subsidies.
Alternative = with public transport improvements or subsidies.
Table 44: School travel plan benefit ($/student/year – 2008)

Location School type
Primary Secondary/intermediate
Auckland 85.35 141.74
Wellington 82.70 121.17
Christchurch/other 74.83 77.97
Based on 55% of changed trips being in peak period
3. Benefits > 3.17 Behavioural change composite benefits
Table 45: Household community-based activity benefits ($/capita/year – 2008)

Location Level of diversion
Standard Low
Auckland 139.11 42.57
Wellington 158.72 49.25
Christchurch/other 129.45 39.19
Based on 15% of changed trips being in peak period
4. Evaluation procedures > 4.1 Introduction to procedures
4. Evaluation procedures
4.1 Introduction to procedures
Predecessors of this manual split evaluation procedures for activities across dedicated chapters for the
simplified and full procedures, and individual appendices. In this manual, evaluation procedures have
been grouped according to the type of activity to be assessed and are designed to be read as a whole.
Relevant information and data has also been transferred from the appendices of the superseded
Economic evaluation manual (EEM) where appropriate.
Before undertaking an evaluation, in-depth consideration must be given to the problem that is to be
solved or mitigated. This initial work to define the problem and consider potential solutions is part of the
Business Case Approach undertaken by an approved organisation and is included in their regional land
transport plans (RLTPs). These procedures do not include this initial problem definitional work but rather
start following the problem definition. As a result of the evaluation the potential solutions and
improvement options may be adjusted or changed during the process due to the availability of additional
or new information as the process develops.
This section includes the evaluation procedures for the following major types of activities:
• walking and cycling
• roading activities
• public transport services
• travel demand management
• education, promotion and marketing
• freight activities
• private sector financing and road tolling.
Each evaluation procedure offers two methods for assessing activities, and the choice of an appropriate
assessment methodology will depend on an activity’s size, risk and complexity.
Table 46 illustrates the relationship between the individual simplified procedures and the types of
improvement activities that are covered by full procedures. While some simplified procedures are directly
relevant to a single type of activity, there are other simplified procedures that may be used to assess
multiple transport and non-transport improvement activities.
The simplified procedures are designed to simultaneously establish the project impacts and the
monetised benefits and costs of undertaking activities that are low-cost and have low levels of risk and
complexity. The full procedures are designed to first establish the impacts of proposed options and then
assign these impacts monetary values, in order to establish the monetary benefits, before calculating the
BCR and other economic indicators.
Back to section 1.8 Costs >>
Table 46: Simplified procedures in relation to full procedures
Education, promotion
Walking and cycling
Types of activities
financing and road

Roading activities
Freight activities
Public transport
Travel demand
and marketing
Private sector
management
services
Simplified procedure
tolling
SP1 Road renewals
SP2 Structural bridge renewals
SP3 General road improvements
SP4 Seal extensions
SP5 Isolated intersection improvements
SP6 High productivity motor vehicle (HPMV)
route improvements
SP8 Freight transport services
SP9 New public transport services
SP10 Existing public transport services
SP11 Walking and cycling facilities
SP12 Travel behaviour change
SP13 Road safety promotion
The two simplified and full assessment methodologies are described below in more detail.
Simplified procedures
The simplified procedures are designed for the appraisal of activities that are low-cost and have low
levels of risk and complexity. Thresholds for activity value also apply (see Table 47).
This manual contains simplified procedures for the following types of sub-activities:
• SP1 Road renewals

• SP2 Structural bridge renewals
• SP3 General road improvements
• SP4 Seal extensions
• SP5 Isolated intersection improvements
• SP6 High productivity motor vehicle (HPMV) route improvements
• SP8 Freight transport services3
• SP9 New public transport services
• SP10 Existing public transport services
• SP11 Walking and cycling facilities
• SP12 Travel behaviour change
• SP13 Road safety promotion.
The criteria and thresholds applicable for deciding whether a proposal is of low-cost, risk or complexity
are described at the beginning of each evaluation procedure.
Each simplified procedure is a stand-alone procedure. They are designed to be applied directly to each
option being considered. Table 47 provides a summary of all 12 simplified procedures covered in this
manual.
3 There is no SP7. A gap has been left to accommodate future simplified procedures for roading or public transport
activities. Since two earlier manuals were combined into the EEM there has only been one new SP developed: SP6,
which was brought into play when the Vehicle Dimensions and Mass (VDAM) Rule 2016 was established and
allowed the use of high productivity motor vehicles (HPMV).
Input values for the simplified procedures may be obtained from:
• the default figures provided, or

• activity specific data collected, or
• the information in the evaluation procedures.
Analysis which alters components of the simplified procedure should not be used as this will
compromise the assumptions on which the procedure is based. In these instances, the full procedures
should be used instead.
Table 47: Simplified procedure summaries

Use If the activity is …
SP1 A road renewal, namely:
• sealed road pavement rehabilitation
• drainage renewals
• preventive maintenance.
Geometric improvements are excluded.
Where the undiscounted whole-of-life cost ≤$15,000,000.
SP2 A structural bridge replacement or renewal, where one of the following:
• undiscounted whole-of-life cost is ≤$15,000,000 and the AADT ≥50 vpd
• undiscounted whole-of-life cost is ≤$1,000,000, the AADT ≤50 vpd and a low-cost
option is not suitable
• undiscounted whole-of-life cost of providing a suitable low-cost option ≥$50,000
cheaper than providing a replacement bridge and the AADT ≤50 vpd.
A decision chart is provided in SP2 to assist selection of the appropriate procedure.
SP3 A general road improvement (including seal widening), where the undiscounted whole-of-life
cost ≤$15,000,000.
Traffic management facilities, new roads, road improvements and property purchases may be
assessed using this SP.
SP4 A seal extension, where the undiscounted whole-of-life cost ≤$15,000,000.
SP5 An isolated intersection improvement where the undiscounted whole-of-life cost ≤$15,000,000.
Traffic management facilities, new roads and road improvements may be assessed using this
SP.
SP6 A roading infrastructure improvement(s) specifically required to establish high productivity
motor vehicle routes and where the undiscounted whole-of-life cost ≤$15,000,000.
Structure component replacements, replacements of bridges and structures, and road
improvements may be assessed using this SP.
SP8 A freight transport service, where the undiscounted funding gap ≤$15,000,000 over the first
three years of operation.
SP9 A new public transport service, where the undiscounted funding gap ≤$15,000,000 over the
first three years of operation.
Bus services, passenger ferry services and passenger rail services may be assessed using this
SP.
SP10 An improvement to an existing public transport service, where the undiscounted funding gap
≤$15,000,000 over the first three years of operation.
Bus services, passenger ferry services and passenger rail services may be assessed using this
SP.
SP11 A walking or cycling facility, where the undiscounted whole-of-life cost ≤$15,000,000.
SP12 A travel behaviour change activity, where the undiscounted whole-of-life cost ≤$15,000,000.
SP13 A road safety promotion activity, where the undiscounted whole-of-life cost ≤$15,000,000.
4. Evaluation procedures > 4.2 Evaluation of walking and cycling activities
Full procedures
The full procedures are to be used to appraise economic efficiency when the assumptions contained in
the simplified procedures, including any cost limits, are exceeded.
The full procedures may be used for all types of land transport activities with appropriate adaptation. The
benefits and costs considered in the evaluation should be adjusted or added to as appropriate for the
activity type.
As much as possible, the full procedures are standardised to follow the same period of analysis and
utilise the Waka Kotahi worksheets, available on the MBCM page on the Waka Kotahi website.
Analysis period
The standard analysis period for improvement activities is 40 years from the year in which significant
benefits or costs commence, unless otherwise agreed with Waka Kotahi in writing. For activities with
short-lived assets, or activities where benefits dissipate quickly, it may only be necessary to assess the
activity over a 5- to 10-year period.
There are three critical times to be set up for the analysis process:
1. time zero – the date that all future cost and benefit streams are discounted to
2. analysis period – the period, starting from time zero, for which all costs and benefits are
included in the BCR calculations
3. base date – the date used as a basis for determining the monetary unit values of costs and
benefits.
Where several options are being evaluated, the analysis period for all options shall be determined by the
option with the earliest benefit or cost. The start of construction/implementation shall be the earliest
feasible date, irrespective of expectations of funding.
Worksheets
The full procedures contain two worksheets to guide the calculations and encourage consistency of
analysis. The two worksheets are Crash cost savings and Transport modelling checks. These worksheets
are to be used as far as is practical when preparing evaluations. Non-standard worksheets may be
submitted with evaluation reports provided the necessary information can be readily obtained from such
worksheets and the information is referenced on the activity checklist.
The worksheets provided in this manual are designed to allow some flexibility in methods of calculation,
since no two activity evaluations are exactly the same.
Summary of the evaluation results will be reflected in the appraisal summary table (AST) and much of the
information required in AST and worksheets contributes to the Waka Kotahi funding allocation process.
The expectation is that the data entered in AST and worksheets can be transferred to Transport
Investment Online (TIO) and vice versa as appropriate.
Blank worksheets can be downloaded in MS Excel format from the MBCM page on the Waka Kotahi
website.
The provided templates must be used when using the simplified procedures. The completed templates
should be attached in TIO. The templates are standardised to allow automated uploading and data
extraction.
4.2 Evaluation of walking and cycling activities

This section describes the specific procedures to be used to evaluate the economic efficiency of walking
and cycling facilities. Activities may be stand-alone interventions, or a component of a wider transport
solution.
Improvements may be of two types:
• route improvements (eg the provision of new or improved paths, lanes or other facilities for
pedestrians or cyclists), or
• improvements at hazardous sites (eg the provision of overbridges, underpasses, bridge widening
or intersection improvements).
Cycling and walking promotion is addressed as part of the evaluation procedures for education,
promotion and marketing in section 4.6.
Activities that involve mode change need to be careful not to claim unrealistic walking distances.
Statistics on walking used in this manual are based on the New Zealand Travel Survey. The average
pedestrian trip length is estimated at 1km.
Statistics on cycling provided are based on the New Zealand Household Travel Survey. The current
average cycle trip length is estimated at 3km. This applies equally to new and existing users.
Integration with other travel demand management initiatives

For walking and cycling activities to be effective, continuous lanes or paths should be provided. Provision
should also be made for secure cycle parking, signage, maps, education, promotion, marketing and
integration of the routes with public transport. All these components should be addressed within a walking
and cycling section of a wider transport strategy that includes an implementation package. Useful sources
include:
• Barnes et al (2005), which sets out a framework and priorities for development of walking and
cycling activities
• Waka Kotahi Cycling network guidance, which provides guidance for cycle network and route
planning
• Waka Kotahi Pedestrian planning and design guide, which is New Zealand's comprehensive
official guide to planning and design for walking. It sets out ways to improve New Zealand’s
walking environment.
Because of synergetic impacts, evaluation of walking and cycling should be done at the package level
rather than just for individual components.
Simplified procedure for walking and cycling activities

The simplified procedure is designed to consider one option at a time. All suitable options for the
proposed activity should be considered in order to select the optimal solution. In most situations this will
involve incremental analysis of the benefits and costs of the different options analysed. In particular,
where a separate dedicated cycleway is proposed, the alternative option of providing wider sealed
shoulders or cycle lanes on the carriageway must be considered.
SP11 for walking and cycling activities
Simplified procedure SP11 provides the appraisal methods for walking and cycling initiatives with an
undiscounted whole-of-life cost of less than $15 million. SP11 assumes that the activity does not include
signalised crossings over roads.
A description of all options considered should be included in SP11 worksheet 1 (SP11-1) and the options
should be tested in the incremental analysis worksheet SP11-8. The worksheets for all the options must
be submitted together with a summary of the incremental analysis.
To use the worksheets, it is necessary to determine the current number of pedestrians and cyclists and
estimate their future growth rate. These must be based on local counts and realistic projections. For
cyclists these can be obtained using SP11-7.
The simplified procedure may be used as part of a multi-modal evaluation. Any such evaluation could
cover travel behaviour change (TBhC) activities, and infrastructure and public transport service
improvements. The procedure uses a 4% discount rate and a 40-year analysis period.
The procedure assumes that activities will be completed in the first year and will be in service by the start
of year two. Where costs are common to the do-minimum and the options, they are not included in the
analysis. All costs are to be exclusive of GST.
Guidance for completing the SP11 Walking and cycling facilities (template worksheets) is provided below
in Table 48 and Table 49.
Table 48: SP11 Walking and cycling facilities procedure template

Worksheet Worksheet purpose Description
SP11-1 Evaluation summary Provides a summary of the general data used for the
evaluation as well as the results of the analysis.
SP11-2 Cost of the do-minimum Used to calculate the PV cost of the do-minimum. The
do-minimum is the minimum level of expenditure
necessary to maintain a minimum acceptable level of
service and generally consists of maintenance work.
SP11-3 Costs of the option Used to calculate the PV costs of the option 1. A
separate worksheet is required for each option
evaluated. Up to 3 options in addition to do-minimum
can be evaluated.
SP11-4 Travel time cost savings Used for calculating travel time cost savings.
SP11-5 Benefits for walking and Used for calculating the health and environmental
cycling facilities benefits of walking and cycling facilities.
SP11-6 Crash cost savings Used for calculating crash cost savings using crash-by-
crash analysis method (method A in Appendix 2: Crash
analysis).
SP11-7 Cycle demand Used for calculating cycle travel demand using the
‘buffers’ methodology
SP11-8 BCR and incremental Used for relative comparison of the options.
analysis
Table 49: Steps in the SP11 evaluation of walking and cycling activities
Step Description
1 Complete items 1 to 3 of Worksheet 1 – Evaluation summary
2 Complete Worksheet 2 – Cost of do-minimum
3 Complete Worksheet 3 – Cost of option(s)
4 Complete Worksheets 4 to 7 for the option(s) being evaluated
5 Complete Worksheet 8 – Incremental analysis (if more than one option is considered)
6 Select the preferred option and finalise Worksheet 1 for the preferred option
Table 50 provides the required benefits factor for different types of cycle facilities.
Table 50: Benefit factors for different types of cycle facilities

Type of cycle facility Relative attractiveness (RA)
On-street with parking, no marked cycle lane 1.0
On-street with parking, marked cycle lane 1.8
On-street without parking, marked cycle lane 1.9
Off-street cycle path 2.0
Full procedure for walking and cycling activities

In cases where the above criteria are not appropriate, the full procedures should be used.
The following table outlines the stages of analysis in the economic efficiency evaluation. The do-minimum
and any other options must be assessed at every stage.
Table 51: Full procedure for walking and cycling activities

Stage Description Refer
1 Complete the activity description including a description of the do- Current section and
minimum, alternatives and options. Section 1.4:
Counterfactuals
Section 1.5:
Alternatives and
options
2 Forecast the demand. Current section
Note: The demand estimate is used for calculating user benefits for new
and existing pedestrians and cyclists, and road traffic reduction benefits.
Care should be taken to ensure assumptions are compatible with
economic evaluation requirements.
3 Calculate the benefits. Current section and
Chapter 3: Benefits
4 Calculate the costs of the proposal. Section 1.8: Costs
5 Discount the monetised benefits and costs over the analysis period to Chapter 5:
obtain present values. Discounting
6 Determine the benefit–cost ratios of the options. Chapter 6: Benefit–
cost ratios
7 Use incremental cost–benefit analysis to select the preferred option for Chapter 6: Benefit–
mutually exclusive options. cost ratios
8 Perform sensitivity tests on the preferred option to determine how robust Chapter 7: Sensitivity
the calculations are and whether a small change in one of the input and risk analysis
parameters has a large change on the evaluation outcome(s).
9 Verify completeness of information, accuracy of calculations and validity Current section
of assumptions.
Stage 1: Describe the do-minimum, options and alternatives

The do-minimum for evaluation of walking and cycling facilities is usually considered as a continuation of
the present transport networks, service levels and facilities in the study area.
The do-minimum shall include any costs and resulting demand implications of committed facility or
service improvements. All committed investment plans that relate to the do-minimum during the analysis
period must be taken into account. Maintenance, replacement and renewal schedules and any planned
service changes must also be included. Improvements are committed if they have been assessed in
accordance with the Waka Kotahi assessment procedures and have been approved for funding.
Any investment plans that are not committed must be included in the evaluation as options.
Where a particular benefit or cost is unchanged among all the alternatives, options and the do-minimum,
it does not require validation or inclusion in the economic analysis.
It is extremely important to:
• not overstate the scope of the do-minimum, and

• only include, as part of the do-minimum, committed and funded transport activities and work
which will preserve a minimum acceptable level of service.
Stage 2: Demand estimates

Evaluators are required to make realistic estimates of the demand for a new or improved walking or
cycling facility, particularly the number of new pedestrians or cyclists.
Factors influencing the demand for walking and cycling include:
• the availability of facilities
• the type and quality of facility, including cycle parking, signage and safety of use
• the location of the facility, the route length, and connectivity of walking and cycling paths or lanes
• the population served by the facilities, and
• any education, promotion and marketing.
Studies have shown a positive correlation between the number and quality of facilities provided and the
percentage of people who use cycle for transportation purposes. It has also been observed that, in
addition to having walking and cycling facilities, they must connect appropriate origins and destinations,
and use of the facilities must be promoted to encourage walking and cycling as alternative commuting
modes.
Education, promotion and marketing are significant drivers for generating demand for walking and cycling,
and any associated mode shift from private vehicles. The methodology for estimating travel impacts in
section 2.2 of this manual should be referenced to estimate the number of private vehicle trips diverted to
new or improved walking and cycling facilities, when this is part of a package that includes travel
behaviour change (TBhC) activities.
Beetham J et al has summarised the methods for walking and cycling demand estimates in five groups,
including a decision tree for determining the right method for demand forecasting based on the size of the
transport proposal and the scale of the change (see Figure 9 below).
Figure 9: Walking and cycling demand methods and selection matrix
Method 1 Method 2 Method 3 Method 4 Method 5

Cycling: Sketch Cycling: Sketch plan Comparison Geospatial Transport model
splan equation equation with approach assessment with locally specific
Walking: Informed informed expert combined with an combined with an data and factors.
expert estimation. calibration. evaluation of level evaluation of level
of service and of service and
Walking: Informed
potential use. potential use.
expert estimation.
Employ modifying Employ modifying
factors based on factors based on
local context. local context.
Where a new or improved walking or cycling facility provides an improved quality of service, then trips in
addition to those diverted from private vehicles may be generated depending on the extent of the quality
change of the improvement.
Table 52 (which is part of simplified procedure SP11) provides a cycle demand model for estimating cycle
trips generated by a new or improved cycling facility.4 This method is based on the number of jobs in
each buffer zone. It includes variables for the quality of service (QoS) of a facility and (optionally) the
directness of a proposed project compared to the most direct route possible. This model falls in the
category of geospatial analysis methods (3 and 4) in Figure 9. Table 52 can be used to estimate the
demand for a new cycle facility if cycle counts are not available or the counts are considered unreliable.
The total demand for a cycle facility may be estimated by referencing the procedures in sections 2.2, 2.3
and 2.4 of this manual.
Table 52: Cycle demand

New and existing cyclists
Buffers (km) <0.4 0.4 to <0.8 0.8 to ≤1.6
1 Jobs in each buffer (solid shape)
2 Jobs in each buffer (annular shape)
3 Likelihood of new cyclist multiplier 1.04 0.54 0.21
4 Weighted jobs [sum of (2) x (3)]
5 Average quality of service (QoS) score
from Auckland Transport QoS method
6 Transformed QoS average score (6) = (4)
– (5)
7 Directness
8 Actual cyclists before intervention –
predicted cyclists before intervention. If (7)
is omitted, leave this blank.
9 Total new and existing cyclists
10 Experienced evaluator judgement of
additional cyclists not modelled
11 Predicted total cyclists after intervention (9)
+ (10)
Table 52 calculates number of jobs within catchments surrounding the facility. It then applies a probability
factor to estimate the number of new cyclists who will use the facility by considering their distance from
the facility and the current mode share of cyclists.
The buffer distances are defined as <0.4km, 0.4km to <0.8km, and 0.8km to ≤1.6km. These represent
the area from the facility which is likely to be affected by the proposal. When calculating the area of each
buffer, the areas of buffers between it and the facility need to be excluded. Therefore, the table only
requires row 1 or 2 to be filled out.
The likelihood multiplier is an adjustment for the likelihood of new cyclists using the facility in each
buffer. Cyclists further from the facility are less likely to use it.
The quality of service (QoS) score is computed using the Auckland Transport QoS method, which is
available on their website.
4The full model and its underlying assumptions are available from the Waka Kotahi Planning and Investment
Knowledge Base (PIKB).
Calculate segment, intersection, and overall average 5 QoS for the proposed project. If the facility is 3.0m
wide, then you can select from two width bands in the QoS tool. Choose the higher (better) score if the
facility is consistently >= 3.0m and/or high-quality, and the lower score if there are pinch points.
Directness is an optional input. Divide the length of the shortest route by the subject route. If there is no
alternative shorter route or the data is not available, then this may be left blank. Table 52 applies the
correct model formula.
If the average daily cyclists before intervention is known, run the model using the existing conditions and
then subtract the known existing average daily riders from the total predicted daily riders to get the
number of new daily riders. If you don't have an existing count or you don't calculate directness, then
leave this value set to zero.
Table 52 enables the analyst to incorporate any additional daily cyclists expected due to other route-
specific conditions not accounted for in the model, for example a large school with high cycling mode
share. The analyst can use bike shed counts (doubled, to account for round-trip travel). Any value
entered must be justified with supporting documentation.
Table 53: Cycling commute share – has been deleted
Back to 2.2 Forecasting demand: procedures for travel behaviour change activities >>
Stage 3: Calculate walking and cycling improvements benefits

The consumer surplus methodology is used to monetise the user benefits of improvements to walking
and cycling facilities.
This manual includes procedures to monetise some of the benefits of walking and cycling projects.
The benefits of walking and cycling monetised in this manual include:
• impact on social cost and incidents of crashes

• impact on physical and mental6 health
• impact on productivity and network utilisation
• wider economic impact (productivity)
• wider economic impact (regional economic development), and
• wider economic impact (land use change).
These benefits arise primarily as a result of changes in road traffic volumes, changes to mode share, and
changes to user benefits from decreased travel time and increased safety.
Cycle operating costs and walking costs are assumed to be included in the perceived costs of changing
to and using these modes.
Activities that combine walking and cycling may claim benefits for both modes, but safety issues arising
from pedestrian or cycle conflicts must be addressed. If there are likely to be additional crashes these
must be accounted for in the evaluation.
Disruption costs to existing users of walking and cycling facilities during the implementation of new or
improved facilities must be included in the evaluation as a disbenefit (negative benefit).
Possible disbenefits include:
• increased travel time, and

• travel discomfort.
Calculate net user benefits for users of a new walking and cycling facility using following procedure to
determine the projected number of new service users. The calculation of net benefits for users of a new
5 Auckland Transport recommends that an average is not used because a single poor-quality segment can render the
rest of the corridor quality a moot point for a prospective rider. However, as long as the analyst clearly acknowledges
such potential failure points, the averaging method has shown to be statistically sound for the modelling purpose.
6 Impact on mental health is not covered in the manual.
walking or cycling facility is based on the maximum benefit value to a potential user. The result is then
divided in half, based on the rule of half.
Net user benefits = Pmax × Qnew × ½
where: Pmax is the willingness to pay value for the new facility
Qnew is the projected number of new users
The value of walking or cycling facility user benefits, other than time saving benefits, such as improved
quality, comfort or security, is usually based on a willingness to pay value derived from a stated
preference (SP) survey or from values derived for similar facility improvements in other areas.
Calculate the facility user benefits for users of an existing walking and cycling facility with a projected
new use level, as follows:
1. Change in net total benefits for existing users

Bfexisting = Pmax × Q1
2. Apply the rule of half for the total benefits for new users
Bfnew = Pmax × (Q2 – Q1) × ½
3. Total facility benefits

Bftotal = Bfexisting + Bfnew
where: Pmax is the willingness to pay value for the improvement of the
facility
Q1 is the existing number of users
Q2 is the projected total number of users
Proposals to improve the quality of an existing walking and cycling facility, such as improving its amenity,
comfort or security, must assess the value of different levels of quality. This assessment must also be
carried out if the proposal is for a new walking and cycling facility. The valuation should be based on a SP
survey or on information from similar improvements to facilities in other areas.
Land Transport NZ (2006), Pedestrian planning and design guide describes an SP methodology and
study to identify preferences for different types of cycling facilities. The study determined the additional
time that cyclists would spend travelling on each type of facility, and the incremental attractiveness of that
type of facility, when compared with a base case of 20 minutes of travel in traffic with road-side parking.
The study is the basis of the values in Table 35.
Stage 4: Calculate costs of walking and cycling do-minimum and options
In general, the costs of walking and cycling activities are limited to:
• planning, investigation and design fees

• costs of property required for the activity
• construction costs
• maintenance and renewal costs, including repair and reinstatement
• facility operating costs.
Stage 5: Discount benefits and costs
Refer to section 1.9 and Chapter 5 of this manual for the detailed information on undertaking discounting.
Benefits and costs generally arise throughout the life of projects, and to calculate their present worth or
present value they need to be discounted back to time zero. Based on a discount rate of 4% and an
analysis period of 40 years, sets of present-worth factors have been calculated to convert future benefits
and costs to their present values. (See Table 102, and Table A132, Table A133 and Table A134 from
Appendix 6: Discount factors. Discount rates of 3% and 6% are also provided in the tables for sensitivity
testing.)
Stage 6: Determine the benefit–cost ratios of the options
Refer to Chapter 6 for detailed information on developing BCRs.
4. Evaluation procedures > 4.3 Evaluation of road renewal and improvement activities
Stage 7: Incremental cost–benefit analysis

Where alternatives and options are mutually exclusive, incremental cost–benefit analysis of the
alternatives and options is used to identify the optimal economic solution.
The incremental BCR indicates whether the incremental cost of higher-cost project alternatives and
options is justified by the incremental benefits gained (all other factors being equal). Conversely,
incremental analysis will identify whether a lower-cost alternative or option that realises proportionally
more benefits is a more optimal solution.
Refer to section 6.3 for detailed information on developing incremental BCRs.
Stage 8: Perform sensitivity and risk analysis
Refer to Chapter 7 of this manual for detailed information on sensitivity and risk analysis.
Assessing the sensitivity of impact evaluations and resulting benefits calculations to critical assumptions
or estimates shall be undertaken using sensitivity testing, which needs to be undertaken for the critical
inputs and assumptions used to choose the preferred option.
Sensitivity testing involves defining a range of potential values for an uncertain variable in evaluation and
reviewing the variation in the evaluation as the variable changes within the range. This will highlight the
sensitivity of the estimated final outcome to changes in input variables.
Inputs to walking and cycling facility evaluations that should be considered for sensitivity testing include:
• demand estimates, and
• major contributors to benefits.
Benefits critical to the outcome of the evaluation may include:
• pedestrian and cyclist volumes particularly those derived from model results
• growth rates and assessments of diverted and generated traffic, and
• crash reductions.
For each significant factor the following must be listed:
• the assumptions and estimates on which the evaluation has been based
• an upper and lower bound of the range of the estimate, and
• the resultant BCR at the upper and lower bound of each estimate.
The results of the sensitivity tests, along with an explanation of any assumptions or choice of test, must
be reported.
Stage 9: Verification of results
Verify completeness of information, accuracy of calculations and validity of assumptions.
4.3 Evaluation of road renewal and improvement activities

The following section describes the procedures that are to be used to evaluate the economic efficiency of
road improvement activities. Activities may be stand-alone improvements or a component of a package or
a wider programme of transport improvements.
For the more simple and relatively standardised improvement activities with an undiscounted whole-of-life
cost of $15 million or less, simplified procedures are provided for the analysis and these are explained
below. For the more complicated projects, and those with an undiscounted whole-of-life cost greater than
$15 million, the full procedures are provided as an alternative to the simplified procedures and are
explained later in this section.
Simplified procedures for road renewal and improvement activities

The following simplified procedures (SPs) for road improvement and related activities use a 4% discount
rate and a 40-year analysis period, and assume that activities will be completed in the first year and will
be in service by the start of year two. Where costs are common to both the do-minimum and the options
they are not included in the analysis. All costs are to be exclusive of GST. The simplified procedures SP1
to SP5 are for the evaluation of road activities that have an undiscounted whole-of-life cost of less than or
equal to $15 million. If any of these criteria are not met then the full procedures (see below) must be
used.
The simplified procedures are designed to consider one option at a time. All suitable options for the
proposed works should be considered in order to select the optimal solution. In most situations this will
involve incremental analysis of the benefits and costs of the different options analysed. A description of all
options considered should be described in worksheet 1 and included in the incremental analysis; for all
other worksheets, only the details of the preferred option need to be included.
It is necessary for all the activities covered by SP1 to SP5 to determine the expected future traffic growth
rate. This can be done either by analysing the traffic count data, following the procedures in Appendix 3:
Traffic data and travel time estimation, for at least the last five years and preferably for the last 10 years,
or by using a default growth rate of zero percent. Simplified procedures SP1 and SP2 are for road
renewals and bridge renewals respectively. These renewal activities are a type of improvement when
compared to the do-nothing or do-minimum, but they are targeted at maintaining the status quo.
Procedure SP3 is for general road improvements, while SP4 is for seal extension works, and SP5 is for
isolated intersection improvements. If an intersection improvement is part of an overall corridor
improvement or is being undertaken with other road improvement works, then it should be considered as
part of a package or programme of works.
Refer to the Waka Kotahi Planning and Investment Knowledge Base (PIKB) for guidance on issues
relating to analysis of road activities, including selection of the preferred option using the Business Case
Approach.
The simplified procedure templates provided must be used when undertaking simplified evaluations. The
completed templates are to be included in Transport Investment Online (TIO). The templates are
standardised to allow automated uploading to and data extraction from TIO.
Each simplified procedure is a stand-alone procedure designed to be applied directly to each option being
considered. Input values may be obtained from:
• the default figures provided
• the information in the appendices.
Analysis that alters components of the simplified procedure should not be used as this will compromise
the assumptions on which the procedures are based and full procedures should be used instead.
If the analyst has any problems with the simplified procedures templates or worksheets, please contact
[email protected].
SP1 for road renewal activities
SP1 provides a simplified method of appraising the economic efficiency of work to be funded under work
categories within the maintenance activity classes, for example pavement rehabilitation.
To be considered eligible for funding under these work categories, the activity must be shown to be the
long-term, least-cost option for the road controlling authority, and must not include geometric
improvements. SP 1 therefore stands apart from all other simplified procedures by solely comparing the
whole-of-life costs of each option and excluding any calculation of benefits.
Under these procedures the present-value cost of the option is determined and compared with the
existing maintenance strategy. An existing maintenance strategy commonly includes pavement
maintenance work such as dig-outs, reseals, and/or other localised repairs needed to ‘hold’ the condition
of an asset.
Guidance for completing the SP1 Road renewals (template worksheets) is provided below in Table 54
and Table 55.
Table 54: SP1 Road renewals procedure template

number
SP1-1 Evaluation summary Used to summarise the general data considered for the
evaluation plus the results of the economic analysis.
SP1-2 Cost of existing maintenance Used to calculate the PV cost of the existing
strategy maintenance strategy.
SP1-3 Cost of the option(s) Used to calculate the PV cost of the option.
Table 55: Steps in the SP1 evaluation of road renewal activities
Step Description
2 Complete Worksheet 2 – Existing maintenance strategy
Select preferred option – refer to work category 214: Sealed road pavement
4
rehabilitation on PIKB
SP2 for bridge renewal activities

SP2 provides a simplified method for appraising the economic efficiency of replacing a bridge for
structural reasons. The benefits analysis focuses on the change in heavy commercial vehicle (HCV)
users’ costs as a result of the activity. Guidance on the application of these procedures is found in the
decision chart in Figure 10.
If road improvements are being considered in conjunction with the bridge renewal, then the improvements
are to be evaluated separately (using SP3, if applicable), when it is confirmed that bridge renewal is the
preferred option.
The procedure for analysing structural bridge renewals is somewhat different from other activities, in that
all options are identified and costed at the outset, including:
• cost of replacement bridge

• average daily traffic
• viability and cost of a concrete ford
• the HCV users of the bridge
• existence of an alternative route, its length and any necessary upgrade costs
• the cost to repair the bridge to a posted limit of 10 tonnes
• revocation costs
• demolition/deconstruction costs.
Once this has been done, the decision chart (Figure 10) can be used to determine the appropriate course
of action and analysis procedure.
Exception to the standard do-minimum

The do-minimum for most road activities shall only include work that is absolutely essential to preserve a
minimum level of service. However, if a bridge serves little traffic and is expensive to replace, a
replacement option should not automatically be taken as the do-minimum, particularly if alternative routes
are available to traffic presently using the bridge. In this case the do-minimum may be to not replace the
existing bridge and to have no bridge. If it is unacceptable to have no bridge at all, then another possible
do-minimum could be rehabilitating the existing bridge. The do-minimum must be clearly determined for
each bridge renewal under consideration.
Note: This procedure does not allow for the possibility of total bridge failure. If this is a real possibility
when certain options are chosen, then account should be taken of the extra costs this would impose on
road users multiplied by the probability of failure occurring. The calculation of these probabilities should
be undertaken by the same engineers who make the decisions regarding posting the bridge.
Figure 10: Decision chart for bridge replacement
Guidance for completing the SP2 Structural bridge renewals (template worksheets) is provided below in
Table 56 and Table 57.
Table 56: SP2 Structural bridge renewals procedure template

number
SP2-1 Building a ford on a low- Used to summarise the general data considered for the
volume road evaluation plus the results of the economic analysis.
SP2-2 Evaluation summary for bridge Used to calculate the PV cost of the do-minimum. The
renewal do-minimum is the minimum level of expenditure
necessary to keep a road open.
SP2-3 Costs of the option(s) Used to calculate the PV costs of the option. A separate
worksheet is required for each option evaluated. Up to
three options in addition to do-minimum can be
evaluated.
SP2-4 HCV user costs when there is Used for calculating travel time cost savings.
an alternative route
SP2-5 HCV user costs when there is Used for calculating vehicle operating cost (VOC)
no alternative route savings.
SP2-6 BCR and incremental analysis Used for comparison of the options considered.
Table 57: Steps in the SP2 evaluation of structural bridge renewal activities
Step Description
1 Complete Worksheet 1 if building a ford is an option – if it is not an option leave blank
3 Complete Worksheet 3 – Cost of option(s) and determine which option is do-minimum
4 Complete Worksheet 4 – HCV user costs – when there is an alternative route
5 Complete Worksheet 5 – HCV user costs – when there is no alternative route
6 Complete Worksheet 6 – Incremental analysis (if more than one option is possible)
7 Select the preferred option and finalise Worksheet 2 for the preferred option*
Table 58 provides freight cost factors for use within this simplified procedure
Table 58: Cost factors for different type of heavy trucks

% Class I HCVI HCVII
100 1.00 1.00
90 1.18 1.22
80 1.44 1.57
70 1.85 2.22
60 2.60 3.67
50 4.33 11.00
SP3 for road improvement activities

SP3 provides a simplified method of appraising the economic efficiency of road improvements, including
road reconstruction, seal widening, new roads and new structures. SP3 specifically excludes road
renewals (SP1), bridge renewals (SP2), seal extension work (SP4), and isolated intersection
improvements (SP5).
Guidance for completing the SP3 Road improvement activities (template worksheets) is provided below in
Table 59 and Table 60.
Table 59: SP3 Road improvement activities procedure template

number
SP3-2 Cost of do-minimum Used to calculate the PV cost of the do-minimum. The
SP3-3 Cost of the option(s) Used to calculate the PV costs of the option. A separate
three options in addition to do-minimum can be
evaluated.
SP3-5 Vehicle operating cost Used for calculating vehicle operating cost (VOC)
savings savings.
crash analysis method (refer Appendix 2: Crash
analysis).
Table 60: Steps in the SP3 evaluation of road improvement activities

Step Description
SP4 for seal extension activities

SP4 provides a simplified method of appraising the economic efficiency of proposed seal extension
works. The method is for the evaluation of activities that have an undiscounted whole-of-life cost less than
or equal to $15 million.
The procedures are designed to consider one option at a time. All suitable options for the proposed works
should be considered in order to select the optimal solution. In most situations this will involve incremental
analysis of the benefits and costs of the different options analysed. A description of all options considered
should be provided in worksheet SP4-1 and included in the incremental analysis; for all other worksheets,
only the details for the preferred option need to be included.
Guidance for completing the SP4 Seal extensions (template worksheets) is provided below in Table 61
and Table 62.
Table 61: SP4 Seal extensions procedure template

number
two options in addition to do-minimum can be
evaluated.
savings savings.
analysis).
Table 62: Steps in the SP4 evaluation of seal extension activities

Step Description
SP5 for isolated intersection activities

SP5 provides a simplified method of appraising the economic efficiency of isolated intersection
improvements. Crash analysis involving an isolated intersection is only to be undertaken where the site
has a crash history of:
• four or more non-injury crashes

• one injury and three or more non-injury crashes, or
• two or more injury crashes.
The most recent five calendar year crash history for the site should be used. Detailed crash listings,
collision diagrams, a description of common factors in the crashes and a diagnosis of the site factors
contributing to the problem should be submitted with the evaluation.
An intersection that does not meet the above criteria may still have a crash analysis carried out using
predictive crash models. In such a case, SP5 does not apply and full procedures must be used.
Guidance for completing the SP5 Isolated intersection improvements (template worksheets) is provided
below in Table 63 and Table 64.
Table 63: SP5 Isolated intersection improvements procedure template

number
worksheet is required for each option evaluated. Up to 4
options in addition to do-minimum can be evaluated.
savings savings.
analysis).
Table 64: Steps in the SP5 evaluation of isolated intersection activities

Step Description
Table 65 provides default values for use within this simplified procedure.
Table 65: Multiplication factors for items with an estimated life of less than 40 years
Construction item Multiplying
factor (MF)
Traffic signs 2.5
Delineation (eg edge market posts, raised pavement markers, sight railing and chevrons) 3.7
Spray plastic 5.7
Road markings 15.5
Full procedures for road improvement activities

The full evaluation procedures for road improvement activities are to be used to appraise the economic
efficiency of activities when the simplified procedures are not appropriate or sufficient.
There are many types of improvements that can be considered when using these procedures, including
specialised improvements such as passing lanes, bridge improvements, etc.
The primary purpose of this section of the manual is to establish the impacts of making road
improvements (ie the changes that occur between the do-minimum and the options) when using the full
procedures. Following on from calculating the impacts, the analyst will need to assign monetary values to
the impacts and then calculate the benefits and the benefit–cost ratios (BCRs).
These procedures cover the range of stages listed above, however, many of the actions for these stages
are covered in greater detail in other sections or appendices of this manual and in external documents for
which links have been provided. A significant focus of the road improvement procedures is on the
calculation of activity impacts, in particular stages 4 to 6 in Table 66.
These procedures are designed to calculate the impacts one at a time and then, after assigning monetary
values to the impacts, they can be added together, including any disbenefits, to establish the total benefit
of the options under consideration. To assist in this process a set of standardised worksheets have been
developed to help guide the analyst through an evaluation and to aid in the process of checking for
completeness and accuracy.
The following table outlines the stages of analysis when undertaking an evaluation of the impacts of road
improvements. The chapters and sections of this manual that apply to each stage of the analysis are
referenced in the table below.
Table 66: Full procedures for evaluation of road improvement activities

1 Consider and describe: Section 1.4:
Counterfactuals
a. the do-minimum
b. improvement alternatives and options Section 1.5:
c. whether the improvement(s) should be part of a package Alternatives and
and/or programme of activities. options
2 Assemble information on the transport route (lengths, road Chapter 2:
classifications, etc) together with current and forecast vehicle, cycle Demand estimation
and pedestrian traffic demand (including suppressed demand and and mode share
induced traffic).
3 Consider the use of transport models and calibration. If a transport Transport model
model is being used, then undertake calibration checks for the development
improvement options as required. guidelines
4 Measure and monetise the impacts (benefits and disbenefits) for the Current section
do-minimum and options, including: and Chapter 3:
Benefits
• impact on social cost and incidence of crashes
• impact of mode on physical and mental health
• impact of air emissions on health
• impact on greenhouse gas emissions
• impact of noise and vibration on health
• impact on network productivity and utilisation
• impact on system reliability
• impact on user experience of the transport system
• wider economic benefits
• other benefits that can be monetised – these are not included
in this manual but can be included if there is sufficient
supporting evidence and the approach is accepted by Waka
Kotahi.
5 Describe and evaluate any mitigation measures necessary for the Section 1.8: Costs
options under consideration.
6 Undertake risk analysis when there are significant unpredictable events Chapter 7:
that may affect or be affected by the improvement activity. Sensitivity and risk
analysis
7 Calculate the costs for the do-minimum and improvement options, Section 1.8: Costs
including (but not exclusively):
• investigation and design
• property
• construction, including preconstruction and supervision
• maintenance, renewal and operation
• risk management
• mitigation of external impacts

• residual values.
obtain present values. Discounting
cost ratios
10 Use incremental cost–benefit analysis to select the preferred option for Chapter 6: Benefit–
mutually exclusive options. cost ratios
11 Perform sensitivity tests on the preferred option to determine how Chapter 7:
robust the calculations are and whether a small change in one of the Sensitivity and risk
input parameters has a large change on the evaluation outcome(s). analysis
12 Verify completeness of information, accuracy of calculations and Current section
validity of assumptions.
Stage 1a: Describe the do-minimum

Generally, the do-minimum for road activities shall only include committed and funded transport activities
and work that is absolutely essential to preserve a minimum level of service. However, in some cases, as
described below, the do-minimum may need to be specified differently. It is important that the do-
minimum is fully described in the evaluation.
For some activities on low volume roads, the existing level of maintenance expenditure may not be the
do-minimum. In such cases, particularly where the existing level of maintenance expenditure is high, the
maintenance expenditure shall be justified as an option along with other improvement options, and the
do-minimum shall only be the work necessary to keep the road open.
Similarly, if a bridge serves little traffic and is expensive to replace, a replacement option should not
automatically be taken as the do-minimum, particularly if alternative routes are available to traffic
presently using the bridge. In this case the do-minimum may be to not replace the existing bridge and to
have no bridge. If it is unacceptable to have no bridge at all, then another possible do-minimum could be
rehabilitating the existing bridge.
The do-minimum generally should not include pavement rehabilitation to an improved standard. The only
exception is when the present value of the cost of the activity and its future maintenance is less than that
of continued maintenance of the existing situation.
For example, on steep unsealed roads, which need frequent grading, to remove corrugations the
continued maintenance of the unsealed road can be more costly than sealing the road. In such a situation
it is possible that sealing the road may be the do-minimum, so long as it is the lowest-cost option
available (eg there is not a realignment option available that is even cheaper).
Most forms of activity evaluation involve choices between different options or courses of action. In theory,
every option should be compared with the option of doing nothing at all, ie the do-nothing.
For many transport activities, it is often not practical to do-nothing. A certain minimum level of expenditure
or activity may be required to maintain a minimum level of service. This minimum level of expenditure or
activity and the resultant performance is known as the do-minimum and should be used as the basis for
evaluation, rather than the do-nothing. It is important not to overstate the scope of the do-minimum.
In some situations, the do-minimum can be the most effective solution to a problem and therefore it can
be the ’preferred option’.
For some situations the best outcomes may be delivered through the do-minimum option, eg lowering the
operating speed to a safe and appropriate level through the use of speed-limit signs and/or minor
infrastructure improvements that go with the new speed limits. In this case, the do-minimum will be the
preferred option.
For safety activities where reducing the speed limit is a potential option, the do-nothing scenario is the
existing baseline conditions of the network, based on the existing speed limit, operating speed,
infrastructure and services.
Where a road-controlling authority decides to introduce one or more interventions to address
unacceptable levels of collective and/or personal risk, to re-set the speed limit, and/or to manage speeds
on a particular piece of road, the do-minimum can include benefits and costs of implementing a new safe
and appropriate operating speed.
In such situations the do-minimum should be compared to both the do-nothing and the other activity
options in order to determine whether the do-minimum is the preferred option (ie the optimal solution), or
whether additional improvements are justified over and above the do-minimum, and if these additional
improvements are therefore the preferred option.
When undertaking safety interventions addressing speed the following information should be referenced:
• Land Transport Rule: Setting of Speed Limits 2017
• the Waka Kotahi Speed management guide, and
• the Waka Kotahi MegaMaps tool, which is used in conjunction with the Speed management
guide.
The crash costs associated with speed management interventions should be calculated using the
predictive crash cost models in Appendix 2: Crash analysis of this manual.
In cases where the do-minimum involves a large future expenditure, the option of undertaking the activity
now should be compared to the option of deferring the activity until this expenditure is due. Similarly, if the
capital cost of the activity is expected to increase for some reason other than normal inflation, again the
option of undertaking the activity now should be compared with the option of deferring construction and
incurring the higher cost.
The activity costs required for determining benefit–cost ratios (BCRs), including incremental benefit–cost
assessment (Chapter 6), and also first-year rate of return (section 1.10) is the difference between the
costs of the activity option and the costs of the do-minimum. The activity benefits are similarly the
differences between the benefit values calculated for the activity option and those of the do-minimum.
It follows that where a particular benefit or cost is unchanged among all the activity options and the do-
minimum, it does not require valuation or inclusion in the economic analysis. For completeness, it should
be noted in any funding application that the benefit or cost is unchanged.
Stage 1b: Describe the alternatives and options
Rigorous consideration of alternatives and options is a requirement of the Land Transport Management
Act 2003 (LTMA). To ensure these obligations are met, evaluators should carefully articulate the problem
or issue that they are seeking to resolve and avoid approaching the analysis with a preconceived solution
in mind.
Alternatives are different means of achieving the same objective as a proposed activity, while options are
variants of a proposed activity. These alternatives and options should not be constrained to a specific
mode, or even to transport solutions, as changes to existing policy may be suitable responses to the
identified problem. As a result, it may be necessary to apply other procedures contained within this
manual as part of the evaluation.
Stage 1c: Consider if an activity is stand-alone, part of a package or part of a programme
Waka Kotahi seeks to encourage, where appropriate, approved organisations to develop packages or
programmes of interrelated and complementary activities, either individually or in association with other
approved organisations.
This is particularly important to ensure that a wide range of options and alternatives are considered and
evaluated in full. Doing so may help avoid issues that arise from narrowing the scope too early such as:
• neglecting options that differ in type or scale, eg a road realignment that may eliminate a bridge
renewal
• neglecting significant externalities, eg the impacts of change in traffic flow upon adjoining
properties
• inconsistencies with wider strategic policies and plans, eg the impacts of improvements to a
major urban arterial on downtown congestion.
Stage 2: Route and network information
Road improvement activities need to be divided into sections with similar geometric and traffic flow
characteristics, and with similar costs of construction and maintenance. In some cases it may be
necessary to separately consider individual traffic movements at intersections. In other cases, benefits
and disbenefits may differ by direction of travel, for example on continuous sections of grade, and in
these cases it will be necessary to consider each direction as a separate section.
For the do-minimum and for each activity option, the route should be divided into sections over which the
terrain, road width, road roughness, speed limit and traffic volumes (for all modes) are essentially
constant, and/or intersections.
For minor activities and for pre-selection studies, all time periods can be considered together. For
significant capital activities, it will be necessary to consider traffic variation with time of day and weekday
versus weekend and holiday periods.
Activity location and layout

Information on location and layout to be provided shall include:
• a location/route map
• a map showing linked activities and/or strategic routes
• a layout plan of the activity.
As is appropriate to the particular activity, the layout plan shall show:
• section end points by name, physical features, including the start and end points of the activity
• intersection approaches and traffic movements
• identifying numbers for each road section, intersection approach and traffic movements
• road section lengths, average gradient and surface type
• speeds, if road sections are determined by speed changes
• locations of traffic survey points
• traffic volumes of intersection movements.
Data
For each route and defined section of the route and/or network the relevant data to be collected should
include, but is not limited to:
• route lengths, average gradients

• route surface condition – ie roughness levels (averages and length specific if available)
• traffic data for all modes for each time period for each route (including volume, composition,
vehicle occupancy and trip purpose, for all modes)
• travel times and speeds
• travel time reliability data
• crash data for each route and for network if required
• resilience, vulnerability, and redundancy data.
For guidance on preparing route and traffic data see Appendix 3: Traffic data and travel time estimation.
Guidance is given on estimating traffic volumes and traffic growth, and measuring travel times and
speeds. Where the traffic growth is likely to vary from the calculated normal traffic growth, future traffic
volumes shall be predicted by taking account of:
• normal traffic growth

• diverted traffic
• intermittent traffic
• suppressed traffic
• induced or generated traffic.

For activities with congested conditions it may be necessary to consider growth suppression or variable
matrix techniques (see Appendix 1: Demand estimation methods and guidance).
Irrespective of their capital cost, the effect of activities on traffic flows in the surrounding network should
also be assessed. For example, a traffic management scheme having a small capital cost may have
significant effects on traffic flows.
Stage 3: Vehicle, cycle and pedestrian demand estimate
There are two approaches that can be used for calculating transport demand for the different modes and
mode shift(s). The first approach is to use a transport model and the second approach is based on
willingness to pay surveys combined with data on current users together with information on existing or
proposed user charges. This second approach is set out in Chapter 2 of this manual.
Where there are congested networks and the potential for induced/generated traffic, refer to Appendix 1:
Demand estimation methods and guidance.
Use of transport models and calibration

When transportation models are used to generate demand forecasts and assign traffic to transportation
networks, documentation should be provided to demonstrate the models have been correctly specified
and produce realistic results. The transport modelling documentation is available on the Waka Kotahi
website. These checklists should be completed for each analysis time period.
Note: if the analyst is not using a transport model to calculate travel times then they must refer to
Appendix 3: Traffic data and travel time estimation for the procedures on how to calculate travel times.
Model validation
The aspects of the models covered by the validation checks are as follows:
• activity model specification – including model type and parameters, data sources, trip matrices,
assignment methodology and forecasting checks
• a base-year assignment validation – comprising checks on link and screen-line flows, intersection
flows, journey times and assignment convergence
• strategic demand model checks – incorporating validation of the models and techniques used to
produce trip matrices.
Model reviewers may also use these checklists to confirm that appropriate documentation has been
provided for review purposes.
All activity benefits calculated using a traffic or transportation model shall be checked to show the results
are reasonable. The checks shall be done and reported at two levels – coarse checks and detailed
checks.
The objective of the course checks is to determine whether the travel time benefits calculated are of the
right order of magnitude. More information on the required coarse model checks is contained in the
Transport model development guidelines.
The objective of the detailed checks is to ensure the travel times on individual road sections, through
critical intersections, and for selected journeys through the network, are reasonable. This analysis shall
be undertaken for the first year of benefits and for a future year, and for both peak and off–peak periods if
appropriate.
Modelling congested networks and induced traffic
Guidelines are provided in Appendix 1: Demand estimation methods and guidance for modelling
situations where very high levels of congestion are anticipated over the economic life of the scheme.
Professional judgement should be used to determine the appropriate procedures to adopt. In cases
where there are excessive or unrealistic levels of congestion in the do-minimum network, a number of
techniques may be used to generate a realistic and stable representation of the do-minimum context.
These commonly involve upgrading the capacity of the do-minimum network or using some form of
growth constraint on the trip matrix, such as matrix capping.
The matrix derived from this process remains the same in both the do-minimum and activity option, and is
then used in the standard fixed trip matrix (FTM) evaluation procedure. Refer to Appendix 1: Demand
estimation methods and guidance for detailed growth constraint techniques.
In some situations, significant levels of congestion may be expected in the activity option across important
parts of the network (spatially) affecting a substantial proportion of the activity life (temporally). The
resulting induced travel may affect benefits as well as the choice of the activity option. The evaluation
should incorporate an analysis of induced traffic effects and Appendix 1: Demand estimation methods
and guidance contains procedures for evaluating these effects.
Stage 4: Calculate road improvement benefits

Calculate the benefits (for each mode) on the route, network, and/or transport system by quantifying, for
the do-minimum and options, the changes that occur for the factors listed in the stages below when an
improvement option is considered. The results of the route and network information collected in stage 2,
and any modelling should be used to estimate the changes needed for benefits valuation:
Note that the benefit calculations should include any negative impacts (disbenefits) during
implementation/construction.
The benefits that have currently (as at 2020) been ascribed standardised monetary values are listed
below. The benefits (ie the differences in the parameter outcomes between the do-minimum and the
options) are ascribed monetary values in Chapter 3 of this manual.
Parameters other than those listed below can be monetised, but the process and values ascribed to these
parameters must be agreed with Waka Kotahi in writing before they are included in the analysis, and
supporting information to validate the inclusion of these parameters must be provided.
Stage 4a: Impact on social cost and incidence of crashes

For the purposes of this manual, a crash is a transport-related event involving one or more road vehicles
that occurs on the transport network and that results in personal physical injury and/or damage to
property.
Crash analysis for project evaluations in related to road improvement activities can be separated into
three groups:
• a safety improvement activity (eg guardrail installation, black-spot upgrade) where most of the
benefits are the result of a reduction in crashes
• other road improvement activities where the key outcome is not to primarily reduce crashes but
where there are safety benefits or disbenefits that need to be included in the economics (eg
installation of a bus transit lane, extension of turn lanes, traffic capacity focused schemes)
• large projects (eg a new motorway link), which will have a network wide impact on crashes.
To undertake a crash analysis the appropriate crash rates, crash prediction models and crash reduction
factors can be found in the Waka Kotahi Crash estimation compendium (2018).
Refer to section 3.1 and Appendix 2: Crash analysis of this manual for detailed information on the
calculation and monetisation of crash numbers, and their severities, for the do-minimum, alternatives and
options. These calculations allow assessment of the crash incidence reductions that can be expected
from the alternatives and options under consideration. Evaluators can also refer to the Waka Kotahi
Standard safety intervention (SSI) toolkit.
Stage 4b: Impact of mode on physical and mental health
Refer to section 3.2 of this manual for detailed information on the calculation and monetisation of impact
of mode on physical and mental health.
Stage 4c: Impact of air emissions on health
Vehicle emissions are a complex mixture of gases and particulates, and in terms of human health the
primary harmful air pollutants that cause adverse health effects and have local impacts are particulate
matter (PM10 and PM2.5), nitrogen dioxide (NO2), carbon monoxide (CO), sulphur dioxide (SO2), and
hydrocarbons (HCs).
Nitrogen dioxide (NO2) is a gas that causes increased susceptibility to infections and asthma. It reduces
lung development in children and has been associated with increasingly more serious health effects,
including reduced life expectancy (Kuschel et al 2022). Particulate matter (PM10, which is smaller than
10µm) impacts predominantly on respiratory and cardiovascular systems. Effects can range from reduced
lung function, increased medication use, and more hospital admissions, through to reduced life
expectancy and death.
Refer to section 3.3 of this manual for information on the calculation and monetisation of the impacts of
vehicle emissions on human health.
Stage 4d: Impact on greenhouse gas (GHG) emissions
Greenhouse gases are pollutants that cause global warming and impact globally, eg carbon dioxide
(CO2), black carbon (BC) and methane (CH 4)
Note: Several harmful pollutants (especially BC) are direct climate pollutants, in that they have a direct
warming effect on the atmosphere. However, many of the remaining harmful pollutants, eg sulphur
dioxide (SO2) and carbon monoxide (CO), are indirect climate pollutants. This means they do not warm
the atmosphere themselves but react with other gases to increase greenhouse gas concentrations.
Therefore, initiatives which address harmful air pollutants typically yield both health and climate change
benefits.
Refer to section 3.4 of this manual for information on the calculation and monetisation of greenhouse gas
emissions.
Stage 4e: Impact of noise and vibration on health
affected by wind and intervening absorbing and reflecting surfaces, and is reduced with distance.

• exhaust systems
psychological impacts such as annoyance reaction and other behavioural impacts. While there is no
noise can cause adverse health effects in people, due mainly to stress-related factors.
noise and vibration on human health.
Stage 4f: Impact on network productivity and utilisation
Changes in travel time (for all modes)
Travel times shall be estimated according to the procedures in Appendix 3: Traffic data and travel time
estimation of this manual. Definitions for classifying traffic data and default traffic data values are also
provided in Table A45, Table A46 and Table A47. Where a specific procedure is not given, the travel time
shall be determined according to a recognised procedure compatible with the manuals and procedures
referred to in Appendix 1: Demand estimation methods and guidance and Appendix 3: Traffic data and
travel time estimation.
The flow chart in Figure 11 shows the basic stages for estimating road section travel time (the stages are
slightly different for intersections).
Figure 11: Stages for estimating travel time
Use of transportation models

When a transportation model is used for activity analysis, the model shall have been satisfactorily
validated on both traffic volumes and travel times. Checklists for validating transportation models are
provided in the modelling checks worksheet.
It is necessary that the travel times used by the model to derive the flows must be consistent with the
travel times estimated by using the procedures in Appendix 3: Traffic data and travel time estimation
during evaluation. To adhere to this, it is suggested that the functions implied by the procedures be used
as a starting point, and modified as necessary to get a satisfactory validation.
Note that, wherever practical, measured travel time information shall be obtained in preference to the
default values given in the tables in this manual.
Refer to Appendix 3: Traffic data and travel time estimation of this manual, which sets out the procedures
for estimating travel times for the do-minimum and the options for various road and intersection types.
Additionally, refer to section 3.6 of this manual for information on the monetisation of travel time impacts.
Vehicle operating costs

Vehicle operating costs (VOC) are categorised into running costs, road surface related costs, speed
change cycle costs, congestion costs and costs while at a stop. Values are provided by vehicle classes
and for standard traffic compositions on four different road categories. VOC for road sections are
functions of the length of the section, traffic volume and composition on the section, and vary by road
roughness condition, gradient and vehicle speed.
Refer to section 3.6 of this manual for information on the calculation and monetisation of vehicle operating
costs.
Traffic composition
Appendix 4: Vehicle operating cost tables also provides VOC for the standard traffic compositions using
the four road categories defined in Table A46, namely: urban arterial, urban other, rural strategic, and
rural other. The road category costs contained in the tables in this appendix are for the ‘all time periods’
traffic mix.
Buses are not included in these standard traffic compositions. If buses form a significant component of
the traffic stream they shall be included in proportion to their representation.
Regression equations
To assist analysts, regression equations are provided (refer to Table 22, Table 23, Table 24 and Table
25) which can be used to predict the VOC when using spreadsheets or transport models. Note that the
regression coefficients vary between vehicle classes and road categories.
The regression equations were used to generate the corresponding VOC tables, so the results will be
consistent, irrespective of which approach is used.
Minor differences will arise when calculating road category costs from individual vehicle class costs due
to the regression equations being developed from the road category data. Where high precision is
required, the vehicle class equations should be summed and used in preference to the road category
equations.
Base vehicle operating costs

The base VOCs comprise fuel, tyres, repairs and maintenance, oil and the proportion of depreciation
related to vehicle use. Standing charges, ie those incurred irrespective of use, are excluded from these
costs. Such charges are included in the travel time costs for vehicle types in Table 15 and the composite
travel time values in Table 16.
The breakdown of the base VOC by component is given in Table 67 below.
Table 67: Breakdown of vehicle operating costs (VOC) by component

Vehicle class Percentage of total base VOC by component
Fuel and oil Tyres Maintenance and Depreciation
repairs
PC 35.7 6.2 37.2 21.0
LCV 39.6 7.2 29.4 23.8
MCV 38.6 4.2 44.2 13.0
HCVI 42.0 8.3 42.1 7.6
HCVII 37.3 10.4 43.0 9.3
BUS 46.1 6.0 36.9 11.0
Road category Fuel and oil Tyres Maintenance and Depreciation
repairs
Urban arterial 36.8 7.3 38.1 17.8
Urban other 36.9 6.9 37.9 18.3
Rural strategic 37.5 7.9 39.4 15.2
Rural other 37.4 7.7 39.1 15.8
Table 68: Recommendations on diversion rates to/from public transport from changes in car
travel costs
Car travel Typical diversion rates Comments on estimates by market segment

cost variable (% of deterred car users
switching to public
transport)
Parking Dependent on market • Regional CBD, work trips: 75%
Charges segment
• Regional CBD, non-work trips and suburban CBD
work trips: 50%
• Other areas: not defined, likely to be much lower
Table based on Wallis (2004)
Stage 4g: Impact on system reliability
between weekdays and weekends. This type of variation is well known to regular drivers and is taken into
account in when calculating the travel time values (including congestion values).
Trip journey time reliability is a different type of variability, which is much less predictable to the driver.
(For example, car drivers who make a particular journey at the same time every day find some days it
takes as little as 20 minutes, and on other days as much as 40 minutes.) Hence, when the car drivers
plan their trips, they have to consider not just the expected travel time but also its variability. Where an
activity improves trip reliability, the benefits apply to both work and non-work trips, and can be calculated
using the in this section.
Journey time reliability is measured by the unpredictable variations in journey times, which are
experienced for a journey undertaken at broadly the same time every day. The impact is related to the
day-to-day variations in traffic congestion, typically as a result of day-to-day variations in flow. This is
distinct from the variations in individual journey times, which occur within a particular period.
Journey time reliability is in principle calculated for a complete journey and the total network variability is
the sum of the travel time variability for all journeys on the network. In practice, models may not represent
the full length of journeys and this is accounted for in the procedure.
Process for evaluating reliability benefits

1. Calculate the standard deviation of travel time on each link between intersections and for each
intersection movement or approach.
Travel journey time variability is expressed as the standard deviation of travel time. The sources of
variability are road sections and intersections. Reduced variability arises from a reduction in
congestion on links and at intersections along a route. For a single section or intersection approach
the standard deviation of travel time can be calculated using that section or intersection movement’s
VC ratio:
s - s0
Standard deviation of journey travel time s0 +
b v - a (min)
1 +e c
where: the VC ratio is represented by s, s0, b and a are taken from Table 69 below
Table 69: Coefficients to calculate standard deviation of travel time

Context S b a S0
Motorway/multi-lane highway (70–100km/h) 0.90 -52 1 0.083
Urban arterial 0.89 -28 1 0.117
Urban retail 0.87 -16 1 0.150
Urban other (50km/h) 1.17 -19 1 0.050
Rural highway (70–100km/h) 1.03 -22 1 0.033
(two lanes in direction of travel)
Signalised intersection 1.25 -32 1 0.120
Unsignalised intersection 1.20 -22 1 0.017
Note: Evaluations of small retail areas on 50km/h sections of a rural highway should use the urban
other (50km/h) context.
2. Square the standard deviations to produce variances.
3. Sum variances along each origin-destination path to obtain the total variance for journeys between
each origin and destination.
4. Take the square root of the aggregated variance for a journey to give the standard deviation of the
journey time.
5. Multiply the total trips for each origin–destination pair by the standard deviation of travel time and
sum over the matrix to give the network-wide estimate of the variability.
6. Calculate the difference in variability between the project and do-minimum networks.
Intersections should be analysed by movement at traffic signals and by movement or by approach
for roundabouts and priority intersections. Variability for the uncontrolled movements at priority
For road sections, the calculation of the standard deviation of travel time assumes there is only one
link between junctions or between changes in link context. If the model has more than one link
between junctions then variability associated with such artificial network nodes should be set to zero.
Network skims compatible with the assigned flows should be used to aggregate travel time variances
(square of standard deviation) along paths to create a matrix (or matrices where multiple paths are
used) of journey time variance for origin-destination pairs. The square root of each cell in the
resulting matrix will provide the variability (standard deviation) of travel time for that journey.
The total network variability is the sum of the products of the number of journeys between origin–
destination pairs and the standard deviation of travel time for that journey.
It is important to note that the process above produces estimates of travel time variability as a
function of VC ratio, reflecting the impact of day-to-day variations in travel demand. This is not the
same as variations in individual journey times within a modelled period, a possible output of micro-
simulation models. The variation in individual journey times from such models will be influenced by
the driver, vehicle type and generation factors used in the stochastic processes used in the model.
For individual intersection upgrades, the turning movements can be used as a proxy origin-
destination matrix with the movement-weighted standard deviation being calculated for the
intersection.
For project areas with more than a single congested intersection or link, an estimate of the proportion
of trips that travel through more than one of these sources of variability must be made in order to
approximate the total study area variability.
For two sources of variability, the reliability estimate for each trip direction is the sum of:
Variability for trips which travel only through source x:
Fx SDx
and, for trips travelling through both source x and y: 2 2
Fx.y SDx + SDy
where: Fx is trips that travel through only source x
Fx.y is trips that travel through both x and y
SDx is standard deviation of travel time for trip at source x
SDy is standard deviation of travel time for trip at source y
For each of the three sources of variability, the reliability estimate is the sum of the individual
components below:
Through source x only: Fx SDx
Through sources x and y only: 2

+
2
Fx.y SDx SDy
Through sources x and z only: 2 2
Fx.z SDx + SDz
2 2 2
Through sources x, y and z only: Fx.y,z SDx + SDy + SDz
where: Fx,y,z = trips that travel through all three sources x, y and z.
If traffic passes through more than three sources of significant congestion in the modelled area
then evaluators must estimate the trip matrix and perform the calculation using the aggregation of
journey variance method.
Rural two-lane roads
Table 70, Table 71 and Table 72 contain travel time variability values for rural two-lane roads of varying
terrain and the volume to capacity (VC) ratio (see Appendix 3: Calculating the volume to capacity ratio).
The time period used to calculate the VC ratio must contain a relatively constant level of traffic volume.
Table 70: Travel time variability – rural two-lane road, level terrain
Standard deviation of travel time (minutes) – percent no-passing for level terrain
VC ratio 0% 20% 40% 60% 80% 100%
0.00 0.01 0.04 0.07 0.11 0.13 0.14
0.10 0.07 0.07 0.08 0.09 0.10 0.11
0.20 0.09 0.08 0.08 0.08 0.08 0.08
0.30 0.09 0.08 0.08 0.07 0.07 0.06
0.40 0.07 0.06 0.06 0.05 0.05 0.04
0.50 0.05 0.05 0.05 0.04 0.04 0.03
0.60 0.03 0.03 0.03 0.03 0.03 0.03
0.70 0.03 0.03 0.03 0.04 0.03 0.03
0.80 0.05 0.05 0.05 0.05 0.04 0.06
0.90 0.10 0.10 0.09 0.09 0.08 0.10
1.00 0.18 0.18 0.15 0.15 0.17 0.18
Table 71: Travel time reliability – rural two-lane road, rolling terrain
Standard deviation of travel time (minutes) – percent no-passing for rolling terrain
VC ratio 0% 20% 40% 60% 80% 100%
0.00 0.03 0.09 0.15 0.17 0.24 0.27
0.10 0.11 0.13 0.15 0.17 0.17 0.18
0.20 0.13 0.13 0.12 0.13 0.12 0.12
0.30 0.12 0.10 0.09 0.09 0.08 0.08
0.40 0.09 0.07 0.06 0.06 0.06 0.05
0.50 0.06 0.05 0.05 0.05 0.06 0.06
0.60 0.05 0.06 0.07 0.08 0.09 0.08
0.70 0.07 0.10 0.12 0.14 0.15 0.14
0.80 0.14 0.18 0.21 0.23 0.23 0.22
0.90 0.26 0.29 0.32 0.34 0.34 0.34
1.00 0.43 0.44 0.47 0.46 0.47 0.49
Table 72: Travel time variability – rural two-lane road, mountainous terrain
Standard deviation of travel time (minutes) – percent no-passing for mountainous terrain
VC ratio 0% 20% 40% 60% 80% 100%
0.00 0.13 0.25 0.32 0.40 0.51 0.65
0.10 0.18 0.21 0.26 0.28 0.32 0.33
0.20 0.17 0.17 0.20 0.21 0.20 0.18
0.30 0.15 0.15 0.17 0.16 0.15 0.13
0.40 0.14 0.15 0.16 0.16 0.15 0.15
0.50 0.15 0.18 0.18 0.18 0.18 0.20
0.60 0.21 0.23 0.22 0.23 0.24 0.26
0.70 0.28 0.30 0.29 0.30 0.32 0.34
0.80 0.37 0.36 0.37 0.38 0.41 0.43
0.90 0.43 0.40 0.44 0.45 0.50 0.55
1.00 0.43 0.39 0.50 0.51 0.59 0.73
7. Assess the percentage of variance occurring outside of the selected study area and select the
adjustment factor.
In many cases, an activity evaluation will consider a defined area that does not represent the full
length of most journeys. As a result, the changes in journey time reliability will be overestimated. In
these cases the variability estimates need to be adjusted. Table 73 below gives some illustrative
contexts where different factors might apply. An estimation of the variance of journey times that
occurs outside of the evaluation area must be made and the appropriate correction factor from Table
73 applied.
The trip time reliability benefit is adjusted by multiplying the calculated variability benefit by the factor.
Table 73: Adjustment factors to apply to variability calculations table

Percentage of variance Factor for benefit calculation Indicative transport network
outside of study area model coverage
<20% 100% Regional model
20% 90% Sub-regional model
50% 70% Area model
75% 50% Corridor model
90% 30% Intersection model,
individual passing lane
8. Calculate the impact of changes in trip reliability using following formula:

0.9 × travel time value ($/h) (Table 14 or Table 16)
× (reduction in the network variability (in min)/60)
× traffic volume for time period (veh/h)
× correction factor (Table 73)
Where the reduction in network variability is the difference between the sums of the variability for all
journeys in the modelled area for the do-minimum and project option. The 0.9 factor is the value of
reliability based on a typical urban traffic mix. For projects with a significantly different vehicle mix,
evaluators should use 0.8 for cars and 1.2 for commercial vehicles.
In addition to the normal day-to-day variation in travel times, there can be occasional large delays
resulting from major incidents (eg crashes or breakdowns). The effect of a major incident will be
related to the amount of spare capacity at the location. A specific analysis should be undertaken to
determine the economic cost of delays from major incidents. Assessing this type of variability is best
handled separately from normal day-to-day variability and could also be part of the calculations of
system vulnerability, which is outside the scope of the procedure just outlined.
A worked example of the trip reliability procedure is provided in Appendix 8: Worked examples.
Stage 4h: Impact on user experience of the transport system

There are two standard values related to this benefit in this manual as follows:
1. Impact on driver frustration derived from ‘time spend passing’. This is an indicator of changes that
passing lane generate in road users experience.
2. Impact on road users’ comfort and productivity due to sealing unsealed roads.
Refer to section 3.8 of this manual for information on the calculation and monetisation of the impact on
user experience of the transport system.
Driver frustration
Vehicle passing options may be provided through the construction of dedicated passing lanes, climbing
lanes, slow vehicle bays, and improved alignments.
Providing passing options releases vehicles from platoons of slower moving vehicles, allowing them to
travel along the road at their desired speed until they are once again constrained by platoons. Typically,
the evaluation of passing options has been undertaken by micro-simulation programmes, which use
various vehicle performance models together with terrain data to establish, in detail, the speeds of
vehicles at each location along the road. These assessments can be excessively complex, particularly
given the general magnitude of such activities.
The demand for passing and consequently the benefits, are a function of a number of parameters
including:
• traffic variables:
o traffic volume
o percentage of HCVs
o initial platooning
o directional split of traffic
o vehicle speed distributions
• road variables:
o terrain/alignment
o grades
o available passing lanes (sight distance)
o passing lane lengths and frequency.
An alternative method based on multiple simulations and the unified passing model is described in
Appendix 5: Passing lanes, and is available in the Provisional passing & overtaking guidelines on the
Waka Kotahi website. This method can be used to identify the most appropriate strategy for providing
improved vehicle passing options over a route, and assess the benefits of individual vehicle passing
options within those strategies.
Road user comfort from seal extension
into account.
Stage 4: Wider economic impacts
Refer to sections 3.9 to 3.13 of this manual for information on the calculation of wider economic impacts.
Wider economic benefits (WEBs) are impacts that can result from transport investment that have been
used internationally to improve transport cost-benefit analysis. They can be thought of as impacts that are
additional to the conventional benefits to transport users (illustrated in the following diagram). WEBs
include changes to productivity, labour supply and imperfect competition, as well as regional economic
development impacts.
Great care is required to ensure that the estimates for wider economic benefits are truly additional to
conventional benefits to avoid double counting. As an example, business travel time savings can result in
productivity and output increases. These are a direct user benefit and any wider economic benefits for
increased productivity have to be additional to these direct user benefits.
In addition to, or in some cases as a consequence of direct impacts, there can be indirect impacts on the
economy. These may cause a redistribution or reallocation of resources or may cause the entry or exit of
firms. These are wider economic impacts and can include:
• economies of scale from improved transport that can encourage agglomeration or specialisation
of economic activity
• mitigating existing market failures by improving accessibility and therefore competition between
spatial markets
• increased output in imperfectly competitive markets by diminishing persistent externalities
• technology and knowledge transfer by connecting people and places and increasing the
interaction between economic actors.
New Zealand application of WEBs
The following wider economic benefits are applicable in the New Zealand context:
• agglomeration, where firms and workers cluster for some activities that are more efficient when
spatially concentrated
• imperfect competition, where a transport improvement causes output to increase in sectors where
there are price-cost margins
• increased labour supply, where a reduction in commuting costs removes a barrier for new
workers accessing areas of employment.
Stage 4j: Other significant impacts that can be monetised
Refer to section 3.15 for a discussion of approaches to monetise impacts not included within this manual.
Resilience
Where system vulnerability and redundancy benefits are expected to comprise a significant proportion of
benefits, due to the renewal or replacement of vulnerable infrastructure, expected costs and benefits may
be calculated using risk analysis and the infrastructure’s probability of failure. Transport analysts are
encouraged to contact Waka Kotahi to discuss the approach by emailing [email protected].
A worked example of the risk analysis procedure for resilience is provided in Appendix 8: Worked
examples.
Stage 5: Describe and evaluate any mitigation measures
Where mitigation measures are required to conform to the Resource Management Act 1991, these
measures must be described and their impact on the economic efficiency evaluation quantified.
Stage 6: Undertake risk analysis for significant unpredictable events

Refer to Chapter 7 of this manual for detailed procedures on risk analysis.
The purpose of considering risk is to develop ways of minimising, mitigating and managing it. Risk
analysis and risk management are continuous processes that start at the project inception stage and
proceed through to project completion and ideally should involve all the relevant parties.
The extent of risk analysis needs to be appropriate to the stages of project development. The critical
project stages are from the rough order cost (ROC) stage through to preliminary assessed cost (PAC)
stage and then to final estimate of cost (FEC) stage. It is intended that the scope and extent of analysis
will progress according to the stage of project development and be most comprehensive at the FEC
stage. The risk identified and evaluated in these various stages needs to be monitored and managed,
particularly in the final construction stage.
Figure 12: Risk analysis process
Start of project stage: During the project stage: At end of project stage:
• Identify risks. • Implement preferred • Report on outcomes of

• Assess risk management strategy. strategy.
strategies (reduction, • Assess implications for
mitigation, avoidance, next stage of project.
quantification through date
collection etc).
• Choose preferred strategy.
Figure 13: Risk analysis steps
Stage 7: Calculate costs for do-minimum and improvement options

For road improvement activities, costs comprise:
• planning, investigation and design fees

• costs of property required for the activity
• construction costs, including preconstruction and supervision
• maintenance and renewal costs, including repair and reinstatement
• operating costs
• risk management costs
• external impact mitigation costs
• provisional costs
• government financing costs
• contingencies
• residual value.
The costs of engineering investigation and design, and the costs of environmental and planning
procedures, shall be included unless they have already been incurred, in which case they are sunk costs
(and are not included in the evaluation).
Land costs
Where land has to be acquired for road development, its resource cost shall be assumed to equate to its
market value for activity evaluation purposes. Similarly, land available for sale due to obsolescence of an
existing road shall be included as a cost saving.
Where land required for an activity is already owned by the road controlling authority, its market value at
the base date shall be included in the analysis. Land shall not be treated as a ‘sunk cost’, as the option of
alternative use nearly always exists.
Market value shall be assessed on the basis that the land is available indefinitely for other use. Small
isolated or irregularly shaped lots of land are often difficult to develop. If amalgamation with adjacent
property is impracticable, the resource cost of the land is its amenity value only. If amalgamation is
possible, the market value of the main property, with and without the addition of the small lot, shall be
assessed. The difference is the resource value of the lot, which in some cases may be considerably more
than the achievable sale price.
Risk management costs
Where there is a quantifiable risk of disruption to traffic, damage to vehicles, the roadway or structures, or
injuries to road users from natural or human-made events, and the activity reduces or eliminates the
impacts compared with the do-minimum, then the appropriate risk-management costs must be included in
the activity evaluation.
The costs of mitigation, repair and reinstatement shall be included for each year of the analysis period
over which they occur, both in the do-minimum and the activity options. These costs and benefits shall be
included either as expected values or as a probability distribution, depending on the size and nature of
the activity.
Mitigation costs
Where a design feature to avoid, remedy or mitigate adverse external impacts is included in an activity
and the feature significantly increases the activity cost, it shall be treated in the following way:
• If the feature is required by the consenting authority in order to conform with the Resource
Management Act 1991 or other legislation, then the cost of the feature shall be treated as an
integral part of the activity cost.
• If the feature is not required by the consenting authority in order to conform with the Resource
Management Act or other legislation, then the feature shall be described and evaluated in terms
of benefits and costs, and the results presented in an incremental BCR calculation.
Where several features are to be included or there are several ways of mitigating an adverse impact, they
should be evaluated separately.
The cost of the preferred mitigation feature should be included in the activity cost calculations.
Provisional costs
Provisional costs shall be included for those costs that are expected to be incurred but are not quantified
at the time of preparing the estimate. For example, it may be known that street lighting is required but
detailed costing for the lighting is yet to be undertaken.
Contingencies
Contingency allowances shall be included in the activity costs to allow for possible cost increases and the
uncertainty of cost estimates. These allowances shall be based on the phase of development of the
activity and the level of accuracy of the estimate and that phase. The following table of default
contingency allowances provides guidance.
This information is to be used when the analyst does not have better information based on road
controlling authority experience:
Table 74: Contingency allowances

Phase Earthworks component (%) Other works (%)
Project feasibility report 30 20
Scheme assessment 25 15
Design and contract estimate 20 10
Contract 10 5
Residual value
The residual value of the investment at the end of 40 years has a very small effect on the evaluation
when discounted at 4% and shall generally be omitted. Where two options have widely differing service
lives, this shall be noted in the activity summary sheet.
Benefits and costs generally arise throughout the life of projects and to calculate their present worth or
analysis period of 40 years, sets of present worth factors have been calculated to convert future benefits
and costs to their present values (see Table 102, and Table A132, Table A133 and Table A134 from
Appendix 6: Discount factors). Discount rates of 3% and 6% are also provided in the tables for sensitivity
testing.
Refer to section 6.3 for detailed information on developing incremental BCRs

Stage 11: Sensitivity testing on the preferred option
Refer to Chapter 7 of this manual for the details on sensitivity testing.
4. Evaluation procedures > 4.4 Evaluation of public transport service activities

Specific types of road improvement activity

There are several types of road improvements that require specialised consideration and there are other
improvements for which the evaluations have been standardised, such as the standard safety intervention
(SSI) toolkit of improvement activities. These later activities, including wide medians, centre and roadside
barriers, etc, can be evaluated using the procedures in the Standard safety intervention toolkit available
on the Waka Kotahi website.
Passing lane procedures

One of the specialised procedures that can be used is for passing lanes and details are provided in
Appendix 5: Passing lanes of this manual on the evaluation of the impacts of passing lanes .
4.4 Evaluation of public transport service activities

public transport (PT) activities. Activities may be stand-alone improvements or a component of a package
or a wider programme of transport improvements.
For simpler and relatively standardised improvement activities with an undiscounted funding gap less
than $15 million, simplified procedures are provided for the analysis, and these are explained below. For
more complicated projects and those with an undiscounted funding gap greater than $15 million, the full
procedures are provided as an alternative to the simplified procedures and are explained later in this
section.
Simplified procedures for public transport services

The following simplified procedures (SPs), for a new PT service (SP9) or for improvements to an existing
PT service (SP10), use a 4% discount rate and a 40-year analysis period. The capital costs of new, and
improved, PT activities may extend into year two if the expected construction duration is longer than 12
months. Where costs are common to both the do-minimum and the options they are not included in the
analysis. All costs are to be exclusive of GST.
Simplified procedures SP9 and SP10 adopt the following approaches:
• Benefits accrue to public transport users and road users.

• Public transport user benefits can include time savings, better reliability and better vehicle and PT
infrastructure quality.
• Road user benefits result from reduction in road traffic, and include travel time savings (including
congestion reduction), vehicle operating cost savings, crash cost savings, and environmental
benefits (including CO2 reduction). The road traffic reduction benefit values assume that the road
corridor has at least one point that operates at greater than 80% capacity during the peak period.
The simplified procedures for PT are for the evaluation of activities that have an undiscounted funding
gap of less than or equal to $15 million over the first three-year period of operation. If this criteria is not
met then the full procedures must be used.
Refer to the Planning and Investment Knowledge Base (PIKB) for guidance on issues relating to analysis
of PT activities, including selection of the preferred option using the Business Case Approach.

Analysis which alters components of the simplified procedure should not be used as this will compromise
the assumptions on which the procedures are based, and full procedures should be used instead.
[email protected].
SP9 for new public transport services
Procedure SP9 provides a simplified method of appraising the economic efficiency of new public transport
services and associated capital infrastructure. The procedure assumes that benefits accrue to new public
transport users and to road users.
A description of all options considered should be described in worksheet SP9-1 and included in the
incremental analysis; for all other worksheets, only the details for the preferred option need to be
included.
Guidance for completing the SP9 New public transport services (template worksheets) is provided below
in Table 75 and Table 76.
Table 75: SP9 New public transport services procedure template

number
SP9-1 Evaluation summary Used to summarise the general data
considered for the evaluation plus the
results of the economic analysis.
SP9-2 Service provider costs Used to calculate the PV cost to the service
provider of a new service. The cost includes
capital, operation and maintenance costs.
SP9-3 Funding gap analysis Used to determine whether the new service
is commercially viable.
SP9-4 Public transport user benefits Used to calculate the PV of public transport
user benefits based on users’ willingness to
pay for the new service.
SP9-5 Road traffic reduction benefits Used to calculate the PV of benefits for
other transport system users
SP9-6 BCR and incremental analysis Used for comparison of the options
considered.
Table 76: Steps in the SP9 evaluation of new public transport service activities
Step Description
2 Complete Worksheet 2 – Service provider costs
3 Complete Worksheet 3 – Funding gap analysis
4 Complete Worksheet 4 – Public transport user benefits
5 Complete Worksheet 5 – Road traffic reduction benefits
SP10 for existing public transport services

Procedure SP10 provides a simplified method of appraising the economic efficiency of improvements to
existing public transport services through service and/or capital infrastructure enhancements.
The procedure includes the following assumptions:
• Benefits accrue to new and existing public transport users and to road users.
• The activity will not generate a drop off in existing passengers (eg as a result of a fare rise).
• Each trip on the improved service is an ‘average’ length for the urban centre.
The benefit may therefore be overestimated where trips are shorter than the average and underestimated
where trips longer than the average. Consider whether this is likely to be significant.
A description of all options considered should be described in worksheet SP10-1 and included in the
incremental analysis; for all other worksheets, only the details for the preferred option need to be
included.
Guidance for completing the SP10 Existing public transport services (template worksheets) is provided
Table 77: SP10 procedure template

number
SP10-4 Net benefits Used to calculate the PV of benefits for new
and existing PT users, as well as benefits
for other transport system users.
considered.
Table 78: Steps in the SP10 evaluation of new public transport service activities
Step Description
4 Complete Worksheet 4 – Net benefits
Full procedures for public transport services

The full evaluation procedures for PT activities are to be used to appraise the economic efficiency of
activities when the simplified procedures are not appropriate or sufficient.
The primary purpose of this section of the manual is to establish the impacts of introducing new PT
services or improving existing services (ie the changes that occur between do-minimum and the options)
when using the full procedures. Following on from calculating the impacts, the analyst will need to assign
monetary values to the impacts and then calculate the benefits and the benefit–cost ratios (BCRs).
which links have been provided. A significant focus of the PT service procedures is on the calculation of
activity impacts and the service provider’s funding gap, in particular stages 3 to 6 in Table 79.
The following table outlines the stages of analysis when undertaking an evaluation of the impacts of
introducing a new PT service or improving an existing one. The chapters and sections of this manual that
apply to each stage of the analysis are referenced in the table below.
Table 79: Stages of analysis for the evaluation of public transport services
Counterfactuals
a. the do-minimum
Section 1.5:
b. improvement alternatives and options
Alternatives and
c. whether the improvement(s) should be part of a package options
and/or programme of activities.
2 Forecast the PT demand either from a transport model or by using Current section and
PT demand elasticities including: Chapter 2: Demand
estimation and mode
• public transport direct elasticities – short run
share
• public transport direct elasticities – by market segment
• public transport direct elasticities – longer run (‘ramp-up’)
effects
• impacts of public transport initiatives on demand for
alternative modes (diversion rate)
• non-public transport cross-modal (diversion rate) effects on
public transport travel.
3 Measure and monetise the impacts (benefits and disbenefits) for the Chapter 3: Benefits
do-minimum and options, including:

• wider economic impact (labour supply)
• wider economic impact (imperfect competition)
• wider economic impact (regional economic development)
• wider economic impact (land use change)
• other impacts that can be monetised – these are not
included in this manual but can be included if there is
sufficient supporting evidence and the approach is accepted
by Waka Kotahi.
4 Undertake risk analysis when there are significant unpredictable Chapter 7: Sensitivity
events that may affect or be affected by the improvement activity. and risk analysis
5 Calculate the costs to the government of services for the do- Current section and
minimum and improvement options, including (but not exclusively): section 1.8: Costs
• land costs
• funding assistance from government
• maintenance, renewal and construction cost savings
• construction costs, including property, for any additional
infrastructure required
• maintenance costs not already included in service contracts.
Bus operating costs must also be calculated either from detailed
operating cost information or the standardised values in this section.
6 Discount the monetised benefits and costs over the analysis period Chapter 5:
to obtain present values. Discounting
7 Calculate the funding gap and any net cost to government by: Current section
• calculating service provider costs
• calculating service provider revenue
• performing cash-flow analysis
• calculating the service provider’s funding gap
• performing sensitivity tests on the funding gap analysis.
cost ratios
9 Use incremental cost–benefit analysis to select the preferred option Chapter 6: Benefit–
for mutually exclusive options. cost ratios
10 Perform sensitivity tests on the preferred option to determine how Chapter 7: Sensitivity
robust the calculations are and whether a small change in one of the and risk analysis
input parameters has a large change on the evaluation outcome(s).

The do-minimum for evaluation of PT services is usually considered as a continuation of the present
transport networks, service levels and the existing PT network in the study area.
The do-minimum must include any costs and resulting demand implications of committed transport
infrastructure and PT service improvements during the analysis period. Activities are committed if they
have been evaluated in accordance with the Waka Kotahi evaluation procedures and have been
approved for funding. Any investment plans that are not committed must be included in the evaluation as
options. Maintenance, renewal/replacement schedules and any planned public transport service changes
must also be included.
be the ‘preferred option’.
in mind.
Stage 1c: Packages and/or programme of activities
• neglecting options that differ in type or scale, eg an extension to an existing bus route may
eliminate the need to introduce a new service
• neglecting significant externalities, eg the impacts of a growth area on future demand for PT
services
• inconsistencies with wider strategic policies and plans, eg generous parking policies in a CBD
that undermine PT use.
If public transport options are part of a wider package, then a multi-modal evaluation may be necessary.
This may involve analysing public transport components and road infrastructure components separately,
using the relevant procedures in this section and section 4.3, and aggregating the results.
General guidance on travel demand forecasting and mode change estimation is provided in Chapter 2 of
this manual.
For significant PT investments, realistic demand estimation is a critical first step in the evaluation process.
Mode change analysis using elasticities of demand may not be sufficient, and therefore it may be more
appropriate to use multi-modal and multi-stage models to approximate future demand. Peer review of
demand estimates will be required in these scenarios. For PT investments that are limited in scope or
scale it may be appropriate to use standardised elasticities of demand. These elasticities may also be
used to calibrate microscale models such as those for corridor improvements. The elasticity of demand is
a measure commonly used to summarise the responsiveness of demand to changes in the factors
determining the level of demand, such as the level of fares or frequency of service provided.
This section describes recommended demand elasticities for public transport travel, drawing on New
Zealand, Australian and international evidence and various literature reviews. In cases of uncertainty
regarding the appropriateness of analytical methods, analysts are asked to seek advice from Waka
Kotahi via [email protected]
This section of the manual includes recommended values for:
• public transport direct (own mode) elasticities, ie the effects on public transport system demand
of changes in public transport attributes
• cross-modal demand effects (’diversion rates‘) resulting from changes in system attributes for
non-PT modes, eg the effects on public transport demand resulting from changes in fuel prices or
parking charges
• cross-modal demand effects (‘diversion rates’) resulting from changes in public transport system
attributes, eg the effects on car travel demand resulting from changes in public transport service
frequency.
Public transport direct elasticities and diversion rates (or cross-elasticities) to/from other modes cover
changes in the following public transport attributes:
• fares
• service changes (including service frequencies, route and network redesign, etc)
• travel time (expected) changes
• travel time reliability changes, and
• overall (‘generalised cost’) changes.
Diversion rates (or cross-elasticities) for public transport demand with respect to changes in the attributes
of other (non-PT) modes include changes in fuel prices and parking charges.
The primary focus of the recommended elasticity and diversion rate values is on ‘short’ run demand
impacts, ie taken as patronage changes within roughly 12 months of any change in service attributes.
Estimates are also provided for ‘long run’ elasticity values, based on expected responses after 5–10
years (or more) following any attribute change.
All evidence indicates that the market responses (represented by elasticities) can differ substantially by
the time period analysed such as the time of day or day of the week. In particular, there are substantial
differences between peak period and off-peak period travel, with off-peak responses themselves then
differing between weekday interpeak, weekday evening and weekend periods. Recommended values for
different time periods are therefore provided where available.
The research evidence indicates that underlying demand elasticities for a given attribute, market segment
and time period etc show a strong similarity across urban areas in most developed countries. Given this,
the elasticity values recommended in this section draw firstly on New Zealand evidence and secondly on
Australian evidence, supplemented by evidence from other developed countries where appropriate
(principally where the New Zealand/Australian evidence is very limited).
The basic expression of elasticity is:
E = Proportional change in demand/proportional change in explanatory variable
= (∆ y/y)/(∆ x/x)
where: ∆ y is the change in the demand y, ∆ x is the change in the explanatory variable x.
Point elasticity: the above definition refers to a change ∆x which is vanishingly small, so may be
expressed mathematically as:
Ep = (∂y/y)/(∂x/x) = (∂y/∂x).x/y
This elasticity represents the slope of the demand curve (∂y/∂x) at a particular point multiplied by the ratio
of the explanatory variable (x) to the level of demand (y) at this point.
This is referred to as a point elasticity measure, representing the elasticity only at a particular point on the
demand curve. In practice, point elasticities cannot be computed from empirical data unless the shape of
the demand curve is known (or postulated) and its parameters may then be estimated from the observed
data. Therefore, other elasticity formulations, which do not require the slope of the demand curve are
often applied.
Arc elasticity: The arc elasticity concept is frequently employed in practical analysis, to estimate the
elasticity from observations for two points on the demand curve: for small changes it approximates the
point elasticity. If we assume a constant elasticity demand function over the range of change, the arc
elasticity can then be calculated as:
EA = ∆ ln y = ln y2 – ln y1 = ln(y2/y1)
∆ ln x ln x2 – ln x1 ln (x2/y2)
This is equivalent to (y2/y1) = (x2/x1)E, consistent with the constant elasticity demand function:
Y = k.(x)E
When applying the elasticity approach to public transport systems, the dependent variable is the demand
or patronage (P), while the independent variable is the attribute of the system that is being varied, eg
service level or fares (S). Hence the formula for the elasticity of patronage with respect to service
frequency is expressed as:
E = ln (P2/P1)/ln (S2/S1)
This (natural) logarithmic (ln) function has a number of advantages over alternative elasticity functions for
analysis and application purposes.7 It also assumes that the demand elasticity is constant over the range
of changes under consideration. While this assumption is open to significant debate, it provides a
reasonable approximation except possibly in situations of very large changes in the independent variable
(eg fares or service levels).8
There are common issues that can arise during the estimation and application of elasticities. Table 80
provides some advice to assist analysts in estimating the patronage impacts, and hence demand
elasticities, from PT initiatives that have been or are being implemented. This advice is also relevant to
forecasting the likely impacts of initiatives being considered for implementation.
Table 80: Issues in elasticity estimation and application

Issue Comments
1. Demand Improvements in services (eg increased service frequencies) on a single route will
effects by generally result in some existing PT passengers switching from broadly parallel
route versus routes, such that the net system patronage increase may be substantially lower than
corridor the increase measured on the route in question. Any patronage and elasticity
analyses need to cover a sufficiently broad corridor to cover all routes likely to be
affected.
7 A significant advantage is that, if S varies from S1 to S2 and then S3, the patronage estimates will be consistent
whether the change from S1 to S3 is calculated directly or via S2; and similarly, if S varies from S1 to S2 and then
back to S1, the formulation will show zero net patronage change.
8 The evidence indicates that fare and service elasticities tend to increase with higher fares and with higher service
headways (lower frequencies).
2. Demand This is a particular case of the issue above. For example, in a case where train fares
effects by PT are increased relative to bus fares in a given corridor, if only the change in train
mode passengers is measured, an apparently very high fares elasticity may well result. But
the overall PT fares elasticity may be much smaller, as a large proportion of the loss
in train patronage may switch to the parallel bus routes.
3. Demand If services are improved at a particular time period, measurement of the patronage
effects by changes only at that time period may under-state or over-state the overall effects on
time period patronage: some of the additional passengers resulting from the service
improvements may switch from travelling at other time periods; but some may also
make additional trips (eg return trips) at other time periods. Generally, the latter effect
is likely to be dominant, but any before versus after surveys need to be of sufficient
scope to assess the overall impacts.
4. Timescale This is discussed in detail in stage 2b and demonstrated in Figure 14. In general, an
of demand ‘after’ survey 6–12 months following introduction of most types of initiative (if
impacts preceded by an appropriate ‘before’ survey, and allowing for any seasonal effects),
should provide a good guide to the shorter-term and potentially the medium- and
longer-term effects of an initiative.
5. Use of In general, any analyses to estimate the patronage impacts of PT initiatives should
control make use of control groups, so as to enable adjustment of results for any patronage
groups changes through the analysis period which may have been independent of the
initiative being analysed. In cases of doubt, the analyst should seek advice on the
selection and analysis of suitable controls.
Stage 2a: Public transport direct elasticities – short run

In the majority of situations, analysts will be interested in applying elasticities to estimate the effects on
public transport demand of changes in a single attribute (eg fares, service frequency, travel time). A set of
overall (short-run) elasticities for this purpose is provided in Table 81.
The ‘generalised costs’ (GC) of a journey represent the weighted sum of all the separate journey
attributes, usually on a door-to-door basis (eg walk to bus stop, wait for bus, time on bus, time walking to
destination, fares paid). For estimating demand impacts, the GC approach is often preferable to the
individual elasticity approach as it gives more consistent results over a range of situations. The empirical
evidence is that GC elasticities appear to be sensibly constant (for a given market) over a wide range of
journeys with different component costs and elasticities; on the other hand, individual component
elasticities tend to vary according to the proportionate contribution of the component to the total
generalised cost (eg in situations where fares are relatively high, the corresponding fares elasticity is
likely to be high).
The following points should be noted:
• Elasticities are generally sensitive to the market segment under consideration, particularly in
terms of the time period (peak, off-peak, etc) to which any attribute changes apply.
Disaggregation of elasticity values by time period and other market segments is provided in
Stage 2b below. The Table 81 values should generally only be used for assessing attribute
changes applying ‘across the board’, or where specific time period changes are not defined.
• The same elasticity estimates (eg as in Table 81) may be applied for all urban public transport
modes: the evidence indicates that there are minimal intrinsic differences between elasticities for
different PT modes, other than those relating to trip lengths, service frequencies etc.
• Fare elasticities should always be applied in real terms, ie after adjusting nominal fares before
and after a fare’s change for any effects of inflation.
One convenient property of generalised costs is that the generalised cost elasticity for a journey is the
absolute sum of all the elasticity estimates for the individual journey components. Table 81 includes a
best estimate GC elasticity of -1.30: it is seen that this is approximately equal to the sum of the absolute
values of the component elasticities.
Table 81: Overall (short run) direct elasticity estimates (at 12 months after service etc change)
Attribute Overall best estimatea Typical rangeb
c,d
Fare levels -0.35 -0.2 to -0.6
e
Service levels +0.45 +0.2 to +0.7
f
In-vehicle time -0.40 -0.1 to -0.7
g
Total generalised cost -1.30 -0.8 to -2.0
Notes:
a. These are best estimate short-run elasticities for each attribute for typical urban public transport
journeys, averaged over all market segments and time periods. More disaggregated estimates,
as given in Table 82 should be used where information is available. Positive values indicate that
demand increases when the attribute increases; negative values indicate the opposite (eg fare
increases result in reduced demand). The ‘short-run’ here refers to the impacts roughly 12
months after the change in the service attribute.
b. Represents the typical range of elasticity values found across different locations and market
segments/time periods (refer Table 82 for further details).
c. All fare elasticity estimates relate to fare changes in real terms (ie after netting off any effects of
inflation on fare levels).
d. In situations with competing PT modes or services, the estimates given here assume that the
fares on all such modes/services are adjusted in the same proportions (ie these are ‘conditional’
elasticities).
e. The service level attribute is often calculated as the number of in-service bus kilometres in the
area of interest. For situations where the route structure is unchanged but the levels of service on
the existing routes are adjusted, the service frequency (number of bus trips per hour) may be
taken as the measure of service level.
f. In-vehicle time may be taken as being the time that the ‘typical’ passenger spends on the service,
between initial vehicle boarding and final vehicle alighting.
In practice, the total generalised cost may not always include all journey attributes, depending on the
attributes of interest (ie the elasticity may be subject to change).
Stage 2b: Public transport direct elasticities – by market segment
This section provides disaggregated information on the variation of typical (short-run) elasticities for fares,
service levels and in-vehicle time (as given in Table 81)Table60 by trip characteristics, service
characteristics and type of service change (refer Table 82 below).
Note: Table 83 provides additional information disaggregating fares and service level elasticities between
peak period and off-peak periods, relative to the all periods average elasticities as in Table 81. Weekday
off-peak elasticities are around twice peak period elasticities and weekend elasticities are generally
higher than weekday off-peak values. It is recommended that the Table 81 relativities be used in the
absence of any segment-specific information.
Table 82: Summary of evidence on component elasticities for key variables

Aspect Elasticity variable
Elasticity variable Aspect Elasticity variable
Trip Off-peak/non-work Off-peak/non-work typically Inconclusive re relative
purpose/ typically about twice about twice peak/work; elasticities, although most
time period peak/work; weekend most weekend most elastic (may be evidence is that off-peak is
elastic partly due to frequency more elastic than peak
differences)
Mode Bus elasticities typically No evidence of significant Bus elasticities typically lower
somewhat greater than rail differences (apart from than rail (reflecting longer trips
(but largely reflects shorter variations with headway) by rail with in-vehicle time a
bus trip lengths) greater proportion of
generalised costs)
Base level of Elasticities increase with Elasticities increase with No firm evidence, although
variable base fare level, but less than headways, but less than expect elasticities to increase
proportionately proportionately. Typical values with proportion of total trip
are around 0.2 for frequent (generalised costs) spent in-
services (10 mins or better), vehicle
increasing to around 0.5–0.6
for infrequent services (hourly
or less)
Trip Highest at very short Highest at short distances Limited evidence – longest
distance distances (walk), lowest at (walk alternative) trips more elastic than
short/medium distances, short/medium distance trips
some increase and then
decrease for longest
distances (beyond urban
area)
City size Lower in larger cities (over Higher in larger cities – EU No evidence
1 million population) – US evidence
evidence
Magnitude No significant differences No significant differences in No evidence
of change in elasticities with elasticities with magnitude of
magnitude of change (most change (most studies)
studies)
Direction of No significant differences Evidence indicates no No evidence
change for fare increases and significant differences between
decreases (most studies) service level increases and
decreases
Notes:
• Trip purpose/time period. Strong systematic variations in elasticities exist between trip
purposes and time periods (these two aspects being strongly correlated for all three variables).
• Mode. The literature indicates some differences between modes, for all three variables. However,
these differences appear largely to reflect differences in other attributes (eg trip length, service
frequency) rather than being intrinsic to the different modes.
• Base level of variable. Both fare elasticity and service elasticity vary strongly, although rather
less than proportionately, with the magnitude of the base fare or service frequency. This is
particularly significant in regard to service frequencies: a typical service elasticity would be
around 0.2 at high frequencies (every 10 minutes or better) increasing to around 0.5 or 0.6 or
more at lower frequencies (hourly or longer). These variations are broadly consistent with a
constant generalised cost elasticity formulation.
• Trip distance. Elasticities vary in a complex way with trip distance: this can be explained, in part,
by the availability of substitutes, with high elasticities for short trips having the alternative of
walking and, in part, by the importance of the component measure in the total trip generalised
cost.
• City size. Elasticities vary with city size, although the fare effect and the service level effect
appear to be opposite. However, data relating to this issue is rather limited.
• Magnitude and direction of changes. Most studies show no significant differences in fare
elasticities between fare increases and decreases, or between large and small fare changes.
Similarly, the limited evidence on service elasticities suggests no significant differences in
elasticities between service increases and decreases, or between large and small service
changes.
Table 83: Typical fare and service level (short-run) relative elasticities by time period
Time period Elasticities relative to overall average
Service levels Fares
All 100 100
Weekday peak 65 75
Weekday interpeak 100
110
Weekday evening 130
Weekend interpeak 150 150

Weekend evening 210 n/a
Sources: Wallis (2004), Wallis (2013)
Notes:
• Figures relative to the all-periods averages given in Table 81.

• The literature indicates some differences between modes, for all three variables. However, these
differences appear largely to reflect differences in other attributes (eg trip length, service
frequency.
Stage 2c: Public transport direct elasticities – shorter and longer run effects
Strategic transport demand models generally assume that travel demand changes to a new equilibrium
level instantaneously in response to changes in the road network, public transport fares, etc. However,
this assumption is far from valid in practice. Typically, in the case of public transport services,
infrastructure, and fare changes, there is a rapid initial patronage response to the change, which then
continues to grow over time but at a gradually decreasing rate. The elasticity values in the prior sections
apply to the situation 12 months after the introduction of the change. This section outlines the extent of
demand responses (relative to the 12-month values) expected within both the initial 0–12 month period
and the ongoing subsequent responses, up to five years or longer.
The term ’ramp-up‘ is often applied, for public transport (and other modes) initiatives, to the pattern of
demand growth over time from the introduction of an initiative until the demand reaches its ‘equilibrium’
state (typically after five years or more). This ’ramp-up‘ effect refers only to the underlying growth in
demand towards equilibrium; any other changes that may occur over the ramp-up period (eg as a result
of changes in demographic or economic factors, or in the transport system) need to be addressed
separately and, as appropriate, added to the ramp-up effect.
Waka Kotahi and Australian research (MRCagney and Ian Wallis Associates 2012; and Wallis 2013)
found that patronage ‘ramp-up’ profiles follow a saturation curve pattern, with the ‘sharpness’ of the curve
being dependent on the type of initiative. The saturation curve that best fits the data in most cases is of
the following form:
Qt = QS * t/(B + t)
Where: t = time since introduction of the initiative
QS = estimated patronage impact (growth) at equilibrium situation (‘saturation’)
Qt = patronage impact at time t
B = constant (reflecting the ‘sharpness’ of the saturation curve, dependent on the type of
initiative). B represents the time at which patronage growth in response to an initiative reaches
50% of its saturation level (when Pt/PS = 0.5, t = B).
Figure 14 shows typical ramp-up profiles for each of the following categories of public transport initiatives
analysed for up to three years (156 weeks) following their introduction:
• route and connectivity changes (all PT modes, including multi-modal) – including new routes,
route variations, new/upgraded stops and stations, park and ride facilities (upper curve in Figure
14: B = 2.2 weeks)
• service frequency changes on existing routes (principally bus mode) – both increases and
reductions (middle curve in Figure 14: B = 6.1 weeks)
• major corridor initiatives (all modes) – including large-scale bus and/or rail corridor improvements,
typically with substantial infrastructure components (lower curve in Figure 14: B = 54.9 weeks).
Figure 14: Typical public transport patronage ramp-up profiles from service changes
Source: MRCagney and Ian Wallis Associates (2012)

For each of the typical profiles Table 84 sets out the numerical data represented in Figure 14:
• The ‘B’ value (ie the number of weeks at which the patronage growth reaches 50% of its
estimated saturation level). This varies from some two weeks for category A schemes up to 55
weeks for category C schemes.
• The proportion of the equilibrium (saturation) patronage growth that occurs by the end of each
quarter or year, for up to 3 years from scheme introduction (data given in the first column under
each initiative category).
• The proportion of the 12 months/52 weeks patronage growth that occurs by the end of each
quarter or year (data given in the second column for each category). These percentages should
be applied to the short run (12 months) elasticity estimates given in Table 82 and Table 83 to
derive estimates required for any specific time horizon following scheme implementation.
• The saturation (long run) patronage changes forecast relative to the 12-month changes (data
given in bottom row of table). It is seen that these are 104% (ie 4% more than the 12-month
figure), 112% and 206% for the three categories.
Table 84: Patronage ramp-up profile data by category of initiative

Initiative category
1. Route and 2. Service frequency 3. Major corridor
connectivity changesb initiativesc
changesa
‘B’ value (weeks) 2.2 6.1 54.9
% of % of end % of % of end % of % of
saturation year 1 saturation year 1 saturation end
year 1
End Q1 (13 weeks) 85 89 69 77 21 43
End Q2 (26 weeks) 91 96 80 90 34 69
End Q3 (39 weeks) 93 98 86 97 43 88
End year 1 (52 weeks) 95 100 89 100 49 100
End year 2 (104 weeks) 97 102 94 106 65 133
End year 3 (156 weeks) 99 104 96 108 73 149
Saturation 100 104 100 112 100 206
Source: MRCagney and Ian Wallis Associates, 2012

Notes:
• Includes new, extended and realigned routes, new/upgraded bus/train/ferry stops, stations and
park and ride.
• Includes service frequency changes (increases and reductions).
• Includes large-scale bus and rail improvements in metropolitan/urban areas.
It is notable that the weight of international evidence in the economic literature is that long-run (ie
saturation) elasticities are typically around twice (in the range 1.5 times to 3.0 times) the short-run
(typically 12 months) values (Balcombe et al 2004; Wallis IP 2004). It is evident from the last point above
that this ratio appears to be valid for the major corridor initiatives (category C – factor 2.06), but that the
long-run: short-run ratios are very much less than this for the other two categories (category A 1.04,
category B 1.12).
The ramp-up profiles recommended in this section are based primarily on experience with bus service
and bus/rail infrastructure changes – very limited evidence is available for other types of PT
enhancements. In the absence of better information, we recommend that for:
• fare changes – adopt the category B ramp-up profile
• major service changes – adopt category C ramp-up profile.
Stage 2d: Impacts of public transport initiatives on demand for alternative modes (diversion rate)
Where public transport services are improved, the additional patronage attracted to the public transport
system may be estimated by various methods, including the use of direct demand elasticities (as outlined
in stages 2a and 2b). This section addresses the sources of this increased patronage (demand), in terms
of the previous mode of travel of the ‘mode switchers’.
Additional patronage may originate from a variety of prior modes and other sources, principally:
• previous car use (as driver or passenger) for the trip in question
• previous active mode use (as pedestrian or cyclist)
• ‘pure’ generated trips also known as ‘induced demand’ (ie the same or a similar trip would not
have been made at all without the public transport improvements).
Two alternative approaches (simple ‘models’) are often used to estimate cross-modal demand effects of
public transport initiatives. These involve the application of:
• diversion rates, or
• cross-elasticity relationships.
The ‘diversion rate’ approach, as described below, is considered more appropriate for application here.9
The ‘diversion rate’ to/from an alternative mode resulting from public transport initiatives is defined as the
proportion of the ‘new’ public transport passengers who previously made the trip in question by the
specified mode (eg as car drivers). In this context, the ‘new’ public transport passengers are those who
did not previously use public transport for their trip.
Research following the implementation of major urban public transport initiatives internationally (including
several projects in New Zealand and Australia) reached the following findings on the previous modes of
travel for users of these new initiatives:
• some 60%–70% of the new initiative users would have previously made the same or similar trip
by public transport
• for the remainder of the new initiative users their previous modes of travel are as summarised in
Table 85.10
These findings relate to typical results for major public transport initiatives. In addition, it should be noted
that:
• For public transport initiatives particularly oriented to attracting motorists, higher car driver
diversion rates are appropriate. These include initiatives such as park and ride facilities and
express bus services, each with diversion rates from car drivers of over 50% and in some cases
as high as 70–80%.11
• For those public transport initiatives with a more ‘social’ focus, lower car driver diversion rates are
appropriate. These include off-peak fare schemes and suburban bus route enhancements. For
such schemes, the diversion rates from car driver may be as low as 20–30%.
Table 85: Prior modes of new public transport passengers resulting from urban public transport
initiatives
Prior mode % of new PT trips Notes
Car (driver/passenger) c.50% Approximately 75% of these previously car
drivers, 25% previously car passengers.
Active modes (walk, cycle, etc) c.10%–15% Depends very much on characteristics of
the PT initiative.
Other modes c.10%–15%
Did not make equivalent trip c.20%–25%
Total new PT trips 100%
Note: The above should be taken as a guide only. Given the considerable range of results round in the
literature, it is recommended that the evaluator should consult with Waka Kotahi staff where good
estimates of the previous modes of new users of a public transport scheme are significant to the overall
evaluation.
More information on the two approaches, their inter-relationships and their relative merits are given in Wallis (2004).
10 More details of international research findings on this topic are given in Australian Transport Assessment and
Planning (2018a).
Initial research following the opening of the Auckland Northern Busway found that 56% of the busway users had
previously used other public transport services in the corridor, 44% were new public transport users for the trip in
question. Of this latter group, around 90% were previous car users (of which three-quarters were car drivers, one-
quarter car passengers).
Stage 2e: Impacts of car travel cost changes on public transport demand (diversion rate)
This section provides advice and recommendations on the proportion of previous car users changing their
travel mode in response to changes in car travel costs who switch to/from public transport. Car travel
costs in this context include fuel prices, other car operating costs, parking charges, toll charges and car
travel (in-vehicle) time.
As in the previous section, the focus is on diversion rates between car and alternative modes rather than
cross-elasticity values, but estimates are provided for the public transport mode only (ie the proportion of
deterred car users who switch to public transport). These estimates can provide the basis for sensitivity
tests on how forecasts of public transport patronage would be affected by plausible changes in car travel
costs.
Table 86 provides recommended values for the diversion rates to/from public transport in response to
changes in the various car travel cost components. The following comments may assist in interpretation
of these recommendations:
• Diversion rates are sensitive to two main factors:

1. The 'competitiveness' of the public transport service offered relative to car travel. For
example, much higher diversion rates apply to CBD-oriented trips than to typical inter-
suburban trips.
2. The ’trip purpose‘, where work trips typically have diversion rates around twice those for non-
work trips. In practice, the trip purpose/time period effect and the public transport service
effect are difficult to separate.
• Diversion rates are lower than average for shorter trips (where walking and cycling are
competitive modes), and higher for longer trips (where any car cost changes, eg resulting from
increases in fuel prices, comprise a relatively high proportion of the total trip generalised costs).
• Diversion rates for time components are believed to be lower than for cost components (although
evidence is limited on this point).
• Long-run and short-run diversion rates are assumed to be similar, although the evidence on this
is inconclusive: it seems likely that, in the longer run, the public transport diversion rates tend to
reduce, as people may well be able to take advantage of a wider range of behavioural changes
(eg change in home or employment location).
It is assumed that these diversion rates are equally applicable to increases and decreases (in real terms)
in car travel time and costs, although there is little evidence on this point.
Table 86: Recommendations on diversion rates to/from public transport from changes in car
travel costs
Car travel cost Typical diversion rates Comments on estimates by market segment
variable (% of deterred car
users switching to PT)
Fuel price/ 30% • Long versus short run: inconclusive, assume equal
vehicle operating
costs • Time period/purpose: peak/work proportion approx.
twice off-peak/ non-work proportion
• PT service quality: higher proportions where high
level/ quality of PT service
• Trip length: higher for longer trips, lower for shorter
trips
Toll charges c.40% • Proportions depend on nature of scheme (all day
versus peak only, etc) and location (primarily CBD
trips, all trips, etc.)
• For area-wide/all-day scheme, would expect similar
diversion rates as for fuel prices/VOC
Parking Dependent on market • Regional CBD, work trips: 75%
charges segment
• Regional CBD, non-work trips and suburban CBD
work trips: 50%
• Other areas: not defined, likely to be much lower.
In-vehicle time 20% • As for fuel prices/VOC (above).
Source: Wallis (2004)
Back to 2.2 Forecasting demand: procedures for travel behaviour change activities >>
Stage 3: Calculate public transport improvements impacts

This section provides guidance on the calculation of impacts on public transport users, arising from
activities that change the attributes of public transport services or infrastructure, and the impact of
diverted road traffic on the wider network. Impacts for all other modes should still be calculated on the
route, network and/or transport system, by quantifying, for the do-minimum and options, the changes that
occur for the factors listed in the stages below when an improvement option is considered.
Note that the impact calculations should include any negative impacts (disbenefits) during
Public transport user impacts can be calculated by estimating the ‘consumer surplus’. This quantifies the
economic benefits or disbenefits experienced by PT users after fare revenues have been accounted for.
Consumer surplus therefore measures the total economic value of the service to passengers.
Consumer surplus benefits can be supplemented or supplanted, where this is justified, to take into
account the calculation of PT user benefits arising from reduced journey time, improved service
frequency, interchange time reductions, and reliability improvements. Public transport service user time
savings are based on the mode-neutral values of value of time (VoT) by trip purpose in Table 14 of this
manual. Depending on circumstances, it is possible for VoT increments for congestion and standing to
both apply to public transport services.
Specific care needs to be taken when calculating reductions in waiting time from increased service
frequencies or changes to transfer times between services, as the VoT is modified according to specific
weights for the perceived effect of these reductions.
For other types of non-price public transport aspects, such as improvements to trip quality, greater
comfort and better facilities, user benefits are based on attribute values which represent the amount of in-
vehicle time (IVT) each attribute represents.
The benefits that have currently (2020) been ascribed standardised monetary values are listed below.
The impacts (ie the differences in the parameter outcomes between the do-minimum and the options) are
ascribed monetary values in Chapter 3 of this manual in the list of stages required to complete an
economic analysis and calculate a BCR.
Parameters other than those listed below can be monetised, but the process and values ascribed to these
parameters must be agreed with Waka Kotahi in writing before they are included in the analysis and
For the purposes of this manual, a crash is a transport related event involving one or more road vehicles
that occurs on the transport network that results in personal physical injury and/or damage to property.
Where road traffic is diverted onto new or improved PT services there is likely to be a reduction in the
quantum of crashes.
To undertake a crash analysis, the appropriate crash rates, crash prediction models and crash reduction
calculation and monetisation of crash numbers and severities for the do-minimum, alternatives and
options, and the crash reductions that can be expected from the alternatives and options under
consideration.
For incremental service improvements, or where a new service is limited in scope, it may be
appropriate to use the compound road traffic reduction benefit value from Table 41 or Table 42, which
includes an allowance for a reduction in crashes per kilometre of vehicle traffic removed from the
network.
Stage 3b: Impact of air emissions on health

hydrocarbons (HCs).
10µm), impacts predominantly on respiratory and cardiovascular systems. Effects can range from
reduced lung function, increased medication use and more hospital admissions, through to reduced life
vehicle emissions, including those generated by PT vehicles with internal combustion engines, on human
health.
Stage 3c: Impact on greenhouse gases (GHG)
Greenhouse gases are pollutants which cause global warming and impact globally eg carbon dioxide
(CO2), black carbon (BC) and methane (CH 4)
dioxide (SO2) and carbon monoxide (CO) are indirect climate pollutants. This means they do not warm
benefits.
emissions.
Stage 3d: Impact of noise and vibration on health

affected by wind, and intervening absorbing and reflecting surfaces, and is reduced with distance.

• exhaust systems
psychological impacts, such as annoyance reaction and other behavioural impacts. While there is no
Stage 3e: Impact on network productivity and utilisation
Changes in travel time estimation
Travel times for PT services may be estimated according to the procedures in Appendix 3: Traffic data
and travel time estimation of this manual. These procedures will be most suitable when the PT services
are expected to share lane capacity with other road users and a model is not available.
Definitions for classifying traffic data and default traffic data values are also provided in Table A45, Table
A46 and Table A47. Where a specific procedure is not given, the travel time shall be determined
according to a recognised procedure compatible with the manuals and procedures referred to in Appendix
1: Demand estimation methods and guidance and Appendix 3: Traffic data and travel time estimation.
The stages for estimating travel time
The flow chart in Figure 15 shows the basic stages for estimating road section travel time when PT
services are expected to share lane capacity with other road users. Note that the stages are slightly
different for intersections.
Figure 15: Flow chart for estimating road section travel time

provided in the Transport modelling checks worksheet.
during evaluation. To adhere to this, it is suggested that the functions implied by the procedures be used
Refer to Appendix 1: Demand estimation methods and guidance and Appendix 3: Traffic data and travel
time estimation of this manual, which sets out the procedures for estimating travel times for the do-
minimum and the options for various road and intersection types.
Changes in time spent waiting for services

In addition to standard travel time savings, PT service users may also experience changes in the time
spent waiting for a service to arrive when service frequencies are improved, or changes in the amount of
time that it takes to transfer between services when facilities are improved. These are known as
‘increased service frequency’ and ’interchange reduction’ benefits respectively.
Both increased service frequency and interchange reduction benefits may be added to the journey time
benefits. Specific procedures for calculating the additional benefits are outlined within section 3.6. These
procedures must be applied to calculate increased service frequency and interchange reduction benefits
as the value of time is weighted to account for users’ perceptions of the time spent waiting.
Refer to section 3.6 of this for information on the calculation and monetisation of vehicle operating costs.
For incremental service improvements, or where a new service is limited in scope, it may be appropriate
to use the compound road traffic reduction benefit value from Table 41 or Table 42 which includes an
allowance for a reduction in vehicle operating costs per kilometre of vehicle traffic removed from the
network.
Stage 3f: Impact on system reliability

Trip journey time reliability is a different type of variability, which is much less predictable to a transport
system user. For example, car drivers who make a particular journey at the same time every day find
some days it takes as little as 20 minutes, and on other days as much as 40 minutes. Hence, when the
car drivers plan their trips, they have to consider not just the expected travel time but also its variability.
PT service users may experience trip journey time reliability impacts while waiting for a service to arrive,
while they are on board a service, or a combination of both. For this reason, specific procedures exist to
calculate PT journey reliability benefits, taking into account the differences in the perception of service
delays when waiting for the service and while on board the service.
Refer to section 3.7 of this manual for information on the calculation and monetisation of reliability
impacts.
Stage 3g: Impact of user experience of the transport system
Public transport users value infrastructure and vehicle features. In addition, a new level of demand as the
result of PT improvement might not be ideal for existing users and should be measured as a disbenefit.
Possible negative effects of demand change on existing transport service users include:
• the proportion of standing passengers is increased

• the probability of being left behind has increased.
Section 3.8 provides a methodology for calculating impacts on PT user experience when vehicles or
facilities are improved. The impact of these improvements is expressed as ‘in-vehicle time’, which is
equivalent to the number of minutes of travel time that would need to be saved to equal the value of the
improvement. This allows monetisation to occur using the value of time.
Refer to section 3.8 of this manual for the monetisation of PT user experience impacts.
Stage 3h: Wider economic impacts

Only the most significant PT infrastructure improvements are likely to generate wider economic impacts
(WEBs). Generally, these would need to change the distribution or density of households and firms within
a major metropolitan area and deliver significant improvements in PT accessibility in order for wider
effects to arise.
WEBs are impacts that can result from transport investment that have been used internationally to
improve transport cost–benefit analysis. They can be thought of as impacts that are additional to the
conventional benefits to transport users. WEBs include productivity, labour supply, imperfect competition,
and land use changes.
Great care is required to ensure that the estimates for WEBs are truly additional to conventional benefits
to avoid double counting. As an example, business travel time savings can result in productivity and
output increases. These are a direct user benefit and any WEBs for increased productivity have to be
additional to these direct user benefits.
In addition to, or in some cases as a consequence of, direct impacts, there can be indirect impacts on the
firms. These are WEBs and can include:
spatial markets
The following wider economic benefits are applicable to PT activities in the New Zealand context:
Stage 4 – Undertake risk analysis for significant unpredictable events
proceed through to project completion, and ideally should involve all the relevant parties.
stage, and then to final estimate of cost (FEC) stage. It is intended that the scope and extent of analysis

collection etc).
Stage 5a: Calculate general costs of public transport do-minimum and options
The costs to government of services include:

• construction costs, including property, for any additional infrastructure required
Land costs
Where land has to be acquired for PT infrastructure, its resource cost shall be assumed to equate to its
market value for activity evaluation purposes.
Road maintenance, renewal and construction cost savings
Some service proposals will provide a cost saving to government if future planned road construction costs
are avoided. Cost savings may also occur if there is a reduction in road maintenance and renewal
expenditure from traffic being removed from the network by the implementation of new PT services.
Government cost savings have the effect of reducing the denominator of the BCR, potentially making a
transport service more attractive.
The proposed transport service and any other options are assessed to determine any planned road
construction savings and any road maintenance and renewal savings that will be made as compared to
the do-minimum roading option.
Care must be taken when claiming a cost saving from future road construction avoided. The year or years
in which the road construction would likely be funded must be assessed.
Note: Normally road construction cost savings should only be claimed if there are significant road traffic
reduction benefits associated with the transport service proposal.
Stage 5b: Calculate bus operating costs:
Where detailed operating cost information is not available, the following standardised bus unit operating
cost rates and guidance may be used to calculate operating costs. The operating costs specifically
exclude any infrastructure costs.
Costing variables and categories
Bus operating costs can be calculated based upon the following three variables, which are summarised in
Table 87:
• the time that the vehicle is in operation – bus hours

• the distance travelled in operation – bus kilometres
• the number of vehicles required to meet peak requirements – buses.
Table 87: Bus operating cost variables

In-service operations Total operations
Bus hours Total time that buses are engaged in All time running between depot and
service operations. start/end of route, and between routes.
In addition to terminus-terminus time, Any extended periods on the road (with
includes short breaks (up to 15 mins driver in charge) additional to in-service
between trips (waiting at termini etc). operations.
May be derived from analysis of vehicle/
driver schedules.
Bus Total distance run by buses in service All distance running between depot and
kilometres operations. start/end of route, and between routes.
May be derived from number of timetabled Any other non-service running (eg to
trips and route lengths. replace broken down buses, driver
training).
Buses Maximum number of buses required in use Additional (‘spare’) buses required in
at any one time on a normal weekday in fleet to allow for operational
order to operate the scheduled services. requirements (breakdowns etc) and
maintenance requirements.
May be derived from analysis of
vehicle/driver schedules.
Cost categories
A range of unit costs can be applied to each operating cost variable to determine the gross operational
costs associated with providing the service. These are exclusive of any administration costs or system
facility costs, such as passenger information and enquiry services, that fall upon any local or regional
authority.
A description of the main bus unit cost categories and their associated variables are set out in Table 88.
Table 88: Unit cost categories and allocation

Unit cost category Cost items included Variable
A. Operating costs – time Drivers – wages and direct on Bus hours (total)
costs
B. Operating costs – distance: Fuel, oil, lubricants Bus kilometres (total) – by
fuel vehicle category
C. Operating costs – distance: Repairs and maintenance, Bus kilometres (total) – by

other wages and direct on-costs, parts, vehicle category
materials and external services,
road user charges, tyres and
tubes
D. Operating costs – vehicles Bus comprehensive insurance, Buses (total) – by vehicle
bus registration and licensing, category
bus cleaning, fuelling, depot
rental and rates
E. Operating costs – overheads Overhead labour – wages/ Percentage mark-up on
salaries and direct on –costs categories A–D
Overheads non-labour
Minor assets (capital charges)
F. Profit margin Profit margin or management fee Percentage mark-up on
categories A–E
G. Capital charges – vehicles Bus assets Buses (total) – by vehicle
category
Unit cost values

Table 89 provides a set of representative unit urban bus operating cost rates, for ‘standard’ size diesel
bus operations. The costs relate to 2009/10 average price levels.
The unit costs given in Table 89 should be regarded as indicative only: it is preferable to use local unit
costs in each region where these are known.
These estimates should also address cost differences:
• between diesel buses and trolley or electric buses, and

• for diesel buses of ‘non-standard’ sizes.
Table 89: Unit cost rates, 2009/10 prices (standard diesel bus)
Cost category Units Cost rate Notes, comments
A. Operating costs – time $/bus hour 22.00
B. Operating costs – distance: $/bus km 0.425 Based on typical diesel
fuel consumption of 37
litres/100km and price of
$1.15/litre.
C. Operating costs – distance: $/bus km 0.452 Includes 0.152 for RUC
other (Type 2 vehicles, 2/11 tonnes
GVM); 0.300 for bus R&M,
tyres and tubes.
D. Operating costs – vehicles $/bus pa 5000
E. Operating costs – overheads % mark-up on 10%
items A–D
F. Profit margin % mark-up on 5% Typical of profit margins on
items A–E competitive urban bus
contracts in Australia.
G. Capital charges – vehicles $/bus pa 36,000 Based on typical new diesel
bus price of $375,000, life 18
years, depreciation rate
12.0% pa (DV), interest rate
7.5% pa (real).
Stage 6 – Discount benefits and costs and calculate BCRs

testing.
For the evaluation of new or improved PT services, a shorter analysis period may be appropriate.
Stage 7: Funding gap analysis
This section provides guidance on the application of funding gap analysis to be used in the appraisal of
public transport options. The funding gap is the level of investment required to ensure that a public
transport service operator obtains a reasonable level of return.
Cash-flow and funding gap analysis is not necessary for determining the BCR of a PT activity, but is a key
component of the decision-making process.
Stage 7a: Calculate service provider costs
Service provider costs are calculated either from industry standard unit costs, or from cost estimates
provided by service providers. The costs include maintenance and operating costs for the new or
increased service.
If costs can be obtained, either from industry standard unit costs or other sources (eg the service
provider) then undertake a full analysis of service provider costs. If the service provider will only disclose
a ‘price’, net of user revenue, for providing the transport service then it can be assumed that the service
provider costs are equal to the ‘price’ plus user revenue for use in the evaluation.
Guidance on the estimation of bus operating costs, excluding infrastructure, is available in stage 5b.
Indicative New Zealand bus industry standard unit operating cost rates are also provided.
Indicative quotes may be used when costs cannot be obtained or calculated, but are most likely to be
used when there is a sole service provider. Estimates of service provider costs are not a commitment to
funding and therefore indicative quotes are acceptable during planning stages.
Service provider costs must be calculated for the do-minimum and for all options considered. These costs
must be exclusive of GST.
Activity costs
Cost details should include any of the following:
• investigation, design and supervision costs

• physical infrastructure construction and land acquisition costs
• vehicle, vessel or rolling stock acquisition costs.
• disruption costs during construction/implementation, if substantial.
• operating and maintenance costs
• costs of decommissioning and salvage values
• environmental mitigation costs
• contingency allowance.
In the case of the do-minimum, costs may include essential rehabilitation.
Where expenditure on an activity has already been incurred, it must still be included in the appraisal if the
item has a market value which can be realised, for example land.
Costs which have been irrevocably committed and have no salvage or realisable value, are termed ‘sunk
costs’ (these may include investigation, design or other costs already incurred), and must not be
accounted for in economic appraisal.
Disruption costs to the service provider during implementation may be included when these are expected
to be substantial.
Operating and maintenance costs
Estimate operating and maintenance costs for the service over the analysis period.
Maintenance costs should include routine and periodic maintenance costs, as well as any refurbishment
and replacement costs that occur in the appraisal period.
Treatment of depreciation
Depreciation is a non-cash item and must not be included as a separate item in the cash flows used to
estimate the net present value of a proposal in the financial analysis. Only actual cash flows associated
with maintenance and asset replacement, which already fully account for the depreciation of capital
assets, are to be included in the analysis.
Treatment of interest
Interest expenses associated with financing an activity often represent an actual cash cost outflow.
Despite this, interest charges should not be included in the annual cash flow as the required rate of return
used in the cash flow analysis already takes account of debt-financing interest.
If interest payments were to be included in discounted cash flows, the interest charges would be double
counted; therefore, the proposal’s funding gap would be overstated and the BCR understated.
Salvage value
In some instances, assets will have a longer lifespan than the appraisal period. The salvage value of
capital assets should be included where:
• items have a market value, and

• there is an alternative use (for example, a bus can provide urban passenger services or could be
used for school services or tours), or
• there is a scrap demand for items.
Stage 7b: Calculate service provider revenue
This section describes the information that should be included in the financial analysis of activity options
that generate revenue. The process for forecasting the revenue of an improved service is different from
that for a new service, and the methods for each type of service are described below.
Existing public transport services

Where there is an existing public transport service, it is the increase in service provider revenue that is
used in calculating the funding gap. Any future funding assistance required will be to facilitate
improvements to the service rather than to fund the existing service.
Using the demand estimate information generated in stage 2, calculate the change in service provider
revenue:
Change in service provider revenue = (Q2 x Pnew) – (Q1 x P1)
where: P1 is the base average fare
Pnew is the proposed average fare
Q1 is the current annual patronage
Q2 is the projected annual patronage.
New public transport services
For a new public transport service, the projected number of new users is multiplied by the proposed
average fare to give the expected annual service provider revenue from a new service.
Using the demand estimate information generated in stage 2, calculate the annual service provider
revenue.
Annual service provider revenue = (Qnew x Pnew)
where: Pnew is the proposed average fare
Qnew is the projected annual patronage.
Stage 7c: Preform cash-flow analysis
A new or improved transport service will usually involve some initial capital expenditure followed by
ongoing annual operating and maintenance costs. These costs are offset to an extent by the annual
revenue. Analysis of this cash flow is used to determine the financial viability of the proposed service.
Net cash-flow
For each year, the net cash flow is calculated as:
Annual net cash flow = (revenue + funding gap) – (capital costs + operating and maintenance costs)
Service provider required rate of return
The annual net cash flows are discounted at the service provider’s desired rate of return.
The weighted average cost of capital (WACC) can be used to estimate the service provider’s desired rate
of return. WACC is the weighted average of the desired return on equity and the (interest) cost of any
debt financing.
The service provider’s WACC should reflect the appropriate risk and norms associated with the industry.
Post-tax rate of return
Analysts should use a post-tax rate of return. Care must be taken to ensure that service provider costs
and revenues are calculated accordingly.
Period of financial analysis
The period of the financial analysis should, if possible, be sufficient to allow projected revenue to offset
the initial capital cost. Care must however be taken to ensure that the analysis period is not unrealistically
long. Uncertainties in demand for the proposed service should also be taken into account when setting
the length of the analysis period.
Stage 7d: Calculate service provider’s funding gap
The funding gap is the deficit in cash flow that needs to be reimbursed by local and central government if
the option is to be financially viable from the service provider’s point of view. This is based on the best
estimate of the service provider’s expected revenue and their desired rate of return.
The funding gap can be defined in a number of different ways:
• as a contribution to the capital cost of the activity (either spread over the construction period or
paid at the end of construction), or
• spread over the first few operating years of the proposal, or

• a combination of these.
Where the funding gap is zero or negative, the activity is commercially viable and no funding assistance
should be required from government.
A positive funding gap is required to operate a subsidised service but does not necessarily mean that
funding assistance is justified from a government (public policy) point of view.
Worksheet 3 of SP 9 New public transport services, contains a table with an inbuilt ‘goal seek’ function
that may be used for determining the funding gap.
A worked example of funding gap analysis is provided in Appendix 8: worked examples.
Funding gap sensitivity tests
The financial analysis required to calculate the funding gap involves making assumptions and estimates
of an uncertain future. Some assumptions may also be subjective in nature. As a result, assessments of
the sensitivity of the funding gap to variations in critical assumptions must be undertaken on the preferred
option.
There are three sensitivity tests that should be performed on the funding gap analysis to estimate upper
and lower bounds. This helps to establish the potential effect of these variations on the present value of
the funding gap of the proposal. The sensitivity tests are set out in Table 90.
Table 90: Sensitivity tests

Sensitivity test variables Description
Service provider An upper and lower bound of the service provider’s required rate of return shall
required rate of return be indicated, along with its effect on the present value of the funding gap of the
proposal.
Timing of capital Where significant capital expenditure is a feature of the proposal, sensitivity
expenditure testing shall include the effect on the present value of the funding gap of varying
the timing of such expenditure.
Period of analysis The effect of varying the length of the period of analysis on the present value of
the funding gap shall be presented.
Stage 8 – Determine the benefit–cost ratios of the options

Refer to Chapter 7 of this manual for the details on sensitivity testing.
4. Evaluation procedures > 4.5 Evaluation of travel demand management activities
4.5 Evaluation of travel demand management activities

This section describes the procedures to be used to evaluate the economic efficiency of travel demand
management (TDM) activities.
Where the road network is working inefficiently, for example where there is congestion, traditionally the
solution has been to provide more road capacity. TDM is about stepping back from this capacity approach
and seeking to achieve the objective of improved mobility by improving the efficiency of the network and
influencing travel behaviour, particularly where, when, how and even if travel is needed at all. It aims to
give people more choice by improving and informing them of their travel options while at the same time
influencing and guiding travel decisions to achieve better outcomes. In effect, it seeks to achieve more
sustainable travel behaviour.
TDM attempts to change travel behaviour and transport demand through provision of options, pricing,
financial incentives, travel information, and enforcement.
There are a range of TDM activities available for use in New Zealand including (but not exclusively):
• transport infrastructure, including for public transport

• intelligent transport systems (ITS)
• traffic management
• priority lanes
• tolling and road pricing
• new or improved public transport services
• freight movement management
• walking and cycling alternatives to car and public transport options
• policing
• traveller information services
• work and study travel policies
• urban parking management
• land use design/management
• improved communications
• education, promotion and marketing.
Most TDM activities will be undertaken as part of a package and/or programme of activities. Many
activities will operate to support and enhance other transport management measures. For instance, if
parking management is introduced at the same time as bus priority measures, it is likely there will be
more use of the enhanced bus services than if bus priority was introduced in isolation. Similarly,
enhanced pedestrian facilities in addition to a bus priority scheme could act as more of an incentive for
public transport use by providing more attractive and direct pedestrian connections to those bus services.
Intelligent transport systems (ITS) activities can include (amongst other options):
• variable message signs, which deliver safety and travel condition messages to vehicle drivers
• real time public transport passenger information systems, which result in travel behaviour
changes, and which in turn potentially result in travel time impacts, decongestion impacts and
other user impacts.
If a TDM package contains substantial infrastructure or public transport components, then a composite
evaluation is necessary. Road infrastructure components of a package should be evaluated using the
procedures in section 4.3 and the public transport and other service based TDM components evaluated
using the procedures in section 4.4. The results are then aggregated, taking care to avoid double
counting of benefits.
For the evaluation of TDM activities involving education, promotion and marketing refer to section 4.6 of
this manual.
Simplified procedures for TDM

A range of simplified procedures available in this manual may be used to evaluate TDM activities with an
undiscounted whole-of-life cost of less than or equal to $15 million. SP11 Walking and cycling, SP12
Travel behaviour, and SP13 Safety promotion are the simplified procedures most relevant to TDM
activities. These procedures have been discussed in detail elsewhere within Chapter 4.
Table 46 (simplified procedure in relation to type of activities) may be referred to select the appropriate
procedure.
Full procedures for TDM

In cases where the simplified procedure assumptions are not appropriate, the full procedures should be
used.
The following table outlines the stages of analysis in the economic efficiency evaluation of TDM activities.
The chapters and sections of this manual that apply to each stage of the analysis are referenced in the
table below.
Table 91: Stages of analysis for TDM

a. the do-minimum Counterfactuals
c. whether the improvement(s) should be part of a package Alternatives and
and/or programme of activities. options
2 Collect data to assess travel impacts: Current section
• target population and Chapter 2:
• demand estimates and modal share Demand estimation
and mode share
• uptake
• level of diversion.
Where possible use transport models to assess the impacts and
undertake calibration of the transport models for the activities under
consideration.
• wider economic impact (employment impact)
by Waka Kotahi.
4 Undertake risk analysis when there are significant unpredictable Chapter 7:
events that may affect or be affected by the improvement activity. Sensitivity and risk
analysis
5 Quantify the costs of the improvement activities including: Section 1.8: Costs
• property
• construction
• risk management; mitigation of external impacts
If there is a service provider, determine service provider costs,
service provider revenue and the funding gap (see section 4.4).
Quantify the net costs to government of the funding gap.
to obtain present values. Then use to calculate the benefit–cost Discounting
ratio(s).
cost ratios
for mutually exclusive options, cost ratios

Generally, the do-minimum for TDM activities shall only include committed and funded transport activities
and work that is absolutely essential to preserve a minimum level of service. However, when TDM
activities are implemented as part of a wider package that includes roading or PT activities, the do-
minimum may need to be specified differently. For more guidance refer to the do-minimum descriptions
within the roading (section 4.3) and public transport (section 4.4) procedures.
Stage 1b: Scale and scope of TDM options
TDM activities, like most economic programmes, will eventually have diminishing marginal benefits. There
is an optimal level of implementation, beyond which incremental costs exceed incremental benefits. TDM
programmes need to track these incremental impacts and limit such programmes before the costs exceed
the benefits.
For example, ridesharing programmes may be extremely cost effective when properly implemented, but
once the potential rideshare market is satisfied there will be little additional benefit from attempting to
expand the rideshare programme with soft measures, eg by sending out more promotional material.
Instead, further expansion may require implementation of additional TDM strategies, such as commuter
financial incentives, to expand the size of the market.
Similarly, cycling improvements can be cost effective where there is latent demand for this mode, but it is
necessary to carefully evaluate investments in cycle paths to ensure that they are cost effective. There
may be better ways to support cycling, such as education and encouragement programmes.
Stage 1c: Packages and/programme activities
Most TDM programmes include a combination of positive and negative incentives. However, there are
cumulative and synergetic impacts, therefore it is important to evaluate TDM activities as a package,
rather than as stand-alone activities or an individual strategy.
The procedure for evaluating road improvement packages involving analysis of the timing of individual
components is not appropriate for TDM packages unless the package contains substantial infrastructure
or public transport components.
There are two types of TDM packages: TDM packages involving substantial infrastructure, and TDM
packages involving travel behaviour change infrastructure.
Stage 2: Demand estimates and modal share

this manual.
TDM activities affect travel behaviour in various ways, including changes in trip scheduling, route, mode,
destination and frequency, plus traffic speed, mode choice and land use patterns. Different types of travel
changes provide different types of impacts, eg a shift from driving to non-motorised travel has significantly
different impacts than a shift to public transport.
In order to evaluate the impacts associated with a TDM activity, it is necessary to estimate the impact that
the activity is likely to have on travel behaviour, including changes in mode share (refer to section 2.1). A
commuter trip reduction activity can reduce vehicle trips to a particular worksite place, if implemented
within a regional TDM strategy that includes components such as road tolling, major public transport
improvements and walking and cycling promotion and facilities improvement. Other types of trips can also
be reduced using appropriate TDM strategies. Land use management strategies such as access
management, smart growth and location efficient planning can reduce per capita vehicle travel in a
specific area. Forecasting these changes in demand is a crucial step in the evaluation of TDM activities.
A well-managed TDM activity can also affect a significant portion of total travel. Comprehensive TDM
activities can achieve cost-effective reductions in private vehicle travel compared with no TDM efforts,
although most activities have only small effects because they focus on particular types of trips (such as
commuting), cover a limited geographic scope, or are limited to strategies that can be implemented by a
particular government agency. The mode share is a function of the difference in generalised costs
between the modes. The relationship can be used in reverse to determine the change in generalised cost
difference that is required to achieve an observed change in mode share.
Because mode share relationships are calibrated to actual behaviour, the generalised cost difference can
be equated to the perceived benefit associated with a given change in mode share. Strategic
transportation planning models contain such mode share relationships.
Stage 3: Calculate TDM impacts
All TDM programmes have the objective of changing travel or travel behaviour. Many TDM analyses
involve users’ perceptions of costs, rather than resource costs as discussed in section 1.7 of this manual.
This requires a consumer surplus approach to measure the value that consumers place on a change in
the price or quality of the goods they consume (travel is considered a ‘consumer good’).
Evaluations of TDM activities must consider not only direct impacts but also the benefits and costs to
individuals and society that may influence transport choice. All impacts should be considered, regardless
of where they occur. Impacts within a particular area or analysis period may be highlighted, but costs and
benefits that occur outside the jurisdiction should not be ignored. For example, a community’s TDM
activity may alleviate traffic congestion and parking demand in adjacent areas. These additional benefits
should be mentioned even if they are not the primary consideration in decision making, since such
benefits may justify support from other levels of government.
Impacts to be considered in the economic efficiency evaluation of TDM activities are (refer to Chapter 3 of
this manual):
• impact on social cost of deaths and serious injuries

• impact on greenhouse gases
• wider economic benefit (productivity)
• wider economic benefit (employment impact)
• wider economic benefit (imperfect competition)
• wider economic benefit (regional economic development)

• other monetised impacts (refer to section 1.7 and Chapter 3 of this manual)
• impacts during implementation/construction (primarily travel time delays).
If a TDM package contains substantial infrastructure or PT components, then a composite evaluation is
necessary. Road infrastructure components of a package should be evaluated using the procedures in
section 4.3 of this manual, while PT components should be evaluated using the procedures in section 4.4.
The results of the component evaluations are to be aggregated taking care to avoid double counting of
benefits.
For any travel behaviour change (TBhC) components in a package, the appropriate composite benefit
value is used to calculate the ‘new user’ benefits for the TBhC target population/area. Section 3.17
provides guidance as to the appropriate evaluation method to calculate benefits for existing users, and for
new users from the population outside the TBhC target population/area, for:

• new or improved walking or cycling facilities
• new or improved roading infrastructure of various types.
The numerator of the BCR for a composite TBhC package is the sum of the TBhC benefits and the non-
TBhC benefits.
Impacts that are financial transfers, such as the impact on business and retail profitability, and property
prices (other than where the change in property price is used as a proxy to value an impact) must not
be included in the economic efficiency calculation.
Impacts to businesses that are not direct travel time or vehicle operating cost impacts, are considered
transfers rather than national economic benefits. However, they can be an important factor in
assembling a strategic case for a TDM activity and obtaining funding for workplace based activity and
they should, therefore, be quantified and reported as part of the non-monetised evaluation (separately
from the economic efficiency calculation).

proceed through to project completion and ideally should involve all the relevant parties.
Costs of TDM activities are the costs to government (Waka Kotahi and local government) and the service
provider costs and revenue (where a service provider is involved). Activity costs include the costs of:

• implementation/construction (including property and supervision)
• disruption costs during implementation/construction
• promotion and education
• maintenance
• operating
• monitoring
• decommissioning.
The estimated costs for investigation and design should be identified separately from those for
implementation. Cost estimates for initial indicative evaluations for TDM activity development funding can
be obtained from past experience or judgement. The cost estimate will need to be refined and the
evaluation reconfirmed based on the completed plan before implementation funding is approved.
The cost of annual expenditure required to maintain the benefits of the TDM package over the analysis
period following completion of the activity should be estimated based on local experience and knowledge.
An activity’s operating cost is the cost of operating the new (or improved) facility or service. This is the
cost to government plus the net cost to the service provider (service provider cost minus service provider
revenue).
The cost of monitoring a TDM activity is not included in the cost–benefit evaluation of an activity, except
where an initial survey is an integral part of the activity and then it should be costed as such.
The marginal cost of carpooling is nearly zero if a vehicle has an extra seat that would otherwise travel
empty (there is a small increase in fuel consumption and emissions). The incremental cost does however
increase if a rideshare vehicle must drive out of its way to pick up passengers, or if a larger vehicle (eg a
van) is purchased just to carry passengers.
Similarly, if a public transport system has excess capacity, transfers from driving to public transport may
have minimal incremental cost. If peak travel results in increased operating costs (including extra
vehicles) then the net cost to government of this must be assessed.
Notes:
• The impact on mode choice of any increase in fare resulting from the purchase of extra vehicles
must also be evaluated.
• If increased patronage results in uncomfortably crowded vehicles, then this disbenefit should be
included in the evaluation.
• Irrespective of the TBhC package composition, the total costs for all components of the package
are included in the denominator of the BCR. Where a new or improved public transport service is
involved, the costs include the ‘funding assistance’ costs (ie the cost that needs to be funded by
local and central government if the activity is to proceed).
• A funding gap analysis is only necessary when a service provider is contracted to deliver a new
or improved service (refer to section 4.4 for the funding gap analysis procedure).
Refer section 1.9 and Chapter 5 of this manual for the details on undertaking discounting.
testing.
For the evaluation of TDM activities, a shorter analysis period is likely to be appropriate.
Note: The numerator of the BCR for a composite TBhC package is the sum of the TBhC benefits and the
non-TBhC benefits.
4. Evaluation procedures > 4.6 Evaluation of education, promotion and marketing activities

Inputs to TDM activity evaluations that should be considered for sensitivity testing include:
• road traffic volumes, particularly model results, growth rates and the assessment of generated
traffic
• transport service patronage or facility users
• maximum user charges estimated from consumer surveys.
For each significant factor the following needs to be listed:
• assumptions and estimates on which the evaluation has been based

• an upper and lower bound of the range of the estimate
• resultant BCR at the upper and lower bound of each estimate.
Refer to Chapter 7 of this manual for detailed information on sensitivity and risk analysis.
4.6 Evaluation of education, promotion and marketing activities

This section describes the procedures to be used to evaluate the economic efficiency of education,
promotion and marketing activities. These activities target travel behaviour change (TBhC) and include
travel planning, education, promotion (including road safety promotion) and marketing related travel
demand management.
TBhC activities tend to result in small impacts to a large number of people. These activities are more
difficult to evaluate than other activities because the impacts tend to be different for each participant,
whereas for road improvement activities most users tend to gain the same benefit. This means there is a
compromise between procedures that accurately reflect all of the different individual responses to TBhC
activities and procedures that are cost effective to use but involve significant approximations and
averaging of the effects on different participants. Waka Kotahi has developed TBhC procedures that
strike a balance between accuracy and cost of application, and these a described in more detail within
this section of the manual.
Simplified procedures for education, promotion and marketing activities

A range of simplified procedures available in this manual may be used to evaluate education, promotion
and marketing activities with an undiscounted whole-of-life cost of less than or equal to $15 million. SP11
Walking and cycling, SP12 Travel behaviour, and SP13 Safety promotion are the simplified procedures
most relevant to TDM activities in general. These procedures have been discussed in detail elsewhere
within Chapter 4.
Table 46 (simplified procedure in relation to type of activities) may be referred to select the appropriate
procedure.
Full procedures for education, promotion and marketing activities

used.
The following table outlines the stages of analysis in the economic efficiency evaluation of education,
promotion and marketing activities. The chapters and sections of this manual that apply to each stage of
the analysis are referenced in the table below.
Table 92: Stages of analysis for evaluation of education, promotion and marketing activities
c. whether the improvement(s) should be part of a package and/or Alternatives and
programme of activities. options
Demand
• demand estimates and modal share estimation and
• uptake mode share
Where possible use transport models to assess the impacts and undertake
calibration of the transport models for the activities under consideration.
3 Measure and monetise the impacts (benefits and disbenefits) for the do- Chapter 3:
minimum and options, including: Benefits
• other impacts that can be monetised – these are not included in
this manual but can be included if there is sufficient supporting
evidence and the approach is accepted by Waka Kotahi.
• property
• construction
If there is a service provider, determine service provider costs, service
provider revenue and the funding gap. Quantify the net costs to
government of the funding gap.
5 If the present value of the total government costs is greater than $1 million Chapter 7:
dollars, undertake a detailed risk analysis. Sensitivity and risk
analysis
obtain present values. Then use to calculate the benefit–cost ratio(s). Discounting
7 Determine the benefit–cost ratios of the options. Chapter 6:
Benefit–cost ratios
8 Where the options being evaluated are mutually exclusive, use incremental Chapter 6:
analysis to select the preferred option. Benefit–cost ratios
9 Perform sensitivity tests on the preferred option to determine how robust Chapter 7:
the calculations are and whether a small change in one of the input Sensitivity and risk
parameters has a large change on the evaluation outcome(s). analysis
10 Verify completeness of information, accuracy of calculations and validity of Current section
assumptions.

Generally, the do-minimum for education, promotion and marketing activities shall only include committed
and funded transport activities and work that is absolutely essential to preserve a minimum level of
service. However, when education, promotion and marketing activities are implemented as part of a wider
package that includes roading or PT activities, the do-minimum may need to be specified differently. For
more guidance refer to the do-minimum descriptions within the roading (section 4.3) and public transport
(section 4.4) procedures.
Stage 1b: Scale and scope of education, promotion and marketing activities options
Overseas experience indicates that the most effective, and lowest cost, method for encouraging people to
change their travel behaviour is to provide them with customised information about what is available
locally. The scale of education, promotion and marketing activities is therefore usually limited, but the
scope of their impacts may be significant. Travel plans targeting workplaces, schools, or households and
communities are one such type of activity where, geographically, the scope is highly targeted but the
resulting travel behaviour changes affect the wider network.
The actual impact on travel is highly dependent on factors such as:
• actual features of the plan

• commitment of the target population
• availability of material that assists people’s understanding of the implications of different forms of
travel behaviour
• availability of suitably trained and experienced people to establish and manage the proposal.
Cost efficiencies and effectiveness are enhanced when school, business, household and community
initiatives are implemented simultaneously rather than separately in an area. These programmes should,
therefore, be implemented by geographic area rather than by type.
Stage 1c: Packages and/programme activities
Education, promotion and marketing activities are most effective at changing travel behaviour when
implemented alongside new or improved PT services or walking and cycling facilities, or in conjunction
with other TDM measures.
Consideration should be given to whether a proposed education, promotion and marketing activity is best
delivered as a part of a package or programme.
this manual.
TBhC activities tend to result in small impacts to a large number of people. These activities are more
difficult to evaluate than other activities because the impacts tend to be different for each participant,
whereas for road improvement activities most users tend to gain the same benefit.
Stage 3: Calculate TBhC impacts

For the TBhC components of a package, the appropriate composite benefit value is used to calculate the
‘new user’ benefits for the TBhC target population/area. Section 3.17 provides guidance as to the
appropriate evaluation method to calculate benefits for existing users, and for new users from the
population outside the TBhC target population/area, for:

• new or improved walking or cycling facilities
• new or improved roading infrastructure of various types.
Impacts to be considered in the economic efficiency evaluation of education, promotion and marketing
activities are (refer to Chapter 3 of this manual):
• impact on social cost of deaths and serious injuries

• impact on greenhouse gases
• wider economic benefit (productivity)
• wider economic benefit (labour supply)
• wider economic benefit (imperfect competition)
• wider economic benefit (regional economic development)
• wider economic benefit (land use change)
• other monetised impacts (refer section 1.7 and Chapter 3 of this manual).
An issue for analyses of travel plans is whether the impacts of TBhC programmes persist or decay after
the completion of the programme.
The following four possibilities exist:
1. impacts decay over time

2. impacts can be maintained by ongoing ‘maintenance’ expenditure
3. impacts are durable without maintenance
4. impacts increase over time.
Impact decay due to reversion to old travel modes is more likely in cases where people are persuaded to
change to a less convenient travel option because it is more environmentally sustainable.
Evaluations of household and community TBhC projects can generally assume that improvements will be
retained in future years with little or no maintenance expenditure. However, workplace travel plans and
school travel plans are more likely to require ongoing maintenance expenditure. The school travel plans
have to estimate staff and student turnover. In the case of workplace travel plans some of this
maintenance expenditure will become part of the companies’ cost of business, but in the case of school
travel plans any ongoing expenditure needs to be estimated and included in the evaluation.
Stage 4: Calculate costs of TBhC activity
If a TBhC activity is implemented as part of a package alongside new or improved PT services (section
4.4) or walking and cycling facilities (section 4.2), or in conjunction with other TDM measures (section
4.5), then the costs of the components must be calculated according to the relevant procedure.
The availability of suitably trained and experienced people to establish and manage travel plans is an
important aspect of TBhC activities and this can be a significant cost which must be included.
The cost of annual expenditure required to maintain the benefits of travel plans over the analysis period
following completion of the activity should be estimated based on local experience and knowledge. For
household-based or community-based activities this is generally zero unless the activity contains specific
plans for follow-up measures. For workplace and school travel plans it is likely that some ongoing
maintenance expenditure will be required to maintain the benefits.
Irrespective of the TBhC package composition, the total costs for all components of the package are
included in the denominator of the BCR. Where a new or improved public transport service is involved,
4. Evaluation procedures > 4.7 Evaluation of freight activities
the costs include the funding assistance costs (ie. the costs that needs to be funded by local and central
government if the activity is to proceed).
Road construction, maintenance and operating cost savings are assumed to be negligible for the number
of private vehicle trips and/or vehicle kilometres that are likely to be removed by the implementation of
education, promotion and marketing activities.
Stage 5: Risk analysis
Detailed risk analysis and risk management are undertaken for education, promotion and marketing
activities that have a net cost to government of $1 million or more. These risk processes start at the
project inception stage and proceed through to project completion and should involve all relevant parties.
Refer to Chapter 7 for detailed information on risk analysis.
testing.
For the evaluation of TBhC activities, a shorter analysis period is likely to be appropriate.
Refer to Chapter 7 of this manual for detailed information on sensitivity on sensitivity testing. Possible
significant factors that should be considered for sensitivity testing include:
• demand estimates (refer to Chapter 2 of this manual for more details on sensitivity testing of
demand estimates)
• impact changes
• major contributors to impacts
• commencement of the proposal.
4.7 Evaluation of freight activities

freight service activities and high productivity motor vehicle (HPMV) routes. Activities may be stand-alone
improvements or a component of a package or a wider programme of transport improvements.
Simple and relatively standardised HPMV route improvements, with an undiscounted whole-of-life cost of
less than or equal to $15 million, may be evaluated using a simplified procedure. Similarly, a separate
simplified procedure may be used to evaluate new rail or sea freight services with an undiscounted
funding gap of less than or equal to $15 million over a three-year period. Both of these simplified
procedures are explained in more detail below. For more complicated activities, and activities that breach
the cost limits of the simplified procedures, the full procedures are provided as an alternative and are
explained later in this section.
Simplified procedures for freight

The following simplified procedures (SPs), for HPMV route improvements (SP6) and new rail or sea
freight services (SP8), use a 4% discount rate and a 40-year analysis period. The procedures assume
that activities will be completed within the first year and will be in service by the start of year two. Where
costs are common to both the do-minimum and the options they are not included in the analysis. All costs
are to be exclusive of GST.
Refer to the Planning and Investment Knowledge Base (PIKB) for guidance on issues relating to analysis
of PT activities, including selection of the preferred option using the Business Case Approach.

Analysis that alters components of the simplified procedure should not be used as this will compromise
the assumptions on which the procedures are based and full procedures should be used instead.
[email protected].
SP6 for high productivity motor vehicle (HPMV) route improvements
Procedure SP6 provides a simplified method of evaluating the economic efficiency of HPMV routes and
associated capital, maintenance and renewal costs. The procedure’s benefits are primarily derived from a
reduction in heavy vehicle trips, with associated reductions in vehicle operating costs, greenhouse gas
emissions, crash costs, and travel time costs.
It is assumed that the route from which heavy vehicles are removed is primarily rural, with a minimal
number of intersections. If the route includes a significant proportion of travel in urban areas, the crash
cost savings procedures described in Appendix 2: Crash analysis should be applied instead.
Only the additional costs required to allow passage of HPMVs on identified routes are included within this
simplified procedure. Where an HPMV activity will bring forward or increase planned maintenance or
bridge work these associated costs are redistributed accordingly within the cost tables. In some cases (eg
where pavements are weak), it may be necessary to compare the costs of the freight transport option with
a road reconstruction option for the affected road network.
Guidance for completing the SP6 High productivity motor vehicles (template worksheets) is provided
Table 93: SP6 HPMV route improvements procedure template

number
considered for the evaluation plus the results
of the economic analysis.
SP6-2 Cost of the option(s) Used to calculate the PV costs of the option.
A separate worksheet is required for each
option evaluated. Up to two options can be
evaluated.
SP6-3 Vehicle operating cost savings, CO2 Used for calculating vehicle operating cost
reductions and travel time cost savings savings, CO2 reduction and travel time cost
savings.
SP6-4 Crash cost savings Used for calculating crash cost savings
using crash rate analysis method (refer
Appendix 2: Crash analysis).
considered.
Table 94: Steps in the SP6 evaluation of HPMV route improvement activities
Step Description
SP8 for freight services

Procedure SP8 provides a simplified method of evaluating the economic efficiency of rail and sea freight
transport services, with or without capital expenditure.
The procedure assumes:
• There are costs to users that are additional to, and offset the difference between, road and rail or
sea freight rates.
• The primary benefits are road maintenance, renewal and improvement cost savings (net of road
user charges), and road traffic reduction benefits (mainly CO 2 and crash cost savings) from the
removal of freight from the road network.
• Other benefits (positive or negative) are not significant. Allowance can be made for additional
benefits if they are found to be significant.
• The route from which heavy vehicles are removed is primarily rural, with a minimal number of
intersections.
If the route includes a significant proportion of travel in urban areas, the crash cost savings procedures
described in Appendix 2: Crash analysis should be applied instead.
It is necessary to complete a funding gap analysis for the proposed service in worksheet SP8-2. A 12%
service provider rate of return is assumed, and the funding gap may be calculated using the embedded
goal seek function.
Worksheet SP8-8 provides a feasibility evaluation using costs that are internalised to the service provider
plus a composite value for non-internalised costs for road freight transport and for sea or rail transport.
This may be used for activities without specific crash or congestion issues on the affected roads.
Guidance for completing the SP8 Freight services (template worksheets) is provided below in Table 95
and Table 96.
Table 95: SP8 Freight services procedure template

number
SP8-4 Freight service user benefits Used to calculate the PV of benefits for
freight service users.
SP8-5 Net cost savings to government Used to calculate the PV of net government
cost savings, including road user charges
forgone.
SP8-6 Road traffic reduction benefits Used to calculate the PV of benefits for
other transport system users.
considered.
SP8-8 Feasibility evaluation Used as an alternative appraisal
methodology in the absence of congestion
or crash cost savings.
Table 96: Steps in the SP8 evaluation of freight service activities
Step Description
4 Complete Worksheet 4 – Freight service user benefits
5 Complete Worksheet 5 – Net cost savings to government
6 Complete Worksheet 6 – Road traffic reduction benefits
6 Complete Worksheet 8 – Feasibility evaluation (optional)
Full procedures for freight services

The full evaluation procedures for freight activities are to be used to appraise the economic efficiency of
activities when the simplified procedures are not appropriate or sufficient.
The primary purpose of this section of the manual is to establish the impacts of introducing new freight
services or improving existing services (ie the changes that occur between do-minimum and the options)
when using the full procedures. Following on from calculating the impacts, the analyst will need to assign
monetary values to the impacts and then calculate the benefits and the benefit–cost ratios (BCRs).
which links have been provided. A significant focus of the freight service procedures is on the calculation
of the net cost savings to the government and the service provider’s funding gap, in particular stages 3 to
6 in Table 97.
The following table outlines the stages of analysis when undertaking an evaluation of the impacts of
introducing a new freight service or improving an existing one. The chapters and sections of this manual
that apply to each stage of the analysis are referenced in the table below.
Table 97: Stages of analysis for freight activities

Stage Description See
b. improvement alternatives and options
Section 1.5:
c. whether the improvement(s) should be part of a package
Alternatives and
and/or programme of activities.
options
2 Forecast the freight demand either from a transport model or by Current section and
using demand elasticities. Chapter 2: Demand
estimation and mode
share
• wider economic impact (labour supply)
• wider economic impact (and use change)
sufficient supporting evidence and the approach is
accepted by Waka Kotahi.
4 Undertake risk analysis when there are significant unpredictable Chapter 7: Sensitivity
events that may affect or be affected by the improvement activity. and risk analysis
5 Calculate the costs to the government of services for the do- Current section and
minimum and improvement options, including (but not exclusively): Section 1.8: Costs
• forgone road use charges (RUC)
• construction costs, including property, for any additional
infrastructure required
• maintenance costs not already included in service
contracts.
to obtain present values. Discounting
7 Calculate the funding gap and any net cost to government by: Current section
• calculating service provider costs
• calculating service provider revenue
Stage Description See
• performing cash-flow analysis

• calculating the service provider’s funding gap
• performing sensitivity tests on the funding gap analysis.
cost ratios
10 Perform sensitivity tests on the preferred option to determine how Chapter 7: Sensitivity
robust the calculations are and whether a small change in one of and risk analysis
the input parameters has a large change on the evaluation
outcome(s).

The do-minimum for evaluation of freight services is usually considered as a continuation of the present
transport networks, service levels and the existing PT network in the study area.
The do-minimum must include any costs and resulting demand implications of committed transport
infrastructure and freight service improvements during the analysis period. Activities are committed if they
have been evaluated in accordance with the Waka Kotahi evaluation procedures and have been
approved for funding. Any investment plans that are not committed must be included in the evaluation as
options. Maintenance, renewal/replacement schedules and any planned freight service changes must
also be included.
be the ‘preferred option’.
in mind.
Stage 1c: Packages and/or programme of activities
• neglecting options that differ in type or scale, eg an extension of an existing HPMV route may
eliminate the need improve alternative routes
• neglecting significant externalities, eg the impacts of a new business park on future demand
for freight services
• inconsistencies with wider strategic policies and plans, eg investing in road improvements or
renewals when significant rail network renewals are planned.
If all transport system users are affected by improvements, then a multi-modal evaluation may be
necessary. This may involve analysing freight components and other transport system users separately,
using the relevant procedures in this section and the rest of Chapter 4, and aggregating the results.
this manual.
For freight service investments that are limited in scope or scale it may be appropriate to use the
standardised cross-price elasticities of demand between road and rail. The values in Table 98 are
indicative only and represent the percentage change in rail volume with respect to the percentage change
in rail to road price.
Table 98: Elasticities for freight commodities

Commodity Range
Food and kindred products -1.04 to -2.58
Lumber and wood products -0.05 to -1.97
Paper products -0.17 to -1.85
Machinery -0.16 to -2.27
The elasticities depend primarily on the level of inter-modal competition. In New Zealand, where inter-
modal competition is likely to be significant, it is considered that freight price elasticities would more likely
be at the higher end of the ranges identified above. However, it should be noted that other factors may
influence a shipper’s decision. Transit time (generally used as a proxy for distance) appears to be a
significant determinant of mode choice, therefore, the greater the distance, the less likely truck transport
will be chosen.
For significant freight service investments, or where there is significant interaction between freight and
other transport system users, the use of models is encouraged.
Stage 3: Calculate freight service improvement impacts
This section provides guidance on the calculation of impacts on freight service users and the impact of
diverted road traffic on the wider network. Impacts for all other modes should still be calculated on the
route, network and/or transport system, by quantifying, for the do-minimum and options, the changes that
occur for the factors listed in the stages below when an improvement option is considered.
Note that the impact calculations should include any negative impacts (disbenefits) during
Freight service user impacts can be calculated by estimating the ‘consumer surplus’. This quantifies the
economic benefits or disbenefits experienced by freight service users after freight haulage rates have
been accounted for. Consumer surplus therefore measures the total economic value of the service to the
service users. Freight haulage rates are usually expressed in $/tonne.
Consumer surplus benefits can be supplemented or supplanted, where this is justified, to take into
account the calculation of freight service user benefits arising from reduced journey time, improved
reliability, and changes to the service quality.
For the calculation of travel time saving benefits, Table 15 contains values of time (VoT), expressed in
$/hour/vehicle, that should be used when vehicles and are used for work purposes to transport freight.
Section 3.6 contains more information on the VoT and adjustment factors that may be applied in highly
congested conditions.
Care needs to be taken to include any additional user costs, such as re-handling or inventory adjustment
costs, as a disbenefit in the evaluation. Additionally, user benefits for freight should also take into account
flexibility in options for frequency of transport and choice of service providers. In some cases, users
transferring freight from road to a rail or sea transport service mode will experience reduced flexibility in
the timing and route of services compared with using a road option. Any such reduced flexibility for the
transport service user must also be included as a disbenefit in evaluations.
The benefits that have currently (2020) been ascribed standardised monetary values are listed below.
The impacts (ie the differences in the parameter outcomes between the do-minimum and the options) are
ascribed monetary values in Chapter 3 of this manual in the list of stages required to complete an
economic analysis and calculate a BCR.
Parameters other than those listed below can be monetised but the process and values ascribed to these
parameters must be agreed with Waka Kotahi in writing before they are included in the analysis, and
For the purposes of this manual, a crash is a transport related event involving one or more road vehicles
that occurs on the transport network that results in personal physical injury and/or damage to property.
Where freight is diverted from road-based services to rail or shipping, there is likely to be a reduction in
the quantum of crashes. Likewise, a new or improved HPMV route may reduce the quantum of crashes if
the number of heavy vehicle trips is reduced.
To undertake a crash analysis, the appropriate crash rates, crash prediction models and crash reduction
calculation and monetisation of crash numbers and severities for the do-minimum, alternatives and
options, and the crash reductions that can be expected from the alternatives and options under
consideration.
For minor route improvements, or where a new service is limited in scope, it may be appropriate to use
the compound road traffic reduction benefit value from Table 41 or Table 42 which includes an
allowance for a reduction in crashes per kilometre of vehicle traffic removed from the network.
Crash rates for rural freight services

For a freight transport service activity, where the road network affected by the activity is primarily rural in
location, crash rate equations for heavy vehicles only are used to estimate the reduction in freight-related
crashes.
Each freight route should be broken down by traffic volume and terrain type. The terrain type can be
selected by analysing the route gradient data. The gradient bands for each terrain type should generally
be maintained throughout each section. Sections of road that are less steep can occur in rolling or
mountainous sections for short lengths. This is allowed provided that the lower gradient length is followed
by another rolling or mountainous gradient. The appropriate crash rate is then used for each section.
Stage 3b: Impact of air emissions on health
hydrocarbons (HCs).
10µm) impacts predominantly on respiratory and cardiovascular systems. Effects can range from reduced
lung function to increased medication use and more hospital admissions through to reduced life
vehicle emissions, including those generated by PT vehicles with internal combustion engines, on human
health.
Stage 3c: Impact on greenhouse gases (GHG)
Greenhouse gases are pollutants which cause global warming and impact globally, eg carbon dioxide
(CO2), black carbon (BC) and methane (CH 4).
dioxide (SO2) and carbon monoxide (CO), are indirect climate pollutants. This means they do not warm
benefits.
emissions.
the compound road traffic reduction benefit value from Table 41 or Table 42 which includes an
allowance for a reduction in GHG emissions per kilometre of vehicle traffic removed from the network.
Stage 3d: Impact of noise and vibration on health

affected by wind, and intervening absorbing and reflecting surfaces, and is reduced with distance.

• exhaust systems
psychological impacts, such as annoyance reaction and other behavioural impacts. While there is no
Stage 3e: Impact on network productivity and utilisation
Changes in travel time estimation
Travel times for freight services may be estimated according to the procedures in Appendix 3: Traffic data
and travel time estimation of this manual. Definitions for classifying traffic data and default traffic data
values are also provided in Appendix 3: Traffic data and travel time estimation. Where a specific
procedure is not given, the travel time shall be determined according to a recognised procedure
compatible with the manuals and procedures referred to in Appendix 1: Demand estimation methods and
guidance and Appendix 3: Traffic data and travel time estimation.
Figure 18 shows the basic stages for estimating road section travel time. Note that the stages are slightly
different for intersections.
Figure 18: Flow chart for estimating road section travel time

provided in the Transport modelling checks worksheet.
during evaluation. To adhere to this it is suggested that the functions implied by the procedures be used
Refer to Appendix 1: Demand estimation methods and guidance and Appendix 3: Traffic data and travel
time estimation of this manual, which sets out the procedures for estimating travel times for the do-
minimum and the options for various road and intersection types.
Refer to section 3.6 of this for information on the calculation and monetisation of vehicle operating costs.
the compound road traffic reduction benefit value from Table 41 or Table 42, which includes an
allowance for a reduction in vehicle operating costs per kilometre of vehicle traffic removed from the
network.
User costs
Where freight service users incur additional handling, inventory or other related costs, or face decreased
flexibility in freight schedules, these must be accounted for in the evaluation as disbenefits. This manual
does not include standardised values for the disbenefits.
Stage 3f: Impact on system reliability
Trip journey time reliability is a different type of variability, which is much less predictable to a transport
system user. For example, car drivers who make a particular journey at the same time every day find
some days it takes as little as 20 minutes, and on other days as much as 40 minutes. Hence, when the
car drivers plan their trips, they have to consider not just the expected travel time but also its variability.
Journey time reliability is measured by the unpredictable variations in journey times, which are
experienced for a journey undertaken at broadly the same time every day. The impact is related to the
day-to-day variations in traffic congestion, typically as a result of day-to-day variations in flow. This is
distinct from the variations in individual journey times, which occur within a particular period.
Journey time reliability is in principle calculated for a complete journey and the total network variability is
the sum of the travel time variability for all journeys on the network. In practice, models may not represent
the full length of journeys and this is accounted for in the reliability procedure for road based activities in
section 4.3.
Stage 3g: Impact of user experience of the transport system
There are two standard values related to this benefit in this manual as follows:
1. Impact on driver frustration derived from ‘time spend passing’. This is an indicator of changes that
passing lane generate in road users experience.
2. Impact on road users’ comfort and productivity due to sealing unsealed roads.
Driver frustration
Vehicle passing options may be provided through the construction of dedicated passing lanes, climbing
lanes, slow vehicle bays, and improved alignments.
Providing passing options releases vehicles from platoons of slower moving vehicles, allowing them to
travel along the road at their desired speed until they are once again constrained by platoons. Typically,
the evaluation of passing options has been undertaken by micro-simulation programmes, which use
various vehicle performance models together with terrain data to establish, in detail, the speeds of
vehicles at each location along the road. These assessments can be excessively complex, particularly
given the general magnitude of such activities.
The demand for passing and consequently the benefits, are a function of a number of parameters
including:
• traffic variables
o traffic volume
• road variables
o terrain/alignment
o grades
An alternative method is based on multiple simulations and the unified passing model described in
Appendix 5: Passing lanes is available in the Provisional passing & overtaking guidelines on the Waka
Kotahi website. This method can be used to identify the most appropriate strategy for providing improved
vehicle passing options over a route and assess the benefits of individual vehicle passing options within
those strategies.
Road user comfort from seal extension
into account. Simplified procedure SP4 Seal extensions provides information on productivity gains. A
value of 10 cents per vehicle per kilometre can be used for road user comfort, which takes account of the
other benefits associated with avoiding unsealed roads.
Stage 3h: Wider economic impacts
Only the most significant infrastructure improvements are likely to generate wider economic impacts
(WEBs). Generally, these would need to change the distribution or density of households and firms within
a major metro area or deliver significant improvements in accessibility between regions in order for wider
effects to arise.
WEBs are impacts that can result from transport investment that have been used internationally to
improve transport cost–benefit analysis. They can be thought of as impacts that are additional to the
conventional benefits to transport users. WEBs include productivity, labour supply, imperfect competition,
and land use changes.
Great care is required to ensure that the estimates for WEBs are truly additional to conventional benefits
to avoid double counting. As an example, business travel time savings can result in productivity and
output increases. These are a direct user benefit and any WEBs for increased productivity have to be
additional to these direct user benefits.
In addition to, or in some cases as a consequence of, direct impacts, there can be indirect impacts on the
firms. These are WEBs and can include:
spatial markets
The following wider economic benefits are applicable to freight activities in the New Zealand context:
Stage 4 – Undertake risk analysis for significant unpredictable events
proceed through to project completion, and ideally should involve all the relevant parties.
stage, and then to final estimate of cost (FEC) stage. It is intended that the scope and extent of analysis

collection etc).
Stage 5a: Calculate general costs of freight service do-minimum and options
The costs to government of services include:

• construction costs, including property, for any additional infrastructure required
Land costs
Where land has to be acquired for infrastructure, its resource cost shall be assumed to equate to its
market value for activity evaluation purposes.
Road maintenance, renewal and construction cost savings
Some service proposals will provide a cost saving to government if future planned road construction costs
are avoided. Cost savings may also occur if there is a reduction in road maintenance and renewal
expenditure from traffic being removed from the network by the implementation of new freight services or
HPMV routes.
Government cost savings have the effect of reducing the denominator of the BCR, potentially making a
transport service more attractive.
The proposed freight service and any other options are assessed to determine any planned road
construction savings and any road maintenance and renewal savings that will be made as compared to
the do-minimum roading option.
Care must be taken when claiming a cost saving from future road construction avoided. The year or years
in which the road construction would likely be funded must be assessed.
Note: Normally road construction cost savings should only be claimed if there are significant road traffic
reduction benefits from freight diverting to rail and sea services.
Road maintenance, renewal and improvement cost savings associated with implementation of a freight
service are calculated by estimating the total annual amount of freight traffic, measured in terms of
equivalent design axles (EDA), removed from the road network. The simplified procedure for freight
services SP8 Freight services provides indicative EDA and $/EDA/km values. However, local values are
to be used for activities where the default values provided in these simplified procedures do not represent
local conditions. If the amount of the freight traffic removed from the road network varies from year to
year, separate calculations are required for each year.
Stage 5b: Calculate road user charges forgone
In New Zealand, road user charges (RUC) are levied against all diesel-vehicles and vehicles over 3.5
tonnes. For the purposes of this manual, it is assumed that all vehicles used in freight services will be
paying RUC.
In the case of a freight service, lost RUC are subtracted from the road maintenance, renewal and
construction cost savings to derive the net savings to government. It is assumed that heavy commercial
vehicles will be removed from the road as a result of new or improved HPMV routes, or a new freight
service. Thus, the loss of RUC as a result of the introduction of a freight transport service will be based on
the weighted average road user charge for the type of vehicle that is removed.
Determine the reduction in RUC revenue as a result of the introduction of a freight service using the
following procedure in Table 99.
Table 99: Calculate reduction of road user charges (RUC) revenue

Step Action
1 From the demand estimate information generated in stage 2, list the following for each travel
time period:
• the existing number of road trips by the vehicle type affected by the transport service
proposal, and
• the predicted new level of road trips by the vehicle type affected by the transport
service proposal.
Note: The travel time period used will depend on the particular freight service being
proposed, but in most cases will probably be an annual figure.
2 Determine the change in road trips by subtracting the existing number of road trips from the
predicted new level of road trips.
3 Using the data from step one and consulting with the industry(ies) affected by the proposed
freight service, determine the average licensed weight of the vehicle type(s) removed.
4 Using the RUC tables published by Waka Kotahi, establish the RUC (in $/1000km) for the
licence weights of the vehicles removed.
5 Determine the length (km) of the road(s) affected by the proposed transport service.
6 Calculate the total number of kilometres of travelling saved:
(Change in road trips per annum) × (km per trip)
Divide this by 1000 to find the annual thousands of kilometres saved.
7 Multiply the road user charge ($/1000km) by the annual thousands of kilometres saved to
derive the total RUC revenue lost.
Stage 6 – Discount benefits and costs and calculate BCRs

testing.
For the evaluation of new or improved freight service, a shorter analysis period may be appropriate if the
freight carried is a non-renewable resource with a finite life.
Stage 7: Funding gap analysis
This section provides guidance on the application of funding gap analysis to be used in the appraisal of
freight service options. The funding gap is the level of investment required to ensure that a freight service
operator obtains a reasonable level of return.
Cash-flow and funding gap analysis is not necessary for determining the BCR of a freight service activity,
but is a key component of the decision-making process.
Stage 7a: Calculate service provider costs
Service provider costs are calculated either from industry standard unit costs, or from cost estimates
provided by service providers. The costs include maintenance and operating costs for the new or
increased service.
If costs can be obtained, either from industry standard unit costs or other sources (eg the service
provider) then undertake a full analysis of service provider costs. If the service provider will only disclose
a ‘price’, net of user revenue, for providing the transport service then it can be assumed that the service
provider costs are equal to the ‘price’ plus user revenue for use in the evaluation.
Indicative quotes may be used when costs cannot be obtained or calculated, but are most likely to be
used when there is a sole service provider. Estimates of service provider costs are not a commitment to
funding and therefore indicative quotes are acceptable during planning stages.
Service provider costs must be calculated for the do-minimum and for all options considered. These costs
must be exclusive of GST.
Activity costs
Cost details should include any of the following:
• investigation, design and supervision costs
• physical infrastructure construction and land acquisition costs
• vehicle, vessel or rolling stock acquisition costs
• disruption costs during construction/implementation, if substantial
• operating and maintenance costs
• costs of decommissioning and salvage values
• environmental mitigation costs
• contingency allowance.
In the case of the do-minimum, costs may include essential rehabilitation.
Where expenditure on an activity has already been incurred, it must still be included in the appraisal if the
item has a market value which can be realised, for example land.
Costs which have been irrevocably committed and have no salvage or realisable value, are termed ‘sunk
costs’ (these may include investigation, design or other costs already incurred), and must not be
accounted for in economic appraisal.
Disruption costs to the service provider during implementation may be included when these are expected
to be substantial.
Operating and maintenance costs
Estimate operating and maintenance costs for the service over the analysis period.
Maintenance costs should include routine and periodic maintenance costs, as well as any refurbishment
and replacement costs that occur in the appraisal period.
Treatment of depreciation
Depreciation is a non-cash item and must not be included as a separate item in the cash flows used to
estimate the net present value of a proposal in the financial analysis. Only actual cash flows associated
with maintenance and asset replacement, which already fully account for the depreciation of capital
assets, are to be included in the analysis.
Treatment of interest
Interest expenses associated with financing an activity often represent an actual cash cost outflow.
Despite this, interest charges should not be included in the annual cash flow as the required rate of return
used in the cash flow analysis already takes account of debt-financing interest.
If interest payments were to be included in discounted cash flows, the interest charges would be double
counted; therefore, the proposal’s funding gap would be overstated and the BCR understated.
Salvage value
In some instances, assets will have a longer lifespan than the appraisal period. The salvage value of
capital assets should be included where:
• items have a market value, and
• there is an alternative use, or
• there is a scrap demand for items.
Stage 7b: Calculate service provider revenue
This section describes the information that should be included in the financial analysis of activity options
that generate revenue. The process for forecasting the revenue of an improved service is different from
that for a new service, and the methods for each type of service are described below.
Existing freight services
Where there is an existing freight service, it is the increase in service provider revenue that is used in
calculating the funding gap. Any future funding assistance required will be to facilitate improvements to
the service rather than to fund the existing service.
Using the demand estimate information generated in stage 2, calculate the change in service provider
revenue:
Change in service provider revenue = (Q2 × Pnew) – (Q1 × P1)
where: P1 is the base average user charge
Pnew is the proposed average user charge
Q1 is the current annual freight volume
Q2 is the projected annual freight volume.
New freight transport services
For a new freight service, the projected annual freight volume is multiplied by the proposed average user
charge to give the expected annual service provider revenue from a new service.
Using the demand estimate information generated in stage 2, calculate the annual service provider
revenue.
Annual service provider revenue = (Qnew × Pnew)
where: Pnew is the proposed average user charge
Qnew is the projected annual freight volume.
Stage 7c: Perform cash-flow analysis
A new or improved freight service will usually involve some initial capital expenditure followed by ongoing
annual operating and maintenance costs. These costs are offset to an extent by the annual revenue.
Analysis of this cash flow is used to determine the financial viability of the proposed service.
Net cash-flow
For each year, the net cash flow is calculated as:
Annual net cash flow = (revenue + funding gap) – (capital costs + operating and maintenance costs)
Service provider required rate of return
The annual net cash flows are discounted at the service provider’s desired rate of return.
The weighted average cost of capital (WACC) can be used to estimate the service provider’s desired rate
of return. WACC is the weighted average of the desired return on equity and the (interest) cost of any
debt financing.
The service provider’s WACC should reflect the appropriate risk and norms associated with the industry.
Post-tax rate of return
Analysts should use a post-tax rate of return. Care must be taken to ensure that service provider costs
and revenues are calculated accordingly.
Period of financial analysis
The period of the financial analysis should, if possible, be sufficient to allow projected revenue to offset
the initial capital cost. Care must however be taken to ensure that the analysis period is not unrealistically
long. Uncertainties in demand for the proposed service should also be taken into account when setting
the length of the analysis period.
Stage 7d: Calculate service provider’s funding gap
The funding gap is the deficit in cash flow that needs to be reimbursed by local and central government if
the option is to be financially viable from the service provider’s point of view. This is based on the best
estimate of the service provider’s expected revenue and their desired rate of return.
The funding gap can be defined in a number of different ways:
• as a contribution to the capital cost of the activity (either spread over the construction period or
paid at the end of construction), or
• spread over the first few operating years of the proposal, or
• a combination of these.
Where the funding gap is zero or negative, the activity is commercially viable and no funding assistance
should be required from government.
A positive funding gap is required to operate a subsidised service but does not necessarily mean that
funding assistance is justified from a government (public policy) point of view.
Worksheet 3 of SP8 Freight services, contains a table with an inbuilt ‘goal seek’ function that may be
used for determining the funding gap.
A worked example of funding gap analysis is provided in Appendix 8: worked examples.
Funding gap sensitivity tests
The financial analysis required to calculate the funding gap involves making assumptions and estimates
of an uncertain future. Some assumptions may also be subjective in nature. As a result, assessments of
the sensitivity of the funding gap to variations in critical assumptions must be undertaken on the preferred
option.
There are three sensitivity tests that should be performed on the funding gap analysis to estimate upper
and lower bounds. This helps to establish the potential effect of these variations on the present value of
the funding gap of the proposal. The sensitivity tests are set out in Table 100.
Table 100: Sensitivity tests

Sensitivity test variables Description
Service provider required An upper and lower bound of the service provider’s required rate of return
rate of return shall be indicated, along with its effect on the present value of the funding
gap of the proposal.
Timing of capital Where significant capital expenditure is a feature of the proposal, sensitivity
expenditure testing shall include the effect on the present value of the funding gap of
varying the timing of such expenditure.
Period of analysis The effect of varying the length of the period of analysis on the present value
of the funding gap shall be presented.
4. Evaluation procedures > 4.8 Evaluation of private sector financing and road tolling

Refer to Chapter 7 of this manual for detailed information on sensitivity testing.
4.8 Evaluation of private sector financing and road tolling

This section describes the specific procedures to be used to evaluate the economic efficiency of activities
involving private sector financing, and road tolling activities.
Private sector financing and tolling provide alternatives to government-funded transport infrastructure.
Wallis (2005) provides guidance on private sector participation in provision of public infrastructure.
Tolls
Tolls are payment by road users for the right to travel on a particular road. In economic efficiency terms
the tolls can be viewed in three ways:
1. If the facility is government funded, the tolls are simply a transfer payment between those
motorists who pay them and the government.
2. If the facility is privately financed and the concessionaire (with its toll level proposal) is selected
by competitive tendering, then the toll charges also represent a true market price, ie the resource
cost, for that part of the activities. Any government contribution or expenditure is also part of the
activity cost.
3. Alternatively, tolls can be related to negative benefits (disbenefits). The effect of the toll is to
reduce overall public benefits. If a road user would achieve a benefit of say $3 by using a new toll
road, but must pay a toll of $2, then the net benefit is only $1 if the tolled road is used. The loss of
benefits by those who continue to use the ‘free’ route will be somewhere between zero (because
there would be no benefit in using the tolled route even if there was no toll) and the cost of the toll
($2).
The present value of gross toll collections is the same, regardless of which way they are viewed. Provided
that tolls are not double counted, the net present value of the activity (present value of benefits minus
present value of costs) is also independent of the way tolls are viewed.
In New Zealand, road tolling can currently only be used in conjunction with a new road and this will
generally be within a network of otherwise ‘free’ roads. This has implications for:
• traffic distribution/assignment
• environmental impacts
• economic efficiency
• financial – toll level and fundability of the new road
• design of the new road and toll facility.
The following are essential steps for consideration of a road tolling proposal:
• ensure that the need for the activity and the benefits to the community have been identified and
maximised
• explore alternative solutions, including non-capital options
• identify risks and returns and determine appropriate allocation among relevant parties
• establish the nature and extent of community support likely to be required through an effective
consultation process.
There are several approaches to setting charges for a toll road where other routes are ‘free’. Three of the
most common approaches are:
• a pricing policy where economic welfare as defined by the benefit–cost ratio (BCR) is maximised
• a revenue maximising pricing policy where service provider revenue is maximised
• a ‘network optimisation’ pricing level which seeks to optimise the performance of the network in
terms of total travel times or average network speeds.
In practice, all these three considerations and possibly others may need be considered in reaching a toll
regime which that meets the overall objectives of the proposal.
Private sector financing

The purpose of private sector activities is to involve private sector funds in community facilities. Treasury
guidance for Public Private Partnerships provides an overview of the New Zealand public private
partnership (PPP) model and policy and is intended to set the scene for procuring entities, potential,
bidders and the public.
When considering private sector financing of a facility, a concession agreement, the following steps
should be taken:
• ensure that any private sector involvement is commercially feasible (see notes on financial
evaluation) and offers a more cost-effective solution that the traditional public sector approach
• only private sector options that reduce public sector costs should remain in the final set of options
under consideration
• ensure that any commercial arrangement with the private sector is appropriate and that any
probity and accountability requirements have been met
• identify the degree to which risks can be shared with, or assumed by, private sector participants.
Options with private sector financing can lead to an earlier start date, depending on the ability of the
private sector to raise funds. Also, there is usually an incentive for early completion of privately financed
activities since revenue starts to accrue upon completion of work. The Procurement manual for activities
funded through the National Land Transport Programme provides transport activity procurement
guidelines for approved organisations and Waka Kotahi.
Concessionaries may propose arrangements where the government provides substantial initial funding
for which repayments are made over time, generally from the activity income. This type of arrangement is,
in effect, a loan and should be identified as such.
In principle, the economic efficiency evaluation of toll options is no different from that for other (non-
pricing) options for any proposal. However, the following issues warrant particular attention:
• the range of options considered
• the treatment of value of time savings

• the composition and application of BCRs.
Consumer surplus methodology must be used for evaluation of road tolling activities because
motorists’ behaviour in response to various levels of tolls (including no toll) must be determined and
therefore a measure of the willingness to pay. Stated preference (SP) surveys or possibly, revealed
preference (RP) data, need to be used to give a general cost equation (combining travel time, VOC
and toll charge).
Simplified procedures for private sector financing, and road tolling activities
A range of simplified procedures available in this manual may be used to evaluate private sector
financing, and road tolling activities with an undiscounted whole-of-life cost of less than or equal to $15
million. SP6 High productivity motor vehicles, SP8 Freight transport services, SP9 New public transport
services, SP10 Existing public transport services, SP12 Travel behaviour, and SP13 Safety promotion are
the simplified procedures most relevant to private sector financing, and road tolling activities. These
procedures have been discussed in detail elsewhere within Chapter 4. Table 46 may be referred to select
the appropriate procedure.
Full procedures for private sector financing, and road tolling activities
used.
The following table outlines the stages of analysis in the economic efficiency evaluation of private sector
financing, and road tolling activities. The chapters and sections of this manual that apply to each stage of
the analysis are referenced in Table 101.
Table 101: Stages of analysis for private sector financing, and road tolling activities
1 Consider and describe: Current section
a. the do-minimum Section 1.4:
b. improvement alternatives and options. Counterfactuals
Section 1.5:
Alternatives and
options
• demand estimates and modal share Demand estimation
and mode share
• uptake
Where possible use transport models to assess the impacts and
undertake calibration of the transport models for the activities under
consideration.

by Waka Kotahi.
4 Undertake risk analysis when there are significant unpredictable Chapter 7:
events that may affect or be affected by the improvement activity. Sensitivity and risk
analysis
• property
• construction
If there is a service provider, determine service provider costs,
service provider revenue and the funding gap. Quantify the net costs
to government of the funding gap.
to obtain present values. Then use to calculate the benefit–cost Discounting
ratio(s).
7 Financial analysis to evaluate the viability of an activity by assessing Current section
its cash flows.
cost ratios

The do-minimum for evaluating activities with public sector financing and/or road tolling shall only include
committed and funded transport activities, the existing road network with minor improvements, and
potentially the provision of the new road at a much later date (if the purpose of tolling is to bring forward
the provision of the new road).
Stage 1b: Scale and scope of private sector financing, and road tolling options
Economic efficiency evaluation of road tolling activities must be undertaken with and without the tolls in
place, as alternatives and options are required to be considered under the Land Transport Management
Act 2003. Additionally, financial analysis of the toll options is required.
Financial analysis is used to determine the optimum tolls, choices of debt financing, optimum borrowing
and timeframe for implementing tolls. The imposition of tolls has consequences in terms of changing the
demand for the facility, diverting traffic onto other facilities, increasing the costs due to toll collection and
other issues.
Tolling must be evaluated as an option compared with the case of no tolls.
A number of other options aimed at optimisation of the transport system should also be assessed,
including:
• revenue maximisation tolls

• level of tolls and other measures maximising social welfare
• level of tolls and other measures maximising traffic diversion from sensitive areas
• level of tolls and other measures to optimise level of service.
When considering private sector financing options, only options that reduce public sector costs should
remain in the final set of options.
Timing of construction start is an important consideration for activities involving private sector financing
and/or road tolling. These strategies are often used to allow an earlier start for the activity than that
which would apply without these funding sources. The analysis period should be extended to capture
the activity benefits over the useful life of all the options.
With activities involving private sector financing, and particularly tolling, there is usually also an
incentive for early completion of the activity as revenue starts to accrue upon completion of the
proposal.

Traffic modelling for a tolled road (and the surrounding road network) is an essential input to evaluation.
The main purpose of the assignment part of the traffic modelling is to forecast traffic volumes (and
corresponding traffic speeds) on each part of the road network and particularly on the toll road. The toll
road traffic volumes in turn determine toll revenues.
For accurate forecasting of route choice between the toll road and alternative routes, it is important to
consider the full range of behavioural preferences of potential users of the toll road. This generally
requires more sophisticated choice models and a better understanding of motorists’ preferences than is
the case in standard traffic models.
Traffic modelling used for road tolling activities should take into account behavioural responses such as:
• peak spreading/contraction
• trip end redistribution
• modal shift
• trip generation/suppression.
The split of traffic between the toll road and alternative routes is likely to be sensitive to the level of
congestion on the road network and the mix of trip purposes by time of day/day of week. Therefore,
detailed traffic modelling must separately consider periods with differing levels of congestion. Expansion
or annualisation factors need to be applied separately to the results for each of these periods based on
the characteristics of the traffic that has the toll route as an option.
Some trips that would use the new route if it was ‘free’ will be deterred from its use by the charges and
will continue to use the existing network. Hence the extent of traffic reduction on existing roads, provided
by the new route is less than would be achieved if the new route were ‘free’.
Stage 3: Calculate private sector financing and road tolling activities impacts
Once traffic impacts have been determined, the calculation of national economic benefits follows in the
normal manner (see section 4.3 for the procedures required to evaluate road renewal and improvement
activities, and Chapter 3 for the relevant benefits) but using the disaggregated willingness-to-pay values
for travel time for benefits or disbenefits (see section 3.6).
Consumer surplus methodology must be used for evaluation of road tolling activities because motorists’
behaviour in response to various levels of tolls (including no toll) must be determined and therefore a
measure of the willingness to pay. Stated preference (SP) surveys or possibly, revealed preference (RP)
data, need to be used to give a general cost equation (combining travel time, VOC and toll charge).
For most transport activities, an average value of time is used in economic efficiency evaluations, ie the
same unit values are used for motorists from more affluent households and for those from less affluent
households. This is essentially an ‘equity’ approach (to avoid favouring activities used by higher income
groups). It also makes the economic evaluation easier. This averaging approach is not of major
consequence for most situations.
However, it has important implications for toll roads, particularly when comparing the economic merits of
tolled versus untolled options. An ‘equity’ value of time will substantially over-estimate the perceived
disbenefits of tolling. The extent of distortion is directly related to the spread of the behavioural value of
travel time.
Evaluation of toll roads (including tolling policies) must use a distribution of values of travel time
consistent with users’ willingness-to-pay values established through SP surveys or other means. A
consistent distribution of values of travel time must be used in both the traffic modelling and economic
efficiency evaluation.
When investigating options and alternatives, behavioural values can be used to calculate initial user
benefits, with the overall results adjusted to the average value of travel time between the behavioural and
equity values for consistency with other activities.
When users are required to pay tolls on a route, some will choose to avoid the toll by using alternative
routes if they are available. The toll charges change the benefits that would otherwise be received by
road users in the following ways:
• For those motorists who continue to use the toll road, benefits are reduced by the extent of the
toll charge, which is added to their generalised cost of travel.
• The benefits to users on the toll road may be increased due to less congestion on the tolled
facility
• For those who would have used the new road if it was not tolled but decide to divert to a ‘free’
road because of the toll, travel time and perhaps vehicle operating costs are likely to increase
• For those who would have continued to use alternative routes, even if the new road was not
tolled, benefits are likely to be reduced because of more congestion.
Environmental and community benefits may also change with a tolled road compared with leaving the
road untolled. Possibilities include:
• overall vehicle use

• use of carpools
• level of public transport use
• options to develop public transport
• overall pollution
• degree of decentralisation
• local area traffic management
• timing of infrastructure provision.
It may not be possible to put values on all these items, but they need to be considered for a tolled facility.
The costs of dislocation and traffic disruption during construction should be included as negative benefits
for all options. These may be different for an untolled road compared to a tolled road (particularly if the
construction period is different).
In assessing the commercial feasibility of private sector funding or a debt facility the following must be
considered.
Risks are different between options with and without private sector financing and/or operation. Technical
capacity, financial backing, business acumen, activity life and government exposure are very important
considerations where there is private sector involvement.
Identification, quantification and assignment of risks among relevant parties are essential for activities
involving private sector financing and for road tolling activities. This should include preparation of a risk
management plan.
For private sector financing, it is essential to ensure that the commercial arrangement with the private
sector is appropriate and that any probity and accountability requirements are met. The degree to which
risks can be shared with, or assumed by, private sector participants must be identified. Details of likely
contractual obligations as they affect pricing, ongoing risk to government, terms of the contract,
termination arrangements and debt and equity contributions of each party should be clearly specified.
Refer to Chapter 7 of this manual for detailed information on risk analysis.
Section 1.8 must be viewed from both an economic and financial point of view.
The public sector financing and/or toll charges reduce the effective activity costs to the government.
Even if an activity is totally funded by the private sector, there will still be some costs to government
agencies, such as contract preparation and ongoing contract management and monitoring. The cost of
these activities should be included in the cost of the option involving private sector financing.
Similarly, the additional cost of toll infrastructure and toll collection must be included in the tolling option.
testing.
For the evaluation of private financing and road tolling, an adjusted analysis period may be most
appropriate especially where there are long-term financial obligations.
Stage 7: Financial evaluation
Where consideration is being given to private sector involvement in financing land transport infrastructure,
it is important to ensure that the involvement is commercially feasible and that it offers a more cost-
effective solution that the traditional public sector funding approach.
Financial analysis is a method to evaluate the viability of an activity by assessing its cash flows. This
differs from economic evaluation in the:
• scope of investigation
• range of input
• methodology used.
Financial analysis views the costs and revenues of the activity from a ‘commercial’ investment point of
view, ie the cash flow impact on government and any private sector party. By contrast economic
efficiency analysis also considers external benefits and costs of the activity whether or not they involve
monetary payments.
Other differences include:
• Market prices and valuations are used in assessing benefits and costs in financial analysis,
instead of measures such as willingness to pay and opportunity cost used in economic analysis.
Market prices include all applicable taxes, tariffs, trade mark-ups and commissions.
• The discount rate used in financial analysis represents the weighted average costs of debt and
equity capital rather than the estimated social opportunity cost of capital.
• The discount rate used in financial analysis and the cash flows to which it is applied are usually
specified in nominal terms (allowing for future inflation), as the costs of debt and equity are
observed only in nominal terms.
Undertaking an economic evaluation does not remove the need for a financial evaluation.
Cash flows to be measured

All incremental costs, revenues and risks associated with an activity and its best alternative should be
identified and measured as nominal cash flows in the period in which they occur. Cash flows should be on
an after-tax basis. An estimate of the asset’s salvage value must be included at the end of the analysis
period to represent the asset’s remaining service potential. The salvage value should not be such as to
bias the viability of the proposal.
Typical inward cash flows to be considered include:
• operating revenues
• subsidies from external parties
• operational savings occurring in other areas as a result of the proposal
• sale of surplus assets
• residual values of assets.
Typical cash outflows to be considered include:
• capital costs (including land, equipment, buildings)

• maintenance and operating costs
• taxes, where appropriate
• operating lease payments
• contract termination payments
• revenue losses to existing operations affected by the proposal
• the opportunity cost of resources (including land) that would otherwise be available for sale or
lease.
Treatment of specific items
Financing costs (interest) should be excluded in the cash flows because the opportunity cost of debt is
accounted for in the weighted average cost of capital (WACC) – the weighted average of the required
return on equity and the (interest) cost of any debt financing reflecting the appropriate risk and norms
associated with the industry.
Accounting, depreciation, economic multiplier effect and sunk costs should be excluded in the financial
analysis. The effect of dividend imputation needs to be considered in the financial analysis.
Operating leases should be evaluated in the form of a series of regular payments and compared to an
outright purchase alternative, with consideration for the value of options such as renewal or purchase
rights if these features are present. Financing leases do not form part of a financial analysis as these are
merely an alternative means of financing the proposal.
Weighted average cost of capital (WACC) is used in financial analysis. The WACC should reflect the
appropriate risk and norms associated with the industry.
Summary measures of commercial merit
The more common measures for evaluating the financial viability of an activity are, for example:
• net present value of cash flows

• net present value per $ of capital invested
• internal rate of return (IRR) of cash flows
• payback period
• profitability indices.
Measures used in commercial evaluations will vary between activities and private sector proponents.
Specialist advice should be sought on financial evaluations and detailed descriptions of these evaluations
are not included here.

The principles of economic appraisal apply to the evaluation of toll road activities and activities involving
private sector financing, however, the BCRs are calculated slightly differently. Refer to Chapter 6 for
detailed information on developing BCRs, including toll road facilities.
In the present value calculations, all costs and benefits are expressed in current dollars and no allowance
is made for price inflation or cost escalation. The discount rate is applied to the costs and benefits in each
year to determine the present value of the costs and benefits which are then summed to get the total
costs and benefits over the analysis period.
Sensitivity testing applies to both financial analysis and economic efficiency analysis. Refer to Chapter 7
of this manual for detailed information on sensitivity testing.
The impact of risks (their probability or likelihood of occurrence and the consequence) on the results must
be tested by sensitivity analysis. Critical assumptions that could be varied should be altered one at a time.
For financial analysis, analyse the sensitivity to variations associated with cash flows for each option, eg
changes to key variables by ±20% and different combinations of key variables which taken together
represent an alternative, plausible and consistent view of the future.
Calculate and present summary financial measures for the best and worst cases and for specific changes
to key variables that are deemed highly probable. Break even points (at which the activity begins to lose
money) should be identified.
5. Discounting > 4.8 Evaluation of private sector financing and road tolling
5. Discounting
present value they need to be discounted back to time zero. The discount rate represents the rate at
which society is willing to trade off present benefits and costs against future benefits and costs.
The discount rate, effective from 1 July 2020, shall be 4% per annum. This is the rate calculated by Waka
Kotahi as being appropriate for transport investment and is subject to ongoing review.
Based on a discount rate of 4%, sets of present worth factors have been calculated to convert future
benefits and costs to their present values. Tables for discount rates of 3% and 6% are also provided for
use in sensitivity testing.
Some benefits and costs occur at a single point in time in which case single payment present worth
factors (SPPWF) shall be used to discount the amounts to their present value. The annual SSPWF table
for discount rates of 3%, 4% and 6% is provided in Table 102. A quarterly SSPWF table for the same
discount rates is provided in Table A132 of Appendix 6: Discount factors.
Other benefits and costs occur continuously over a number of years, in which case either uniform series
(USPWF) or arithmetic growth present worth factors (AGPWF) should be used to discount the amounts to
a present value, depending on whether the amounts are uniform or increase arithmetically over time (eg
traffic and patronage growth). USPWF and AGPWF tables for discount rates of 3%, 4% and 6% are
provided in Table A133 and Table A134 respectively of Appendix 6: Discount factors.
When discounting benefits or costs determined from a transportation model, the present worth factors
specified in this manual must be used. If necessary, adjust values to time zero equivalents. Traffic growth
rates may also require a similar adjustment to time zero.
When discounting crash benefits the traffic growth rate will need to be adjusted in accordance with the
procedures in Appendix 2: Crash analysis to determine the appropriate arithmetic growth rate to apply.
External impacts are assumed to remain constant so the uniform present worth series should be used to
obtain the present value of monetised impacts.
Worked examples of discounting using the SSPWF, USPWF and AGPWF are provided in Appendix 8:
Worked examples.
5.1 Single payment present worth factor

Where a single benefit or cost arises at some future time, a SPPWF shall be applied to calculate its
present value.
The formula for determining SPPWF factors is:
1 1
SPPWFin = =
(1 + i)n 1.04n
for a 4% discount rate, where: n is time in years after time zero, and
i is the discount rate expressed as a decimal, ie for 4% i = 0.04.
Table 102: Annual single payment present worth factors

Time 4% Discount rate SPPWF 3% Discount rate 6% Discount rate SPPWF
(years from time zero) SPPWF (sensitivity test)
(sensitivity test)
0 1.0000 1.0000 1.0000
1 0.9615 0.9709 0.9434
2 0.9246 0.9426 0.8900
3 0.8890 0.9151 0.8396
4 0.8548 0.8885 0.7921
5 0.8219 0.8626 0.7473
6 0.7903 0.8375 0.7050
7 0.7599 0.8131 0.6651
8 0.7307 0.7894 0.6274
9 0.7026 0.7664 0.5919
10 0.6756 0.7441 0.5584
11 0.6496 0.7224 0.5268
12 0.6246 0.7014 0.4970
13 0.6006 0.6810 0.4688
14 0.5775 0.6611 0.4423
15 0.5553 0.6419 0.4173
16 0.5339 0.6232 0.3936
17 0.5134 0.6050 0.3714
18 0.4936 0.5874 0.3503
19 0.4746 0.5703 0.3305
20 0.4564 0.5537 0.3118
21 0.4388 0.5375 0.2942
22 0.4220 0.5219 0.2775
23 0.4057 0.5067 0.2618
24 0.3901 0.4919 0.2470
25 0.3751 0.4776 0.2330
26 0.3607 0.4637 0.2198
27 0.3468 0.4502 0.2074
28 0.3335 0.4371 0.1956
29 0.3207 0.4243 0.1846
30 0.3083 0.4120 0.1741
31 0.2965 0.4000 0.1643
32 0.2851 0.3883 0.1550
33 0.2741 0.3770 0.1462
34 0.2636 0.3660 0.1379
35 0.2534 0.3554 0.1301
36 0.2437 0.3450 0.1227
37 0.2343 0.3350 0.1158
38 0.2253 0.3252 0.1092
39 0.2166 0.3158 0.1031
40 0.2083 0.3066 0.0972
41 0.2003 0.2976 0.0917
42 0.1926 0.2890 0.0865
43 0.1852 0.2805 0.0816
44 0.1780 0.2724 0.0770
45 0.1712 0.2644 0.0727
46 0.1646 0.2567 0.0685
47 0.1583 0.2493 0.0647
48 0.1522 0.2420 0.0610
49 0.1463 0.2350 0.0575
50 0.1407 0.2281 0.0543
51 0.1353 0.2215 0.0512
52 0.1301 0.2150 0.0483
53 0.1251 0.2088 0.0456

54 0.1203 0.2027 0.0430
55 0.1157 0.1968 0.0406
56 0.1112 0.1910 0.0383
57 0.1069 0.1855 0.0361
58 0.1028 0.1801 0.0341
59 0.0989 0.1748 0.0321
60 0.0951 0.1697 0.0303
5.2 Uniform series present worth factor

Where a series of equal benefits or costs arise each year or continuously over a period, USPWF should
be applied to calculate the present value of these costs and benefits.
The formula for determining USPWF factors is:
(1- (1 + i)-n)
USPWFin =
loge(1 + i)
where: n is the time in years after time zero

i is the discount rate expressed as a decimal i.e. for 4% i = 0.04.
The present value of a time stream of equal annual benefits or costs shall be calculated as follows:
Present value = annual benefit (or cost) × (USPWFe - USPWFs)
where: s is the start year
e is the end year of the cost or benefit stream.
The USPWF factors in Table A133 assume that the annual benefits or costs are evenly spread over each
year and are continuously compounded.
5.3 Arithmetic growth present worth factors

Where costs or benefits increase (or decrease) each year arithmetically, arithmetic growth present worth
factors (AGPWF), together with the corresponding USPWF factors, should be applied to calculate the
present values of these costs and benefits.
It is assumed in this manual that traffic growth is arithmetic.
The formula for determining AGPWF factors is:
AGPWFin = [loge (1 + i)]-2 - n.(1 + i)-n. [loge (1 + i)]-1 - (1 + i)-n. [loge (1 + i)]-2
where: n is time in years after time zero
i is the discount rate in percent
The present value of a time stream of benefits or costs which increase or decrease arithmetically shall be
calculated as follows:
Present value = annual benefits × {(USPWFe - USPWFs) + (R x (AGPWFe - AGPWFs))}
where: R is the arithmetic growth rate at time zero
s is the start year
e is the end year of the cost or benefit stream.
The AGPWF factors in Table A134 assume that the annual benefits or costs occur continuously
throughout the year and are continuously compounded.
Back to 1.9 Discounting: Present value >>
Back to 4. Evaluation procedures >>
6. Benefit–cost ratios > 6.1 National benefit–cost ratio
6. Benefit–cost ratios
The activity costs required for determining benefit–cost ratios (BCRs), incremental benefit–cost ratios,
and the first-year rate of return, are the difference between the costs of the activity option and the costs of
the do-minimum. The activity benefits are similarly the differences between the benefit values calculated
for the activity option and those of the do-minimum. In this sense, all of the BCRs calculated for transport
activities are incremental BCRs
It follows that where a particular benefit or cost is unchanged among all the activity options and the do-
minimum, it does not require valuation or inclusion in the economic analysis. For completeness, it should
be noted in any funding application that the benefit or cost is unchanged.
6.1 National benefit–cost ratio

Waka Kotahi uses the national benefit cost ratio as a measure of economic efficiency from a national
perspective. A national benefit cost ratio must be calculated for all the short-listed options.
The formula for determining the national benefit cost ratio is:
National benefit cost ratio is the present values of national economic benefits divided by present value of
national economic costs
𝐵
𝐵𝐶𝑅𝑁 = 𝐶
National economic benefits are defined as the net direct benefits and disbenefits experienced by transport
users, the net indirect benefits and disbenefits experienced as externalities by the population outside the
transport system, and all other monetised impacts.
Where the national economic benefits include wider economic benefits (WEBs) it is a requirement that the
national benefit cost ratio is calculated with and without the WEBs included. This is described in more
detail below.
National economic costs are defined in one of two ways. Where there is no service provider, national
economic costs are the net cost to Waka Kotahi and approved organisations. Otherwise, national
economic costs are the net cost to Waka Kotahi and approved organisations, plus net service provider
costs.
National benefit–cost ratio with WEBs

If a transport activity generates static or dynamic agglomeration or labour supply benefits, or reduces
imperfect competition, these benefits may be monetised and included in the calculation of national
economic benefits. However, when these WEBs are present, it is a requirement that two national benefit–
cost ratios are calculated and reported.
Firstly, a national benefit–cost ratio excluding WEBs must be calculated. For the BCR excluding WEBs,
the present value of national economic benefits less the present value of WEBs is divided by the full
national economic costs of the activity.
Secondly a national benefit–cost ratio including WEBs must be calculated. For the BCR including WEBs
the full present value of national economic benefits is divided by the full national economic costs of the
activity. The BCR including WEBs is treated as a sensitivity test.
At all times, any benefits that arise from increased international tourism activity are excluded from BCR
calculations. These benefits must only be reported within the appraisal summary table.
6. Benefit–cost ratios > 6.2 Government benefit–cost ratio
Presenting the national benefit–cost ratio

When presenting the national benefit–cost ratio, the BCR should be rounded to one decimal place if the
BCR is greater than 1.0 but less than 10.0. If the BCR is greater than 10.0 the BCR should be rounded to
the nearest whole number.
If the national benefit–cost ratio is below 1.0 it should be reported to two decimal places and must not be
rounded.
6.2 Government benefit–cost ratio

The government benefit–cost ratio (BCRG) is used to indicate the level of benefits obtained from
investment of local and central government funds in situations where government funding is
supplemented by the availability of third-party funding or tolling revenue, or where it is necessary to cover
service provider costs in the event of a funding gap for the operation of PT services.
The BCRG is not an alternative to the BCRN and it will not replace the BCRN in the Waka Kotahi
Investment Prioritisation Method (IPM). Rather the BCRG is additional information that is helpful when
considering both the business case and the financing of an activity.
Note:
• All costs are exclusive of GST, including tolls, farebox revenue, private sector contributions.
• Net cost to government should include cost incurred to enable private sector contributions (for
example. tolling infrastructure and transaction costs).
The BCRG formula is:
𝐵−𝐶𝑝 𝐵−𝐶𝑝
𝐵𝐶𝑅𝐺 = 𝐶−𝐶 =
𝑝 𝐶𝑔
Where:
B is the PV of national economic benefits (including disbenefits). In the estimate of the total
benefit the suppression impact (for example) of tolling on demand should be taken into account.
C is the PV of national economic costs. It includes costs incurred to enable private sector
contributions (for example tolling infrastructure and transaction costs) and the maintenance costs
associated with maintaining a facility (for example toll road facilities).
Cp are the costs borne by parties outside the NLTF system (for example the PV of gross toll
revenue in the case of toll roads, farebox revenue in the case of PT services, or private sector
contributions to enable an activity to proceed that otherwise would not proceed).
Cg is the PV of net costs to government = C – Cp (that is, PV central and local government costs
less PV private sector contributions).
Examples of BCRG
BCRG for a toll road is:
𝑝𝑟𝑒𝑠𝑒𝑛𝑡 𝑣𝑎𝑙𝑢𝑒𝑠 𝑜𝑓 𝑛𝑎𝑡𝑖𝑜𝑛𝑎𝑙 𝑒𝑐𝑜𝑛𝑜𝑚𝑖𝑐 𝑏𝑒𝑛𝑒𝑓𝑖𝑡𝑠 − 𝑝𝑟𝑒𝑠𝑒𝑛𝑡 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑡𝑜𝑙𝑙𝑠
𝐵𝐶𝑅𝐺 =
𝑝𝑟𝑒𝑠𝑒𝑛𝑡 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑛𝑒𝑡 𝑔𝑜𝑣𝑒𝑟𝑛𝑚𝑒𝑛𝑡 𝑐𝑜𝑠𝑡𝑠
Where:
National economic benefits = net direct and indirect benefits and disbenefits to all affected
transport users plus all other monetised impacts
Tolls = total toll collections (excl. GST)
Net government costs = PV costs to Waka Kotahi and approved organisations.
6. Benefit–cost ratios > 6.3 Incremental cost–benefit analysis
BCRG for public transport services with fare revenue included is:
𝑝𝑟𝑒𝑠𝑒𝑛𝑡 𝑣𝑎𝑙𝑢𝑒𝑠 𝑜𝑓 𝑛𝑎𝑡𝑖𝑜𝑛𝑎𝑙 𝑒𝑐𝑜𝑛𝑜𝑚𝑖𝑐 𝑏𝑒𝑛𝑒𝑓𝑖𝑡𝑠 − 𝑝𝑟𝑒𝑠𝑒𝑛𝑡 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑓𝑎𝑟𝑒𝑏𝑜𝑥 𝑟𝑒𝑣𝑒𝑛𝑢𝑒𝑠
𝐵𝐶𝑅𝐺 =
Where:
Farebox revenue = total revenue from tickets (excl. GST)
BCRG for a private sector contribution is:

𝑝𝑟𝑒𝑠𝑒𝑛𝑡 𝑣𝑎𝑙𝑢𝑒𝑠 𝑜𝑓 𝑛𝑎𝑡𝑖𝑜𝑛𝑎𝑙 𝑒𝑐𝑜𝑛𝑜𝑚𝑖𝑐 𝑏𝑒𝑛𝑒𝑓𝑖𝑡𝑠 − 𝑝𝑟𝑒𝑠𝑒𝑛𝑡 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑝𝑟𝑖𝑣𝑎𝑡𝑒 𝑠𝑒𝑡𝑜𝑟 𝑐𝑜𝑛𝑡𝑟𝑖𝑏𝑢𝑡𝑖𝑜𝑛𝑠
𝐵𝐶𝑅𝐺 =
Where:
Private sector contributions = total private sector contributions (excl. GST) over and above
the normal central and local government contributions (for
example contributions made in line with the IFF Act 2020 12)
Note: for the purposes of calculating the BCRG the normal Waka Kotahi discount rate of 4% is used
together with sensitivity testing using a discount rate of 6%. For known high-risk projects sensitivity
testing using an 8% discount rate is recommended as well as using the 4% and 6% rates.
Three worked examples of the BCRG procedure are provided in Appendix 8: Worked examples.
6.3 Incremental cost–benefit analysis

Where activity alternatives and options are mutually exclusive, incremental cost–benefit analysis of the
alternatives and options must be used to identify the optimal activity from an economic efficiency
standpoint.
The incremental BCR indicates whether the incremental cost of higher-cost alternatives and options is
justified by the incremental benefits gained, all other factors being equal. Conversely, incremental
analysis will identify whether a lower-cost alternative or option that realises proportionally more benefits is
more economically efficient.
Incremental BCR is defined as the incremental benefits per dollar of incremental cost.
The formula for determining the incremental BCR is:
incremental benefits
Incremental BCR =
incremental costs
Procedure for calculating the incremental BCR

The following procedure should be used to calculate the incremental BCR of mutually exclusive options:
1. Rank the options in order of increasing cost.
12Infrastructure Funding and Financing Act 2020. This act provides for a new way to fund and finance infrastructure
projects that support housing and urban development. The act enables a long-term property tax to be pledged to
special project debts (called ‘revenue bonds’) to provide alternatives to governments’ general obligations (that is, tax-
backed) debts.
6. Benefit–cost ratios > 6.4 First year rate of return
2. Starting at the lowest-cost option, consider the second to lowest-cost option and calculate the
difference between the present value of the benefits of the lowest cost-option and the second to
lowest-cost option. These are the incremental benefits.
3. Next, calculate the difference between the present value of the costs of the lowest cost-option
and the second to lowest-cost option. These are the incremental costs.
4. Calculate the incremental BCR by dividing the incremental benefits by the incremental costs.
5. If the incremental BCR is equal to or greater than the target incremental BCR, discard the lower-
cost option and use the second to lowest-cost option as the comparison basis with the next
higher-cost option.
6. If the incremental BCR is less than the target incremental BCR, discard the higher-cost option
and use the lower-cost option as the basis for comparison with the next higher-cost option.
7. Repeat the procedure from steps 2 to 6 until all options have been analysed.
8. Finally, select the option with the highest cost which has an incremental BCR equal to or greater
than the target incremental BCR.
A worked example of the incremental BCR procedure is provided in Appendix 8: Worked examples.
Target incremental BCR

The analyst shall choose and report the target incremental BCR used when undertaking incremental
analysis of project options. Where the selected target incremental ratio differs from the guidance below,
the analyst must provide a detailed explanation supporting the chosen value. The following guidance is
provided:
• The minimum incremental BCR shall be 1.0, in order to ensure that the additional spending to
invest in a higher cost project option rather than a lower cost option is economically efficient.
• Where the BCR of the preferred option is greater than 3.0 but less than 5.0, the target
incremental BCR shall be 3.0.
• Where the BCR of the preferred option is greater than or equal to 5.0, the target incremental BCR
shall be 5.0.
Sensitivity testing of incremental analysis

The results of the incremental BCR analysis should be sensitivity tested using a target incremental BCR
that is 1.0 higher than the chosen target incremental BCR. If this affects the choice of preferred
alternative or option, the results of this sensitivity test must be described and included in the activity’s
economic case. For example, if the target incremental ratio is 2.0, the choice of alternative or option
should also be tested by using a target incremental ratio of 3.0, and how this affects the choice of option
should be reported.
6.4 First year rate of return

First year rate of return (FYRR) is used to indicate the optimal start date of an activity.
FYRR, expressed as a percentage, is calculated by dividing the present value of benefits in the first full
year following completion of construction by the activity’s full present value of net costs. The formula for
determining the FYRR is:
FYRR = present value of the activity benefits in first full year following completion × 100
present value of the activity costs over the analysis period
The FYRR is useful for sequencing activities when funding is constrained, but it should not be used to
evaluate whether an activity is economically efficient. The FYRR indicates the extent to which the benefits
of an activity arise immediately, or are dependent on future growth, but the overall economic efficiency
cannot be evaluated on the basis of the activity’s benefits in the first year of operation.
It is a requirement that the FYRR is calculated for the preferred option.
7. Sensitivity and risk analysis > 7.1 Sensitivity analysis overview
7. Sensitivity and risk analysis

The forecasting of future costs and benefits always involves some degree of uncertainty, and in some
situations the resulting measures of economic efficiency (the BCR and FYRR) may be particularly
sensitive to assumptions or predictions inherent in the analysis.
Two types of uncertainty may occur in a transport activity:
• uncertainty about the size or extent of inputs to an analysis, such as the variation in construction,
maintenance or operating costs, future traffic volumes, particularly due to model results, growth
rates and the assessment of diverted and induced traffic, travel speeds, road roughness or crash
reductions
• uncertainty about the timing and scale of unpredictable events, either from natural causes (such
as earthquakes, flooding and landslips) or from human-made causes (such as accidental damage
and injury from vehicle collisions).
Assessing the sensitivity of evaluations to critical assumptions or estimates must be undertaken using
either a sensitivity analysis or risk analysis, or both, as appropriate.
Sensitivity analysis involves defining a range of potential values for an uncertain variable in evaluation
and reviewing the variation in the evaluation as the variable changes within the range. This will highlight
the sensitivity of the estimated final outcome to changes in input variables. Sensitivity analysis is an
important tool for testing the veracity of the analysis and contributes to confidence at the decision-making
level.
Risk analysis is a more detailed type of sensitivity analysis that involves describing the probability
distributions of the input variables and those of the resulting estimates of benefits. For a risk analysis to
be possible, both the impact arising from each of the possible outcomes and their probability of
occurrence have to be estimated.
A recent development in the consideration of risk and uncertainty is the adaptive decision-making
approach, a procedure that provides the analyst with additional flexibility during the assessment process
where there is deep uncertainty, by considering all possible outcomes when selecting options for further
investigation. This approach is referred to in section 7.6.
The use of sensitivity and risk analysis can support development of ways of minimising, mitigating and
managing uncertainties.
7.1 Sensitivity analysis overview

The application of the benefit quantification procedures in this manual will generally result in point
estimates. This implies a level of precision and accuracy that is not realisable in real world settings due to
the inherent uncertainty in forecasting future conditions.
While the real BCR of an activity is uncertain and unknowable ex-ante, this does not invalidate the use of
BCRs to forecast efficiency. The BCR is a decision support tool, and sensitivity analysis, when properly
applied, can improve the quality of decision-making by highlighting the critical assumptions and conditions
required for an activity to be efficient ex-post.
Sensitivity analysis is useful for quickly testing the veracity of the analysis and demonstrating to decision-
makers the robustness of the BCR to often extreme changes in key assumptions. For transport analysts,
it can also indicate where more effort is required to quantify the uncertainty affecting the estimation of
benefits and as a precursor to full risk analysis.
At a minimum, sensitivity analysis can be conducted simply and quickly by varying a single assumption
variable while holding all others constant. It is however recommended that multiple assumptions,
representing a range of future scenarios, are varied at the same time to identify the extreme bounds of
the BCR.
7. Sensitivity and risk analysis > 7.2 Sensitivity tests
7.2 Sensitivity tests

This section lists a selection of sensitivity tests that should be considered during appraisal. These lists are
not exhaustive and professional judgement should be applied to select the most appropriate sensitivity
tests based upon the type of activity under consideration and any uncertainties identified during activity
development.
The results of the sensitivity tests, along with explanation of any assumptions or choice of test, should be
reported as upper and lower bounds on the BCR as a component of an activity’s economic case.
Discount rate
While the base evaluation uses the standard 4% discount rate, sensitivity testing should be carried out at
discount rates of 3% and 6%. In particular, sensitivity testing at the lower rate of 3% can be used for
activities with long-term future benefits that cannot be adequately captured with the standard discount
rate. Discounting at these other rates should be applied and reported as a standard sensitivity test for full
procedures using Table 102 (SPPWF) from Chapter 5 as well as Table A133 (USPWF) and Table A134
(AGPWF) from Appendix 6: Discount factors.
Demand estimation
As all forecast benefits and disbenefits are derived from future demand estimates, particular emphasis
should be placed on testing the drivers or demand. This is described in detail in section 7.3.
Benefit and cost estimation

Inputs to the quantification of benefits, disbenefits and costs that should be tested include:
• maintenance costs, particularly where there are significant savings

• cost overruns
• traffic volumes
• service or infrastructure quality
• travel speeds
• road roughness
• crash reductions.
For sensitivity tests of benefits, disbenefits and costs it is not acceptable to limit analysis to variations of
variables by ±20%. Appropriate bounds should be selected based upon professional judgement and
activity specific conditions. These bounds need not be symmetrical and may have long tails in a single
direction.
Safety improvements
Safety improvement activities are undertaken where a route or site (eg curve, railway crossing, bridge
etc) has a high occurrence of crashes, or when the risk of crashes is considered high.
Given the majority of benefits (and hence benefit–cost ratio) for such schemes arise from a reduction in
crashes, it is important that a robust assessment is undertaken. Analysts should avoid basing their
assessment on a small number of historical crashes or using unsuitable crash rates, crash prediction
models or crash reduction factors. It is also important to undertake sensitivity testing to understand how
sensitive the benefit–cost ratio is to the crash history, crash prediction and crash reduction factors.
7.3 Demand estimation sensitivity tests

Demand estimates are important in economic assessments because the amount of predicted use of the
facility, service or mode is often a key, if not critical, driver of the potential benefits. Many of the
parameters used in the development of demand estimates are averages from a wide range of potential
values. Sensitivity tests of demand estimates are therefore an important consideration, particularly in
situations where:
7. Sensitivity and risk analysis > 7.3 Demand estimation sensitivity tests
• the estimated use of the activity is a critical factor; for example, projected volumes on toll roads,
estimated mode share for larger PT schemes, and estimated volume of cyclists on a new
significant dedicated cycleway
• the economic analysis includes longer-term (20 years into the future or longer) demand
estimates, which implicitly means there is more uncertainty
• the analysis and outcomes are sensitive to the level of congestion, delays, and queueing
(particularly parts of the network being highly congested for significant lengths of time) – this can
relate to both the do-minimum and activity scenarios, but is often most pertinent in the do-
minimum scenario where the activity may reduce congestion
• economic outcomes are particularly sensitive to smaller changes in demand estimates; for
example, the level of development anticipated by certain years on a key approach to an
intersection improvement assessment.
Forecast horizons and uncertainty, provides some context around different scenarios, and where and
when the number and range of sensitivity tests may be important. This relates to producing short-,
medium- and long-term forecasts in different locations when assessing different scales of activities.
Appendix 1: Demand estimation methods and guidance provides guidance on techniques, methods and
considerations for developing demand estimates. Where these approaches and techniques can be used
to develop sensitivities, this is described and noted.
Scenario testing and demand estimate sensitivities

As described in section 1.11: Scenario testing, scenarios are plausible states of the future and are a
powerful approach to sensitivity analysis. Plausible future states relating to transport demand may be
developed through a combination of adjustments to several sensitivity parameters; for example, to
produce a ‘high private vehicle mode share’ scenario or a ‘high public transport mode share’ scenario. As
described in section 1.11, this concept extends to the potential development of multiple do-minimum and
activity scenarios. For example, a ‘high private vehicle mode share’ future demand scenario may have
associated with it a do-minimum network that features additional vehicle optimisation improvements.
Guidance on demand estimate sensitivity tests

Table 103 and Table 104 below provide overview guidance on when certain elements may have a greater
influence on demand estimation and or have higher levels of uncertainty needing further consideration.
This leads to considerations of how to determine which elements may be important to alter as sensitivity
tests in order to produce a range of demand estimates, particularly where the demand estimation is a key
driver of the economic benefits.
This should not be considered a comprehensive list of all the potential elements influencing demand
estimation, or a sensitivity test check list, rather it provides context on where and when sensitivity tests
may be important in relation to demand estimation aspects.
Table 103 is focused on project models that do not have demand estimates fed by regional models,
calculations, spreadsheet methodologies, trend analyses and similar. Table 104 is focused on regional
transport modelling and, by association, project models that have demand estimates fed by regional
models.
Table 103: Guidance on importance of sensitivity tests – project model and calculation focus
Factors affecting demand estimates Project model/calculation approaches
Network Short Spreadsheet Straightforward
project corridor/ or similar calculations
model (not intersection equations/
linked to/fed model (not models/
by regional linked to/fed calculations
model) by regional
model)
Elasticity methods, relationships and I I P U
values
Local land use changes P P P U
Local transport system and supply P P P U

changes
Application of trip rates P P P U
Application of distribution analysis P P P U
Factors/trends selected in factoring P P I U

methods
Application of trend analysis P P I I
Application of engineering estimate U U P I

methods of predicted facility use
I = Several sensitivity tests likely to be important
P = Small number of sensitivity tests potentially or partially important
U = Sensitivity tests unlikely to be important or critical/less applicable
Section 2.12, below Table 2 provides some expanded information on each of the elements, methods and
approaches in the above table. The points below relate these considerations to potential methods and
approaches for developing sensitivity tests. In some cases, several of these elements may be varied
separately or together to produce a number of demand estimate sensitivities.
• Elasticity methods, relationships and values: sensitivity tests varying the elasticity value(s)
used are likely to be a good way to examine the response and suitability in relation to the specific
activity in the local context.
• Local land use changes: sensitivity tests could include running scenarios with and without
specific land use developments that are important to the activity demand scenario, faster and
slower rates of land use development uptake, etc.
• Local transport system and supply changes: sensitivity tests may be straightforward, for
example, testing with and without a local transport system change that effects the economic
assessment.
• Trip rates: increasing and decreasing trip rates produces straightforward demand estimate
sensitivity tests, particularly where an assumed trip rate has a direct effect on the economic
assessment.
• Distribution analysis: alterations to the distribution assumptions produce straightforward
demand estimate sensitivity tests, particularly where an assumed trip distribution has a direct
effect on the economic assessment.
• Factoring methods: varying factors is a straightforward approach to sensitivity tests and
analyses.
• Trend analysis: where medium- and longer-term forecasts are developed from trend analysis,
and the resulting demand estimates are important to the economic assessment, straightforward
sensitivity tests may be developed by varying the trend factor(s), and/or adjusting the magnitude
of growth that occurs within certain timeframes (for example, large proportion of growth occurring
within an early timeframe and lower growth following this, and vice-versa).
• Engineering estimates of predicted facility use: varying the level of predicted facility use
produces sensitivity tests on the predicted outcomes.
Table 104 provides some high-level guidance on which elements may have a greater influence on and/or
produce greater levels of uncertainty in demand estimates in regional transport models. These
characteristics lead to guidance on altering these elements and inputs as sensitivity tests to produce a
range of demand estimates.
One, two, or several of these elements may be adjusted in isolation or combination to produce demand
estimate sensitivities.
Table 104: Guidance on importance of sensitivity tests – regional transport modelling focus
Factors affecting demand estimates Geographic context/transport environment
Major urban Moderate Small Township,
centre urban urban rural
(population centre centre corridor/
roughly greater (population (population area
than 500,000) roughly ~ 30,000– (population
between 100,000) roughly
100,000– less than
500,000) 30,000)
Population structure/make-up (particularly I I I P
age)
Household/family structure (retired, school- I I I P
age children, in workforce, etc)
Vehicle availability/access to a vehicle I I P P
Access to alternatives modes and I I P U

infrastructure (public transport, cycling, etc)
Public transport – service coverage, service I I P U
frequency, charges
Residential density – accessibility to activities I P P U
Parking – charge and availability of supply I P P U
Road congestion/delay I P P U
Road pricing/tolling I I I I
Route choice P P U U
Technology influencing behaviour (online P P P U

shopping, work/school travel plans)
I = Important driver, several sensitivity tests liked to be important
P = Moderate impact, small number of sensitivity tests potentially required
U = Sensitivity tests unlikely to be important or critical/less applicable
Section 2.12, below Table 3, provides some expanded information on each of the elements, methods and
approaches in the above table. The points below relate these considerations to potential methods and
approaches for developing sensitivity tests. In some cases, several of these elements may be varied
separately or together to produce a number of demand estimate sensitivities. In all cases, the likelihood of
these elements changing should be assessed so that sensitivity tests are realistic.
• Population structure/make-up: as well as varying the overall population projection for a region
(for example, through Statistics NZ’s low, medium and high projections), the make-up and
structure of the population could be altered to produce sensitivities.
• Household/family structure: the ease of varying these input assumptions as sensitivity tests will
be dependent on the structure of the model. However, if significant changes in household/family
structure are anticipated, then sensitivity tests should be carried out varying the number of
shopping trips, work trips, etc.
• Vehicle availability/access to a vehicle: varying (for example, by reducing) vehicle
availability/access for households is a method for providing a sensitivity test on potential
changing future travel patterns (for example, less private vehicle trip-making as a result of climate
change policy, behaviour changes and/or technology changes).
• Access to alternative modes and infrastructure: sensitivity tests may involve factoring the
modal demand down; for example, reducing the car demand estimated for a road-based activity
to quantify the potential reduction in economic benefits.
• Public transport coverage/frequency/charges: sensitivity tests here may involve adjustment to
walk times (for example, due to increased micro-mobility), wait/transfer times (for example, due to
technology improvements that improve the reliability and accuracy of PT arrival times), and fares
(for example, due to climate change policy or behaviour changes).
• Residential density: sensitivity tests could involve changing where there are employment
opportunities and/or schools to represent closer opportunities to shop and work, and for
education. Realistic patterns should be considered.
• Parking: varying parking charges may produce a change in mode choice or a change in
destination choice, and may be an approach to developing sensitivity tests.
• Road pricing/tolling: sensitivity tests could involve altering monetary charges and elements that
influence the human response to cost changes. In Australia, there have been legal challenges on
toll road forecasts, particularly where consortia were bidding to secure the project. Practical
engineering considerations, quality reviews, and sensitivity tests are critical in these scenarios
There are several scenarios to consider in relation to developing demand sensitivity tests from regional
transport models:
• Specific activity economic assessment: when carrying out an economic assessment of a

specific type of transport activity (for example, public transport improvement or a new toll road), it
is sensible and advisable to carry out targeted sensitivity tests that will directly influence the
demand estimates associated with the activity (for example, parameters and assumptions
effecting PT mode choice and people’s responses to road pricing/tolling).
• General demand scenario development: a region may develop a set of demand estimates for
ongoing consideration and use in assessments, or more generalised demand scenarios may
need to be developed for significant projects in key locations (for example, a new bridge in a
major urban centre). These are grouped below in ‘themed’ demand scenarios, purely as
illustrative examples:
o Historical trip-making scenario: parameters, relationships, models and equations are
‘held’ as per the validated base model to produce medium- and long-term demand
forecasts.
o High PT uptake: parameters and assumptions that effect the relative attractiveness
(generalised cost) of PT travel versus private vehicle travel are adjusted to increase PT
mode share.
o Reduced overall trip-making: parameters and assumptions that effect the overall number
of trips made by trip purpose are adjusted to reflect significantly increased working and/or
shopping from home and/or higher uptake of e-mobility, etc. The result being fewer
overall trips being made across the network.
The NZ Modelling User Group (NZMUGS) has been developing considerations and information relating to
sensitivity tests, the content above draws from this work. Further background, references and information,
there are several discussion papers on forecasting and sensitivity tests available via the NZMUGS
webpage.
One recommendation from the NZMUGS work is the use of an uncertainty log, which is described further
below.
Uncertainty log
Various agencies around the world have introduced the concept of an uncertainty log in relation to
transport analysis. This is a record of the assumptions made in the model (and/or any calculations,
mathematical models, etc) that will affect demand and supply.
In the United Kingdom (UK), an uncertainty log is included in the Department for Transport guidance. The
UK guidance uses subtlety different definitions, particularly for forecasts and scenarios, than the MBCM.
For more information and background, see TAG Unit M4: Forecasting and uncertainty (UK Department of
Transport 2019).
The broad UK approach is to develop a core scenario, which is based on the most unbiased and realistic
set of assumptions. This forms a central case, and alternative scenarios are developed around this. This
concept could be applied to both the do-minimum and, in the case where the activity is anticipated to
have a more significant effect on travel demand (for example, a large PT scheme), the activity scenario.
Alternative scenarios are then developed around the core scenario. Importantly, the alternative scenarios
are developed based on key uncertainties in the core scenario.
The uncertainty log is used to summarise the significant assumptions and uncertainties in the
modelling/analysis and forecasting approach. The purpose of the log is to document the central
assumptions that underpin the core scenario and record the degree of uncertainty around these central
assumptions. These assumptions and uncertainties can then be used as the basis for the development of
alternative forecasts and/or model parameter sensitivity tests.
The uncertainty log extends to cover both elements of uncertainty in demand estimation (including base
and future years) and elements of uncertainty in the specification/settings/parameters in the model (for
example, vehicle route choice parameters). Another way to describe this is that broadly there are two
sources of uncertainty:
1. inputs (such as size of new housing development, future population, etc)

2. error in the model parameters and specifications (how these inputs propagate through the
model).
Along with the elements, considerations and methods that relate more directly to demand estimation
sensitivity tests as described in the sections above, the uncertainty log and associated sensitivities could
include factors such as:
• concerns or issues with the observed data that has been used to develop a model or carry out
analysis
• currency of the base year models, robustness of the base year calibration/validation, and
comments by the peer reviewer
• any specific identified areas of weakness in the base model; for example, a specific location in
the network where the modelled versus observed comparison is poor
• currency of the inputs used to develop travel demand estimations; for example, the population
projections, regional land use development plans or strategies used
• appropriateness and design of the model available for the assessment
7. Sensitivity and risk analysis > 7.4 Risk analysis overview
• key parameters or settings that directly influence the model’s ability to predict a key outcome; for
example, balance of traffic volumes on competing routes, the operation (delays and queues) of a
particular network feature (for example, a key multi-lane roundabout), queues blocking back into
critical areas of the network, traffic signal coordination and associated measurement of delays
between scenarios, consistency and reliability of public transport travel times.
Whether these components are used to develop sensitivity tests or not in relation to an economic
assessment depends on how they are anticipated to affect the outcomes of the assessment. Greater
uncertainty in relation to these elements could result in an increased number of sensitivity tests being run;
for example, further future-year demand tests and/or changes to key model parameters.
7.4 Risk analysis overview

Risk is an extensive concept that involves different aspects and levels of magnitude. The risk procedures
contained within this section are not exhaustive but are designed to cover the most common types of
risks that arise in transport activities.
The risk procedures are set forth with two audiences in mind:
1. Transport analysts should consider the risks that arise from the inaccurate estimation of benefits,
and costs. The cost–benefit analysis and appraisal concepts in Chapter 1 of this manual should
be used to guide the development of proposals and options. There are a range of uncertainties
(such as the accuracy of forecasting) and limitations (such as the availability of data) that can
affect the estimation of economic benefits and costs.
2. Decision makers should be aware of the risk of making sub-optimal or poor investment decisions.
Waka Kotahi has legislative obligations to ensure that activities are efficient and effective (as per
section 20 of the Land Transport Management Act 2003). The primary source of risk for decision
making is the possibility of investing in non-efficient activities or failing to invest in efficient
activities.
The risk procedures are designed to follow the principles set out in the ISO31000 standard on risk
management. These principles are set out below:
• Integrated: Risk management is an integral part of all organisational activities.

• Structured and comprehensive: A structured and comprehensive approach to risk
management contributes to consistent and comparable results.
• Customised: The risk management framework and process are customised and proportionate to
the organisation’s external and internal context related to its objectives.
• Inclusive: Appropriate and timely involvement of stakeholders enables their knowledge, views
and perceptions to be considered. This results in improved awareness and informed risk
management.
• Dynamic: Risks can emerge, change or disappear as an organisation’s external and internal
context changes. Risk management anticipates, detects, acknowledges and responds to those
changes and events in an appropriate and timely manner.
• Best available information: The inputs to risk management are based on historical and current
information, as well as on future expectations. Risk management explicitly considers any
limitations and uncertainties associated with such information and expectations. Information
should be timely, clear and available to relevant stakeholders.
• Human and cultural factors: Human behaviour and culture significantly influence all aspects of
risk management at each level and stage.
• Continual improvement: Risk management is continually improved through learning and
experience.
Risk procedure coverage

Activities
The risk procedures provide guidance on risks at an activity level rather than organisational risks. The
primary purpose of this manual is to establish consistency, transparency and comparability between
activities to aid the evaluation of their economic efficiency.
For organisation-wide risk management guidance and requirements please refer to Minimum standard
Z/44 – Risk management practice guide.
Economic impact
The risk procedures focus on the economic impact risks of activities rather than risks associated with
project delivery. Details of construction and project management risks are required as part of a business
case but out of scope for this manual. The risk procedures in this manual are designed to assess the risks
related to the benefits and costs of an activity.
Risk reduction benefits
Risk reduction benefits are monetisable benefits that may arise from improvements to resilience, safety or
travel time reliability. A worked example of risk reduction using the Risk analysis procedure for resilience
is provided in Appendix 8: Worked examples.
Risk concepts
The risk procedures are designed around a common set of concepts, which are described in more detail
below.
Risk
The effect of uncertainty on objectives. An effect is a deviation from the expected. It can be positive,
negative or both, and can address, create or result in opportunities and threats. Risk is usually expressed
in terms of risk sources, potential events, their consequences, and their likelihood.
Risk source
An element which alone, or in combination with others, has the potential to give rise to risk.
Event
An occurrence or change of a particular set of circumstances. Events included changes that are expected
to occur, but do not, or changes that are not expected to occur but do so. Additionally, an event can be a
risk source.
Consequence
The outcome of an event affecting objectives. Consequences can be expressed qualitatively or
quantitatively. Any consequence can escalate through cascading and cumulative effects.
Likelihood
The chance of something happening. In risk management terminology, the word ‘likelihood’ is used to
refer to the chance of an event occurring, whether defined, measured or determined objectively or
subjectively, qualitatively or quantitatively, and described using general terms or mathematically (such as
a probability or a frequency over a given time period).
Risk rating definitions

The risk procedures refer to a common set of definitions drawn from Z/44.
Table 105: Likelihood rating (Z/44)

Rare Unlikely Possible Likely Almost certain
Probability <= 5% >5%–30% >30%–55% >55%–85% >85%
(applicable to
capital
projects)
Frequency Less than At least once At least once At least once At least once
(applicable to once in 10 in a period of in a period of in a period of in a period of
M&O years 6–10 years 2–6 years 1–2 years 12 months
contracts)
Consequence rating
Generally speaking, the consequence is rated according to the extent of the impact on objectives. The
consequence rating includes five categories:
• insignificant
• minor
• moderate
• severe
• extreme.
Each risk is likely to have an individual definition and measurement of its consequences (both positive
and negative). Consequences can be expressed qualitatively or quantitatively.
The criteria used to distinguish different consequence ratings are based upon professional experience of
the key factors which affect levels of risk. Where there is any doubt as to the appropriate classification,
the general quantitative rule expressed in Table 106 should be used:
Table 106: Consequence ratings

Extreme Severe Moderate Minor Insignificant
Cost deviation >±20% ±10%–20% ±5%–10% ±1%–5% <±1%
Benefit deviation >±20% ±10%–20% ±5%–10% ±1%–5% <±1%
The consequence thresholds apply separately to each single benefit or cost risks. At a project level, the
interaction of risks may result in a higher risk threshold than a simple summation of benefit and cost risks.
Qualitative consequence ratings and critical thinking are recommended when determining overall project
risk consequence ratings.
Table 107 provides an example of high level project risk consequence using qualitative rating criteria
(Z/44).
Table 107: Risk consequence – qualitative rating criteria (Z/44)

Deliverables and Scope (products, Financial Benefits Product quality
milestones organisational) impact
(variance in
working days
Extreme Slippage to Complete failure Variance (+) Complete Quality of more
project deliverable of project to from failure to than one
greater than 40% deliver currently currently realise product on
approved scope approved agreed critical path
Slippage to
life of does not meet
milestone on
critical path project cost baseline quality criteria

greater than 40% of greater benefits for product
than 30% acceptance, and
specified quality
is not
achievable
Severe Slippage to Delivery of a Variance (+) Significant Quality of a
project deliverable significant portion from reduction product on
of between 20%– of approved currently and delay in critical path
40% scope regarded approved realising does not meet
as essential by life of agreed quality criteria
Slippage to
project executive project cost, baseline for product
milestone on
is seriously of between benefits acceptance, and
critical path of
impaired 20%–30% specified quality
between 20%–
is not
40%
achievable
Moderate Slippage to Delivery of a Variance (+) Significant Quality of more
project deliverable component of from delay, but than one
of between 10%– currently currently no, or minor, product on
20% approved scope approved reduction in critical path
regarded as life of realising does not meet
Slippage to
essential by project cost, agreed quality criteria
milestone on
project executive of between baseline for product
critical path of
is impaired 10%–20% benefits acceptance, but
between 10%–
specified quality
20%
is achievable
Minor Slippage to Delivery of a Variance (+) Minor Quality of a
project deliverable component of from reduction product on
of between 5%– currently currently and delay in critical path
10% approved scope approved realising does not meet
regarded as non- life of agreed quality criteria
Slippage to
essential by project cost, baseline for product
milestone on
project executive of between benefits acceptance, but
critical path of
is impaired 5%–10% specified quality
between 5%–10%
is achievable
Insignificant Slippage to Delivery of a Variance (+) No reduction Quality of one or
project deliverable component of from but minor more products
of up to 5% currently currently delay in not on critical
approved scope approved realising path does not
Slippage to
regarded as life of agreed meet quality
milestone on
insignificant by project cost baseline criteria for
critical path of up
project executive of up to 5% benefits product
to 5%
is impaired acceptance, but
specified quality
is achievable
7. Sensitivity and risk analysis > 7.5 Risk analysis procedures
Risk matrix (Z/44)

The classification of risks by likelihood and consequence allows a qualitative risk rating to be assigned to
the benefit or cost rating according to the matrix in Table 108.
Table 108: Risk matrix
7.5 Risk analysis procedures

The risk analysis of an activity should be developed according to the following procedural steps.
Risk management and analysis

Process
The risk management process involves the systematic application of policies, procedures and practices to
the activities of communicating and consulting, establishing the context and assessing, treating,
monitoring, reviewing, recording and reporting risk.
Figure 21: Risk management process
Risk assessment
A risk assessment is the overall process of risk identification, risk analysis and risk evaluation. Risk
assessments are to be conducted using risk analysis worksheet 1 and risk analysis worksheet 2. Risk
treatments are to be captured in risk analysis worksheets 3.
Risk analysis worksheet 1 is used for both an abbreviated summary of risks for activities that are in the
early stages of evaluation, and for detailed reporting of risks for activities that are at the detailed (DBC) or
single-stage business case (SSBC) stage.
Risk analysis worksheet 2 is to be used to provide additional detailed information on the high and critical
risks identified in risk analysis worksheet 1 and an estimated quantifiable risk calculation.
Risk analysis worksheet 3 is the risk-adjusted benefits and costs and BCR risk tool.
All three risk analysis worksheets, and their associated instructions, are contained within this manual in
Appendix 7: Risk analysis worksheets.
Risk identification
The purpose of risk identification is to find, recognise and describe risks that might help or prevent an
activity achieving its objectives. Transport analysts may use a range of techniques to identify
uncertainties that affect one or more objectives. The following factors, and the relationship between these
factors, should be considered:
• tangible and intangible sources of risk

• causes and events
• threats and opportunities
• vulnerabilities and capabilities
• changes in the external and internal context
• indicators of emerging risks

• the nature and value of assets and resources
• consequences and their impact on objectives
• limitations of knowledge and reliability of information
• time-related factors
• biases, assumptions and beliefs of those involved.
Risk analysis
The purpose of risk analysis is to comprehend the nature of risk and its characteristics including, where
appropriate, the level of risk. Risk analysis can be undertaken with varying degrees of detail and
complexity, depending on the purpose of the analysis, the availability and reliability of information, and the
resources available. Analysis techniques can be qualitative, quantitative or a combination of these,
depending on the circumstances and intended use.
Risk analysis should consider factors such as:
• the likelihood of events and consequences

• the nature and magnitude of consequences
• complexity and connectivity
• time-related factors and volatility
• the effectiveness of existing controls
• sensitivity and confidence levels.
Highly uncertain events can be difficult to quantify. This can be an issue when analysing events with
severe consequences. In such cases, using a combination of techniques generally provides greater
insight.
Risk analysis provides an input to risk evaluation, to decisions on whether risk needs to be treated and
how, and on the most appropriate risk treatment strategy and methods. The results provide insight for
decisions, where choices are being made, and the options involve different types and levels of risk.
Risk evaluation
The purpose of risk evaluation is to support decisions. Risk evaluation involves comparing the results of
the risk analysis with the established risk criteria to determine where additional action is required. This
can lead to a decision to:
• do nothing further
• consider risk treatment options
• undertake further analysis to better understand the risk
• maintain existing controls
• reconsider objectives.
Decisions should take account of the wider context and the actual and perceived consequences to
external and internal stakeholders. The outcomes of risk evaluation should be recorded, communicated
and then validated at appropriate levels of decision maker.
Analysing benefit estimation risks from demand estimation

Benefit estimation may involve analysis that includes estimating the demand for the activity in current or
base years, and forecasting future-year demand. Forecasting future-year demands is often linked to, or
based on, base year models and historical data sources. The benefit estimation risks below are focused
on the risks in any demand estimate base year data used and in the future-year demand estimation
forecasts. This risk covers three main categories:
• base year/model/source data used to develop demand estimations

• growth forecasting/future-year demand estimate projections
• estimation of benefits specific to the activity.
All risks listed in Table 109, Table 110 and Table 111 are mode neutral. Risks covered in this section are
not necessarily exhaustive. Transport analysts are encouraged to identify benefit risks on a case-by-case
basis. Section 7.3 discusses the concept of an uncertainty log in relation to transport modelling, base year
and future year demand estimation. The uncertainty log is likely to feed this risk analysis.
For each possible risk, a brief description is provided as well as possible risk sources. As a general
principle, the consequence rating is extreme, severe, moderate, minor, or insignificant when the benefit
estimation deviation is >±50%, ±25%–50%, ±10%–25%, ±5%–10%, or <±5%.
Table 109: Benefit risks from base year models/source data used to develop demand estimations
Source data Description Possible risk source

related to base
demand estimate
Age of data Data source used to generate
• Older traffic counts and other data (for
demand estimate is out of date,
example travel times).
particularly where significant
changes in the study area or travel • Older household travel survey data
behaviour have occurred since (more than 10 years old) used as basis
data source was collected. for regional model.
• Significant changes in modes
available/used since the data was
collected.
• Other older input data (PT, signal
timing, pedestrian activity etc).
Scope/coverage/ Data is not extensively relevant to
• Data is not close to and/or doesn’t
relevance of data the proposed activity’s travel
capture areas/locations and information
mode, study area, and/or critical
locations within study area. which the activity is likely to have a
more significant influence on.
• Travel pattern data/parameters
imported from other areas/countries and
is not relevant.
Robustness and Data is not representative of
• Traffic data from a single day sample
statistical reliability typical conditions in the study area
survey, not checked against typical
of data and/or statistically robust.
average values across multiple
days/large sample.
• Traffic data effected by incident.
• Traffic data surveyed on day/period not
most appropriate/relevant to activity
assessment due to seasonality effects,
holiday effects, day-to-day patterns,
peak period coverage, etc.
• The survey has a low sample or there is
bias in the data collected. This includes
surveys of counts, travel times,
household interview/diaries, PT usages,
vehicle occupancy, OD patterns, etc.
Weakness in Base model has overly prescribed
• No independent peer review of base
model description, parameters and inputs,
model, and/or review of suitability of
parameters and inappropriate/unrepresentative
settings parameters and settings, lacks key model to specific economic
functionality for assessment. assessment.
• Poor documentation/reporting related to

base model, leading to model
weaknesses not being understood.
• Lack of understanding of model quality,
strengths, weaknesses and capabilities
in application to economic assessment
and analyses.
Model calibration/ Model calibration/validation weak
• Weaker overall calibration/validation.
validation in the areas relating to the activity
• Specific calibration/validation weakness
assessment (geographic and/or
model predictive capability). in vicinity or key location related to the
activity scenario.
• Other specific calibration/validation
weaknesses in relation to activity
scenario assessment; for example, if
key time periods, vehicle composition
etc are not well represented in the base
year model.
• Specific calibration/validation weakness
in model predictive capability important
to activity assessment; for example, PT
mode choice prediction, change in
travel time (delay/queue/block back)
prediction etc.
• Available regional tools and/or models
have some or several of the above
weaknesses, and if the appropriateness
of application is not checked or
confirmed.
Note: Base demand data sources may be counts, historical surveyed data, research and guideline
values, and information from transport models (particularly relates to base/base-year models).
Table 110: Benefit risks from future year forecast demand estimations
Future-year Description Possible risk source

demand estimation
input/source
Elasticities, trend The assumed elasticity value,
• Elasticity parameter may not be
analysis, trend calculation, factor,
appropriate in local context, may have
mathematical mathematical model etc is not
models etc realistic, not appropriate for increased uncertainty/weakness/
assessment, or has fundamental appropriateness when applied forward
weaknesses. in time.
• Trend analysis carried out on low
sample or unrepresentative sample of
historical data.
• Mathematical models/factors not
appropriate in local context, may have
increased uncertainty/weakness/
appropriateness when applied forward
in time.
Trip rates, trip Trip rates and/or assumed trip
• Trip rates not appropriate to local
distribution distributions are not realistic, not
context, may have increased
analyses appropriate for assessment, or
have significant weaknesses. uncertainty/weakness/appropriateness
when applied forward in time.
• Distribution analysis may have
increased uncertainty/weakness/
appropriateness when applied forward
in time.
• Application of matrix estimation in the
base year may have distorted the trip
distribution.
• Combined trip rate/distribution analysis
to estimate future year demands is not
representative and/or doesn’t account
for appropriate trip-making activity/
response across whole study area.
Population The assumed population growth
• Optimism bias in population projection;
and/or mix is not realistic, not
for example, high growth assumed in
appropriate for assessment, or has
significant weaknesses. local vicinity of activity.
• Outdated population projections.
• Predicted do-minimum operation
doesn’t support assumed population
growth levels.
• New growth/development areas
missing.
Other land use The assumed land use
• Regional land use development plans
development is not realistic, not
and strategies outdated, or effected by
appropriate for assessment, or has
significant weaknesses. optimism bias.
• Land use development not balanced
across wider region.
• Weaknesses in detail/content of land

use data inputs.
• New growth/development areas
missing.
Local land use/ Local land use development areas
• Development-related travel (or other
transport supply and/or local transport supply has a
land use changes, such as density
changes large effect on activity scenario
(including walking, cycling, PT changes) contributes a notable
travel related to local proportion of, or has a significant
development(s)). influence on, predicted future-year travel
volumes and/or direction of travel in the
vicinity of the activity. Associated
weaknesses/uncertainty in these
assumptions.
• Local transport supply changes
contribute to a notable proportion of, or
have a significant influence on,
predicted future-year travel volumes
and/or network operation in the vicinity
of the activity. Associated weaknesses/
uncertainty in these assumptions.
Other key trip Vehicle availability/access,
• The assumption that historical trip-
making household/family structure, pricing
making behaviours continue into future
assumptions/ (fuel, parking, tolls etc), technology
relationships influences etc are likely to effect may be important to test/review/
future travel and trip-making. consider in relation to the assessment.
• Key assumptions (for example, parking
charges, tolls, road pricing and similar)
play an important role in relation to the
activity scenario and/or overall travel in
the region in the future.
Transport mode Diversion of trips from other
• Activity may cause diversion of trips
diversion transport modes is a large portion
from other transport modes (vehicle,
assumption of the activity demand estimation
projection. walking, cycling, PT) and redistribution
of travel demand. If such diversions are
a significant part of demand estimates,
the risks are likely to be higher. These
risks will be further increased if there is
potential variability in the extent of
transport capacity to be provided as part
of the activity (for example, bus
frequency using a new PT corridor).
Note: Future-year demand estimation forecasts may be from calculations, trend analysis factors etc,
project models taking inputs from other sources/methods, and regional transport models.
Analysing benefit realisation risks

The analysis of benefit realisation risks is focused on the uncertainty of realising the proposed activity’s
estimated benefits after construction is completed. Even if the estimation of benefits is accurate, most
benefits are dependent on external factors in terms of their actual realisation in the future. If future
conditions are unfavourable there is a benefit realisation risk.
As a general principle, the consequence rating is extreme, severe, moderate, minor, or insignificant when
the benefit estimation deviation is >±20%, ±10%–20%, ±5%–10%, ± 1%–5%, or <±1%.
Table 111: Benefit realisation risks

Benefit Examine all relevant dependencies for benefit realisation and assess
realisation risks corresponding risks. Below is only an example of possible benefit realisation risks.
Dependency on If the benefit assumptions include economic growth that is higher than the historical
future economic average, then the overall risk to the benefits of lower economic growth must be
conditions assessed.
Dependency on Any activities or technologies that have not been implemented at the time of
future activities assessment, and are not included in the activity’s scope, but whose successful
or technology implementation is a prerequisite for benefit realisation are a risk to successful
benefit realisation. If the successful implementation of other activities is assumed,
these risks must be analysed.
Extreme events Extreme events, such as earthquakes, can affect an activity’s benefit realisation. If
an activity is likely to be affected by a periodic or once-off extreme event, then the
impact of the extreme event on the activity’s benefit realisation must be considered
and assessed.
Other future If future activities are likely to significantly affect the activity’s user volumes (greater
activities than 10%) then the analyst must conduct sensitivity tests to determine possible
future effects.
Diversion from Mode change benefits are likely to constitute a significant proportion of benefits for
private vehicle public transport service, walking and cycling, or travel behaviour change (TBhC)
activities. These benefits are difficult to estimate with precision, being sensitive to
the assumed elasticities and/or model coefficients. Stable iterative modelling
processes are required, linked to assignment procedures able to measure
accurately the impacts of small traffic changes.
Consequently, the risk associated with diversion from private vehicles, and the
associated benefits, should be noted, unless it can be convincingly demonstrated
that these risks are reduced by the particular modelling processes adopted.
Supply Supply relationships will generally include link capacities, free-flow speeds and
relationships speed-flow relationships (in the context of a traffic assignment). Parts of the
network can be significantly congested.
In this case transport analysts should conduct sensitivity tests allowing for a
uniform matrix change of ±5% or a uniform change in all saturated junction and link
capacities of ±5%.
Routing The routing parameters control the relative effects of time and distance (and any
parameters other factors) on the choice of route. If an activity is a longer distance than the
existing infrastructure and of a much higher standard than existing route then the
transport analyst should conduct sensitivity tests allowing the nominal parameter
value to vary by ±50% or some equivalent increment.
Analysing cost risks

The cost risks set forth in Table 112 are not exhaustive but should be treated as a guide. These are
expected to be managed throughout the whole project lifecycle according to the risk management
methodology. At each stage the level of detail available will differ and any new or emerging risks should
be managed and reported as soon as possible. Any cost risks that are specific to the activity should be
managed under the same approach and methodology.
The cost risks include economic costs and financial (delivery) costs. The economic costs will generally
not affect cash flow and their estimation may be less accurate than the financial cost risks. Both types of
costs can affect the BCR.
As a general principle, the consequence rating is extreme, severe, moderate, minor, or insignificant when
the cost deviation is >±20%, ±10%–20%, ±5%–10%, ±1%–5%, or <±1%.
Table 112: Cost risks

Environmental and Concerning each of the issues, the tests of risks are the same, and
planning concern issue identification, tractability and sensitivity and consultation.
Tangata whenua Identification: no environmental surveys or little consultation.
Emissions Tractability: contentious issues with conflicting requirements.
Landscape and visual
Sensitivity of the activity to extreme events: extreme events disrupt
Ecological effects delivery of the activity and are likely to affect its viability.
Archaeological and historic Consultation: significant consultation is required, but its extent cannot be
sites predicted.
Social networks and Parties involved: no prior contact and parties have no prior experience in
severance consultation process.
Economic/amenity impacts
New or changed designation and/or resource consents to be applied for.
on land users
Natural hazards
Land and property
Property acquisition Property still to be acquired from several owners with opposition
expected.
Property economic value No recent market valuation; approximate valuation established on an
area basis by zoning; land where change of use is possible in short to
medium term (such as rural land on urban periphery).
Earthworks
Knowledge of ground Required: high density of sampling; variety of techniques and data
conditions available; good exposure of conditions; data interpreted by two parties
(peer review).
Risk source: no or very little subsurface investigation or site exposure.
Complex/ Swamps, marine sediments, rock masses with steeply dipping clay-filled
unpredictable conditions seams, or moisture sensitive clays; high water table or pressurised
aquifers.
Road design form High cuts/fills, tunnels, bridges or viaducts.
Extent of topographical Hilly, mountainous terrain, heavily vegetated and little topographical data.
data
Source and disposal of High volume requirements, uncertain sourcing and resource consent
material ramifications.
Other engineering costs
Engineering complexity Complex solutions to difficult engineering issues.
Signalling and Signalling and communications infrastructure should generally be
communications considered a high-risk element of engineering costs.
Transport service Unless a transport service operating surplus/deficit (the balance of
operating surplus/deficit revenue and operating costs) forms a large part of total costs, it would
normally be classified as low risk.
Services Underground and overhead services may include (but not be limited
to) telecommunications or electricity cables, gas mains, water
mains and sewers.
Existence, location and Service authorities not contacted, or services data unreliable,
condition engineering details and condition unknown or poorly defined.
Site flexibility Constrained (normally urban) corridor with few options to accommodate
changes.
Cooperation of utilities Several authorities to be coordinated in the same work area and/or
poorly resourced and organised authority, or an authority in a state of
major organisational change.
Contingencies
Significant cost risks that cannot be realistically reduced by other means are covered by contingencies in
the cost estimate. The purpose of contingency is to increase the accuracy of cost estimates according to
the transport analyst’s best effort.
The overall contingency allocated should be specified and an indication given of the confidence attached
to the contingency, in terms of the likelihood of a cost over-run greater than the contingency.
Concerning the relevant contingencies, if the following six types are distinguished, then it is expected that
the contingency table focuses on items 4 to 6, while for most activities, items 1 and 3 would be allowed
for in uniform factors on costs; item 2 is excluded:
1. changes in scope definition arising from omissions

2. changes in scope definition arising from client instruction
3. estimating inaccuracy
4. identified risks which are not managed
5. known but undefined risks
6. unknown risks.
For more information on contingency please refer to SM014 – Cost estimation manual.
The reasonable contingency amount should be determined and justified according to the activity’s
uncertainty. An assessment of the required contingency is required.
As a general principle, a contingency of between 10%–20% of the total capital cost of an activity is
considered reasonable. If an activity has significant uncertainties, and a contingency above 20%, further
investigation, with a focus on resolving cost estimation uncertainties, should be prioritised rather than
proceeding to implementation with a higher contingency.
Risk treatments
Risk treatment involves an iterative process of:
• formulating and selecting risk treatment options

• planning and implementing risk treatment
• assessing the effectiveness of that treatment
• reporting and managing residual risks.
Risk treatments should be documented within an activity’s risk register according to SM030 Minimum
standard Z/44 – Risk management practice guide. The risk register template is available for use on all
transport activities. Some content of the risk management guide has been written to specifically apply to
Waka Kotahi state highway activities, but the concepts are generalisable to all transport activities.
Risk treatment options
The strategic directions available to treat risk will depend on the lifecycle of an activity. These include:
• avoiding the risk by deciding not to start or continue with the activity that gives rise to the risk
• taking or increasing the risk in order to pursue an opportunity
• removing the risk source
• changing the likelihood
• changing the consequences;
• sharing the risk (eg through contracts, buying insurance)
• retaining the risk by informed decision.
These strategic directions may in turn lead to the following actions:
• abandon the activity
• reformulate the activity to capture the majority of the benefits at reduced cost
• conduct further investigation to reduce one or more of the identified uncertainties (either physical
investigations of more detailed assessment of risks)
• defer further processing of the activity until information comes available that assists in reducing
the uncertainties
• defer further processing of the activity until the FYRR increases to the required cut-off level
• proceed to the next stage of processing, or to tender
• other justified actions.
Examples of risk treatment options are provided in Table 113.
When selecting risk treatment options, an organisation should consider the values, perceptions and
potential involvement of stakeholders and the most appropriate ways to communicate and consult with
them. Risk treatments, even if carefully designed and implemented, may not produce the expected
outcomes and can result in unintended consequences. Monitoring and review need to be an integral part
of the risk treatment implementation to provide assurance that the different forms of treatment become,
and remain, effective.
Monitoring and review
The purpose of monitoring and review is to provide assurance and continuously improve the quality and
effectiveness of process design, implementation and outcomes. Ongoing monitoring and periodic review
of the risk management process, and its outcomes, should be a planned part of the risk management
process with responsibilities clearly defined.
Monitoring and review should take place in all stages of the process. Monitoring and review includes
planning, gathering and analysing information, recording results and providing feedback.
Table 113: Risk treatment options example

Risk Examples of risk No Do more work on Purpose of risk Defer
treatment actions action, issue in: treatment
accept investment is to:
risk this later quantify reduce
phase phase risk risk
Base matrix Short-term emphasis on X X X - X -
matrix estimation,
validation and additional
validation data collection
Medium-term model X - X - X X
improvement/updating
Longer-term data X - - - - X
collection
Growth Ensure that planning X X X - X X
forecasts estimates are reliably
based on best practice
procedures
Assignment Collect more validation X X X X - -
data
Improve model X X X - X X
Crashes Collect more crash data X X X - X -
Defer activity until crash X - - - X X
rates can be determined
with greater confidence
Services Surveys X X X X X -
Relocation of services X - X - X -
Alternative road design X X - - X X
Geotechnical Surveys; increase X X X X X -
sampling density
7. Sensitivity and risk analysis > 7.6 Adaptive decision making
Environment Scheme selection X X - - X X

and planning Redesign/extend X X X - X -
consultation procedure
Natural hazard X X X X X -
Base Alternative design X X - - X X
engineering Can more be done to X X X - X -
reduce complexity risks?
Land and Scheme selection X X - - X X
property Early acquisition X X X - X -
Back to 1.11 Sensitivity analysis >>
7.6 Adaptive decision making

The cost–benefit analysis procedures in this manual consider risk and uncertainty as set out in the
preceding sections. While risk and uncertainty are features of all activity appraisals, it is particularly
important when dealing with disruption (in terms of the likelihood of outages, damage to infrastructure,
and economic loss from such events), and how transport users respond to such situations. Where
appropriate, these risks and uncertainties need to be considered in transport economic appraisals.
Adaptive decision making may in some circumstances offer a chance to consider risk and uncertainty
where there is deep uncertainty regarding future risks or occurrences.
Waka Kotahi has received a report ‘Climate change adaptation and investment decision making’
(Torshizian 2022), which in turn references Decision making under deep uncertainty: from theory to
practice (Marchau 2019). The first report identifies available methods for adaptive decision making in
relation to climate change and the pros and cons of the different methods. The recommendations of that
report have not been adopted due to the complexity and costs in carrying out the processes, and the
application of adaptive decision making may be limited to investments that involve deep uncertainty
relating to risks and benefits. However, adaptive decision making may offer scope for consideration of
how to deal with deep uncertainty and in such circumstances analysts are advised to seek input from
Waka Kotahi investment advisors on the use of adaptive decision making.
The focus of an adaptive investment decision-making process is to allow for flexibility by considering all
possible outcomes where there is deep uncertainty when selecting options for further investigation. Under
scenarios of deep uncertainty, adaptive decision making relies on plans that are designed to be adaptive
over time in response to how the future unfolds as deep uncertainties are resolved. A wide range of future
scenarios is explored, with a plan of action to respond to the need for adaptation, so that there is greater
confidence in whether an option can contribute to the outcomes and objectives that are sought.
Caveat on use of adaptive decision making

Waka Kotahi will consider the use of adaptive decision making where deep uncertainty exists. However,
its use is to be preceded by discussion with, and feedback from, Waka Kotahi investment advisors.
List of commonly used acronyms

AADT Annual average daily traffic
AGPWF Arithmetic growth present worth factors
AST Appraisal summary table
BCA Business Case Approach
BCR Benefit–cost ratio
CBA Cost–benefit analysis
FEC Final estimate of cost
FTM Fixed trip matrix
FYRR First year rate of return
GC Generalised costs
GHG Greenhouse gases
GIS Geographical information system
HCV Heavy commercial vehicle
HPMV High productivity motor vehicles
IDMF Investment Decision-Making Framework
IRI International roughness index (for roads)
ITS Intelligent transport system
MBCM Monetised benefits and costs manual
MoT Ministry of Transport
MoT TOF Ministry of Transport’s Transport Outcomes Framework
NLTF National Land Transport Fund
NLTP National Land Transport Programme
NPV Net present value
PAC Preliminary assessed cost
PIKB Planning and Investment Knowledge Base
PPP Public private partnership
PSD Passing sight distance
PT Public transport
PTD Percentage of time delayed
PV Present value
ROC Rough order cost
RP Revealed preference (surveys)
RUC Road user charges
SIA Social impact assessment
SP Stated preference (surveys)
SPPWF Single payment present worth factors

SSI Standard safety intervention toolkit
TBhC Travel behaviour change
TDM Travel demand management
TIO Transport Investment Online
USPWF Uniform series present worth factor
VC Volume to capacity
VEPM Vehicle Emissions Prediction Model
VOC Vehicle operating costs
VoSL Value of statistical life (VoSL)
VoT Value of time
WACC Weighted average cost of capital
WEB Wider economic benefit
References
References
Akcelik R (1981). Traffic signals: Capacity and timing analysis. ARRB research report 123.
Australian Transport Assessment and Planning (2016). Australian Transport Assessment and Planning
guidelines. T5 Distributional (equity) effects.
https://www.atap.gov.au/sites/default/files/t5_distributional_equity_effects.pdf
Australian Transport Assessment and Planning (2018a). Australian Transport Assessment and Planning
guidelines. M1 – Public transport. https://www.atap.gov.au/sites/default/files/M1_Public_transport.pdf
Australian Transport Assessment and Planning (2018b). Australian Transport Assessment and Planning
guidelines. T2 Cost benefit analysis.
https://www.atap.gov.au/sites/default/files/t2_cost_benefit_analysis.pdf
Australian Transport Assessment and Planning (2018c). Australian Transport Assessment and Planning
guidelines. O2 – Optimism bias. https://www.atap.gov.au/sites/default/files/o2-optimsim-bias.pdf
Australian Transport Council (2006). National guidelines for transport system management in Australia 5:
background material.
https://www.transportinfrastructurecouncil.gov.au/sites/default/files/National_Guidelines_Volume_5.pdf
Austroads (1988). Guide to traffic engineering practice part 2: roadway capacity.
Austroads (1989). Rural road design – guide to the design of rural roads.
Austroads (1993). Guide to traffic engineering practice part 6: roundabouts.
Austroads (2003). Rural road design – guide to the geometric design of rural roads.
Austroads (2012). Guide to project evaluation part 4: project evaluation data.
https://austroads.com.au/publications/economics-and-financing/agpe04-12
Austroads (2020). Guide to traffic management part 3: traffic studies and analysis.
https://austroads.com.au/publications/traffic-management/agtm03
Balcombe et al (2004). The demand for public transport: a practical guide. TRL report 593.
https://trl.co.uk/sites/default/files/TRL593%20-%20The%20Demand%20for%20Public%20Transport.pdf
Barnes G, KJ Krizek, P Mogush and G Poindexter (2005). Guidelines for analysing the benefits and costs
of bicycle facilities.
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.522.6714&rep=rep1&type=pdf
Bein P, CJ Johnson and T Litman (1995). Monetization of environmental impacts on roads. Planning
services branch. Ministry of Transportation and Highways, Victoria, British Columbia.
Beetham J et al (2021). Latent demand for walking and cycling. Waka Kotahi NZ Transport Agency
research report 676. https://www.nzta.govt.nz/resources/research/reports/676/
Charles P (2011). ‘Optimism bias in transport planning’. Transport Futures Institute.
https://transportfutures.institute/optimism-bias/
Committee on the Medical Effects of Air Pollutants (COMEAP) (2015). Statement on the evidence for the
effects of nitrogen dioxide on health. https://www.gov.uk/government/publications/nitrogen-dioxide-health-
effects-of-exposure
DEFRA (2015). Valuing impacts on air quality: updates in valuing changes in emission of oxides of
nitrogen (NOx) and concentrations of nitrogen dioxide (NO2). http://qna.files.parliament.uk/qna-
attachments/419202%5Coriginal%5C676%20-%20HL2275%20Air-quality-econanalysis-nitrogen-interim-
guidance.pdf
Denne T et al (2023). Monetised benefits and costs manual (MBCM) parameter values: results of a
survey to derive values for road safety, travel time and reliability. Waka Kotahi NZ Transport Agency
References
Dravitzki V et al (2001). Traffic noise guidelines for low noise areas in New Zealand. Transfund New
Zealand research report 190. https://www.nzta.govt.nz/resources/research/reports/190/
Douglas Consulting Services and Abley Transportation Consultants 2011. Trips and parking related to
land use. NZ Transport Agency research report 453:
https://www.nzta.govt.nz/resources/research/reports/453/
Duranton G and D Puga (2003). Micro-foundations of urban agglomeration economies. National Bureau
of Economic Research working paper 9931. https://www.nber.org/papers/w9931.pdf
Francis T, AP Roozenburg and SA Turner (2006). Predicting accident rates for cyclists and pedestrians.
Land Transport NZ research report 289. https://www.nzta.govt.nz/resources/research/reports/289/
Graham DJ and DC Mare (2009). Agglomeration elasticities in New Zealand. NZ Transport Agency
Heerdegen W (2013). The discount rate for the Economic Evaluation Manual 2013. NZ Transport Agency
EEM Technical Paper. https://www.nzta.govt.nz/assets/planning/investment/docs/research-paper-on-
discount-rate-revisions.pdf
Highways England (1996). Design manual for roads and bridges. Volume 12, Section 2, Part 1, Appendix
F. https://assets.highwaysengland.co.uk/roads/road-projects/a2-bean-ebbsfleet-junction-
improvements/Orders/I.8+DMRB+Part+1+Traffic+Appraisal.pdf
Hoban CJ et al (1991). A model for simulating traffic on two-lane roads: user guide and manual for
TRARR version 3.2, technical manual ATM 10B. Australian Road Research Board, Victoria.
Hughes J, A Wild and C Muzyk (2020). Developing a method for quantifying transport interdependencies.
Waka Kotahi NZ Transport Agency research report 671.
Kennaird A (1995). Accident Trends in New Zealand. NZ Transport Agency Research Report 47.
Kernohan D and L Rognlien (2011). Wider economic impacts of transport investments in New Zealand.
NZ Transport Agency research report 448. https://www.nzta.govt.nz/resources/research/reports/448
Koorey GF et al (1999). Assessing passing lanes – stage 2. Transfund NZ research report 146.
Koorey G (2003). Assessment of rural road simulation modelling tools. Transfund NZ research report 245.
Kuschel G et al (2022). Health and air pollution in New Zealand 2016 (HAPINZ 3.0) He rangi hauora he
iwi ora: Volume 1 – Finding and implications. Ministry for the Environment, Ministry of Health, Te Manatū
Waka Ministry of Transport and Waka Kotahi NZ Transport Agency.
https://environment.govt.nz/publications/health-and-air-pollution-in-new-zealand-2016-findings-and-
implications/
Land Transport NZ (2006). Pedestrian planning and design guide.
https://www.nzta.govt.nz/assets/resources/pedestrian-planning-guide/docs/pedestrian-planning-guide.pdf
Land Transport NZ/Energy Efficiency and Conservation Authority (2004a). Travel behaviour change
evaluation procedures – technical report.
Land Transport NZ/Energy Efficiency and Conservation Authority (2004b). Travel behaviour change
guidance handbook.
Land Transport Safety Authority (2004). Cycle network and route planning guide.
https://www.nzta.govt.nz/assets/resources/cycle-network-and-route-planning/docs/cycle-network-and-
route-planning.pdf
Marchau, VAWJ et al (2019). Decision making under deep uncertainty: from theory to practice.
https://library.oapen.org/handle/20.500.12657/22900
References
McWha V and R Tooth (2020). Better measurement of the direct and indirect costs and benefits of
resilience. Waka Kotahi NZ Transport Agency research report 670.
Ministry of Transport (2005a). Getting there – on foot, by cycle.
https://www.beehive.govt.nz/sites/default/files/Getting%20there.pdf
Ministry of Transport (2005b). Surface transport costs and charges: main report.
Ministry of Transport (2016). The social costs of road crashes and injuries 2015 update.
https://www.transport.govt.nz/assets/Uploads/Research/Documents/5eb1efb361/Social-cost-of-road-
crashes-and-injuries-2015-update.pdf
Ministry of Transport (2017). The social costs of road crashes and injuries 2016 update.
https://www.transport.govt.nz/assets/Uploads/Research/Documents/da762d8ad1/Social-cost-of-road-
crashes-and-injuries-2016-update-final.pdf
MRCagney and Ian Wallis Associates (2012). TransLink service and infrastructure evaluation framework
(SIEF), Part 1 – ramp-up profiles. Report to TransLink Transit Authority.
New South Wales Department of State and Regional Development (1997). Guidelines for private sector
participation in the provision of public infrastructure.
New Zealand Modelling User Group (2017). ‘NZMUGs micro time-of-day choice research’.
https://drive.google.com/file/d/1j9DleWOsSIIrdPTqxY1q1ZV6UxKU1Hqq/view
NZ Transport Agency (2009). Pedestrian planning and design guide.
https://nzta.govt.nz/resources/pedestrian-planning-guide/
NZ Transport Agency (2011). High-risk rural roads guide. https://www.nzta.govt.nz/resources/high-risk-
rural-roads-guide/
NZ Transport Agency (2013). High-risk intersection guide. https://www.nzta.govt.nz/resources/high-risk-
intersections-guide/
NZ Transport Agency (2016). Crash estimation compendium.
https://www.nzta.govt.nz/assets/resources/economic-evaluation-manual/economic-evaluation-
manual/docs/crash-risk-factors-guidelines-compendium.pdf
NZ Transport Agency (2019). Transport model development guidelines.
https://www.nzta.govt.nz/resources/transport-model-development-guidelines/
O’Fallon C and C Sullivan (2006). Increasing cycling and walking: an analysis of readiness to change.
Land Transport NZ research report 294. https://www.nzta.govt.nz/resources/research/reports/294
Ortúzar JD and LG Willumsen (2011). Modelling transport. Fourth edition.
Raux C and S Souche (2004). ‘The acceptability of urban road pricing: a theoretical analysis applied to
experience in Lyon’. Journal of Transport Economics and Policy, 38(2).
https://www.ingentaconnect.com/content/lse/jtep/2004/00000038/00000002/art00002
Smith MG (2019). Urban transport modelling in New Zealand – data, practice and resourcing. NZ
Transport Agency research report 659. https://www.nzta.govt.nz/resources/research/reports/659
Statistics New Zealand (2007). Regional GDP concepts, sources and methods.
http://archive.stats.govt.nz/~/media/Statistics/browse-categories/economic-indicators/gross-domestic-
product/regional-gdp/regional-gdp-concepts-sources-methods.pdf
Tate FN (1995). Assessing passing lanes – stage 1. Transit NZ research project PR3-0097
Tinch R (1995). The valuation of environmental externalities. Report for the Department of Transport,
London.
Traffic Design Group (2001): Guide to estimation and monitoring of traffic counting and traffic growth.
References
Thrush MJ (1996). Assessing passing opportunities. Transit NZ research report 60.

https://www.nzta.govt.nz/resources/research/reports/60
Tomecki AB, K Yushenko and A Ashford (2016). Considering a cost–benefit analysis framework for
intelligent transport systems. NZ Transport Agency research report 584.
Torshizian E, E Isack and A Fehling (2022). Climate change adaptation and investment decision making.
Report by Principle Economics Limited for Waka Kotahi NZ Transport Agency.
https://www.nzta.govt.nz/assets/resources/Monetised-benefits-and-costs-manual-technical-notes/Climate-
change-adaptation-and-investment-decision-making.pdf
Transit NZ (2003). Finance and toll projects: implementation guide.
Transit NZ (1994). Guide to estimating AADT and traffic growth.
Transportation Research Board (1994). Highway capacity manual. Third edition.
Transportation Research Board (2006). Guidelines for analysis of investments in bicycle facilities.
NCHRP report 552.
https://books.google.co.nz/books?hl=en&lr=&id=FTtEy3UwVRoC&oi=fnd&pg=PA1&dq=NCHRP+report+5
52&ots=D9rM_hsljH&sig=6Ow3zo0pvF4iP9Ja31Aj51Zg4II#v=onepage&q=NCHRP%20report%20552&f=
false
Treasury (2015a). Guide to social cost benefit analysis. https://treasury.govt.nz/sites/default/files/2015-
07/cba-guide-jul15.pdf
Treasury (2015b). The New Zealand PPP model and policy: setting the scene: a guide for public sector
entities. https://treasury.govt.nz/sites/default/files/2015-10/ppp-public-model-and-policy-sep15.pdf
UK Department of Transport (1988). Calculation of road traffic noise. http://bailey.persona-pi.com/Public-
Inquiries/M4-Newport/C%20-%20Core%20Documents/14.%20Noise%20and%20Vibration/14.2.1%20-
%20Department%20of%20Transport%20and%20Welsh%20Office%20Calculation%20of%20Road%20Tr
affic%20Noise.%201988.pdf
UK Department of Transport (2019). TAG Unit M4: Forecasting and uncertainty.
https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/93887
8/tag-m4-forecasting-and-uncertainty.pdf
US Department of Transportation, Federal Highway Administration (2000). Toolbox for regional policy
analysis. https://www.fhwa.dot.gov/planning/processes/tools/toolbox/
Vanclay F (2003). International principles for social impact assessment. Impact Assessment and Project
Appraisal, 21:1. https://www.tandfonline.com/doi/pdf/10.3152/147154603781766491
Venables AJ (2016). Incorporating wider economic impacts within cost–benefit appraisal. OECD
International Transport Forum discussion paper 2016-05. https://www.itf-
oecd.org/sites/default/files/docs/incorporating-wider-economic-impacts-cba.pdf
Victoria Transport Policy Institute. Online TDM encyclopaedia. www.vtpi.org/tdm/tdm12.htm
Waka Kotahi NZ Transport Agency (2019a). Discount rate and analysis period. A technical paper
prepared for the Investment Decision-Making Framework Review.
https://www.nzta.govt.nz/assets/planning-and-investment/docs/investment-decision-making-framework-
review-discount-rate-november-2019.pdf
Waka Kotahi NZ Transport Agency (2019b). Estimating economic benefits from increased international
tourist activity. A technical paper prepared for the Investment Decision-Making Framework Review.
https://www.nzta.govt.nz/assets/planning-and-investment/docs/investment-decision-making-framework-
review-tourism-benefits-november-2019.pdf
Waka Kotahi NZ Transport Agency (2020a). Health and active modes impacts. A technical paper
prepared for the Investment Decision-Making Framework Review.
References
https://www.nzta.govt.nz/assets/resources/MBCM-technical-papers-and-reports/health-and-active-modes-
impacts-march-2020.pdf
Waka Kotahi NZ Transport Agency (2020b). Impact on urban amenity in pedestrian environments. A
technical paper prepared for the Investment Decision-Making Framework Review.
https://www.nzta.govt.nz/assets/resources/MBCM-technical-papers-and-reports/impact-on-urban-
amenity-in-pedestrian-environments-march-2020.pdf
Waka Kotahi NZ Transport Agency (2020c). Transformative transport projects (dynamic WEBS and land
use benefits and costs). A technical paper prepared for the Investment Decision-Making Framework
Review. https://www.nzta.govt.nz/assets/resources/MBCM-technical-papers-and-reports/transformative-
transport-projects-may-2020.pdf
Waka Kotahi NZ Transport Agency (2021). Economic valuation of greenhouse gas emissions.
https://www.nzta.govt.nz/assets/resources/MBCM-technical-papers-and-reports/Technical-report-
Economic-evaluation-of-GHG-emissions-FINAL.pdf
Wallis IP (2004). Review of passenger transport demand elasticities. Transfund NZ research report 248.
Wallis I (2005). Implications of selected urban road tolling policies for New Zealand. Land Transport NZ
research report 270. https://www.nzta.govt.nz/resources/research/reports/270
Wallis IP (2013). Experience with the development of off-peak bus services. NZ Transport Agency
research report 487. https://www.nzta.govt.nz/resources/research/reports/487
Appendix 1: Demand estimation methods and guidance > Broad approaches to demand estimation
Appendices
Appendix 1: Demand estimation methods and guidance
Chapter 2: Demand estimation and mode share, provides background and guidance on different
elements relating to travel demand estimation.
This appendix provides guidance on techniques, methods and considerations for developing demand
estimates. This includes estimating and forecasting facility use by specific modes, general techniques and
guidance on developing demand estimates and forecasts, guidance on traffic modelling forecasts, the
specific areas of elasticity techniques and values, and evaluating congested networks (using fixed and
variable matrix techniques).
This appendix also contains guidance on developing a benefit–cost ratio (BCR) after variable trip matrix
methodologies have been used, and suggested checks to validate the methodology applied.
Broad approaches to demand estimation

Broadly there are four general techniques for developing demand estimates, these are described further
in section 2.4 and are briefly:
• first principle estimates, such as factoring, daily traffic volume estimates, and predicted facility
use estimates.
• simple mathematical models and approaches, such as growth trends, trip generation, and
application of elasticities.
• development of demand estimates for project transport models, models which do not have the
capability to estimate travel demands from land use
• application of regional transport models that do estimate travel demands from land use.
The sections below provide guidance, methodologies and in some cases procedures for developing
demand estimates for economic assessments. For guidance on situations when it may be appropriate to
use the approaches discussed below, see section 2.4 and Table 1.
Demand estimation methodologies and considerations

Estimating and forecasting facility use
There are three distinct procedures for forecasting the demand for transport services or facilities,
depending on whether the proposed activity is for a new service or facility, an improvement to an existing
service or facility, or a travel behaviour change activity.
The estimated future demand for the do-minimum and each option, including the proposal, must be
calculated.
Table 52 can be used to assess demand for a cycle facility when traffic counts have not been carried out
in the area.
Procedure for a new service or facility
Where a new transport service or facility is proposed, and there are no comparable services or facilities,
an evaluator could undertake a consumer preference survey. A demand estimate may be drawn from the
survey using a methodology appropriate to the proposed service or facility.
It is common practice to use preference values derived in other locations, including internationally, and
this is a valid alternative to undertaking specific stated preference or revealed preference surveys. In this
case, it is important to consider if the imported/transferred values are representative of local conditions
and adjust or sensitivity test accordingly.
The basis of the survey and demand estimate, and any underlying assumptions – particularly those
related to traffic growth rates – must be clearly stated in the evaluation report.
Appendix 1: Demand estimation methods and guidance > Demand estimation methodologies and considerations
Procedure for improvements to a service or facility

Forecasting demand for improvements to transport services or facilities involves the following procedure.
Table A1: Forecasting demand for improvements to services or facilities

Steps Action
1 Estimate the willingness to pay and elasticity of demand for the relevant quality
improvement using one of the following two methods:
• If the activity is for a major improvement to an existing service or facility, then a

specially commissioned stated preference survey could be undertaken to estimate
the willingness to pay and the elasticity of demand. Alternatively, import
appropriate values from other locations (within New Zealand or internationally).
• If the activity is for a relatively small change to an existing service or facility,
then inference of the willingness to pay for the specific service quality, and its
elasticity of demand, may be drawn from other comparable services or facilities.
Note: Where information from a comparable service or facility is used, full details of the
comparison must be provided.
2 Identify the relevant elasticity and cross-elasticity values for the user charges and service
quality changes. This may either be from the stated preference survey or using values from
other sources.
Note: Some values applicable to New Zealand are provided in section 4.4 and section 4.7.
3 If there is an increase in user charges, calculate the new demand for the service.
The total number of new and existing users is calculated according to the following
formula:
Qprice = [((P1- Pnew) / P1) x UCE x Q1] + Q1
where: Qprice is the total demand at the new average user charge
Q1 is the existing number of users
P1 is the existing average user charge
Pnew is the new average user charge
UCE is the user charge elasticity
4 Using the relevant elasticity value derived from the stated preference survey, or from an
alternative source, calculate the demand for the service or facility based on the change in
service quality.
Multiply the elasticity value by the number of new and existing users (Qprice) as calculated
in step 3, to derive the total demand for the improved service (Qquality).
5 Determine the proportion of new users transferring from private vehicles and other
sources.
Use cross-elasticity values that have been estimated for people transferring from private
vehicle to alternate services where available, or use other sources such as surveys.
Alternatively, section 4.4 contains appropriate indicative values.
The diversion rates given in this chapter for workplace travel plans with public transport
improvements may also be applicable.
6 Test the result’s impact on demand by varying the user charge levels and service quality
elasticity. From this testing, a more complete demand curve can be derived.
7 Compare the results of the demand estimate with other similar services, where feasible, to
check that the estimate is credible.
Travel volume growth trend analysis

Trend analysis is the practice of collecting, collating and analysing information (typically historical travel
volumes) to attempt to identify patterns to use to estimate unknown or uncertain values (typically
forecasting future year travel volumes). As described in section 2.12, some care needs to be taken with
trend analysis, as seemingly small inaccuracies can have a significant effect on future projections. The
following are important considerations when carrying out trend analyses using historical data to predict
future transport demand numbers:
• A suitably large sample of historical data points is required for robust trend analysis. Generally,
the longer the forecast horizon the more historical data points that are required to be examined.
• Ideally, seasonal effects should be excluded from a trend analysis. For example, predicting year-
on-year growth should be based on historical data points which are taken from periods of the
year where transport volume numbers are similar to other times of the year (for example, non-
holiday weekday autumn and spring periods). In some cases, where seasonality effects are more
significant (for example, locations with significant tourist and holiday activities), trend analysis
may need to concentrate on a specific month or time-of-year.
• Macro effects, such as national and international economic cycles, may need to be considered.
For example, the global financial crisis was known to reduce traffic volume growth in many
locations from 2007 to 2009. Trends calculated from datasets that include these years should be
treated with care. The Covid-19 effects, starting from March 2020 onwards, are a similar if not
more significant consideration, and in the Canterbury region the Christchurch earthquakes and
recovery should be considered from 2011 onwards.
• Local effects may need to be checked to ensure data points are comparable year to year. For
example, a larger land use development effecting nearby count sites, or a more significant
change to the transport system such as a new road opening, effecting local travel patterns.
• The data used in trend analysis should not be constrained by the transport supply. For example,
in the case of using road traffic counts, the data analysed should not be affected by queuing and
delays. This may be accounted for by extending the time period reviewed (for example, weekday
peak periods of three to five hours or considering daily traffic volumes. Re-routeing at peak times
may also be a consideration – volumes across multiple links may need to be examined
(commonly referred to as a screenline).
• When applied to vehicle counts, ideally light and heavy vehicle volumes should be separated and
trend analysis carried out separately. Light and heavy vehicle growth rates are likely to differ in
many locations.
A range of mathematical trend analysis methods are possible, such as:
• regression analysis
• annual change analysis, which is examining the growth rates between each subsequent year and
performing statistical analysis on this set of results
• point-in-time to point-in-time analysis, that is examining growth over a longer period of time.
There are noted risks in this approach, as it requires robust knowledge and understanding of both
points analysed.
Where trend analysis is informing key inputs to an economic assessment (for example, medium- to
longer-term forecast travel volumes), it may be appropriate to carry out several different trend analysis
methods and compare the results. As described in section 7.3, sensitivity tests on the outcomes (for
example, growth rates) from trend analyses are likely to be important and are generally straightforward to
carry out.
Trip rate and distribution analysis
Trip rates and distribution analysis can be used to develop demand estimates for specific land use
activities and areas as a first-principles style approach. The trip rate uses estimates of the travel volume
generated by an activity (often related to peak hours or times), and the distribution analysis estimates
where these trips travel to and from in the study area (usually expressed as percentages).
The traffic generation of any activity will be influenced by its location on the transport network and the
practical level of access to sustainable transport modes, such as public transport, cycle or walking, as
well as the time of day and day of the week. The peak period of traffic generation for the activity also
needs to be taken into account, as this will not necessarily coincide with the peak period of movement on
the adjacent road network. Together, these factors mean that some degree of engineering judgement is
required when forecasting the likely level of traffic generation for a new activity.
There are a range of sources that can be used to inform decisions on expected traffic generation rates.
These include but are not limited to:
• NZ Transport Agency research report 453: Trips and parking related to land use. (Douglas
Consulting Services and Abley Transportation Consultants 2011)
• Institute of Traffic Engineers, Trip generation manual
• Transport for New South Wales, Guide to traffic generating developments
• TRICS (Trip Rate Information Computer System, UK based) and TDB (Trips Database Bureau,
NZ Transport Group)
• New Zealand census data.
Where it is practical to do so, it is recommended that reference to recent surveys for similar activities and
locations are adopted for any assessment. The TRICS/TDB source includes the most recent information
for New Zealand.
Trip rate and generation considerations
Some care needs to be taken when applying trip rate and/or generation methodologies across a study
area to estimate future demands for a network. Trip rates may be estimated from driveway surveys – as
such across a network they may not account for trip chaining (travel involving multiple purposes and
multiple destinations) and pass-by trips (such as an A to B trip becomes two trips A to C, C to B and the
original A to B trip is removed). Because of these behaviours and factors, wide-spread application of trip
rate and/or generation methods for multiple land-use developments in a network may overestimate traffic
volumes.
In section 2.15, Sense checking forecasts includes recommendations for reporting key information. This
includes the total study area demand growth and per annum demand growth rates by time period and,
where appropriate, vehicle classification and travel mode. These are useful checks to confirm the
appropriateness of trip rate applications, particularly when applied to multiple land use activities across a
study area.
Trip distribution considerations
There are several approaches to developing trip distributions associated with trips generated by a specific
land use activity; for example:
• develop from nearby location, ideally with similar land use

• first principle analysis using geo-located data; for example, census journey-to-work or journey-to-
education data, electronic tracking data such as from Bluetooth sensors/mobile phones,
commercial transaction data, or similar
• potentially can be developed from travel volume information (for example, a series of traffic
counts), although this is noted as being unlikely to be robust as the specific activity information
may not be easily isolated from movements associated with other activities in the network and/or
it can be difficult to identify the actual desired origin and destinations in this form of data.
In section 2.15, Sense checking forecasts includes recommendations on checks and reporting associated
with trip distributions. This includes summaries by time period and potentially vehicle classification and
travel mode of sector-to-sector growth, trip length distributions, and flow difference plots and/or
geographic location growth figures.
Elasticities
As described in section 2.1: Demand estimation and mode share: Key concepts, in transport demand
estimation elasticities typically describe a percentage change associated with a demand response linked
or related to a change in the transport system or environment.
Cross elasticities refer to the percentage change in the consumption of a good resulting from a price
change in another, related, good. For example, an increase in the cost of driving tends to reduce demand
for parking and increases the demand for public transport travel.
Transport elasticities tend to increase over time, as consumers have more opportunities to take prices
into account when making long-term decisions. For example, if consumers anticipate that the future cost
of private vehicle use will be low, they are more likely to choose a suburban home located where there is
more dependency on using a private vehicle. Alternatively, if they anticipate significant increases in
driving costs, they may place a greater premium on having alternatives to private vehicle use, such as
access to public transport and shops within convenient walking distance.
These long-term decisions, in turn, affect the options that are available. It may take many years for the full
effect of a price change to be felt.
Long-run travel demand elasticities are typically two to three times short-run elasticities.
Calculating the potential demand for a new or improved service or facility using elasticities will generally
be based on willingness-to-pay values (derived from a stated preference survey) combined with data on
current users, and existing and proposed user charges.
Section 4.4: Evaluation of public transport service activities and section 4.7: Evaluation of freight activities
contain elasticity and cross-elasticity values that may be used for public transport or freight services
respectively.
Nature of demand
The demand for a new or improved service or facility depends on a number of factors such as:
• the current or base average user charge

• the nature of the change in service
• existing users’ willingness to pay for the new or changed service/facility, and
• the responsiveness of demand to changes in user charges (the user charge elasticity) or another
journey attribute (for example, in vehicle or walking time).
Factors affecting price elasticities
Even if stated preference surveys have been specifically conducted for an activity, caution needs to be
exercised when extrapolating the elasticities. Surveys may only cover a small range of price and quality
variations and therefore the calculated elasticities may not be valid for extreme changes of price or
quality.
The following factors can also affect how much a change in prices impacts travel activity.
Type of price change
• Vehicle purchase prices and registration fees can affect the number and type of vehicles
purchased.
• Fuel prices, emission fees, and government rebates for sustainable vehicles affect the type of
vehicle used.
• A toll on a road may shift some trips to other routes and destinations.
• Congestion pricing may shift travel times, encourage people to change mode or destination, and
reduce the total number of trips that occur.
• Residential parking fees are likely to affect vehicle ownership. A time-variable parking fee can
affect when trips occur.
Type of trip and traveller
• Commuting trips tend to be less sensitive to changes in prices than shopping or recreational trips.
• Weekday trips may have very different elasticities than weekend trips.
• Urban peak period trips tend to be price insensitive because congestion already discourages
lower-value trips.
• Travellers with higher incomes tend to be less price sensitive than lower-income travellers.
Quality and price of alternative routes, modes and destinations
• Price sensitivity tends to increase if alternative routes, modes and destinations are of good quality
and affordable.
Scale and scope of pricing
• In general, demand is more sensitive to the pricing of a narrowly defined transport attribute – for
example, peak period travel on major arterials – than more broadly defined transport attributes
such as total personal travel. This is because consumers have more alternatives in the narrowly
defined case.
Willingness to pay surveys
The following points are noted around the potential requirements to carry out specific user willingness to
pay surveys:
• Proposed activities for new transport services and or facilities or for major improvements to an
existing service/facility, and any activities entailing a subsidy or price change, may require a
specially commissioned study to establish users’ willingness to pay and the elasticity of demand.
• For small alterations to existing services or facilities, or where the required amount of financial
assistance is small, the demand estimates may be produced using willingness to pay values
drawn from other comparable services.
Independently developed project models
Independently developed project/assignment transport models are not linked to or fed by a regional
transport model (a model response to changes in land use and transport supply and predicts travel
demand). Due to the relative prevalence of regional transport models in New Zealand (see section 2.10),
independent project models are not all that common.
These models may be developed specifically for assessing a particular economic activity, or may have
been developed for other transport planning or traffic engineering work in an area and may be utilised for
an economic assessment
Demand estimates and forecasts in this form of model may be developed through one, or a combination,
of the methodologies described above. For example, trend analysis may be used to develop growth rates
on external trips passing through a network in combination with a trip rate/distribution approach to
develop future demand estimates for internal trips in the study area.
Elasticities may be applied within models of this nature to estimate travel demand responses. For
example, PT mode share changes resulting from significant PT service and facility improvements.
The considerations and guidance that are discussed in the sections above in relation to methodologies
and approaches used to develop demand estimates apply if and where these approaches are used in
independent project models.
Applying regional models
Regional transport models exist in all of the major cities and many of the larger towns around New
Zealand. A list of existing transport models is provided in the first table in Appendix A in Urban transport
modelling in New Zealand – data and practice and resourcing (Smith 2019). Generalised guidance on
using regional transport models to produce demand estimates and travel forecasts is not provided as
these models are complex tools and should be operated by experienced transport modellers/planners.
Some commentary and considerations are noted in relation to the application of regional models to
economic assessments below.
For economic assessments, regional transport models may be used in isolation, primarily in the early
stages of establishing the feasibility of an activity, or applied to estimate demands that are then fed into a
project model. Extended use of existing regional models through into the later stages of assessing an
activity may apply to significant schemes in major urban centres effecting mode choice, such as large-
scale PT activities in Auckland and Wellington. In later assessment stages, often project models that are
specifically developed to assess the activity are applied. Processing the demands from a regional model
for a project model is described in Project models fed by regional models.
Depending on how the regional transport model is applied, it may produce either fixed or variable trip
matrices. Fixed trip matrices are where the demand for each mode, for a given time period and year, are
the same for the do-minimum and activity options. Variable trip matrices are where the demand is
different for the do-minimum and activity options. The default mode for almost all regional transport
models in New Zealand will be to produce variable trip matrices. Changes to the demand, and whether
each change will require a variable or fixed trip matrix benefit procedure, are shown in Table A2. More
background on these issues is provided in section 2.13: Fixed trip matrix and variable trip matrix
assessments.
Table A2: Fixed or variable trip matrices from a regional transport model
Changes Variable trip matrix? Fixed trip matrix?
Change in total number of trips produced (pure induced Yes
traffic)?
Change in mode choice? Yes
Change in destination choice (distribution)? Yes
Change in macro time choice? Yes
Change in route choice? No
There may be situations where a fixed trip matrix approach is adopted using a regional transport model.
In this case, a demand is estimated, and the same demand is assigned to both the do-minimum and
activity. It is more conservative to estimate this demand using the do-minimum, and then assign it to the
activity. If the activity is used to estimate the demand, and this demand is assigned to the do-minimum,
there is a higher likelihood that the do-minimum may be unrealistically congested and as a result,
inappropriately high benefits are calculated.
Project models fed by regional models
As described in section 2.10 regional models are relatively prevalent in New Zealand and, as such, it is
common for project models to be developed with links to a regional model. Project models with these
links/systems are available in many of New Zealand’s major urban areas and regional centres. The
analyst should check the availability and suitability of any existing project models for a specific economic
assessment.
For certain assessments (for example, single-stage and detailed business cases) it is common to develop
a specific project model for the purpose of analysing the economic benefits of a certain activity. A model
of this nature is also likely to be used for other components of the assessment, such as option testing,
informing the design, and providing a variety of technical and detailed information to inform the
assessment.
If a regional model (or regional modelling system/structure) is available, then it would be generally be
expected to be used to inform demand estimates (for example, base-year travel patterns and future-year
forecast growth and travel changes) in any project modelling carried out in the region and in the example
of a project model developed for a specific assessment.
Rarely, if ever, will it be appropriate to use the regional future-year origin–destination (OD) demand
directly within a project model. This is because in most cases the development of the project model will
include refinement and adjustment to OD demand inputs inherited from a regional model, and these
refinements/adjustments should be carried through in some manner to the future-year project model
demands. There are a number of methods for developing growth forecasts in a project model which
account for this, several are described below.
In the descriptions below, RM is regional model and PM is project model.
• OD additive growth method: the difference between the regional model base year (or closest
forecast year to the project model base year) and regional model forecast year is added to the
project model base year for each OD value. Where the base years are different, often a linear
growth rate is assumed between the base and future years. For example, for a regional model
base year of 2018 and a future year of 2028, 8/10ths of the growth is added to the project model
2020 base year to create the project model 2028 forecast.
The possibility of this resulting in a negative value on an OD volume needs to be considered and
accounted for. One simple approach is to use the multiplicative method (described below) as a
substitute in any OD cells that result in negative values from the additive calculation. That is, if
the additive growth method results in negative values, multiply the project model OD base volume
by the percentage reduction predicted through the regional model.
An additive method is often a straightforward and low-risk approach. It is simple to review and
check, and in particular, reflects changing future travel patterns across the study area predicted
by the regional model (for example, larger levels of growth in one specific geographic area due to
greenfield land use development).
One weakness of the additive method is that it disconnects the growth change from an
associated trip rate basis.
• OD multiplicative growth method: the regional model future-year OD value is divided by the
regional model base-year OD value producing a factor/percentage growth which is applied to the
project model base year OD value. In the same manner as the additive method, if the project
model and regional model base years are different a proportion of the regional model future-year
growth is applied. For OD cells where there are zero trips in the regional model base year and
trips in the regional model future year, the additive approach will need to be used.
Generally, more care is required with this approach than the additive method. In particular, large
factor increases can be applied to high project model OD values resulting in inappropriately high
project model future-year OD values.
• OD combined additive and multiplicative growth method: combining the additive and
multiplicative methods is possible; for example, averaging the additive and multiplicative growth
estimates or a weighted average – purely as an illustrative example: 75% additive growth and
25% multiplicative growth.
• More complex methods, including trip-end approaches: other growth calculations are
possible, examples include:
o Furnessing (Furness iterative balancing) the predicted regional model trip end (matrix
row and column totals) growth with the row/column distribution in the base year OD
matrix
o applying the row or column growth only, and using the project model base year OD
distribution to distribute the trip growth through the matrix.
These approaches may be considered when there is an understood, or established, notable strength in
the distribution in the project model base year OD demand (for example, if the development of the project
model and base year OD demand made more extensive use of OD surveys and/or additional OD-style
observed data) and an understood or clear local weakness in the regional model distribution.
In all cases, as outlined Project/assignment transport demand forecast checks and reporting, various
checks and reporting would be anticipated in the development of growth forecasts for a project model.
Checking the sector-to-sector growth distribution (comparing the regional model to the project model)
would be particularly important for a more complex method, such as growth Furnessing.
Appendix 1: Demand estimation methods and guidance > Methods to calculate/adjust demand in transport models
Walking and cycling demand estimation

Procedures for estimating bicycle demands can be found in section 4.2. Table 52 may be used for
estimating the demands of a new cycle facility.
In addition, some regional transport models will estimate bicycle and walking demands and may be
appropriate to estimate active mode facility use. The regional transport models forecast aggregate
demand and may be too coarse for bicycle or pedestrian facility assessment. Therefore more detail is
provided in relation to walking and cycling in these sections than other aspects such as public transport,
where regional models may be more robust.
Several New Zealand urban centres have specific “cycle models”. If assessing a new facility, or a
significant upgrade to an existing facility, the suitability, ability and appropriateness of these tools to
provide cycle volume estimates for an assessment should be checked.
Methods to calculate/adjust demand in transport models

This section provides guidance on the evaluation of transportation activities where networks are highly
congested or where the demand in the do-minimum and the activity are different due to a change in
supply introduced by the activity. The procedures describe a range of modelling techniques that can be
used to estimate demand when transport models do not converge to equilibrium due to significant levels
of congestion, models gridlock, or when there are significant changes to the distribution of trips or modal
share.
This section also contains guidance on developing a benefit–cost ratio (BCR) after variable trip matrix
methodologies have been used, and suggested checks to validate the methodology applied.
Travel behaviour change demand estimation
Procedures for travel behaviour change activities
Active mode facilities
Education, promotion and marketing can be prime drivers for generating demand for walking and cycling
(and change from use of private motor vehicles). The methodology for estimating travel impacts in section
4.6 should be used to estimate the number of trips using other modes (primarily private vehicles) diverted
to new or improved walking and cycling facilities, where this is part of a package including travel
behaviour change (TBhC) activities.
Where a new or improved walking or cycling facility provides a significantly improved quality of service,
there may be an induced demand, which is trips in addition to those diverted from other modes being
generated. The total demand for the facility may be estimated using the procedures in section 4.2.
Target population
The target population is the total population of the workplace, school, or community in which the
programme is being implemented. It includes the people who do not participate in the programme and
those who participate but do not change their behaviour.
Table A3: Travel behaviour change target population

Type of travel plan Definition of target population
programme
Workplace The total workforce (number of employees) at the workplace covered by the
travel plan. Make appropriate adjustment if a significant proportion of
employees work more or less than five days per week or if there is a
significant uptake of working from home.
School The total school student roll. If this is expected to vary significantly in the
next few years use an appropriate average.
Household and The total population of the community/suburb/area in which the household-
community based or community-based programme is being implemented.
Mode shift and mode share

Gaining the full mode shift impact usually takes around three years. Maintaining this mode shift then
requires constant investment of staff time and marketing resources in support of the activities.
Table A4: Composite evaluation of TBhC packages

Non-TBhC Benefits to existing users and non-TBhC target Comments
component population new users
New or improved Use the appropriate public transport service There is potential for
public transport evaluation procedure to: double counting of new
service • calculate benefits for existing users (whether user benefits. Care must
inside or outside the TBhC target population be taken not to count the
area) TBhC benefits of the
• calculate benefits for new users and target population twice.
associated externality (remaining road user)
benefits for the population located outside the
TBhC target population area.
New or improved Use the walking and cycling simplified procedure
cycle infrastructure to:
• calculate the cycling benefits for existing users
(whether inside or outside the TBhC target
population area)
• calculate the cycling benefits for any new
users from the population located outside the
TBhC target population area.
New or improved Consider if more walking trips will be created than There is potential for
walking infrastructure is given by the TBhC evaluation diversion rates, double counting of new
the walking and cycling simplified procedure can user benefits.
be used to estimate the additional benefits
associated with the extra trips.
New or improved Use the relevant procedure to calculate all benefits Minor road
road infrastructure, associated with the road infrastructure component. improvements including
including: improvements such as
• bus priority intersection treatment,
lane/high parking changes, road
occupancy crossings.
vehicle lane There is potential for
• road capacity double counting new
improvements user benefits where a
• minor road bus priority lane is
improvements proposed – see ‘New or
• traffic calming. improved public
transport service’ above.
Level of diversion
The level of diversion is used for calculating user benefits for new and existing pedestrians/cyclists,
assuming the main change is a reduction/transfer from private vehicle, and hence road traffic reduction
benefits. Benefits arising from changing travel mode from private vehicle to PT are also considered,
depending on the PT measures included in the travel plan. Care should be taken to ensure that any
assumptions are compatible with the economic evaluation requirements.
When conducting initial evaluations for workplace and school travel plans the diversion rate should be
selected based on the analyst’s knowledge of the organisations involved and the area. For the final
evaluation for implementation funding the diversion rate will be based on the actual features of the
completed plan.
Workplace travel plans
There are two sets of diversion rates for workplace travel plans:
• standard – where no public transport improvements are proposed

• alternative – where there are proposed public transport improvements.
Within these two sets of diversion rates, a scoring system is used to select the appropriate profile for a
given workplace travel plan. The score, out of six, is assigned based on the responses to the questions in
Table A4.
Table A5: Scoring system for workplace travel plan diversion rates
Travel plan questions Yes No

Is parking availability constrained at the workplace? 1 0
Does the proposed workplace travel plan include:
One or more parking management strategies? 1 0
Improvements to cycling/walking facilities? 1 0
Ridesharing matching service? 1 0
Public transport service improvements or company transport? 1 0
Public transport subsidies? 1 0
Total score:
Strategies for managing parking demand include activities such as parking charges, reduced supply of
parking spaces, parking ‘cash-out’ schemes, etc. Use the total score from above in Table A4. First, obtain
the reduction in the target population of car drivers assigned across the other modes of transport.
Table A6: Workplace diversion rates

Reduction in target population Mode share of the mode change
Score Reduction in Car as Public Cycling Walking
car as driver passenger transport
Standard – without public transport measures
Low 1 or 2 0.0% 0% 0% 0% 0%
Medium 3 or 4 -5.0% 26% 26% 12% 36%
Alternative – with public transport/company measures or improvements
Low 1 or 2 0.0% 0% 0% 0% 0%
Medium 3 or 4 -5.0% 26% 52% 6% 16%
High 5 or 6 -12.9% 26% 57% 8% 9%
The standard diversion rate values are applicable in most situations where no significant public transport
measures are included in the workplace travel plan. The alternative ‘with public transport service
improvements’ diversion rate values are applicable when significant public transport service
improvements (including company provided transport), subsidy schemes, or other similar measures
(covered by the last two questions in Table A4) are part of the workplace travel plan.
School travel plans
There are two default diversion rate profiles for schools, one for primary and another for intermediate and
secondary schools. Assign the change in car passengers across public transport, cycling and walking.
Table A7: School diversion rates

School type Car as driver Car as passenger Public transport Cycling Walking
Primary 0.0% -9.0% 0% 17% 83%
Secondary/ 0.0% -9.0% 55% 6% 39%
intermediate
Household and community-based activities

The standard diversion rate value is applicable for most activities, and situations where the activity will
implement fewer measures than ‘usual’ household-based programmes. For example, where a community
travel awareness campaign on its own would not achieve the standard diversion rate, public transport
services and cycling and/or walking facilities in the area are poor, and no significant changes to these are
envisaged as part of the travel behaviour change proposal. Assign the changes in car drivers and car
passengers across public transport, cycling and walking.
Table A8: Household and community diversion rates

Car as driver Car as passenger Public transport Cycling Walking
Low -1.0% -0.2% 42% 25% 33%
Standard -3.1% -0.5% 39% 25% 36%
Fixed trip matrices

This section relates to adjusting demand in congested networks to ensure demand does not exceed
supply where the same adjusted demand is then applied consistently for the do-minimum and
activity/option. This is referred to as a fixed trip matrix (FTM). In almost all cases, the demand will be
adjusted in the do-minimum (or the network with the least capacity) and then applied to the activity. This
is to ensure the network with the least capacity is not inappropriately congested.
Growth constraint techniques are to be considered where high levels of congestion are present in the do-
minimum network and/or where a stable network representation in which supply and demand are in broad
equilibrium cannot be achieved. Growth constraint techniques constrain traffic growth in peak period
matrices in highly congested conditions.
Procedure
Any one of the procedures listed below is available for traffic growth constraint; however, it is advised that
the shadow network technique be used with caution.
Peak spreading may be used on its own, or with any of the other procedures detailed here.
Having decided that there is insufficient capacity in the do-minimum to appropriately accommodate the
forecast travel demands, and that a realistic forecast of the future scenario requires the use of a matrix
growth constraint technique, follow the steps in Table A9 to apply growth constraint to the trip matrix.
Table A9: Steps to apply growth constraint to the do-minimum/activity trip matrix
Step Action
1 Determine whether to consider peak spreading (Table A67). If so, apply peak spreading
to modify the matrix and peak period.
If the matrix results in a realistic assignment to the do-minimum network, no further
capping need be considered. Otherwise go to step 2.
2 Select an appropriate method to cap the matrix:
Selected method Go to
Matrix scaling Applying the matrix scaling method
Incremental matrix capping Applying the incremental matrix capping method
Shadow network Applying the shadow network method
Elasticity methods (FTM) Applying fixed trip matrix with elasticity methods
Demand models (FTM) Applying fixed trip matrix with demand response
models
Automated growth constraint methods, such as the ME2 matrix capping technique
contained in the SATURN modelling package, may also be used.
Peak spreading and matrix scaling growth constraint techniques can be applied to the do-minimum and/or
the activity demand matrices to address an imbalance between supply and demand and produce specific
and varied demands. Additional analysis is likely to be required if different adjustments are applied to the
do-minimum and activity option, and advice on this is provided in the relevant sections.
Applying peak spreading
Peak spreading procedures may be used to spread traffic from the busiest part of the peak period to the
peak shoulders. Peak spreading is also called ‘micro time of day choice’. It is distinct from ‘macro time of
day choice’ where trips shift from a peak (morning or evening) to the interpeak or off peak due to
congestion.
Micro time of day choice methods may also be considered in assessing the activity, particularly in
situations where the activity enables destinations to be reached distinctly more quickly. This can result in
a peak contraction, and not accounting for this may misrepresent the travel time savings between the do-
minimum and the activity.
The procedure below is concentrated towards using peak spreading to reduce congestion in the do-
minimum.
Procedure
There are various existing procedures to calculate peak spreading. References include:
• Design manual for roads and bridges, Volume 12, Section 2, Part 1, Appendix F, ‘The application
of peak spreading’ (Highways England 1996)
• a UK Government-funded project that looked at demand and assignment modelling with a focus
on departure time choice resulting in a model called HADES (heterogeneous arrival and
departure times based on equilibrium scheduling theory)
• ‘NZMUGs micro time-of-day choice research’ (New Zealand Modelling User Group 2017).
Broadly there are two styles of transport modelling that effect the application of peak-spreading:
• Models where the total OD demand is considered over a longer peak period (at least two hours,
but often three to four hours) and the movement of traffic through these periods is controlled with
‘profiles’. Typically a number of profiles are developed and specified, often by vehicle
classification, in smaller time intervals (usually 5 to 15-minute increments). Microsimulation is one
example.
• Models where the OD demand is reflected in ‘peak periods’, such as peak one-hour or peak two-
hour OD matrices. Static assignment is one example.
For models where OD demand is profiled in shorter intervals across longer periods, the NZMUGs
research is notable as it demonstrates valid application of a micro time-of-day methodology in the New
Zealand context against observed data. Section 9.4 of the NZMUGs research provides a set of practical
considerations for applying peak spreading methods.
As a general guide, the following points should be kept in mind if peak spreading is applied to peak period
models:
• Decide whether to apply peak spreading uniformly or only to specific parts of the trip matrix. This
decision will depend on the extent and location of congestion in the network, and how realistic it
is for specific movements through the network to respond to congestion by adjusting their
departure times.
• Unless evidence suggests otherwise, it is recommended that the transfer of trips from the peak to
interpeak or off-peak periods be not more than 10% of the total peak period traffic, although this
will depend on the length of the time periods modelled, with shorter periods (for example, one
hour) likely to have greater volumes changing their travel times due to congestion.
• If appropriate, the traffic profile during the peak period may be adjusted, but it is advisable that
the reduction of the peak traffic intensity be no more than 10% unless evidence or justification
can be provided.
• It is recommended that information on local traffic profiles and trends in traffic growth for different
time periods, such as peak shoulder and business periods, be sought to support assumptions.
• It is preferable to apply the same consistent peak spreading adjustment to the demands applied
to both the do-minimum and activity. If different adjustments are warranted, then evidence and
justification should be reported. Checks should be carried out with the do-minimum peak period
adjustment, or profile, applied to both the do-minimum and the activity, and then the converse
(the activity peak period profile applied to both the do-minimum and activity). The resulting
benefits for each test should be compared to confirm the peak period adjustment is not
inappropriately inflating project benefits.
• If the peak spreading adjustment produces a change in total demand across the entire modelled
period for either the do-minimum or the activity, then variable trip matrix calculations will be
required (see Variable trip matrices). This will occur if the peak shoulders are not included in the
modelled period.
Applying the matrix scaling method
Matrix scaling procedures may be used to constrain growth in the trip matrix. If congestion is widespread,
the entire matrix may be scaled or, if congestion is confined to a particular area, only the corresponding
sections of the matrix need be scaled.
The final levels of congestion in the network due to the capped matrix should be sensible. When capping
the matrix with this procedure, only cap the matrix as much as needed. Excess capping is likely to reduce
computed project benefits unnecessarily.
Procedure
Follow the steps below to apply matrix scaling.
Table A10: Steps to apply matrix scaling

Step Action
1 Choose a scaling factor to reduce congestion to acceptable levels in affected parts of the
do-minimum network. As a general guide, link saturation ratios should normally be less
than 1.1 after the new matrix is assigned. Unless evidence suggests otherwise, the
scaling factor would typically be between 0.95 and 1.0 for scaling of the entire matrix, or
between 0.9 and 1.0 for scaling of selected sections of the matrix.
See also: calculating the volume to capacity ratio in Table A64.
2 Multiply the chosen elements of the matrix by the scaling factor. This matrix will be used
with the do-minimum and project options.
3 Assign the matrix to the network
This should be applied consistently to the do-minimum and the activity such that the demand matrix is the
same in order for a fixed trip matrix technique to be appropriate. Varying the matrix scaling between the
do-minimum and activity will require a variable trip matrix technique described in Variable trip matrices. If
different scaling is applied to the do-minimum and activity demands, justification and evidence will be
required and must be reported.
Applying the incremental matrix capping method
The incremental matrix capping method may be used to constrain growth in selected cells of the matrix.
This method is also known as the ‘incremental loading’ method and should not be confused with
incremental assignment techniques.
Procedure
In the incremental matrix capping method, choose a series of forecast year matrices and assign these to
the do-minimum network in chronological order. Once an assignment results in average travel speeds
dropping below acceptable limits for a matrix cell (or group of cells), further traffic growth is prevented in
the affected cells as later matrices are applied.
This process effectively restricts the growth rate in selected matrix cells to levels corresponding to some
earlier year (at which an acceptably realistic traffic assignment could be obtained).
Follow the steps below to apply incremental matrix capping.
Table A11: Steps to apply incremental matrix capping

Step Action
1 Choose a series of forecast years (say, at five-year intervals) and generate initial forecast
matrices for each of these years.
2 Select minimum allowable overall journey speeds for each origin-destination pair. As a
guide, minimum speeds will be in the range 15–25km/h, depending on the quality of the
route and the trip length.
3 Assign the first forecast-year matrix to the do-minimum network.
4 Update each matrix cell for the next future year, except those where the speed for the
origin–destination pair (obtained from the assignment run) has fallen below the minimum
allowable speed. Assign the new matrix to the do-minimum network.
5 Repeat step 4 until all future years have been assigned.
Applying the shadow network method

The shadow network method may be used to provide location-specific capping for a trip matrix.
The shadow network technique should be used with care. It may take more effort to implement and may
risk counter-intuitive results (for example, negative growth in some parts of the matrix).
Procedure
Follow the steps below to apply the shadow network technique.
Table A12: Steps to apply the shadow network technique

Step Action
1 Construct a duplicate ‘shadow’ network and connect it to the ‘real’ network at the zone
centroids.
2 Select minimum allowable speeds for the links of the shadow network. The choice of this
speed will affect the number of trips that are suppressed. As a general guide, minimum
speeds will be in the vicinity of 10km/h for links of average length (on very short road links,
intersection delays may realistically lead to very low overall link speeds), but this limit may
be varied to suit the particular network context.
3 Assign the matrix to the dual network.
4 Check the results and readjust the shadow network speeds if the results are
unreasonable. If the speeds are changed, repeat steps 3 and 4.
5 The real network will now contain normal trips and the shadow network trips considered to
be suppressed. To obtain a matrix for economic evaluation, cordon off the matrix assigned
to the real network.
Applying fixed trip matrix with elasticity methods

Fixed trip matrix (FTM) elasticity methods may be used to constrain growth in the trip matrix. As with
other fixed trip methods, the matrix produced by an FTM elasticity approach will be used for the do-
minimum and activity options.
Elasticity methods are based on the principle that the demand for travel between two zones varies
according to the cost of travel between the zones. An elasticity method iteratively adjusts the trip matrix
by assigning it to the network, measuring the change in costs between the assignment and a reference
case, then adjusting the demand according to the cost change.
Procedure
The inputs to an elasticity approach are:
• a pivot travel cost matrix from which changes in cost are measured – this is derived by assigning
the appropriate trip matrix to the network
• an initial estimate of the do-minimum matrix for the forecast year – this will usually be derived
either using a growth factor applied to a base matrix or from an external regional transport model
• an elasticity parameter that specifies the sensitivity of travel demand with respect to travel cost
• an elasticity formulation that expresses the necessary adjustment to the trip matrix as a result of
cost changes.
The pivot matrix and network will commonly be those for the base year, but it would be equally
appropriate to use the project opening year (if the network was expected to be relatively uncongested at
that time) as a pivot for forecasting trip matrices for later years in the project’s economic life.
Follow the steps below to apply elasticity methods.
Table A13: Steps to apply elasticity methods

Step Action
1 Assign the trip matrix from the base year to the base network. Obtain a pivot travel cost
matrix from the assignment results (𝑐𝑖𝑗𝑃 ).
2 Take an initial estimate (using suitable prediction methods) of the forecast year matrix 𝑇𝑖𝑗𝐹 and
1
assign it to the appropriate do-minimum network. Obtain an initial cost matrix 𝑐𝑖𝑗 from the
assignment results.
3 Derive a new matrix 𝑇𝑖𝑗1 by adjusting each cell in the matrix 𝑇𝑖𝑗𝐹 according to an elasticity
formulation. The power formula is advised for this purpose as follows:
1 𝐸
1 𝐹
𝑐𝑖𝑗
𝑇𝑖𝑗 = 𝑇𝑖𝑗 ( 𝑃 )
𝑐𝑖𝑗
where:
𝑇𝑖𝑗1 is the adjusted number of trips between origin i and destination j
𝑇𝑖𝑗𝐹 is the initial estimate of the number of trips between i and j
1
𝑐𝑖𝑗 is the forecast journey cost (or time) between i and j
𝑃
𝑐𝑖𝑗 is the pivot journey cost (or time) between i and j
E is the elasticity of demand with respect to journey cost (or time).
Note that the elasticity, E, will be negative.
Convergence may be assisted by using a damping process, and taking the average of the
matrices produced by the two previous iterations: ie, replace 𝑇𝑖𝑗1 by
1 𝐹
(𝑇 + 𝑇𝑖𝑗1 )
2 𝑖𝑗
4 Assign the new matrix 𝑇𝑖𝑗1 to the network, producing a new cost 𝑐𝑖𝑗2 matrix. Ensure that the
assignment converges satisfactorily.
5 Using the power formula, compute a new trip matrix 𝑇𝑖𝑗2 equal to:
𝐸
𝑐𝑖𝑗2
𝑇𝑖𝑗2 = 𝑇𝑖𝑗𝐹 ( )
𝑐𝑖𝑗𝑃
1
Damp as required, by replacing 𝑇𝑖𝑗2 by (𝑇𝑖𝑗1 + 𝑇𝑖𝑗2 ).
2
6 Repeat steps 4 and 5 until the process converges, that is, trip and cost matrices produced on
successive iterations are sufficiently similar.
As with all approaches to demand forecasting, the final matrix produced by the elasticity formulation must
reasonably represent the future year demand. It may be appropriate to exclude some matrix cells from
the elasticity adjustments – for example, those that exhibit negative growth (generally it is undesirable to
have cases where traffic volumes between an origin and destination pair decrease between successive
forecast years), unreasonably high growth or those that represent external trips.
Elasticities
Elasticities used with an elasticity method should reflect the sensitivity of demand to the user’s perceived
costs of travel, that is as used in the demand modelling process (not the resource costs, which typically
will be different – refer to Table A17).
The elasticities should also be consistent with the basis on which the user costs are expressed. It is
preferable that user costs and elasticities are expressed in terms of generalised costs (a combination of
time costs and money costs), rather than in terms of time or money alone (but see below). The
generalised cost approach allows demand to respond to both time and money changes and is found to
give more consistent results over a range of situations.
The application of elasticity methods depends on the transport model being able to model travel costs
realistically, and elasticities consistent with these travel costs being able to be estimated. In general,
elasticities specific to a study area will not be available and values from other locations need to be used.
Table A14 provides a set of default long-run generalised cost elasticity values for use in New Zealand
(principally urban) situations.
Table A14: Long-run generalised cost elasticities

Model competition Peak period Off-peak period
Low -0.4 -0.7
High -0.6 -1.0
It should be noted that:
• These elasticity values are constant, for use with a power function formula (as outlined earlier).
• These values essentially represent long-run responses, which may take some time (5+ years) to
materialise (short-run values would be significantly lower than these values, but are not usually
appropriate for activity evaluation purposes).
• The ‘low’ modal competition values should generally be used. However, in corridors to/from major
city central business districts where public transport has a substantial modal share, the ‘high’
modal competition values may be more appropriate.
• The values given do not allow for any significant time period switching effects, such as might
occur with a road pricing scheme involving differential prices by time of day. For such situations,
advice should be sought from Waka Kotahi and/or specific research undertaken.
If for any reason the model costs are expressed in terms of travel times only (that is, no distance
component included) rather than generalised costs, then equivalent travel time elasticities may be
calculated and applied. If the ‘true’ generalised cost function is t + k.d, but the model assigns on the basis
of travel time t, then the equivalent elasticity is obtained by dividing the generalised cost elasticity by the
factor (1 + k.v), where v is the average study area journey speed (in units of kilometres per minute).
Applying fixed trip matrix with demand response models
Regional transport models are commonly used to derive matrices or matrix growth factors that are
sensitive to transport system travel times.
Regional transport models refer to one or more of the standard generation, distribution and (optional)
mode split models generally handled by proprietary transport modelling software. In the main urban
centres, regional transport models are commonly used to generate matrices. Some project models are
also capable of modelling variable travel demands (for example by using trip distribution models).
Procedure
The forecast matrices derived from regional transport models are modified appropriately for the project
model and used in standard FTM project evaluation procedures.
Project models with demand responses can be applied in a similar way to elasticity methods (see
Applying fixed trip matrix with elasticity methods) using the demand response component of the model to
adjust the trip matrix, rather than an elasticity formulation.
In both options, and with all approaches to demand forecasting, the resulting matrix should be a
reasonable representation of demand, and the demand models should be properly validated (see the
Transport modelling checks worksheet. For a fixed trip matrix technique to be appropriate, the demands
must also be the same for the do-minimum and the activity.
Variable trip matrices
This section relates to cases where the demand is different in the do-minimum and activity/option for a
given forecast year. Where the activity introduces more capacity for a particular mode, the demand for
that mode with the activity will be higher than for the do-minimum. This is referred to as variable trip
matrices (VTM), that is, the demand varies between the do-minimum and the activity.
There are specific processes to calculate economic benefits where there are variable trip matrices, which
is described within this section.
Variable trip matrix methods are to be used for all complex improvements, unless:
• it can be demonstrated that:

o the congestion level expected throughout the analysis period in the do-minimum or option
will not be substantial, and
o the peak period public transport mode share changes (in the activity compared with the
do-minimum) by less than 5% based on total travel mode share -this will depend on
location and current mode shares, or
• preliminary evaluation shows that the fixed trip matrix benefits are unlikely to differ by more than
10% from those from a variable trip matrix approach, or
• Waka Kotahi approves the use of a fixed trip matrix approach for other reasons.
If the volume change between the do-minimum and activity increases typical peak period travel times by
more than 10%, then this is an indicator that variable trip matrix techniques may be required.
Variable matrix methods provide estimates of the effects of an activity on travel patterns (that is, the
difference between the do-minimum and option matrices) and on the benefits of the activity.
Procedure
Three variable matrix methods based on analytical techniques are recommended. These are:
• elasticity methods
• application of growth constraint techniques reducing the demand in the do-minimum and/or the
activity to produce equilibrium of supply and demand
• demand response models.
For demand response modelling approaches, where the source of data is from a regional transport
model, there may be the possibility that the regional transport model will not have sufficient sensitivity to
measure the impact on the trip matrix of a single scheme (unless the single scheme is significant in
nature), and the use of such models will therefore generally not generate a demand response that is
considered realistic. In examples where the scheme is considered significant enough to generate a
demand response, elasticity methods can be used to supplement the regional transport model.
Whatever method is applied, its results should be verified by comparison with an FTM evaluation based
on the do-minimum trip matrix.
Having decided that congestion will be significant in both the do-minimum and project option for a
forecast year, follow the steps below to apply variable matrix methods. If a variable trip matrix model (for
example, a regional transport model or a project model with demand response) is available, it should be
used.
Table A15: Steps to apply variable matrix methods

Step Action
1 Select an appropriate method to adjust the do-minimum and project option matrices:
Method Description Go to
A Use elasticity methods for both the do- Applying variable trip matrix
minimum and option matrices with elasticity methods
B Use other growth constraint techniques Applying variable trip matrix

(Table A9) for the do-minimum matrix with elasticity methods
and elasticity techniques to estimate
the effects of the activity on the trip
matrix
C Use a demand response model Applying variable trip matrix
(regional transport model or project with demand response
model with demand response) to models
produce the do-minimum and activity
demand matrices
D Use a demand response model to Applying variable trip matrix
produce the do-minimum and activity with growth constraint
demand matrices, then apply growth techniques
constraint techniques in a project
model to balance demand to supply
Alternatively, use a fixed matrix approach, then apply a predetermined correction factor to
adjust benefits for variable matrix effects. Any predetermined correction factor would
require robust justification, particularly if a significant increase in benefits resulted.
Note that project benefits will need to be calculated using a consumer surplus evaluation
and reported in the appraisal summary table.
2 Conduct a fixed matrix analysis (see Fixed trip matrices) and compare the results with
those obtained from the variable matrix analysis.
Applying variable trip matrix with elasticity methods

Variable trip matrix (VTM) elasticity methods are referenced in Table A15 (methods A and B). The two
recommended applications are:
1. where the do-minimum and activity option matrices are both estimated using elasticity methods,
or
2. where the do-minimum matrix is first established using growth constraint techniques and elasticity
methods are used to estimate the effect on this matrix of the activity option.
Elasticity methods are based on the principle that the demand for travel between two zones varies
according to the cost of travel between the zones. An elasticity method iteratively adjusts a trip matrix by
assigning it to the network, measuring the change in costs between the assignment and a reference case,
then adjusting the demand according to the cost change.
The inputs to an elasticity approach are:
• a pivot travel cost matrix from which changes in cost are measured – this is generally derived by
assigning the appropriate matrix to the network
• an initial estimate of the trip matrix for the forecast year
• an elasticity parameter that specifies the sensitivity of travel demand with respect to travel cost
• an elasticity formulation that expresses the necessary adjustment to the trip matrix as a result of
cost changes.
See Applying fixed trip matrix with elasticity methods for a full description of elasticity methods,
emphasising the estimation of the do-minimum matrix. The process is illustrated using the base matrix
and network as the pivot point, and the unconstrained forecast matrix (produced by growth factor
techniques or an external model) as the initial matrix estimate.
Refer also to Applying fixed trip matrix with elasticity methods for a discussion of suggested elasticities.
Method A procedure
For method A, the processes described in Table A13 are applied separately but consistently for the do-
minimum and activity option matrices. For example, if the method is pivoted on the base year matrices,
1
then steps 1–6 in Table A13 are applied first using the do-minimum network (in step 2 for 𝑐𝑖𝑗 and
1
subsequent steps) and then repeated using the activity option network (in step 2 for 𝑐𝑖𝑗 and subsequent
steps).
Method B procedure
Table A16: Steps to apply variable matrix method B

Step Action
1 Assign the do minimum matrix to the do-minimum network for the relevant forecast year.
Obtain a pivot travel cost matrix from the assignment results 𝑐𝑖𝑗𝑃 .
2 Use the do-minimum matrix as the initial estimate of the forecast year matrix 𝑇𝑖𝑗𝐹 and assign
1
it to the activity option network. Obtain an initial cost matrix 𝑐𝑖𝑗 from the assignment results.
3 Derive a new matrix 𝑇𝑖𝑗1 by adjusting each cell in the matrix 𝑇𝑖𝑗𝐹 according to an elasticity
formulation. The power formula is advised for this purpose as follows:
1 𝐸
𝑐𝑖𝑗
𝑇𝑖𝑗1 = 𝑇𝑖𝑗𝐹 ( 𝑃 )
𝑐𝑖𝑗
where:
𝑇𝑖𝑗1 is the adjusted number of trips between origin i and destination j
𝑇𝑖𝑗𝐹 is the initial estimate of the number of trips between i and j
1
𝑐𝑖𝑗 is the forecast journey cost (or time) between i and j
𝑐𝑖𝑗𝑃 is the pivot journey cost (or time) between i and j
E is the elasticity of demand with respect to journey cost (or time).
Note that the elasticity, E, will be negative.
Convergence may be assisted by using a damping process, and taking the average of the
1
matrices produced by the two previous iterations: ie, replace 𝑇𝑖𝑗1 by (𝑇𝑖𝑗𝐹 + 𝑇𝑖𝑗1 ).
2
4 Assign the new matrix 𝑇𝑖𝑗1
to the activity option network, producing a new cost 𝑐𝑖𝑗2 matrix.
Ensure that the assignment converges satisfactorily.
5 Using the power formula, compute a new trip matrix 𝑇𝑖𝑗2 equal to:
𝐸
𝑐𝑖𝑗2
𝑇𝑖𝑗2 = 𝑇𝑖𝑗𝐹 ( )
𝑐𝑖𝑗𝑃
1
Damp as required, by replacing 𝑇𝑖𝑗2 by (𝑇𝑖𝑗1 + 𝑇𝑖𝑗2 ).
2
Automated application of elasticity methods (for example, some traffic modelling software has built-in
capabilities such as ‘elastic assignment’ methods) may be used as an alternative to the manual method
given above.
For method B, the do-minimum matrix may be determined using any of the fixed trip matrix techniques in
Table A9.
As for FTM elasticity methods, the final matrix produced by the elasticity formulation (in either method A
or B) should be a reasonable representation of demand. It may be appropriate to exclude some matrix
cells from the elasticity adjustments – for example, those that exhibit negative growth, unreasonably high
growth or those that represent external trips. The convergence requirements for VTM methods are,
however, significantly more onerous: the stability and convergence requirements of the combined
VTM/assignment procedures are the same as for the simpler FTM assignment-only procedures (see the
Transport modelling checks worksheet, Base year assignment validation).
Applying variable trip matrix with demand response models
VTM activity demand response models may be used to estimate trip matrices differentiated between the
do-minimum and activity. As with other VTM approaches, these guidelines should be used only when
high levels of congestion exist in both the do-minimum and activity options. Applying a regional transport
model or a project model with a demand response component are options to develop VTM demand
response matrices. If the regional transport model is used alone (that is, without a more detailed project
model), the activity should have a significant impact, and/or the analysis should be in the earlier stages of
the planning process. In determining appropriateness, it would be necessary to demonstrate that the
model could be reliably applied to the appraisal of individual schemes.
In such cases, VTM evaluation procedures would be used. The stability and convergence requirements
are the same as for VTM elasticity methods (see Applying variable trip matrix with elasticity methods).
The validation of such models is discussed in the Transport modelling checks worksheet, Strategic
demand model check.
Applying variable trip matrix with growth constraint techniques
The final situation is where a demand response model (regional transport model or project model with a
demand response component) is applied to produce separate demand matrices for the do-minimum and
the activity, and these demands are then fed into a project model which is applied directly the economic
assessment.
While the demand response model may be in equilibrium for demand and supply, a demand/supply
imbalance may occur in the project model. This typically happens when the demand forecast by the
demand response model is significantly greater than the supply in the project model. In this situation,
growth restraint techniques would be applied in the project model to the do-minimum and/or activity
demands. Appropriate methodologies to constrain growth are listed in Table A9 and are most likely to
involve the application of peak spreading and/or matrix scaling.
In all cases, these techniques will be applied to reduce demand to be more aligned with supply. These
techniques should not be applied to increase demands in the do-minimum, which would result in greater
levels of congestion and an inappropriate increase in benefits. Similarly, the activity demands should not
be increased or decreased to artificially increase the benefits. These approaches are to reduce demand
forecast by one model so that it does not significantly exceed supply when used in a second model.
Conducting cost–benefit analyses using variable trip matrix methods
Where the demand matrix for the activity is different to that of the do-minimum, VTM methods (refer
Variable trip matrices) need to be applied. VTMs require more complex economic calculations than FTM
methods in order to determine activity benefits. This appendix gives advice on the calculations required,
and shall be used as a guide to summarising the net benefits and costs of the project options in the
appraisal summary table.
Background
For fixed matrix evaluations (where only the route of travel, at most, may change), the benefits are the
change in resource costs between the do-minimum network and the option. Where variable matrices are
involved, the benefits of the additional journeys must be included. These additional journeys may be
completely new trips (induced), or occur due to a change in destination (redistribution), a change in modal
choice, or a change in macro time choice associated with the activity. Since the decision to make
additional journeys is based on the costs perceived by users of the transport system, the measure of the
benefits is also based on perceived user costs, and is usually computed as the change in road user
surplus. It is also necessary to include a correction term to compute the total social benefits, since
transport system users do not take full account of the effects of their decisions on resource consumption.
This additional term is often referred to as the resource cost correction.
The resulting formula for the net project benefit is computed for each cell of the matrix individually (for a
given time period) and is:
1
Benefit = (𝑇 + 𝑇𝐷𝑀 ) × (𝑈𝐷𝑀 − 𝑈𝑂𝑃𝑇 ) + 𝑇𝑂𝑃𝑇 (𝑈𝑂𝑃𝑇 − 𝑅𝑂𝑃𝑇 ) − 𝑇𝐷𝑀 (𝑈𝐷𝑀 − 𝑅𝐷𝑀 )
2 𝑂𝑃𝑇
change in road user surplus resource cost correction

Or, rearranging terms:
1
Benefit = (𝑅𝐷𝑀 𝑇𝐷𝑀 − 𝑅𝑂𝑃𝑇 𝑇𝑂𝑃𝑇 ) + (𝑈 + 𝑈𝑂𝑃𝑇 ) × (𝑇𝑂𝑃𝑇 − 𝑇𝐷𝑀 )
2 𝐷𝑀
change in resource costs adjustment for variable trip matrix

where:
TDM is the number of trips in the do minimum
TOPT is the number of trips in the project option
UDM is the user cost of travel in the do minimum
UOPT is the user cost of travel in the project option
RDM is the resource cost of travel in the do minimum
ROPT is the resource cost of travel in the project option.
The implied subscripts i and j have been omitted for clarity.
For a fixed matrix evaluation when TDM equals TOPT, the second term is zero and this formula becomes
the simple difference in resource costs (the first term in the formula). While this first term can be
computed using matrix manipulations, it is standard practice to accumulate the resource costs over the
network links and use network statistics to estimate total network-wide resource costs in both the do-
minimum (the term RDM TDM) and option (the term ROPT TOPT). This is termed a link-based evaluation.
The values of time and vehicle operating costs given in the appendices are resource costs (which are the
actual costs of travel excluding taxation and other non-resource costs). Estimate user costs directly from
resource costs according to Table A17.
Table A17: Guidelines for estimating user time and vehicle operating costs
Cost component Obtain resource costs from … To derive the user cost …
Value of time Table 13, Table 14, Table 15 and User cost = resource cost
(working) Table 16
Value of time (non- Table 13, Table 14, Table 15 and User cost = resource cost × 1.15
working) Table 16
Vehicle operating cost (in urban networks):
Tables and graphs Table A79 to Table A88 User cost = resource cost × 1.2
of cost by average
speed and gradient
Tables and graphs Table 25, Table A89, Table A90 and User cost = resource cost × 1.125
of additional costs Table A91
for roughness
Tables of fuel costs Table A92, Table A93, Table A94 and User cost = resource cost × 2.0
due to bottleneck Table A95
delay
Graphs of Table A96 to Table A115 User cost = resource cost × 1.9
additional costs for
speed change
cycles
Matrix-based computation
For a variable matrix evaluation, adopt either of the following two methods to accumulate the net benefits
of project options:
1. a matrix-based analysis, where an average cost is computed for each origin–destination pair, or
2. a link-based analysis, where costs are computed separately for each link (or groups of links).
The first of these approaches enables benefits to be identified for particular travel movements, which may
be useful in identifying gainers and losers. The matrix-based analysis must be used where the model
includes time spent waiting queueing inside zones (for example, microsimulation models) and such
models must be run so that all trips complete. This is so that any differences in trip volumes and trip travel
times between the do-minimum and activity are not missed due to time spent waiting in the zone. This
applies for both FTM and VTM.
The second approach has the advantage that it allows benefits to be estimated for a region in the network
that is relatively self-contained, which can be useful for planning purposes.
Most network demand modelling software will allow benefits to be derived on a matrix (origin–destination)
basis without the need for the additional model runs needed for the second approach.
Create the matrices of trips and costs required to compute the benefits as itemised in Table A18.
Using matrix manipulations, compute the benefit matrix (for a single time period).
For a road activity with no tolls or a public transport activity with no fares, the formula for estimating net
benefits for any origin – destination (ij) pair will be:
𝐷𝑀 𝐷𝑀 𝑂𝑃𝑇 𝑂𝑃𝑇
1 𝐷𝑀
𝐵𝑖𝑗 = (𝑅𝑖𝑗 𝑇𝑖𝑗 − 𝑅𝑖𝑗 𝑇𝑖𝑗 ) + (𝑈 + 𝑈𝑖𝑗𝑂𝑃𝑇 ) × (𝑇𝑖𝑗𝑂𝑃𝑇 − 𝑇𝑖𝑗𝐷𝑀 )
2 𝑖𝑗
The total project benefit B is then given by the matrix total summed over all matrix cells.
Table A18: Required cost and trip matrices

Data Symbol Comment
Trip matrices 𝑇𝑖𝑗𝐷𝑀 , 𝑇𝑖𝑗𝑂𝑃𝑇 Available from the model
Resource and user 𝐷𝑀 𝑂𝑃𝑇
𝑅𝑖𝑗 , 𝑅𝑖𝑗 The constituent times and distances by link type are skimmed
cost matrices from the networks and the costs subsequently computed.
𝑈𝑖𝑗𝐷𝑀 , 𝑈𝑖𝑗𝑂𝑃𝑇
The same paths (and link speeds) should be used for both
resource and user costs.
In the case of public transport where a fare is paid by users, the net benefit for each ij pair will be:
Bij = [1/2(TDM + TOPT) (UDM - UOPT)] (perceived user benefits)
+ [(TDM PTRDM - TOPT PTROPT)] (change in public transport supply resource cost)
+ [(TOPT (OUOPT - OROPT) - TDM (OUDM - ORDM)] (change in other resource costs)
+ [TOPT FOPT - TDM FDM] (fare resource correction)
Where, for each ij pair:
T is the number of trips
U is the perceived cost/trip
F is the fare/trip (as included in the perceived cost of travel)
OU is the other perceived user cost/trip (eg generalised cost of travel time)
PTR is the resource cost of providing public transport/trip
OR is the other resource travel costs (eg travel time and environment)/trip.
Subscripts:
DM = do-minimum, OPT = option, U = F + OU, and R = PTR + OR
Perceived user benefits are calculated on an origin–destination basis (ie for each ij pair in the transport
matrix), with the total perceived user benefit being the sum of perceived benefits for all ij pairs. Other
benefit components can be calculated on a network basis. Calculation of the change in public transport
supply resource costs will generally be based on changes in the service quantity provided across a
network between the do-minimum and option, rather than on a cost per passenger trip. Usually the
change in public transport supply costs will be treated as a cost, in which case the item should be
removed from the formula above.
The equivalent formula applies in road tolling activity, where tolls are part of the perceived cost of travel,
with the value of F being the toll rather than the public transport fare. In addition to tolls, the value of U
includes the perceived value of travel time and the motorists’ perceived vehicle operating costs when
making travel decisions. The equivalent to PTR will be the direct resource cost of vehicle use, and the OR
counterpart will be the resource value of travel time, environmental and social externalities of vehicle use.
Again, the total change in perceived user benefits will be the sum of the benefit for each ij pair. Other
impacts can be estimated drawing on aggregate resources used in the network (eg total vehicle – km and
person – hours of travel) and total toll revenue. Unlike changes in public transport supply resource cost,
changes in the resource cost of vehicle use are treated as a benefit and so should be included as part of
the benefit formula.
Link-based computation
Link-based computation of activity benefits is possible, with the change in resource costs determined by
calculating resource costs multiplied by trips summed over the network (‘change in resource costs’), with
the component ‘adjustment for variable trip matrix’ calculated based on network statistics that require
some extra analysis as detailed below.
First, the extra term can be expanded to four terms to read:
1
(𝑈𝑂𝑃𝑇 𝑇𝑂𝑃𝑇 − 𝑈𝐷𝑀 𝑇𝐷𝑀 + 𝑈𝐷𝑀 𝑇𝑂𝑃𝑇 − 𝑈𝑂𝑃𝑇 𝑇𝐷𝑀 )
2
I II III IV
where each of these four terms (I–IV) may be computed from network statistics.
• I: This is the total user cost for the option network, and may be calculated in the same manner as
the resource costs but using the cost weights in Table A17.
• II: This is the total user cost for the do-minimum network, and may be calculated in the same
manner as the resource costs but using the cost weights in Table A17.
Terms III and IV require a particular network/assignment procedure called a ‘crossload’.
• III: This term uses the do-minimum network, but the user costs must be weighted by the trips in
the activity option matrix; this is achieved by loading the activity option matrix on the do-minimum
network keeping the paths and link speeds unchanged (that is, there are no speed or path-
building iterations and the paths and speeds are those determined from assigning the do-
minimum matrix); network statistics are then extracted and processed using standard techniques.
• IV: This term uses the activity option network, but the user costs must be weighted by the trips in
the do-minimum matrix; this is achieved by loading the do-minimum matrix on the activity option
network keeping the paths and speeds unchanged; network statistics are then extracted and
processed using standard techniques.
For the computation of variable matrix benefits using link-based evaluation, the assignment software must
be able to handle ‘crossloading’.
Having summed items I–IV and halved the result to obtain the ‘adjustment for variable trip matrix’, then
add the change in resource costs, (𝑅𝐷𝑀 𝑇𝐷𝑀 − 𝑅𝑂𝑃𝑇 𝑇𝑂𝑃𝑇 ) as described above. The result should be
recorded as VOC savings. Note that for use with this procedure, the road user surplus and resource cost
formulas should be applied to travel time and vehicle operating costs only (other benefits are assumed to
be unaffected by road user surplus issues). The remaining resource costs associated with crashes and
vehicle emissions will be recorded separately as Crash cost savings and Vehicle emission reductions.
Checking fixed or variable trip matrices
These checks are related to the procedures in Determining traffic volumes and may be used to check the
appropriateness of fixed trip matrix adjustments or variable trip matrix calculations for dealing with
suppressed and induced traffic, as well as changes in destination, modal, or macro time period choice.
The checks supplement the general model validation guidelines given in the Transport modelling checks
worksheet.
Suggested checks
Suggested checks include:
Table A19: Suggested checks for fixed and variable trip matrix calculations/adjustments
Method used Suggested information
The capacity of the do-minimum Demonstration that the capital cost of do-minimum
network was upgraded improvements is less than 10–15% of the project option cost.
Indication of adequate capacity (see below).
A growth suppression technique was Indication of adequate capacity (see below).
used (eg matrix scaling, incremental
matrix capping, shadow network, Details on the size and location of the suppressed travel.
elasticity method) Evidence, where feasible, of network performance before and
after growth suppression.
Details of the methodology applied.
Peak spreading was used Evidence of current variations in peak proportions:
• within the study area, in the base year and historically
• between cities or across New Zealand.
Based on this evidence, an indication that current traffic
profiles in the study area are more peaked than in other
locations or during other time periods.
Forecasts of a decline in peak period speeds relative to the
interpeak (because peak spreading is more likely to occur
when peak speeds deteriorate faster than interpeak speeds).
Alternatively, for models that cover longer time periods (more
than one hour) and predict behaviour for small time
increments, comparison of speeds at congested times
compared with uncongested where the peak spreading is to be
applied.
A variable matrix technique was used Indication of adequate capacity.
(eg, elasticity method on both the do-
minimum and activity option or a Differences between the do-minimum and activity option
demand response model was matrices.
applied) Evidence of the convergence of the method (ie stable
estimates of costs and matrices), or other evidence to justify
reliance on forecasts (see Transport modelling checks, Base
year assignment validation).
Details of the methodology applied.
Checking capacity in do-minimum and activity option

To check the do-minimum and activity option capacity, the following performance indices may be used. If
the indices suggest congestion over large or significant parts of the network, judged on the basis of at
least one hour of flow, then the network should be considered as congested. If, however, the congestion
occurs only in the later years of the economic life of the scheme (which contribute very little to the BCR),
these effects may be ignored where reasonable. For models that gridlock and will not produce viable
outputs, growth constraint techniques (peak spreading or matrix scaling) will need to be applied to
estimate benefits.
Table A20: Suggested checks for capacity

Performance indices Indicator of significant congestion
Level of service Level of service F*.
Matrix feasibility Network model is unable to achieve a stable
realistic assignment.
Demand response model is unable to produce
stable results, based on comparing outputs (for
example, person-hours travelled and person-
kilometres travelled) between successive
iterations of estimating demand.
Plots of link volume to capacity ratios or manual Ratios consistently higher than 1.0.
calculation of the ratio (see Calculating the
volume to capacity ratio)
Link speed plots Speeds consistently below realistic values (15–
25km/h) for links of average length.
Junction delay statistics Delays consistently longer than five minutes per
junction or queues ‘blocking back’ to upstream
links.
* Level of service (LOS) F is when forced or breakdown flow occurs or has reached a point that most
users would consider unsatisfactory. At LOS F, the amount of traffic trying to pass a point exceeds that
which can pass it. Queuing, delays and flow breakdown occur at these flow levels (Source: Guide to
traffic management part 3 (Austroads 2020)).
Back to 3.6 Impact on network productivity and utilisation: Travel time estimation procedures >>
Appendix 2: Crash analysis > Choosing the type of analysis
Appendix 2: Crash analysis

This appendix gives guidance on calculating crash costs for the do-minimum and option scenarios for a
route or site. For the purposes of this manual, a crash is an event involving one or more road vehicles
that results in personal physical injury and/or damage to property. Crash rates, crash prediction models
and crash reduction factors can be found in the Waka Kotahi Crash estimation compendium.
Choosing the type of analysis

This section of the manual provides further guidance on the selection of different analysis methods,
depending on the availability and quantum of crash data for the site. Evaluation of sites may require a
combination of analysis methods to assess the do-minimum and option cases.
Selecting the crash analysis method
There are three types of crash analysis methods available:
Follow the flowchart steps below to determine the need for a crash analysis and the selection of crash
analysis method(s). Details on each step in the process are discussed in more detail following the flow
chart in Figure A1.
Figure A1: Decision process for selecting crash analysis methods
Availability of crash history data

A reliable five-year crash history is not always available for a route or site. This can occur when there has
been a major change during the last five years, for example when a new intersection has been
constructed, or an intersection has been modified. If the change is likely to have altered the number and
pattern of crashes then the full five year crash history is not a reliable record of the current crash risk at
the site. Routes and sites that have been constructed during the last five years will also not have a full
crash history.
As shown in the decision process diagram, the historical crash period can be shortened to three or four
years if this is all that is available post changes. A period below three years cannot be used due to the
potential for regression-to-the-mean (see Definitions). For method A the criterion in terms of number of
crashes must still be met albeit with a shorter time period.
In some situations it is beneficial to use a longer crash history, as long as there has not been a major
change on a route or at a site, including a period of high traffic growth. A longer crash period of up to 10
years should be considered when traffic volumes are below 1500 vehicles per day and when looking at
crashes involving less common modes, such as pedestrians and cyclists.
Accuracy of crash data
The latest data available in the Waka Kotahi Crash Analysis System (CAS) should be used for crash
analysis. As there is typically a lag (up to three months) between the time when a crash occurs and when
it is entered into CAS, care should be taken to ensure that the data being used is complete.
When establishing the crash history, it is considered good practice to check all the traffic crash reports
(TCR) along the length of the site and up to one kilometre either side. Where possible, the location of
serious and fatal crashes should be discussed with the local police to confirm the location, particularly
along roads where it is suspected that crashes may have incorrect locations noted in the TCR. At sites
with low crash occurrence, the impact of an incorrectly coded crash in the TCR, particularly a serious or
fatal crash, can have a major impact on crash benefits (both positive and negative). It can also be useful
to look at other sources of data such as public and contractor reported crashes and, where available,
hospital records.
Minimum number of crashes required for method A
To use method A for crash analysis, there needs to be a minimum number of crashes at a site. In method
A (crash by crash analysis) future benefits are estimated from a reduction in historical crashes. For each
option, a reduction from 0 to 100% is applied to each historical crash.
When using method A the historical crash data from the last five years is used to predict the number of
crashes that will occur at the site over the next 40 years. When the crash record has few injury crashes,
or has a single serious or fatal crash, then this is often not an ideal basis on which to predict the number
of crashes that might occur over such a long period. In these circumstances the benefit of a safety or
roading improvement project can be overstated as the crash reduction may be applied to an over-
estimated number of future do-minimum crashes or crash costs. For example, when using five years of
data it can be unclear whether a single serious or fatal crash is a one in five year event or a one in 50
year event. When there are two or more serious and/or fatal crashes then it is a lot more likely that such
crashes occur frequently. Hence to reduce the likelihood of crash benefits being overstated, a minimum
number of crashes is specified for method A.
The criterion for method A is as follows:
At intersections or sites less than 1 kilometre in length, within the last five years there have been:
a. five or more injury crashes; and/or

b. two or more serious or fatal crashes.
At sites longer than one kilometre in length, within the last five years there have been:
c. three or more injury crashes per kilometre; and/or

d. one or more serious or fatal crash per kilometre.
Note: The number of injury, or fatal and serious crashes, is calculated by dividing the number of crashes
by the length.
For sites on low volume roads, with an AADT of fewer than 1500 vehicles per day (vpd) then a longer
historical crash period of ten years can be used. The 10-year history must be divided by 2 to obtain an
Appendix 2: Crash analysis > Applying the analysis methods
equivalent five-year history. If the crashes in this equivalent five-year period meet the criterion above,
then method A may be used.
Where a site does not meet these minimum requirements, then method C (the weighted procedure)
should generally be used. Method C still gives some consideration to the historical crash record but this is
combined with a prediction from rate or crash prediction model which has been developed for similar
sites.
An issue arises when there are no suitable crash rates or crash prediction models available to use
method C. In such circumstances there needs to be a discussion with Waka Kotahi, who may agree to
method A being used even though the threshold for crash numbers have not been meet. A primary
consideration in this situation is whether a recognised crash investigation specialist considers that the site
has significant safety deficiency (eg high-crash-risk sites) and therefore needs to be improved.
Fundamental change
When there is a fundamental change along a route or at a site, method B is generally used for analysis of
the option, while method C or A can be used for the do-minimum depending on the number of crashes
that have occurred at the site.
Where there is a fundamental change at a site but no crash rates or crash prediction models are available
for the do-minimum, method A may be used (subject to Waka Kotahi approval in writing) for the do-
minimum. While method B is used for the option cases, providing that models are available. Refer to
Definitions for more information on a fundamental change.
Availability of crash rates and models
Details on the available crash rates and crash prediction models are found in the Crash estimation
compendium. Crash rates and crash prediction models, other than those specified in this compendium,
may be used if the robustness of these rates or models can be demonstrated to Waka Kotahi or their
nominated peer reviewer.
Applying the analysis methods

This section describes the general process for how to determine future annual crash numbers and costs
for the do-minimum and option cases using the three analysis methods:
Worked examples of methods B and C are provided in Appendix 8: Worked examples.
Categorisation by speed limit
Crashes are categorised according to the speed limit areas in which they occur:
• 50km/h speed limit areas (including 30 km/h, 40 km/h and 60 km/h areas)
• 70km/h speed limit areas
• 100km/h speed limit areas (including 80 km/h and 90 km/h areas).
Change in traffic volume
If there is a change in traffic volume for the option compared with the do-minimum, then the crash
numbers must be scaled in proportion to this change.
Method A: Crash by crash
Figure A2: Method A flow chart
Method A overview
Crash-by-crash analysis is based on the crash history of the site and is dependent on the number of
reported crashes. The analysis uses the individual crash severity (see Definitions) categories (fatal,
serious, minor, non-injury) and these can be further disaggregated by movement category and/or type of
vehicle involved.
In the first stage of the analysis, using the Crash cost savings worksheet in the full procedures, the do-
minimum total estimated number of crashes per annum is calculated. Costs are assigned using the crash
costs from Table A28 to Table A31 for 50km/h speed limit areas and from Table A32 to Table A35 for
100km/h speed limit areas.
The number of crashes predicted for a project option is determined from an expected reduction in the do-
minimum crash numbers, based on guidance provided in the Crash estimation compendium. The forecast
percentage of crash reductions for the project option can be applied either globally or varied for each
crash type and severity (eg fatal, serious, minor and non-injury).
Costs are taken from Table A29 to Table A35, as appropriate to the site. Where the mean speed of traffic
for the do-minimum and/or options differs from that provided in the tables, an adjustment should be made
to the costs using the formula found in Adjusting crash costs to reflect mean speeds.
Changes in crash severity
Options, such as crash barriers, in some cases can reduce the crash severity at a site. In this situation
different crash reductions are applied to each historical crash depending on type and severity (eg fatal,
serious, minor and non-injury).
Use of local crash data
Waka Kotahi and local authorities have set up systems that involve the collection of local contact crash
data (also called ‘contractor reported’ or ‘unreported to police’ crashes) from contractors, local residents
and network management personnel. The quality of this data varies and caution should be taken when
using it in crash analysis.
Local contact crash data can be used in a crash-by-crash analysis (method A) where the data is
supported by sufficient evidence to be audited and there is reasoned justification provided as to why it
should be used to supplement information from CAS. Evidence might include a second independent
report of the crash, confirmation of crashes by the local police or by local network contractors or
consultants.
If local contact crash information is used for an analysis then under-reporting factors must not be
included in the calculations of injury or non-injury crash costs.
Redistribution of fatal and serious crash costs
The difference between occurrences of a fatal or serious crash at a site is influenced by random chance.
The severity of a crash can be influenced by various factors, including the roadside environment and the
location of major hazards like large trees and power poles. Given fatal crashes are rare events that have
a high cost, fatal and serious crashes are redistributed in accordance with the fatal to serious ratios in
Table A23, Table A24 and Table A25 for each crash type. This method applies for up to two fatal crashes
and unlimited serious crashes at each site. The exception is when three or more fatal crashes occur at a
site where the crash costs do not need to be redistributed at the site.
Vehicle involvement
In assigning costs to crashes using method A, crashes are classified by ‘vehicle involvement’ according to
the highest ranked ‘vehicle’ involved in a crash (refer to Definitions for further details).
Adjustment for under-reporting
Only a proportion of non-fatal crashes that occur are recorded on TCR and in CAS. This is referred to as
under-reporting. It is generally assumed that all fatal crashes are reported.
To counteract the effect of under-reporting when using method A, factors are applied to reported crash
numbers (TCR numbers) to estimate the total number of crashes that actually occur. Table A26 provides
factors for converting from reported injury to total injury crashes, while Table A27 provides factors for
converting from reported non-injury to total non-injury crashes.
If local contact crash information has been used, then under-reporting factors must not be included in the
calculations of injury or non-injury crash costs.
Method B: Crash rate analysis
Figure A3: Method B flow chart
Method B overview
Method B crash rate analysis involves determining a typical crash rate (refer to Definitions) per annum as
the basis for calculating the crash cost savings for a project. Typical crash rates have been calculated
using either a crash rate or crash prediction model provided in the Crash estimation compendium, which
have been derived using information from similar types of sites elsewhere.
In some cases, the rates and models used for the do-minimum and the option scenarios already account
for the proposed improvement/treatment of the site (eg an intersection treatment to change from priority
or a roundabout to signalised; the construction of a two-lane rural bridge to replace a single-lane bridge).
In others, it may be necessary to apply a crash modifying factor (CMF) from the Crash estimation
compendium to the option crash rate or model to take account of the site treatment/improvement (eg
various mid-block pedestrian treatments; construction of a cycle lane).
In crash rate analysis, it is not possible to differentiate crashes other than by speed limit category,
therefore the crash costs are taken from Table A36, Table A37 and Table A38, and are for ‘all vehicles
and all movements combined’. Where the mean speed of traffic for the do-minimum and/or options differs
from that provided in the tables, an adjustment should be made to the costs using the formula found in
Adjusting crash costs to reflect mean speeds.
Only reported injury crashes are considered when using crash rate analysis because of the inconsistency
in non-injury reporting rates between districts. The crash costs in Table A36, Table A37 and Table A38
take into account the typical number of unreported injury crashes, the number of non-injury crashes, and
the proportion of crashes of each severity (refer to Definitions) per reported injury crash.
Refer to the calculation of future crash benefits section below for details on calculating future safety
benefits when using crash prediction models. Use the Crash cost savings worksheet.
Changes in crash severity
Changes in crash severity can be calculated using method B when methods A and C are not appropriate
for the option case. Refer to the Crash estimation compendium for the crash modifying factors for
treatments that impact on crash severity (eg safety barriers).
Calculation of future crash benefits
In most crash prediction models the relationship between traffic volume and number of crashes is non-
linear. When using crash prediction models, a prediction should be produced every five years through to
the end of the analysis period. Intermediate crash costs can be interpolated. If traffic volumes fall above
or below the traffic volume ranges specified by the model, the predictions must be capped at the lowest or
highest flow allowed for analysis purposes. The Crash cost savings worksheet should be used.
When using crash rates, future predictions are not required as the relationship between crash numbers
and traffic volumes is linear. In such circumstances, only future traffic volumes need to be checked that
they are within any ranges specified; otherwise, the benefits need to be capped. The Crash cost savings
worksheet.
Method C: Weighted crash procedure
Figure A4: Method C flow chart
Method C overview
The weighted crash procedure uses both historical crash data relating to a particular site, and the typical
crash rate (refer to Definitions) for the site. The typical crash rate is calculated using the appropriate crash
rates or crash prediction models and crash modifying factors (CMFs) in the Crash estimation
compendium.
The historical data is converted into a site-specific crash rate (refer to Definitions) by dividing the reported
crashes by the number of years of data. The site-specific crash rate is then combined with the typical
crash rate, resulting in a weighted crash rate (refer to Definitions) for the do-minimum and the option(s).
Crash cost savings for the do-minimum and option(s) are calculated using the costs provided in Table
A36, Table A37 and Table A38. Where the mean speed of traffic for the do-minimum and options differs
from that provided in the tables, an adjustment should be made to the costs using the formula found in
Adjusting crash costs to reflect mean speeds.
The weighted crash procedure also allows analysis of sites with no crash history (refer to Definitions), if
the site has been in existence for more than three years with no major changes and the site is assessed
to have a high crash risk.
Weighted crash rate for the do-minimum
The do-minimum weighted crash rate (refer to Definitions) is calculated using the following equation:
AW,dm = w × AT + (1 – w) × AS
where: AW,dm is the do-minimum weighted crash rate
AT is the typical crash rate calculated from the appropriate crash rate or crash prediction model
for the do-minimum
AS is the site-specific crash rate (from historical crash data)
w is the weighting factor
Weighting factor (w)
When w = 1, the method simplifies to a crash rate or crash prediction model (method B).
When w = 0, the method simplifies to a crash-by-crash analysis (method A). w is calculated using the
following equation; where k is specified in the Crash estimation compendium:
K
w=
k + AT(km) × Y
Where: k is a dispersion parameter (refer to Definitions)
AT(km) is typical annual crash rate per site or kilometre (for mid-blocks)
Y is the number of years of crash records
For mid-block sections, the typical crash rate (AT) must be divided by the length of the mid-block because
the mid-block k values provided in the Crash estimation compendium are on a per kilometre basis. In all
other situations AT is for the full length of the mid-block section.
Reliability of crash history
An assessment of the reliability of both the site-specific crash rate and the typical crash rate is required
for method C. The reliability factor for the site-specific crash rate is αX and the reliability factor for the
typical crash rate is αM.
The main factor influencing the reliability of the site-specific crash rate is whether crashes are correctly
coded at the site. Crashes may be missing or incorrectly coded within the site. For example, a crash may
be incorrectly coded within a series of back-to-back curves, where it is not always easy to accurately
locate the exact curve the crash occurred on.
When the historical crash data is reliable, αX should equal 1.0 (this is the default setting). When it is
unreliable, αX should be between 1.0 and 2.0, with 2.0 being very unreliable data.
Reliability factors (αX, αM)
The reliability of the typical crash rate information presented in the Crash estimation compendium is an
issue when a crash rate or crash prediction model is used for:
• a different type of site, or part of a site, than that the rate or model was derived for – for example,
a four-arm traffic signal model might be used for a five-arm traffic signalised intersection (the
prediction would then be approximately 125% of that given by the model), and
• a ‘non-standard’ intersection, mid-block or other site or part of a site – an example of a ‘non-
standard’ intersection would be one with many traffic signal phases (say five or six) or greater
than four approach lanes or a priority seagull.
Appendix 2: Crash analysis > Crash trends
In both situations αM should be increased above 1.0 (the default value). A value of 2.0 would represent
poor reliability.
Weighted crash rate for project option
Method C can only be used for the project option when it does not bring about a fundamental change in a
site (refer to Definitions). In this case, the site-specific historic crash data is still relevant for the project
option analysis. The project option weighted crash rate is calculated by increasing or decreasing the
typical crash rate of the project option by the same proportion used to adjust the do-minimum typical
crash rate to the do- minimum weighted crash rate.
AW,opt = AT,opt × AW,dm / AT,dm
where: AW,opt is the weighted crash rate for the option case
AW,dm is the weighted crash rate for the do-minimum
AT,opt is the typical crash rate calculated from the crash rate or crash prediction model
for the option case. Note: It may be necessary to apply a crash modification factor (CMF)
from the Crash estimation compendium if the crash rate or crash prediction model does
not already take the treatment / improvement into account
AT,dm is the typical crash rate calculated from crash rate or crash prediction model for
the do-minimum
Crash trends
This section provides guidance on the adjustment of crash numbers for general crash trends.
General crash trends
Since 1985 there has been a downward trend in reported traffic crashes. At the same time crash numbers
have decreased and traffic volumes have increased, indicating that crash rates per vehicle have
decreased at a greater rate than crash numbers (Kennaird, 1995).
The combination of these two factors means that typical crash rates (refer to Definitions) established from
past research and site-specific crash numbers need to be adjusted in order to give a realistic estimate of
the likely crash situation at a project site in the future.
The adjustment to crash numbers is a two-stage procedure, with the first being to modify the crash
numbers at time zero. The second being to modify the growth rate used for discounting crash benefits to
take account of the forecast continued trend after time zero.
There have been differences between the crash trends in 50km/h areas compared with 70km/h and
above areas. Therefore, different factors are used to modify the crash numbers for the different posted
speed limit areas.
Table A21 provides factors to convert historic average crash numbers to time zero for method A. For
method B, an equation is provided to adjust the rate to time zero.
Table A22 provides factors to modify the predicted future traffic growth rate when discounting the crash
cost savings.
Adjustment to time zero
Crash numbers and rates for project evaluation are to be determined for time zero. This requires
adjusting the observed or predicted number of crashes assessed at the mid-point of the crash analysis
period to time zero (normally five years). The procedure differs if using the crash history (method A and
C) or crash rate analysis (methods B and C).
Method A adjustment
This procedure should be followed if using method A and C. From Table A21, select the appropriate
adjustment factor for the site based on its traffic growth rate and posted speed limit. For example, for a
project where the posted speed limit is 50km/h and the traffic growth rate is 2% at time zero, the crash
numbers will be factored by 0.90 to adjust the crash numbers to time zero.
Appendix 2: Crash analysis > Crash trends
Table A21: Crash trend adjustments factors

Speed limit Traffic growth rate
0% 1% 2% 3% 4% 5% 6% 7%
50 and 60km/h 0.83 0.86 0.90 0.93 0.96 0.99 1.03 1.06
70km/h and above 0.95 0.98 1.02 1.06 1.10 1.14 1.17 1.21
Method B adjustment
This procedure should be followed if using method B and C. As the crash rates and crash prediction
models in the Crash estimation compendium use historical crash data, the predicted number of crashes
needs to be adjusted for crash trends:
A = AT × (1 + ft (yz – 2006))
where: A is the crash rate adjusted for crash trends
AT is the typical rate found from models or rates
ft is the factor for adjusting the typical rate:
• -0.01 for sites with speed limits 60km/h and below
• -0.02 for sites with speed limits 70km/h and above
yz is year zero of the analysis period
Adjusting traffic growth rate for discounting
When discounting the crash cost savings from time zero forwards, the predicted growth rate is adjusted to
reflect the predicted continued trend in crashes. Table A22 provides the adjustments to use for the
different speed limit areas.
Using the factors in Table A22 it is possible for the crash growth rate used for discounting to be negative
if the predicted traffic growth rate at the site is less than 1% in 50km/h areas or 2% in 70km/h and above
areas. For example, if the site is in a 50km/h posted speed area and the traffic growth rate for the site is
1.5% then the growth rate to use for discounting crash costs is 1.5 - 1 = 0.5, ie 0.5% is entered in the
discounting equation.
Table A22: Growth adjustment factors

Posted speed limit
50 and 60km/h 70km/h and above
Modification to traffic growth rate -1% -2%
Adjusting crash costs to reflect mean speeds

Effect of speed on crash costs
Evidence indicates that injuries per crash and injury severity (refer to Definitions) increase linearly with
speed. To account for this in a crash analysis, the crash costs for the do-minimum and the option(s) are
calculated using mean traffic speeds.
Adjusting crash costs by movement and vehicle involvement
Crash costs by movement and vehicle involvement for use in method A are provided for 50km/h and
100km/h speed limits in Table A28 to Table A35.
Where the mean speed of the do-minimum and/or project options differ from these speeds, the crash
costs are adjusted using the following formula:
CV = C50 + (C100 - C50)(V - 50)/50
where: CV is the cost of crashes for the mean speed V
C50 is the cost of crashes in 50km/h speed limit areas
C100 is the cost of crashes in 100km/h speed limit area
V is the mean speed of traffic in km/h
Adjusting reported injury crash costs
Costs per reported injury crash for use in method B or C are provided for 50km/h, 70km/h and 100km/h
speed limits in Table A36, Table A37 and Table A38.
Appendix 2: Crash analysis > Tables
Where the mean speed of the do-minimum and/or project options differ from these speeds, the crash
costs are adjusted using the one of the following formulae:
for 50 < V < 70km/h: CV = C50 + (C70 - C50)(V - 50)/20
for 70 < V < 100km/h: CV = C70 + (C100 - C70)(V - 70)/30
where: CV is the cost of crashes for the mean speed V
C100 is the cost of crashes in 100km/h speed limit area
V is the mean speed of traffic in km/h
Calculation of mean speed
If the road section has a design speed based on the 85th percentile speed, then to convert the design
speed to the mean speed use the approximation of dividing the 85th percentile speed by 1.13 (or
multiplying by 0.885) and round the result to the nearest whole kilometre per hour.
Mean speed should be established over a section length of at least one kilometre.
Tables
Introduction
Table A23 through to Table A44 are for use in the Crash cost savings worksheet provided on the Waka
Kotahi website. These tables are used for calculating annual crash costs, depending on which of the
crash analysis procedures are used.
Table A23 through to Table A35 and Table A39 to Table A44 are for use with method A crash-by-crash
analysis, while Table A36 to Table A38 are for use with methods B and C crash rate analysis and the
weighted crash procedure.
• Table A23, Table A24 and Table A25 – ratio of fatal to serious crash severities by movement for
different speed limits.
• Table A26 – factors for converting from reported injury crashes to total injury crashes.
• Table A27 – factors for converting from reported minor injury crashes to total non-injury crashes.
• Table A28, Table A29, Table A30 and Table A31 – cost per crash by movement and vehicle
involvement for fatal, serious, minor and non-injury crashes in 50km/h speed limit areas for use
with method A, crash-by-crash analysis.
• Table A32, Table A33, Table A34 and Table A35 – cost per crash by movement and vehicle
involvement for fatal, serious, minor and non-injury crashes in 100km/h speed limit areas for use
with method A, crash-by-crash analysis.
• Table A36, Table A37 and Table A38 – cost per reported injury crash for methods B and C.
• Table A39 – ratio of fatal and serious/all injury factors by crash type, ONRC grouped and by
alignment type 80–100km/h.
• Table A42 – ratio of fatal and serious/all injury as a function of crash type by ONRC for 80–
100km/h.
70km/h.
50km/h.
Refer to Definitions for more information.
Table A23: Ratio of fatal to serious crash severities by movement for 50km/h speed limit areas
Movement category CAS movement codes Fatal/ Serious/
(fatal + serious) (fatal + serious)
Head on AB, B 0.08 0.92
Hit object E 0.06 0.94
Lost control off road AD, CB, CC, CO, D 0.13 0.87
Lost control on road CA 0.05 0.95
Miscellaneous Q 0.13 0.87
Overtaking AA, AC, AE-AO, GE 0.04 0.96
Pedestrian N, P 0.08 0.92
Rear end, crossing FB, FC, GD 0.07 0.93
Rear end, queuing FD, FE, FF, FO 0.07 0.93
Rear end, slow vehicle FA, GA-GC, GO 0.05 0.95
Crossing, direct H 0.05 0.95
Crossing, turning J, K, L, M 0.03 0.97
All movements 0.07 0.93
Head on AB, B 0.24 0.76
Crossing, turning J, K, L, M 0.10 0.90
Head on AB, B 0.36 0.64
Crossing, turning J,K,L,M 0.16 0.84
Table A26: Factors for converting from reported injury crashes to total injury crash
Speed-limit area Injured person category Fatal Serious Minor
50, 60 and 70km/h speed limit Pedestrian 1.0 1.5 4.5
Other 2.75
80 and 100km/h speed limit Pedestrian 1.0 1.9 7.5
(excluding motorways) Other 4.5
100km/h speed limit remote Pedestrian 1.0 2.3 13.0
rural area Other 7.5
Motorway All 1.0 1.9 1.9
All All 1.0 1.7 3.6
Table A27: Factor for converting from reported non-injury crashes to total non-injury crashes
Speed-limit area 50, 60 or 70km/h 80 or 100km/h Motorway
All movements 7 18.5 7
Table A28: Cost per crash by movement and vehicle involvement for fatal injury crashes in
50km/h speed limit areas
50km/h speed limit fatal injury crashes Total cost per crash by vehicle type ($M July 2021)
Movement category CAS movement Cycle Motor- Bus Truck Car, All
codes cycle van, vehicles
other
Head on AB, B 12.7 13.0 12.9 12.9 16.7 14.1
Hit object E 12.7 13.0 12.9 12.9 14.9 13.6
Lost control off road AD, CB, CC, CO, D 12.7 13.0 12.9 12.9 15.9 15.5
Lost control on road CA 12.7 13.0 12.9 12.9 14.9 14.2
Miscellaneous Q 12.7 13.0 12.9 12.9 12.7 12.7
Overtaking AA, AC, AE-AO, GE 12.7 13.0 12.9 12.9 14.9 12.7
Pedestrian N, P 12.7 13.0 12.9 12.7 12.7 12.7
Rear end, crossing FB, FC, GD 12.7 13.0 12.9 12.9 14.9 14.2
Rear end, queuing FD, FE, FF, FO 12.7 13.0 12.9 12.9 14.9 14.2
Rear end, slow vehicle FA, GA-GC, GO 12.7 13.0 12.9 12.9 14.9 14.2
Crossing, direct H 12.7 13.0 12.9 12.9 15.3 14.2
Crossing, turning J, K, L, M 12.7 12.9 12.9 12.9 14.9 13.9
All movements 12.7 13.0 12.9 12.9 14.9 14.2
50km/h speed limit serious injury crashes Total cost per crash by vehicle type ($000 July 2021)
other
Head on AB, B 684 684 707 731 1033 930
Hit object E 684 731 755 739 707 707
Lost control off road AD, CB, CC, CO, D 684 707 755 803 835 803
Lost control on road CA 684 684 755 771 827 739
Miscellaneous Q 692 684 755 684 787 755
Overtaking AA, AC, AE-AO, GE 684 684 755 771 819 700
Pedestrian N, P 755 787 684 739 692 700
Rear end, crossing FB, FC, GD 692 684 755 707 731 715
Rear end, queuing FD, FE, FF, FO 692 684 755 803 731 731
Rear end, slow vehicle FA, GA-GC, GO 684 707 755 771 835 739
Crossing, direct H 692 715 755 779 827 779

Crossing, turning J, K, L, M 700 707 747 851 779 755
All movements 692 707 755 771 787 755
Table A30: Cost per crash by movement and vehicle involvement for minor injury crashes in
50km/h speed limit Total cost per crash by vehicle type ($000 July 2021)
minor injury crashes
other
Head on AB, B 68 74 68 80 100 91
Hit object E 68 68 71 74 71 71
Miscellaneous Q 65 74 74 71 74 74
Pedestrian N, P 103 88 68 68 68 71
All movements 68 68 80 77 83 80
speed limit areas
50km/h speed limit non-injury crashes Total cost per crash by vehicle type ($000 July 2021)
other
Head on AB, B 1.4 1.5 6.3 8.7 3.0 3.5
Hit object E 1.4 1.5 7.3 8.6 2.9 3.7
Miscellaneous Q 1.4 1.5 8.0 7.8 2.3 3.7
Pedestrian N, P 0.8 1.7 0.3 7.2 1.7 1.8
All movements 1.5 1.7 4.1 8.6 2.8 3.1
Table A32: Cost per crash by movement and vehicle involvement for fatal injury crashes in
100km/h speed limit fatal injury crashes Total cost per crash by vehicle type ($M July 2021)
other
Head on AB, B 13.6 14.4 19.9 15.5 17.5 16.6
Hit object E 12.7 13.5 18.1 14.6 14.9 13.6
Miscellaneous Q 12.7 13.5 18.1 14.3 14.0 14.0
Pedestrian N, P 12.7 13.5 18.1 13.3 12.7 13.0
All movements 12.7 13.5 18.1 14.6 15.2 14.9
serious injury crashes
other
Head on AB, B 746 722 1075 882 1083 979
Hit object E 690 746 987 810 842 802
Miscellaneous Q 746 778 987 786 802 794
Pedestrian N, P 755 787 987 810 706 700
All movements 746 730 987 810 882 842
Table A34: Cost per crash by movement and vehicle involvement for minor injury crashes in
minor injury crashes
other
Head on AB, B 77 77 74 89 106 103
Hit object E 66 69 94 77 77 77
Miscellaneous Q 72 74 94 80 77 77
Pedestrian N, P 103 88 94 68 69 71
All movements 72 72 94 80 89 86
Table A35: Cost per crash by movement and vehicle involvement for non-injury crashes in
non-injury crashes
other
Head on AB, B 1.9 2.4 6.4 11.3 3.6 5.2
Hit object E 1.9 2.4 7.3 10.0 2.9 3.7
Miscellaneous Q 1.9 2.0 9.9 9.6 2.4 5.5
Pedestrian N, P 1.9 2.2 4.3 9.9 2.1 4.1
All movements 1.9 2.2 4.3 10.5 2.8 3.5
Table A36: Cost per reported injury crash

Crash site/type Cost per reported injury crash by speed limit area ($000 July 2021)
50km/h 70km/h 100km/h rural
Mid-block crashes 654 1,036 1,457
Intersection crashes:
Uncontrolled T 464 893 1,320
Roundabout 500 809 1,358
Priority T & Y 571 702 1,408
Priority X 547 726 1,433
Signalised T, Y 559 667 1,358
Signalised X 524 631 1,545
All other sites 524 809 1,445
Table A37: Cost per reported injury crash for specific sites
Crash site/type Cost per reported injury crash by speed limit area ($000 July
2021)
Motorway and four-lane divided N/A N/A 722
road crashes
Rural railway crossing crashes N/A N/A 3,276
Rural bridge crashes N/A N/A 1,420
Table A38: Cost per reported injury crash by mode

Crash site/type Cost per reported injury crash by speed limit area ($000 July
2021)
Heavy vehicle crashes (excludes 785 1,309 1,893
LCV and MCV)
Cycle crashes 559 1,238 1,607
Pedestrian crashes 807 2,338 3,510
Table A39: Ratio of fatal and serious/all injury factors by crash type, ONRC grouped and by alignment type, 80–100km/h
80km/h– Crossing Crossing Head Hit Loc Loc Misc Over- Pedestrian Rear end Rear end Rear
100km/h direct turning on object off on taking crossing queueing end
road road slowing
vehicle
National Straight 0.27 0.26 0.56 0.15 0.18 0.24 0.11 0.13 0.64 0.16 0.05 0.12
strategic (high and
volume) curved
Winding 0.38
and
tortuous
National and Straight 0.3 0.31 0.58 0.24 0.24 0.23 0.27 0.25 0.73 0.29 0.08 0.23
regional and
strategic curved
Winding 0.32 0.41 0.25 0.23 0.25 0.36
and
tortuous
Arterial and Straight 0.35 0.27 0.55 0.20 0.25 0.30 0.39 0.26 0.59 0.20 0.08 0.26
primary collector and
curved
Winding 0.22 0.43 0.22 0.24 0.36 0.33 0.62 0.14
and
tortuous
Secondary Straight 0.38 0.31 0.47 0.30 0.3 0.38 0.49 0.22 0.56 0.18 0.06 0.30
collector and and
access curved
Winding 0.28 0.37 0.27 0.31 0.31 0.61 0.28 0.59 0.50
and
tortuous
All 0.35 0.29 0.51 0.23 0.26 0.29 0.35 0.21 0.62 0.22 0.06 0.23
Table A40: Ratio of fatal and serious/all injury factors by crash type, ONRC grouped and by alignment type, 60–70km/h
60km/h–70km/h Crossing Crossing Head Hit Loc Loc Misc Over- Pedestrian Rear end Rear end Rear
direct turning on object off on taking crossing queueing end
road road slowing
vehicle
National strategic Straight 0.18 0.17 0.53 0.19 0.31 0.29 0.24 0.43 0.50 0.1 0.23
(high volume) and
curved
Winding 0.30
and
tortuous
National and Straight 0.09 0.16 0.51 0.21 0.44 0.11 0.50 0.05 0.22
regional strategic and
curved
Winding 0.17 0.16 0.29 0.60 0.41
and
tortuous
Arterial and Straight 0.19 0.36 0.16 0.24 0.29 0.47 0.50 0.02 0.10
primary collector and
curved
Winding 0.35 0.25 0.46 0.22 0.36 0.33 0.60 0.50
and
tortuous
Secondary Straight 0.17 0.25 0.38 0.21 0.24 0.13 0.46 0.50 0.17
collector and and
access curved
Winding 0.26 0.40 0.25 0.35
and
tortuous
All 0.18 0.43 0.17 0.24 0.22 0.45 0.50 0.05 0.17
Table A41: Ratio of fatal and serious/all injury factors by crash type, ONRC grouped and by alignment type 40–50km
40km/h–50km/h Crossing Crossing Head Hit Loc Loc Misc Over- Pedestrian Rear end Rear end Rear
direct turning on object off on taking crossing queueing end
road road slowing
vehicle
National strategic Straight 0.14 0.15 0.24 0.19 0.15 0.22 0.13 0.27 0.08 0.03 0.12
(high volume) and curved
Winding 0.27 0.67 0.25
and
tortuous
National and Straight 0.16 0.13 0.23 0.14 0.21 0.14 0.30 0.15 0.5 0.04 0.03 0.11
regional strategic and curved
Winding 0.13 0.17 0.33 0.22 0.26 0.15
and
tortuous
Arterial and Straight 0.15 0.14 0.24 0.14 0.19 0.22 0.35 0.15 0.29 0.05 0.04 0.10
primary collector and curved
Winding 0.17 0.17 0.26 0.21 0.22 0.23 0.22 0.25 0.25
and
tortuous
Secondary Straight 0.13 0.14 0.23 0.14 0.22 0.27 0.43 0.15 0.29 0.06 0.04 0.15
collector and and curved
access
Winding 0.15 0.23 0.14 0.17 0.24 0.32 0.32 0.35 0.29 0.10
and
tortuous
All 0.14 0.14 0.23 0.14 0.21 0.23 0.38 0.16 0.28 0.05 0.04 0.12
Note: blanks indicate insufficient data and that ‘all’ value for the applicable speed should be used.
80–100km/h National National Regional Arterial Primary Secondary Access All
strategic (high strategic strategic collector collector
volume)
Crossing direct 0.26 0.39 0.26 0.31 0.42 0.38 0.37 0.35
Crossing 0.27 0.35 0.30 0.26 0.28 0.31 0.31 0.29
turning
Head on 0.56 0.57 0.55 0.53 0.51 0.45 0.39 0.51
Hit object 0.15 0.27 0.23 0.17 0.23 0.28 0.32 0.23
Loc off road 0.18 0.25 0.23 0.24 0.26 0.30 0.32 0.26
Loc on road 0.24 0.22 0.24 0.26 0.35 0.33 0.40 0.29
Misc 0.11 0.25 0.28 0.35 0.39 0.45 0.59 0.35
Overtaking 0.13 0.30 0.22 0.24 0.30 0.27 0.15 0.21
Pedestrian 0.64 0.80 0.66 0.60 0.58 0.53 0.65 0.62
Rear end 0.16 0.27 0.29 0.18 0.22 0.21 0.00 0.22
crossing
Rear end 0.05 0.09 0.08 0.07 0.09 0.07 0.06 0.06
queuing
Rear end slow 0.12 0.22 0.25 0.24 0.27 0.36 0.19 0.23
vehicle
All 0.15 0.30 0.28 0.26 0.29 0.31 0.31 0.27
60–70 km/h National National Regional Arterial Primary Secondary Access All
volume)
Crossing direct 0.18 0.00 0.11 0.19 0.14 0.35 0.17
Crossing 0.17 0.13 0.16 0.19 0.22 0.23 0.30 0.18
turning
Head on 0.52 0.71 0.41 0.37 0.41 0.32 0.48 0.43
Hit object 0.18 0.18 0.13 0.21 0.16 0.33 0.17
Loc off road 0.30 0.21 0.21 0.24 0.24 0.22 0.29 0.24
Loc on road 0.28 0.33 0.46 0.34 0.14 0.50 0.33 0.35
Misc 0.40 0.00 0.67 0.25 0.40 0.33 0.33
Overtaking 0.25 0.18 0.11 0.27 0.35 0.15 0.22
Pedestrian 0.44 0.63 0.39 0.41 0.58 0.42 0.58 0.45
Rear end 0.12 0.09 0.10 0.10 0.06 0.11 0.20 0.10
crossing
Rear end 0.10 0.05 0.05 0.04 0.02 0.05 0.05
queuing
Rear end slow 0.21 0.25 0.22 0.10 0.09 0.19 0.29 0.17
vehicle
All 0.21 0.21 0.18 0.20 0.24 0.23 0.30 0.21
40–50 km/h National National Regional Arterial Primary Secondary Access All
volume)
Crossing direct 0.13 0.18 0.15 0.15 0.13 0.12 0.16 0.14
Crossing 0.15 0.14 0.13 0.14 0.15 0.14 0.15 0.14
turning
Head on 0.30 0.35 0.24 0.24 0.25 0.19 0.23 0.23
Hit object 0.19 0.15 0.14 0.14 0.14 0.16 0.13 0.14
Loc off road 0.16 0.24 0.20 0.19 0.20 0.23 0.23 0.21
Loc on road 0.21 0.30 0.08 0.20 0.24 0.28 0.27 0.23
Misc 0.22 0.60 0.25 0.35 0.30 0.35 0.48 0.38
Overtaking 0.12 0.18 0.15 0.13 0.23 0.16 0.18 0.16
Pedestrian 0.27 0.28 0.25 0.25 0.28 0.28 0.29 0.26
Rear end 0.08 0.04 0.04 0.04 0.06 0.06 0.06 0.05
crossing
Rear end 0.03 0.03 0.03 0.04 0.04 0.04 0.04 0.04
queuing
Rear end slow 0.11 0.05 0.12 0.11 0.12 0.16 0.12 0.12
vehicle
All 0.15 0.16 0.14 0.15 0.18 0.18 0.20 0.17
Appendix 2: Crash analysis > Definitions
Definitions
Regression to For the purpose of crash analysis, generally a minimum of the past five years of
the mean reported crash history is used. This reduces the error caused by regression to the
mean.
The principle of regression to the mean states that when an earlier measurement is
either extremely high or extremely low, then the expected value of later
measurements will be closer to the true mean than the observed value of the first.
The effect of regression to the mean can be reduced by using a longer crash
history when investigating crashes at a site, and by ensuring that there is a
commonality amongst crashes at the site.
Defining In assigning costs to crashes using method A, crashes are classified by ‘vehicle
crashes by involvement’ according to the highest ranked ‘vehicle’ involved in a crash. The
vehicle ranking from highest vehicle to lowest vehicle is:
involvement
• pedestrian
• bicycle
• motorcycle including moped
• bus
• truck
• cars, light commercial vehicles and any other.
For example, a crash involving a truck and a bicycle is categorised as a ‘cycle
crash’.
Dispersion ‘k’ is a dispersion parameter of the negative binomial distribution, which is the
parameter probability distribution assumed for the crash data. ‘k’ values for different sites are
provided in the Crash estimation compendium.
Generally, the higher the value of k the higher the accuracy of a crash prediction
model (and vice versa). The accuracy is, however, also relative to the typical crash
rate at a site (ie a low k value) may be acceptable at a site with a low typical crash
rate but unacceptable at a site with a high typical crash rate.
Fundamental An option results in a fundamental change in a site when the types of crash or the
change in a site level of crash severity is expected to change significantly. The following list gives
examples of site changes that would result in a fundamental change:
• a completely new site is being provided (such as a new road or
intersection)
• realignment of a road (other than an isolated curve)
• removal or significant modification of road elements (eg grade separation
of a railway crossing and conversion of a single-lane bridge to a two-lane
bridge)
• change in intersection form of control
• flush median installed on an urban road with multiple accesses
• adding lanes, including passing lanes.
Options that are not normally regarded as resulting in fundamental changes
include:
• upgrade of a single or S-bend to a higher-design speed curve or S-bend
• shoulder widening on rural roads (in the absence of road realignment)
• signage and delineation improvements, including lighting
• traffic volume changes (in the absence of other improvements)
• road resurfacing and shape corrections
• minor improvement works.
Intersection Crashes occurring within the area of priority controlled intersections, roundabouts
crashes and traffic signals on the primary road network, and up to 50 metres from the
Appendix 2: Crash analysis > Definitions
influence of the intersection in a 50km/h speed limit area and up to 200 metres in
an 80km/h and above area.
Mid-block Crashes occurring on a road section excluding crashes at major intersections, or

crashes 50 metres from the influence of the intersection in a 50km/h speed limit area and
up to 200 metres in an 80km/h and above area. Crashes at minor intersections are
sometimes included.
Remote and Remote rural roads are sites carrying less than 1000 vpd and more than 20
near rural roads kilometres away from a town with a population of 3000 or more. Other rural sites
are considered to be ‘near rural’.
Severity In method A, crashes are categorised by the most severe injury sustained. The four
severity categories are:
• Fatal: when death ensues within 30 days of the crash.
• Serious: injuries requiring medical attention or admission to hospital,
including fractures, concussion and severe cuts.
• Minor: injuries other than serious, which require first aid or cause
discomfort or pain, including bruising and sprains.
• Non-injury: when no injuries occur, sometimes referred to as ‘property
damage only’ (PDO) crashes.
The crash reports from police officers recorded in CAS are to be used to classify
crash severity in preference to hospital records.
Site A site is the specific road infrastructure for which an evaluation is carried out. A site
can be a bridge, intersection, mid-block, curve, S-bend etc, or any combination of
these (eg a mid-block and an intersection). In the case of combinations, a site may
have to be broken into parts for the purpose of evaluation.
Types of crash A crash rate is the average number of injury crashes per year, measured over a
rate period of time (normally five calendar years). Caution is required when using the
latest three to six months CAS data as the data set may not be complete.
Site-specific crash rate (AS)
The crash rate for a specific site based on reported injury crashes on the record of
TCRs prepared by the police and compiled by Waka Kotahi (normally five years of
data). These are available from CAS.
Typical crash rate (AT)
The crash rate for a typical or generic site (eg a bridge with characteristics similar
to the site being evaluated). Typical crash rates are determined using either a
crash rate or a crash prediction model from the Crash estimation compendium,
depending on the type of site, or part of a site, being evaluated.
Weighted crash rate (AW)
The crash rate produced when using the weighted crash procedure.
Back to 3.1 Impact on social cost of deaths and serious injuries >>
Appendix 3: Traffic data and travel time estimation > Traffic composition
Appendix 3: Traffic data and travel time estimation

This appendix begins by defining standard values for traffic composition (based on the vehicle
classes), vehicle occupancy and trip purpose. Guidance is also provided on measuring and estimating
traffic volumes, traffic growth and speed.
The traffic data generated by these methods can be used:
• in the procedures for estimating travel time

• in the absence of measured data, or
• in the absence of data from calibrated and validated transportation models.
Following the traffic data methods, procedures for estimating travel time are detailed.
The travel time estimation procedures are capable of application by hand, spreadsheet and within
transportation models. The methodology gives a reasonable approximation for travel time without
having to analyse dynamic queuing situations, though more precise methods are not precluded.
Where a specific procedure is not given, the travel time shall be determined according to a recognised
procedure compatible with the manuals and procedures referred to in this appendix.
validated on both traffic volumes and travel times It is necessary that the travel times used by a model
to derive the flows must be consistent with the travel times estimated by the procedures. To adhere to
this, it is suggested that the functions implied by the procedures in this appendix be used as a starting
point and modified as necessary to get a satisfactory validation.
Traffic composition
Vehicle classes
The definitions for vehicle classes are provided in Table A45.
Table A45: Vehicle classes

Vehicle classes Vehicle class composition
Passenger cars Cars and station wagons, with a wheelbase of less than 3.2 metres
Light commercial vehicles Vans, utilities and light trucks up to 3.5 tonnes gross laden weight.
(LCV) LCVs mainly have single rear tyres but include some small trucks
with dual rear tyres
Medium commercial vehicle Two axle heavy trucks without a trailer, over 3.5 tonnes gross laden
(MCV) weight
Heavy commercial vehicle I Rigid trucks with or without a trailer, or articulated vehicle with three
(HCVI) or four axles in total
Heavy commercial vehicle II Trucks and trailers and articulated vehicles with or without trailers
(HCVII) with five or more axles in total
Buses Buses, excluding minibuses
Back to 3.6 Impact on network productivity and utilisation: Vehicle operating costs
>>
Appendix 3: Traffic data and travel time estimation > Traffic composition
Road categories
Road categories for the traffic data classifications in this appendix are provided in Table A46.
Table A46: Road categories

Road categories Definition
Urban arterial Arterial and collector roads within urban areas carrying traffic volumes
of greater than 7000 vehicles/day
Urban other Other urban roads, carrying fewer than 7000 vehicles/day
Rural strategic Arterial or collector roads, connecting main centres of population and
carrying over 2500 vehicles/day
Rural other Other roads outside urban areas
Back to 3.6 Impact on network productivity and utilisation: Vehicle operating costs
>>
Standard traffic composition
Table A47 provides standard traffic compositions. For larger projects or sites with unusual traffic
characteristics, classification counts are required. Bus numbers are site dependent and are not
included in the standard traffic composition.
Table A47: Traffic composition (%)
Road category and time period Traffic composition by vehicle class (%)
Car LCV MCV HCVI HCVII
Urban arterial
Morning commuter peak 85 10 2 1 2
Daytime inter-peak 84 11 2 1 2
Afternoon commuter peak 84 11 2 2 1
Evening/night-time 85 9 2 1 3
Weekday all periods 85 10 2 1 2
Weekend/holiday 87 8 3 1 1
All periods 85 10 2 1 2
Urban other
Weekday 86 8 3 2 1
Rural strategic
Weekday 75 12 4 4 5
Rural other
Weekday 78 11 3 4 4
Traffic composition data is not provided for strategic routes on the fringes of large population centres
(ie populations greater than 40,000). Such routes are characterised by predominantly rural strategic
traffic mixes but with high commuter peaks more typical of an urban arterial road. On these routes
individual surveys of traffic composition will normally be required. Also traffic stream compositions are
likely to vary throughout the day, and the result of a single period survey may not accurately reflect
the daily traffic composition – if this is the case more surveys through the day will be required.
Appendix 3: Traffic data and travel time estimation > Separating an activity into its component sections
Separating an activity into its component sections

Follow the steps below to separate the activity into its component sections.
Table A48: Steps to separate an activity into its component sections

Step Action
1 Separate the activity into:
• motorway sections
• multi-lane roads
• two-lane rural roads
• other urban roads
• signalised intersections
• priority intersections
• roundabouts.
2 Identify any bottleneck locations
Sections must be chosen so as to ensure conservation of vehicle movements (ie the sum of the flows
into a section must equal the sum of the flows out).
Section lengths may be divided into sub-sections when it comes to calculating vehicle operating
costs.
Guidance for motorways and multi-lane roads
Each motorway section or multi-lane road section shall consist of a length of road with:
• uniform design speed

• one direction of travel
• uniform number of through lanes
• boundaries which generally extend between major interchanges where significant flows leave
or join the section.
Guidance for two-lane rural roads
Each two-lane rural road section shall be at least 1km and not more than 5km in length. The two-lane
rural road section to be analysed may be longer than the activity length.
Dividing the year into time periods

Each year is defined as having 365 days comprising:
• 245 weekdays
• 52 Saturdays
• 68 Sundays and public holidays.
Weekends and holiday periods cover Saturday and Sunday, all public holidays and two weeks over
Christmas and New Year. These account for 120 days per year.
The default weekday time periods are:
• morning commuter peak (0700–0900)

• daytime interpeak (0900–1600)
• evening commuter peak (1600–1800)
• evening/night-time (1800–0700).
Saturdays and Sundays do not usually need to be divided into time periods unless there are
substantial demands.
Appendix 3: Traffic data and travel time estimation > Vehicle occupancy and travel purpose
Procedure
Follow the steps below to divide the year into time periods.
Table A49: Steps to divide the year into time periods

Step Action
1 Divide the year into the days specified above
2 Divide each day type into time periods as follows:
If there… Then…
are only very low levels of vehicle no division of the day is necessary.
interaction throughout any day
are significant levels of vehicle interaction divide each day into a number of time
periods to allow analysis at different flow
levels, so that:
• operating conditions (such as
proportion of traffic turning, percent
working and vehicle composition) are
essentially constant
• the period is long enough to ensure
sufficient total capacity is available,
even though for some of the time the
capacity is exceeded.
Vehicle occupancy and travel purpose

Standard vehicle occupancy and travel purpose figures are provided in Table A50. For large activities
or sites with unusual traffic characteristics, vehicle occupancy surveys shall be conducted by roadside
observation of the traffic stream in conjunction with classification counts. Vehicle occupancy counts
shall include drivers and passengers.
'Working' refers to trips carried out in the course of paid employment, 'commuting' refers to trips
between home and work, while ‘other’ refers to all other non-work trips (ie other than commuting).
Travel purposes is a difficult characteristic to survey and recourse to the standard values provided in
Table A50 will be required in most cases. At present there is no accepted method for differentiating
between work and non-work trips by observing moving traffic stream. Field surveys of trip purpose
require roadside interviews. Survey results from urban transportation studies can be used where
appropriate. The values in Table A50 have been derived from the New Zealand Household Travel
Survey.
Table A50: Vehicle occupancy and travel purpose
Road Car LCV MCV and HCV

category Occupancy Travel purpose % Occupancy Travel purpose % Occupancy Travel purpose %
Work Commute Other Work Commute Other Work Commute Other
Urban arterial
AM peak 1.4 10 50 40 1.4 65 20 15 1.2 90 5 5
Daytime 1.3 30 10 60 1.4 65 5 30 1.2 90 0 10
interpeak
PM peak 1.4 10 30 60 1.4 65 15 20 1.2 90 5 5
Evening/ 1.4 10 5 85 1.4 65 15 20 1.2 90 5 5
night-time
Weekday all 1.4 20 20 60 1.4 65 10 25 1.2 90 5 5
periods
Weekend 1.7 5 5 90 1.7 10 10 80 1.6 75 5 20
All periods 1.5 15 15 70 1.5 50 10 40 1.3 85 5 10
Urban other
Weekday 1.4 20 20 60 1.6 65 10 25 1.2 90 5 5
Weekend 1.7 5 5 90 2.0 10 10 80 1.6 75 5 20
All periods 1.5 15 15 70 1.7 45 10 45 1.3 85 5 10
Rural strategic and rural other roads
Weekday 1.6 40 10 50 1.6 75 5 20 1.3 90 5 5
Weekend 2.2 5 5 90 2.0 10 10 80 1.8 75 5 20
All periods 1.7 30 10 60 1.7 55 5 40 1.4 85 5 10
Traffic volumes are generally expressed in terms of annual average daily traffic (AADT), annual average
weekday, average weekend/holiday, average hour, or average quarter hour volumes. The methods given
below for determining traffic volumes based on traffic counts are derived from:
• Guide to estimating AADT and traffic growth (Transit NZ 1994)

• Guide to estimation and monitoring of traffic counting and traffic growth (Traffic Design Group
2001).
General information and background on demand estimation and forecasting can be found in Chapter 2:
Demand estimation and mode share. Information on transport model availability in New Zealand and high
level background on model capability can be found in section 2.10. Wherever properly
calibrated/validated transportation models are available in a study area, they should generally be used to
assess the effects of the activity on traffic volumes and predict future traffic volumes.
As well as the normal calibration/validation required to ensure that the models are operating satisfactorily,
they should also be calibrated/validated in the local area containing the activity. See the Transport model
development guidelines (NZ Transport Agency 2019), and particularly reference to model type D:
Transport Agency scheme assessment/project evaluation, which outlines calibration/validation principles
within the area of influence/focus for an assessment of an activity.
For background on when differences between the do-minimum and activity demands may occur, see
section 2.13: Fixed trip matrix and variable trip matrix assessments and for procedures for developing
demands and applying fixed trip matrix and variable trip matrix techniques see Fixed trip matrices and
Variable trip matrices in Appendix 1.
Method for estimating AADT
To estimate AADT from a sample count it is necessary to adjust the count data for a number of factors.
Count data shall be checked for consistency and reasonableness and axle pair counts (eg from tube
counters) shall be corrected by applying an adjustment factor to convert from axle pair counts vehicle
counts.
Daily counts for less than a week shall be adjusted by applying day factors (for the appropriate typical
traffic pattern) to derive weekly average daily traffic. Weekly average daily traffic figures shall then be
adjusted by applying the appropriate week factors to derive AADTs. If more than one week is counted,
the AADT shall be determined for each week, and then averaged.
To determine day and week factors, the appropriate traffic pattern control group shall be identified from
the Guide to estimating AADT and traffic growth. Alternatively, these factors may be derived from rigorous
local traffic counting programmes.
Method for estimating weekday or weekend/holiday volumes
The weekday, Saturday and Sunday/holiday volumes shall be derived from AADTs by applying locally
derived day factors where these are available, or the factors in the Guide to estimating AADT and traffic
growth if local data is not available. The Saturday and Sunday/holiday volumes so obtained shall be
averaged to derive an average weekend/holiday daily volume.
Method for estimating hourly or quarter hourly directional volumes
Where traffic volumes are required for shorter time periods than a day, these shall be obtained from
directional counts.
Counts done to produce estimates of the AADT will usually have been obtained from traffic counters that
record volumes by 60 or 15 minute intervals. Week factors shall be applied to these counts to obtain
estimates of 60 or 15 minute traffic volumes.
For intersection volumes, manual counts of turning movements should be consistent with the
requirements of NZS 5431:1973 clause 5.4.
Appendix 3: Traffic data and travel time estimation > Traffic growth rates
Axel pair adjustment factors

Wherever possible measured data shall be used to determine the axle pair adjustment factors, but in
absence of such data the following factors shall be used. To convert axle pairs to vehicles, multiply by the
appropriate factor.
Table A51: Axle pair adjustment factors

Road category Axle pair adjustment factor
Urban 0.91
Rural 0.83
Traffic growth rates

Guidance for developing growth rates using trend analysis can be found in Travel volume growth trend
analysis in Appendix 1. Traffic growth rates shall be arithmetic growth rates (not geometric growth rates,
which compound year-on-year) and expressed as a percentage of the predicted traffic volume at the time
zero.
It might not be appropriate to assume continuation of current traffic growth rates over the whole project
analysis period. The current traffic growth rate shall be adjusted, as appropriate, to account for the future
traffic volume influences described below.
Future traffic volumes

In predicting future traffic volumes, normal traffic growth, diverted traffic, generated and redistributed
traffic shall be taken into account. More information on developing demand forecasts can be found in
Chapter 2: Demand estimation and mode share. This includes background on the equilibrium of demand
and supply, factors and considerations influencing demand estimation, and information on forecast
horizons and uncertainty.
Normal traffic growth
Traffic growth, either developed from transport models or from procedures and processes described in
this manual (for example, facility use, trend analysis, and trip generation methods) can be considered
normal when the estimated demand does not result in excessive or an unrealistically long time spent in
stationary queues (for further background, see Equilibrium of transport demand and supply). Provided
estimates are robust , localised effects are accounted for (for example, localised land use development,
and transport supply change), and other checks are carried out (see Sense checking forecasts), normal
traffic growth is likely to be considered to provide a sound basis for predicting future traffic demands.
Re-assigned traffic
A re-assigned trip is a trip travelling from A to B in the same period and by the same mode that takes a
different route to get there. Re-assigned traffic to or from the route(s) served by the activity, nearby and
surrounding the activity, occurs when:
• traffic re-routes from another route because the activity (or another activity on the route) now
makes this a more attractive route
• traffic re-routes to another route because an activity on that route now makes it a more attractive
route
• delays/changes to the user travel cost at the activity site or elsewhere on the route cause traffic to
re-route to other routes
• delays/changes to the user travel cost on other routes cause traffic to re-route to the route
• estimated changes in travel demands in the future affect one, or several, of the elements above.
These effects shall be taken into account in estimating future traffic volumes. Typically, when re-routeing
traffic is, or has the potential to be, an issue, traffic models are applied. As described in section 2.13:
Fixed trip matrix and variable trip matrix assessments, re-assigned trips should not result in a difference
in the OD demands matrices used in the do-minimum and activity scenarios, and as such the study area
Appendix 3: Traffic data and travel time estimation > Travel times and speed
for the transport model would need to cover any potential significant re-assignment changes from the
effects described in the bullet points above.
Induced demand, redistributed trips, mode-shifted trips, or macro-time shifted trips
Activities that reduce the cost of travel can induce new trips, redistribute trips, result in a change of travel
mode, or macro-time shift trips from one discrete time period to another. See section 2.13: Fixed trip
matrix and variable trip matrix assessments. In cases where these potential effects are expected to
significantly affect the evaluation, then a variable matrix approach should be adopted (see Variable trip
matrices in Appendix 1).
Travel times and speed

Travel time and/or speeds shall be measured where required. Suitable methods for measuring average
travel times or speed depending on circumstances include:
• floating car survey

• number plate survey
• spot measurement of speed.
The floating car and number plate survey methods measure the average travel time over a length of road.
The floating car survey method is a relatively cheap and convenient method but will not readily
differentiate the average travel times of light and heavy traffic. It is only suitable for higher traffic volumes
in excess of 500 vehicles/hour/lane.
The number plate method is a larger undertaking but potentially more accurate and has the ability to give
data on the average travel times of individual or categories of vehicle. Several software packages are
available for analysing number plate survey data, as are electronic field-book programmes for facilitating
the data input.
The average travel time over a section of road may not provide sufficient information for calculating
vehicle operating costs if one or more speed change cycles occur within the section. Speed change
cycles should be separately identified in urban areas where speeds reduce to below 20km/h and for rural
areas where vehicles slow down for example to negotiate a sharp bend or at an intersection.
In such cases, spot measurement of speed will be required at a sufficient number of other locations to
establish the average cruise speed for the road section and at the points of minimum speed. If vehicles
stop at any point on the road section, then the average length of stopped time will also be required for the
operating cost calculations. An alternative to spot measurements of speed will be to arrange number plate
survey points so they do not contain speed change cycles within their length.
When averaging the results of speed spot measurements, the space mean speed should be calculated
using the following formula:
n
v =
1 1 1 1 1
 + + + ++ 
 v1 v2 v3 v 4 vn 
where: vi is spot speed measurement

N is total number of spot speed measurements
Appendix 3: Traffic data and travel time estimation > The stages for estimating travel time

The flow chart below shows the basic stages for estimating road section travel time (the stages are
slightly different for intersections).
Figure A5: Stages for estimating road section travel time
Start
Separate the activity into its component road sections

Divide the year into time periods as appropriate
Determine traffic volumes for each time period Spread the peak
Calculate the free speed travel times
Yes
Is the section an ‘other’ urban road?
No
Determine the capacity
Yes
Is the section a two-lane rural road?
No
No
Are vehicles interactions significant?
Yes
Calculate congestion delay

Calculate bottleneck delay if capacity is exceeded
Calculate the time period total average travel time
Yes
Should peak spreading be considered?
No
End
Appendix 3: Traffic data and travel time estimation > Determining traffic volumes
Determining traffic volumes

This procedure details the base and future year traffic volumes that need to be determined for estimating
travel times.
In some cases, growth constraint methods may be needed to estimate the do-minimum and activity
option matrices where high future levels of congestion are anticipated, usually because the network(s)
has insufficient capacity to meet unrestrained travel demands and/or because of high level of forecast
future demand.
In some cases, variable matrix methods may be needed to estimate the do-minimum and activity option
matrices (Variable trip matrices in Appendix 1).
The base traffic volumes are the traffic volumes at the point-in-time closest to time zero. It is noted that in
many cases time zero values may not be explicitly estimated and provided, for example, when the
transport model used for the assessment has been calibrated/validated to a year which is several years
earlier than time zero. The base traffic volumes are from:
• a recent census year, or near-future forecast year, for which travel demand estimates exist
• a year at which the transport model has been calibrated/validated to
• a year at which robust traffic count data exists.
Procedure
Follow the steps below to determine traffic volumes.
Table A52: Steps to determine traffic volumes

Step Action
1 Determine the base traffic volumes for each section using the procedure outlined previously
in this appendix, or by means of a transportation model.
2 Estimate the traffic volumes for each section for at least two future years using a suitable
prediction method.
Note: The method adopted for estimating future traffic volumes should not result in excessive
or an unrealistically long time spent in stationary queues (for further background, see
Equilibrium of transport demand and supply).
3 Judge whether future year capacity problems occur.
Note: This step requires an estimate of the capacity that is not determined until Determining
the capacity of road sections. A first iteration of this whole procedure may be used before
judging whether this step is relevant.
If there … Then …
is sufficient capacity for future year generally apply standard fixed trip matrices and
traffic volumes in the do-minimum evaluation procedures.
and activity option
are adequate levels of service for generally improve the capacity of the do-minimum
future year traffic volumes in the network and/or apply growth constraint techniques to
activity option, but not in the do- the do-minimum matrix (Fixed trip matrices in
minimum (depends on local study Appendix 1).
area context, ranges from level of
service E to F through to excessive
congestion and long time periods
with stationary queues)
is high congestion (depends on local consider the use of variable matrix methods, see
study area context, ranges from level section 2.13: Fixed trip matrix and variable trip matrix
of service E to F through to assessments.
excessive congestion and long time
periods with stationary queues) in For verification purposes, carry out a fixed matrix
both the do-minimum and activity analysis using growth constraint techniques (Fixed
options trip matrices in Appendix 1).
Appendix 3: Traffic data and travel time estimation > Calculating free speed travel time
Calculating free speed travel time

This procedure may be used for all road section types.
Procedure
Follow the steps below to calculate the free speed travel time.
Table A53: Steps to calculate the free speed travel time

Step Action
1 Take measurements of free speed in the field at flow rates below 600veh/h/lane.
Alternatively, measurements of free speed from a similar road section in the locality, with
similar characteristics, can be used.
Note: To proceed with a preliminary value of free speed before measurements have been
collected or if the road section is part of a proposed facility, then follow step 2.
2 If measured speeds are not available, then determine the free speed using the appropriate
procedure as follows:
If the road section is … Then use the procedure …
a motorway section 105km/h
where design speed >110km/h
a multi-lane road Determining the free speed of multi-lane roads.
a two-lane rural road Determining the free speed of two-lane rural
roads.
other urban road Determining the free speed of other rural roads.
3 Using the free speed determined in either step 1 or 2, calculate the travel time in minutes per
kilometre.
Example:
Free speed = 100km/h
Free speed travel time = 60/100
= 0.600 min/km
4 Determine the capacity from Determining the capacity of road sections.
Other urban road capacity is not required for calculating travel time but is used in determining
additional vehicle operating cost of congestion.
Determining the free speed of multi-lane roads

This procedure is required for analysis of activities to which Table A53 applies.
The free speed of proposed or existing facilities for which there is no measured data is estimated by
adjusting the basic free speed under ideal conditions.
Adjustments to the basic free speed are made for:
• dividing medians
• lane width
• lateral clearance, and
• density of access points.
Lateral clearance is the sum of any median shoulder and sealed left hand shoulder widths beyond the
edge of the through lanes that are continuously available.
Procedure
Follow the steps below to determine the free speed of a multi-lane road section.
Appendix 3: Traffic data and travel time estimation > Determining the free speed of multi-lane roads
Table A54: Steps to determine the free speed of a multi-lane road

Step Action
1 If measured speeds are not available, then determine the basic free speed for the multi-
lane road section as follows:
If the section has a posted speed limit of … Then use a basic free speed of…
100km/h 105km/h
80km/h 90km/h
70km/h 80km/h
50km/h 60km/h
2 Adjust the basic free speed to account for dividing medians as follows:
Dividing median Adjustment to basic free speed
Has a dividing median No reduction
No dividing median Reduce by 3km/h
3 Adjust the basic free speed to account for lane widths as follows:
If lane widths are… Adjustment to basic free speed
3.5m or greater No reduction
less than 3.5m Reduce by 3km/h
4 Adjust the basic free speed to account for lateral clearance as follows:
If the section has lateral clearance of … Adjustment to basic free speed
3m or greater No reduction
less than 3m but at least 2m Reduce by 2km/h
less than 2m but at least 1m Reduce by 4km/h
less than 1m Reduce by 9km/h
5 Adjust the basic free speed to account for density of access points along the section as
follows:
If the section has a density of access points Adjustment to basic free speed
per km of …
less than 40 0.4km/h per access point
40 or more 16km/h
Example calculation
Below is an example calculation for the free speed of a multi-lane road section where measured speeds
are not available.
Example:
Posted speed limit = 70km/h
Median divided = yes
Lane width = 3.5m
Lateral clearance = 1.0m
Access points density = 10 per km
Basic free speed = 80km/h
Dividing median speed reduction = 0km/h
Lane width speed reduction = 0km/h
Lateral clearance speed reduction = 4km/h
Access point speed reduction = 10 × 0.4 = 4km/h
Free speed = 80 – 0 – 0 – 4 – 4 = 72km/h
Appendix 3: Traffic data and travel time estimation > Determining the free speed of two-lane rural roads
Determining the free speed of two-lane rural roads

This procedure is required for analysis of activities to which Table A53 applies and should be used if no
measured speeds are available.
The procedure adopted in this section provides a realistic but approximate method for assessing travel
times. Alternatively, the Transportation Research Board (1994) Highway capacity manual (HCM) provides
a more detailed methodology for the evaluation of local improvements, such as design speed increases,
and climbing and passing lanes, and the computer programme TRARR may be used for detailed
analyses.
The definition of design speed used in this section is that used by the HCM and the Austroads (1988)
Guide to traffic engineering practice part 2 roadway capacity.
Procedure
The free speed of a two-lane rural road is determined by the speed environment that can be
approximated by the average design speed of the road section under consideration and the associated
approaches.
Follow the steps below to determine the free speed of a two-lane rural road section.
Table A55: Steps to determine the free speed of a two-lane rural road
Step Action
1 Obtain the following basic data for the road section:
• length of road section
• centreline length of each curve including transitions
• length of each straight (tangent)
• design speed of the straights (tangents)
• design speed of the curves.
2 Calculate the travel time for each curve and straight, as per steps 3 and 4.
Note: It is acceptable to assume an abrupt change in speed where straights and curves meet.
3 Calculate the travel time on curves (including transitions).
Example:
Curve 1 length = 0.200km
Curve 1 design speed = 80km/h
Curve 1 travel time = 0.2/80 × 60 = 0.150 minutes
Total curve travel times = 0.150 + 0.129 + 0.086 = 0.365 minutes
4 Calculate the travel time on the straights (tangents)
Note: Unless constrained by other design criteria the design speed for straights (tangents)
should be assumed to be 100km/h in severe terrain and a maximum of 120km/h in gentler
country (Austroads (1989) Rural road design).
Example:
Tangent length = 0.550km
Tangent design speed = 120km/h
Tangent travel time = 0.550/120 × 60
= 0.275 minutes
5 Calculate the total travel time on the road section.
Example:
Travel time on curves = 0.365 minutes
Appendix 3: Traffic data and travel time estimation > Determining the free speed of other rural roads
Step Action
Travel time on straights = 0.275 minutes
Total travel time = 0.365 + 0.275
= 0.640 minutes
6 Calculate the average design speed for the road section.
Example:
Road section length = 1km
Total travel time = 0.640 minutes
Average design speed = 1.000/0.640 × 60
= 93.75km/h
7 Determine the free speed as follows:
If the average design speed is … Then the free speed is…
above 100km/h 105km/h
below 100km/h 105km/h minus 13km/h for every 18km/h
reduction in design speed below 100km/h
Example:
Average design speed = 93.75km/h
Free speed = 105 - ([(100 – 93.75) / 18] × 13)
= 100.5km/h
Determining the free speed of other rural roads

This procedure is required for analysis of activities to which Table A53 applies and should be used if no
measured speeds are available.
Procedure
Follow the steps below to determine the free speed of an ‘other urban road’.
Table A56: Steps to determine the free speed of an 'other urban road'
Step Action
1 Determine the classification of the other urban road section as follows:
If the design category of the And the functional category Then the road classification
road section is … is … is …
suburban principal Class I
suburban minor Class II
intermediate principal Class II
intermediate minor Class II or III
urban principal Class II or III
urban minor Class III
Design category
Criterion Suburban Intermediate Urban
Driveway/access Low density Moderate density High density
density
Arterial type Multi-lane divided, Multi-lane divided or Undivided one-
undivided or two-lane undivided, one-way, way, two-way,
with shoulders two-lane two or more
lanes
Parking No Some Significant
Separate right-turn Yes Usually Some
lanes
Signals/km 0.6–3.0 2–6 4–8
Pedestrian activity Little Some Usually
Appendix 3: Traffic data and travel time estimation > Determining the capacity of road sections
Step Action
Roadside Low to medium Medium to moderate High
development density
Functional category
Criterion Principal Minor
Mobility function Very important Important
Access function Very minor Substantial
Points connected Motorways, important Principal arterials
activity centres, major traffic
generators
Predominant trips served Relatively long trips Trips of moderate length
between major points and within relatively small
through-trips entering, geographical areas
leaving, and passing
through the city
2 Determine the free speed for the road section as follows:
If the road classification is … Then the range of likely free And a typical free speed
speeds are between … would be …
Class I 60 and 65km/h 63km/h
Class II 50 and 60km/h 55km/h
Class III 45 and 55km/h 50km/h
Determining the capacity of road sections

In the absence of measured capacities, the capacity of a road section shall be determined by the
methods specified in this appendix for each facility type according to the conditions that prevail during the
time interval. For example, when estimating capacity, the proportion of commercial vehicles, the average
intensity of conflicting flows, and the performance of traffic control devices during the time interval shall be
taken into account.
For other road types not covered by these procedures refer to the HCM.
In fulfilling the requirement that demand is in approximate equilibrium with supply, the procedure adopted
for estimating future traffic volumes must ensure that, in particular, the estimated traffic volume over any
time period is less than the total available capacity for the time period of all road sections and
intersections located within and near the project under analysis.
Where traffic volumes exceed capacity, the resulting queues may block back onto upstream links. In such
circumstances care must be taken that the delays arising on the under-capacity section are not double
counted on any upstream section.
Selecting the appropriate procedure
Follow the steps below to select the appropriate procedure for determining the capacity of each road
section.
Appendix 3: Traffic data and travel time estimation > Determining the capacity of motorways
Table A57: Steps to select the appropriate procedure for determining the capacity of road
sections
Step Action
1 Select the appropriate procedure for determining the capacity of each road section as
follows:
If the road section is … Then go to …
a motorway section Determining the capacity of motorways
a multi-lane road Determining the capacity of multi-lane roads
a two-lane rural road Determining the capacity of two-lane rural
roads
other urban road Calculating the time period total average
travel time
It is not necessary to determine capacity for
travel time. However, the capacities below
are required when determining the additional
congestion vehicle operating cost.
Road class Capacity
Class I 1,200 veh/lane/hour
Class II 900 veh/lane/hour
Class III 600 veh/lane/hour
2 Once the capacity has been determined go to Determining whether vehicle interactions are
significant.
Determining the capacity of motorways

Procedure
Follow the steps below to determine the capacity of a motorway section where each direction of travel is a
separate motorway section component (see Separating an activity into its component sections).
Capacities are expressed as passenger car units (pcu).
Table A58: Steps to determine the capacity of a motorway section with separate motorway
components in each direction of travel
Step Action
1 Determine the basic capacity for the motorway section as follows:
If the road section has ... Then use a basic capacity of …
2 through lanes 4,500 pcu/h
2 Determine the passenger car equivalent to be used for trucks for the motorway section as
follows:
If the terrain type is … Then use a passenger car equivalent for
trucks (Et) of …
level 1.7 pcu
rolling 4.0 pcu
mountainous 8.0 pcu
3 Calculate the adjustment factor for trucks using the passenger car equivalent for trucks (E t)
determined in step 2.
Adjustment factor (ft) = 1/ (1 + Pt × (Et - 1))
where Pt = the proportion of trucks in the traffic
stream during the peak period.
Example:
Terrain type = rolling
Appendix 3: Traffic data and travel time estimation > Determining the capacity of multi-lane roads
Step Action
Proportion of trucks (Pt) = 0.12
Pcu for trucks (Et) = 4.0 pcu
Adjustment factor (ft) = 1/(1 + 0.12 × (4.0 - 1 ))
= 0.735
4 Calculate the motorway section capacity by multiplying the basic capacity, determined in step
1, by the adjustment factor for trucks (ft) determined in step 3.
Motorway section capacity = basic capacity × ft
Example:
Through lanes = 3 lanes
Basic capacity = 6,900 pcu/h
Adjustment factor (ft) = 0.735
Motorway section capacity = 6,900 × 0.735
= 5,072veh/h
Using field measurements
If actual field measurements at the site give a different capacity from that which is determined above, then
the field measurements should be used. However, if field measurements are used, then the analyst must
prove that the measurements are representative of the average capacity in a variety of conditions.
Accounting for auxiliary lanes
Auxiliary lanes within road sections may contribute to the road’s capacity in which case the detailed
procedures of the HCM shall be used. Otherwise the auxiliary lanes shall be considered not to contribute
to the capacity.
Determining the capacity of multi-lane roads

Procedure
Follow the steps below to determine the capacity of a multi-lane road.
Table A59: Steps to determine the capacity of a multi-lane road

Step Action
1 Obtain ‘the sum of the basic free speed reductions’ for the multi-lane road section, as
determined in Table A57.
Example:
Free speed reductions for:
dividing median = 0km/h
lane width = 0km/h
lateral clearance = 4km/h
access points = 4km/h
Sum of the basic free speed reductions
= 8km/h
Note: If the free speed for the multi-lane road section was measured rather than estimated,
then use step 1 of the procedure in Table A57 to determine the multi-lane road basic free
speed, and subtract the measured free speed to obtain the equivalent of ‘the sum of the
basic free speed reductions’.
2 Determine the capacity of the multi-lane road section as follows:
If the sum of the basic free speed reduction Then use a capacity of …
is …
zero 2,200 veh/h per lane
2,200 veh/h per lane minus 10veh/h per lane
between 0 and 30km/h for every km/h of basic free speed
reductions
above 30km/h 1,900 veh/h per lane
Appendix 3: Traffic data and travel time estimation > Determining the capacity of two-lane rural roads
Example:
Sum of the basic free speed reductions = 8km/h
Road section capacity = 2,200 - 8 × 10
= 2,120 veh/h per lane
Determining the capacity of two-lane rural roads

The capacity of the road section shall be calculated by adjusting the ideal capacity of 2800veh/h (total in
both directions of travel) to account for the following factors:
• directional distribution of traffic during the time period

• the presence of narrow lanes and restricted shoulders
• the proportion of heavy vehicles in the flow.
Procedure
Follow the steps below to determine the capacity of a two-lane rural road section.
Table A60: Steps to determine the capacity of a two-lane rural road
Step Action
1 Determine the adjustment factor for traffic directional distribution during the time period as
follows:
If the directional distribution is … Then use an adjustment factor of:
100/0 0.71
90/10 0.77
80/20 0.83
70/30 0.89
60/40 0.94
50/50 1.00
2 Determine the total roadway width. The total roadway width equals the lane width(s) plus
sealed shoulder width. Round to the nearest metre.
3 With the total roadway width determined in step 2 determine the adjustment factor for
trafficable width as follows:
If the total roadway width is… Then use an adjustment factor of:
8m or greater 1.00
7m 0.91
6m 0.82
5m 0.73
4m 0.65
less than 4m 0.60
4 Determine the passenger car equivalent for trucks for the road section as follows:
If the terrain type is… Then use a passenger car equivalent for
trucks (Et ) of:
level 2.2 pcu
rolling 5.0 pcu
mountainous 10.0 pcu
5 Calculate the adjustment factor for trucks using the passenger car equivalent for trucks (E t)
determined in step 4.
Adjustment factor (ft) = 1/(1 + Pt × (Et- 1 ))
Where Pt is the proportion of trucks in the traffic stream during the time period
Example:
Proportion of trucks (Pt) = 0.10
pcu for trucks (Et) = 5.0 pcu
Appendix 3: Traffic data and travel time estimation > Determining whether vehicle interactions are significant
Step Action
Adjustment factor (ft) = 1/[1 + 0.10 × (5.0 – 1))]
= 0.714
6 Calculate the road section capacity by multiplying the ideal two-way capacity of 2,800veh/h
by the adjustment factors determined in steps 1, 3 and 5.
Road section capacity = Ideal capacity × adjustment factor for directional
distribution × adjustment factor for trafficable width × ft
Example:
Directional distribution = 70/30
Trafficable width = 7m
Adjustment factors:
directional distribution = 0.89
trafficable width = 0.91
trucks = 0.714
Road section capacity = 2,800 × 0.89 × 0.91 × 0.714
= 1,620 veh/h
7 Calculate the peak direction capacity using the road section capacity determined in step 6.
Peak direction capacity = road section capacity x proportion of traffic in the
peak direction
Example:
Proportion of traffic in
peak direction = 0.7
Peak direction capacity = 1,620 × 0.7
= 1,134 veh/h
Determining whether vehicle interactions are significant

When the effects of vehicle interactions are significant on road sections it is necessary to calculate the
additional travel time caused by those interactions. Vehicle interactions do not apply to other urban roads.
Procedure
Follow the steps below to determine whether the effects of vehicle interactions are significant.
Table A61: Steps to determine whether the effects of vehicle interactions are significant
Step Action
1 Use the capacity for the road section determined in Table A57.
2 Take a time period with its corresponding traffic volume (demand) as determined in Table A52.
3 Calculate the volume to capacity ratio.
Example:
Time period = 0700 to 0900
Time period traffic vol = 6202 vehicles
Traffic flow = 6202/2
= 3101 veh/h
Capacity = 4300 veh/h
Volume to capacity ratio = 3101/4300
= 0.72
4 Determine whether the effects of vehicle interactions are significant as follows:
If the road section is a … And the volume to Then vehicle interactions …
capacity ratio is …
motorway section greater than 0.7 shall be considered (continue to
Types of delays)
motorway section 0.7 or less are not considered (go to Table
A68)
multi-lane road greater than 0.7 shall be considered (continue to
Types of delays)
Appendix 3: Traffic data and travel time estimation > Types of delays
multi-lane road 0.7 or less are not considered (go to Table

A68)
two-lane rural road greater than 0.7 shall be considered (continue to
Types of delays)
two-lane rural road 0.7 or less are not considered (go to Table
A68)
5 Repeat steps 2 to 4 for any other time periods in which traffic volumes are likely to result in
significant vehicle interactions.
Types of delays
This section describes the difference between vehicle interaction delay and bottleneck delay, explaining
why the two types of delay require different procedures to calculate their levels.
The diagram below shows approximately when vehicle interaction (or random) delay and bottleneck (or
over-saturation) delay occur.
Figure A6: Vehicle interaction delay and bottleneck delay
0 0.5 1.0 1.5
Definition of vehicle interaction delay

Vehicle interaction delay is the delay that occurs as demand approaches capacity, and each vehicle’s
progress is impeded by the proximity of other vehicles.
Ideally, no delay would occur when demand was below capacity, but variations in driver behaviour and
differences in speed between individual vehicles mean that delay does occur. Because the actual delay
depends on the many variable factors, vehicle interaction delay is also known as random delay.
Definition of bottleneck delay
Bottleneck delay is the delay that is experienced when the demand at some location exceeds the
capacity of the road at the location. Such delays occur at a point on the road section where the capacity is
below that of the upstream capacity, and equal to or less than the downstream capacity.
Appendix 3: Traffic data and travel time estimation > Average peak interval traffic intensity
Because bottleneck delay occurs when demand exceeds capacity (ie when the volume to capacity ratio
exceeds 1.0), it is also known as over-saturation delay.
Average peak interval traffic intensity

As traffic volumes on a road increase, vehicle interactions increase, and as a result the average travel
time per vehicle increases. The additional travel time that results from vehicle interactions is a function of
the volume to capacity flow ratio (VC ratio), where VC ratio is the ratio of demand volume to road capacity
averaged over a period of time. When predicting the average travel time to traverse a section of road, the
extent to which averaging smooths the flow profile will affect the accuracy of the estimate of the additional
travel time due to vehicle interactions. Peak interval analysis is one method of correcting for potential loss
of accuracy.
The diagram below shows the relationship between the time period and the peak interval, and the
relationship between the average traffic intensities for the time period and the peak interval.
Figure A7: Average peak interval traffic intensity
Time period
Time
Average time period traffic intensity

The average time period traffic intensity is the average traffic flow for the time period under analysis. It is
generally reported as vehicles per hour, or vehicles per x minutes.
Peak interval
The peak interval (in minutes) is that portion of the time period over which the demand is greater than the
average time period traffic intensity.
Average peak interval traffic intensity
The average peak interval traffic intensity is the average traffic flow for the peak interval. The average
peak interval traffic intensity is used in the analysis to determine delays. Generally average peak interval
traffic intensity is reported in vehicles per hour.
Determining the peak interval

This procedure should be used if the conclusion from the procedure in Table A61 was that vehicle
interactions shall be considered.
Appendix 3: Traffic data and travel time estimation > Determining the peak interval
Procedure
Follow the steps below to determine the peak interval.
Table A62: Steps to determine the peak interval

Step Action
1 Select a time period to be analysed (usually the weekday morning or evening commuter
peak). See Dividing the year into time periods.
Note: The time period must be long enough to ensure sufficient capacity, even though for
some time that capacity is exceeded.
2 Identify the time interval that traffic data for the time period has been collected (usually 5, 10
or 15 minute intervals).
3 Set out the traffic data for the time period.
Example:
Time Observed traffic volume
7:00–7:15 800
7:15–7:30 1,040
7:30–7:45 1,200
7:45–8:00 1,280
8:00–8:15 1,240
8:15–8:30 1,140
8:30–8:45 1,020
8:45–9:00 840
4 Calculate the average time period traffic intensity (Ftp) (see Average time period traffic
intensity definition)
Example:
Time period traffic volume = 8,560 vehicles
Length of time period = 2 hours
Traffic data time interval = 15 minutes
Average time period traffic intensity (Ftp)
= 8,560 / (2 × 60 / 15)
= 1,070 per 15 minutes
5 Identify when the observed traffic volume rose above the average time period traffic intensity
(Ftp)
Example:
From step 3, the interval 7:30–7:45 was the first interval with an observed traffic volume
greater than the average time period traffic intensity (Ftp)
Start time of interval (ti) = 7:30
Volume in interval (vi) = 1,200 vehicles
Volume in prior interval (vi-1) = 1,040 vehicles
6 Calculate the peak interval start, which is the notional time at which the flow rate rose above
the average time period traffic intensity (Ftp).
Peak interval start = ti + (Ftp- vi-1)/vi - vi-1) × interval from step 2
Example:
Peak interval start = 7:30 + (1,070 – 1,040)/(1,200 – 1,040) × 15
= 7:32.8
Appendix 3: Traffic data and travel time estimation > Calculating the average peak interval traffic intensity
Step Action
7 Identify when the observed traffic volume fell below the average time period traffic intensity
(Ftp).
Example:
From step 3, the interval 8:30–8:45 was the first interval after the peak with an observed traffic
volume lower than the average time period traffic intensity (Ftp).
Start time of interval (ti) = 8:30
Volume in interval (vi) = 1,020 vehicles
Volume in prior interval (vi-1) = 1,140 vehicles
8 Calculate the peak interval end, which is the notional time at which the flow rate fell below the
average time period traffic intensity (Ftp).
Peak interval end = ti + (vi-1 - Ftp)/(vi-1 - vi) x interval
Example:
Peak interval end = 8:30 + (1,140 – 1,070)/(1,140 – 1,020) × 15
= 8:38.8
9 Calculate the length of the peak interval.
Example:
Peak interval start = 7:32.8
Peak interval end = 8:38.8
Length of peak interval = 8:38.8 - 7:32.8
= 66.0 minutes
Calculating the average peak interval traffic intensity

This procedure should only be applied after having calculated the peak interval in Table A62.
Procedure
Follow the steps below to calculate the average peak interval traffic intensity.
Table A63: Steps to calculate the average peak interval traffic intensity
Step Action
1 Calculate the peak interval traffic volume.
Example:
Peak interval start =
7:32.8
Peak interval end =
8:38.8
Volume 7:30–7:45 =
1,200 vehicles
Volume 7:45–8:00 =
1,280 vehicles
Volume 8.00–8.15 =
1,240 vehicles
Volume 8:15–8:30 =
1,140 vehicles
Volume 8:30–8:45 =
1,020 vehicles
Peak interval traffic vol =
(7:45 - 7:32.8)/15 × 1,200 + 1,280 +
1,240 + 1140 + (8:38.8 - 8:30)/15 × 1,020
= 5,234 vehicles
2 Calculate the average peak interval traffic intensity (Fpi).
Example:
Length of peak interval = 66.0 minutes
Average peak interval traffic intensity (Fpi)
= 5,234 × 60/66.0
= 4,758 veh/h
Appendix 3: Traffic data and travel time estimation > Calculating the volume to capacity ratio
Calculating the volume to capacity ratio

The volume to capacity (VC) ratio is also known as the saturation ratio.
Procedure
Follow the steps below to determine the VC ratio.
Table A64: Steps to determine the volume to capacity (VC) ratio
Step Action
1 Determine the appropriate capacity for calculating the VC ratio as follows:
If the road section is a … Then use the …
motorway section capacity determined in Table A58
multi-lane highway capacity determined in Table A59
two-lane rural road peak direction capacity determined in Table A60
other urban road capacity specified in Table A57
2 Obtain the average peak interval traffic intensity (Fpi) as determined in Table A63 and use
this volume in step 3.
Note: If the VC ratio is being calculated for a time period for which it is not appropriate to
calculate Fpi, then use an appropriate peak volume.
3 Calculate the VC ratio using the appropriate capacity and traffic volume determined in steps
1 and 2.
Example:
VC ratio = volume/capacity
= 4,758/5,072
= 0.938
Calculating the additional travel time

The average additional travel time above that experienced when travelling at the free speed shall be
determined as a function of the VC ratio during the peak interval of a given time period.
The additional travel time calculated for the peak interval is also used as the value for time period
additional travel time.
Procedure
Follow the steps below to calculate the additional travel time.
Table A65: Steps to calculate additional travel time

Step Action
1 Determine the appropriate procedure for the road section as follows
If the road section is a … Then go to …
motorway section step 2, and then step 4
multi-lane highway step 2, and then step 4
two-lane rural road step 3, and then step 4
2 Calculate the peak interval additional travel time factor, using the VC ratio determined in
Table A64, as follows (for motorways and multi-lane roads only):
If the peak interval VC ratio is … Then the peak interval additional travel time
factor (Fdr) equals …
less than or equal to 0.7 0
between 0.7 and 1.0 0.27 × (VC ratio - 0.70)
equal to or greater than 1.0 0.081
Go to step 4.
Appendix 3: Traffic data and travel time estimation > Calculating the additional travel time
Step Action
3 Determine the peak interval additional travel time factor from the tables below, using the VC
ratio determined in Table A64 for two-lane rural roads only.
Additional travel time factor for level terrain
VC ratio Percent no-passing
0 20 40 60 80 100
0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.10 0.04 0.04 0.05 0.05 0.06 0.06
0.20 0.08 0.08 0.09 0.10 0.10 0.11
0.30 0.11 0.12 0.12 0.13 0.14 0.14
0.40 0.14 0.14 0.15 0.16 0.16 0.17
0.50 0.16 0.16 0.17 0.18 0.18 0.19
0.60 0.18 0.19 0.19 0.20 0.20 0.21
0.70 0.21 0.21 0.21 0.22 0.22 0.23
0.80 0.24 0.24 0.24 0.25 0.25 0.25
0.90 0.27 0.27 0.28 0.28 0.28 0.28
1.00 0.32 0.32 0.32 0.32 0.32 0.32
Additional travel time factor for rolling terrain
0 20 40 60 80 100
0.00 0.00 0.00 0.00 0.00 0.00 0.02
0.10 0.06 0.06 0.07 0.08 0.09 0.09
0.20 0.11 0.12 0.13 0.13 0.14 0.15
0.30 0.14 0.15 0.16 0.17 0.18 0.18
0.40 0.16 0.17 0.19 0.20 0.20 0.20
0.50 0.18 0.19 0.21 0.22 0.23 0.23
0.60 0.20 0.22 0.24 0.25 0.26 0.26
0.70 0.23 0.26 0.28 0.30 0.31 0.31
0.80 0.29 0.32 0.35 0.37 0.38 0.39
0.90 0.38 0.42 0.45 0.47 0.49 0.50
1.00 0.50 0.55 0.59 0.62 0.64 0.65
Additional travel time factor for mountainous terrain
0 20 40 60 80 100
0.00 0.00 0.00 0.01 0.02 0.03 0.03
0.10 0.06 0.09 0.11 0.12 0.13 0.14
0.20 0.13 0.16 0.19 0.20 0.22 0.23
0.30 0.19 0.22 0.25 0.27 0.29 0.30
0.40 0.24 0.28 0.31 0.33 0.35 0.37
0.50 0.29 0.33 0.36 0.39 0.42 0.44
0.60 0.35 0.40 0.43 0.47 0.50 0.53
0.70 0.43 0.48 0.52 0.56 0.59 0.63
0.80 0.54 0.59 0.64 0.68 0.72 0.75
0.90 0.68 0.73 0.78 0.83 0.87 0.92
1.00 0.86 0.92 0.98 1.03 1.07 1.12
Appendix 3: Traffic data and travel time estimation > Calculating bottleneck delay
Step Action
Alternatively calculate Fdr directly using the expression:
Fdr = min(a + b.PNP + d.PNp2 g.PNP3 + c.VC ratio + e.VC ratio2 + h.VC ratio3 + f.PNP.VC ratio
+ i.PNP.VC ratio2 + j.PNP2.VC ratio ,0)
where:VC ratio is the volume to capacity flow ratio
PNP is the percent no-passing
And the coefficients a to j are given below.
Coefficient Level terrain Rolling terrain Mountainous terrain
a -1.906 × 10-2 -2.658 × 10-2 -3.039 × 10-2
b 1.420 × 10-4 1.640 × 10-4 1.480 × 10-3
c 0.617 1.008 1.059
d 3.260 × 10-6 3.610 × 10-6 1.378 × 10-5
e -0.771 -1.918 -1.515
f 6.43 × 10-4 6.220 × 10-4 1.570 × 10-3
g -2.42 × 10-8 -9.470 × 10-9 5.260 × 10-8
h 0.496 1.440 1.346
i -8.70 × 10-4 -1.748 × 10-3 2.897 × 10-4
j -6.49 × 10-7 -1.320 × 10-5 -1.379 × 10-6
4 Calculate the peak interval additional travel time by multiplying the free speed travel time in
Table A53 by the factor from step 2 or 3.
Peak interval additional = free speed travel time x peak interval
travel time additional travel time factor (Fdr)
Example 1: (motorway or multi-lane highway):
Free speed travel time = 0.571 mins/km
VC ratio = 0.938
Fdr (from step 2) = 0.27 × (0.938 - 0.70)
= 0.0643
Peak interval additional travel time = 0.571 × 0.0643
= 0.037 mins/km
Time period additional travel time = peak interval additional travel time
= 0.037 mins/km
Example 2: (two-lane rural road):
Free speed travel time = 0.636 mins/km
Percent no-passing = 60%
VC ratio = 1.10
Fdr (from tables in step 3) = 0.62
Peak interval additional travel time = 0.636 × 0.62
= 0.394 mins/km
Time period additional travel time = peak interval additional travel time
= 0.394 mins/km
Calculating bottleneck delay

This procedure should be used for all time periods during which demand exceeds capacity (VC ratio
greater than one) at some time.
Where traffic volumes exceed capacity, the resulting queues may block back onto upstream links. In such
circumstances care must be taken to ensure that the delays that arise on the under-capacity section are
not double counted on any upstream section.
Procedure
Follow the steps below to calculate bottleneck delay.
Appendix 3: Traffic data and travel time estimation > Calculating bottleneck delay
Table A66: Steps to calculate bottleneck delay

Step Action
1 Select a time period to be analysed (usually the weekday morning or evening commuter
peak).
2 Determine the capacity of the road section. See Table A57.
3 Identify the time interval step that traffic data for the time period has been collected (usually
5, 10 or 15-minute periods).
4 Set out the traffic data for the time period.
Example:
Time interval Observed traffic volume
7:00–7:15 264
7:15–7:30 475
7:30–7:45 591
7:45–8:00 600
8:00–8:15 591
8:15–8:30 475
8:30–8:45 264
8:45–9:00 250
9:00–9:15 234
5 At each time interval, calculate the cumulative demand with a running total of observed
traffic volume since the time period start.
Cumulative demand at time interval = sum of observed traffic volume since time period
start.
Example from step 4:
Cumulative demand for time interval 8:00–8:15 = 264 + 475 + 591 + 600 + 591
= 2,521
6 At each time interval, calculate the vehicles discharged. If the traffic volume for the time
interval is below the road section capacity then all the traffic is discharged. Only the number
of vehicles equivalent to the road section capacity is discharged if the traffic volume
exceeds capacity.
Time interval =
8:00–8:15
Capacity =
500 vehicles
Traffic volume =
591 vehicles
Vehicles discharged =
minimum of traffic volume or
capacity
= minimum (591, 500)
= 500
7 At each time interval, calculate the cumulative discharge with a running total of vehicles
discharged since the time period start.
Cumulative discharge at time interval = sum of vehicles discharged since
time period start
Appendix 3: Traffic data and travel time estimation > Determining whether to consider peak spreading
Step Action
8 At each time interval, calculate the queue at the end of the interval when traffic volume
exceeds capacity.
Time interval = 7:30–7:45
Traffic volume = 591 vehicles
Capacity = 500 vehicles
Queue at end of interval
= traffic volume - capacity, if traffic volume >
capacity
= 0, if traffic volume ≤ capacity
= 591 - 500
= 91 vehicles
9 At each time interval, calculate the queue at the start of the interval. This is the queue at the
end of the previous interval.
Time interval = 7:30–7:45
Queue at start of interval
= queue at end of previous interval
= 91 vehicles
10 At each time interval, calculate the average delay in vehicle minutes.
Average delay =interval time step × (queue at end of interval
+ queue at start of interval)/2
11 Sum the average delays over the entire time period to obtain the time period total delay.
12 Calculate the time period average delay per vehicle from the time period total delay divided
by the cumulative discharge of vehicles at the time period end.
Average delay per vehicle = total delay/cumulative discharge of vehicles
at the time period end
A worked example of the bottleneck delay procedure is provided in Appendix 8: Worked examples.
Determining whether to consider peak spreading

Some peak spreading may occur at low levels of bottleneck delay, but in general, drivers will only begin to
refine their trips when bottleneck delays are severe.
Appendix 3: Traffic data and travel time estimation > Determining the additional travel time resulting from speed change cycles
Procedure
Follow the steps below to determine whether peak spreading should be considered.
Table A67: Steps to determine whether to consider peak spreading

Step Action
1 Calculate the average delay per delayed vehicle, using the time period average delay per
vehicle determined in Table A66.
Average delay per delayed vehicle
= Time period average delay per vehicle x
(Time period traffic volume/sum of traffic volumes of intervals with an end queue)
Example (using the calculating bottleneck delay example in Appendix 8: Worked examples):
Average delay per delayed vehicle
= 3.37 × (3744 / (591 + 600 + 591 + 475 + 264))
= 3.37 × (3744 / 2521)
= 5.0mins/veh
2 Determine whether peak spreading should be considered as follows:
If the average minutes delay And there is … Then peak spreading …
per delayed vehicle is …
between 0 and 15 does not need to be
considered
between 15 and 25 an alternative route does not need to be
considered
between 15 and 25 no alternative route shall be considered, see
Applying variable trip matrix
with growth constraint
techniques
25 or greater shall be considered, see
Applying variable trip matrix
with growth constraint
techniques
Determining the additional travel time resulting from speed change cycles
If vehicles are required to slow to negotiate some isolated feature and then accelerate back to cruise
speed the travel time estimated above must be increased to account for the time lost during this speed
change cycle. Where the initial cruise speed and the minimum speed are available, tables in Appendix 4:
Vehicle operating cost tables provide the amount of additional travel time in seconds for speed change
cycles.
In the absence of measured data, the additional travel time that occurs as a result of having to slow for
substandard horizontal curves can be approximated using this procedure.
Procedure
Follow the steps below to determine the additional travel time resulting from speed change cycles
associated with substandard curves.
Appendix 3: Traffic data and travel time estimation > Determining the additional travel time resulting from speed change cycles
Table A68: Steps to determine the additional travel time of speed change cycles from substandard
curves
Step Action
1 Determine the curve negotiating speed for each vehicle type in the traffic mix.
The desired negotiation speed for an isolated curve (S c) is related to the ideal approach
speed (Sa) and the curve radius (R) by the following equation:
Sc = a0 + a1.Sa + a2 / R
where: Sa = f1.FS
Fs is the average free speed determined from Table A53 and Table A56 and the
coefficients f1,a0, a1, and a2 are as follows:
Vehicle type f1 a0 a1 a2
Car 1.00 45.21 0.5833 -3,892
LCV 0.97 54.51 0.4531 -3,337
MCV 0.89 51.77 0.4744 -3,245
HCVI 0.91 59.16 0.4068 -3,506
HCVII 0.91 69.57 0.3085 -3,768
Bus 0.91 59.16 0.4068 -3,506
Example:
A horizontal curve of radius 100m exists within a road section where the free speed is
estimated at 94.33km/h.
Ideal approach speed = 0.89 × 94.33
For MCV = 84km/h
Desired negotiation speed for MCV
= 51.77 + 0.4744 × 84 – 3,245/100
= 59km/h
2 Determine the initial operating speed of the road section. The operating speed is the sum
of the free speed travel time and the time period additional travel time all divided by the
section length. This accounts for the reduction in the ideal approach speed as a result of
traffic interactions.
Initial operating speed = length/(TTFS + TTATT)
Example:
1km at free speed travel time
= 0.636mins/km
1km additional travel time for vehicle interactions (from Table A65)
= 0.636 × 0.2
= 0.127 mins/km
Initial operating speed = 1.00/(0.636 + 0.127) × 60
= 1.00/0.763 × 60
= 79km/h
3 The additional travel time associated with speed change cycles is then determined from
the appropriate table in Appendix 4: Vehicle operating cost tables.
Note: Where the desired negotiating speed is greater than the operating speed no speed
change will occur.
Example:
Using Table A100
Initial cruise speed for all vehicles = 79km/h
Curve speed for MCV = 59km/h
MCV additional travel time per speed change = 2.0 seconds
Appendix 3: Traffic data and travel time estimation > Calculating the time period total average travel time
Step Action
4 Calculate the total speed change cycle travel time for a road section with the additional
following information.
Traffic volume for the time period
Traffic composition (default values available in Table A47)
For each vehicle type:
proportion in traffic from traffic composition
number of vehicles = traffic volume × proportion in traffic
additional travel time = number of vehicles × additional travel time for
speed change cycles
Sum over all vehicle types to obtain the total additional travel time.
Calculating the time period total average travel time

Use this procedure once free speed and delays caused by vehicle interactions and speed changes have
been calculated.
When evaluating 'other urban roads', this procedure is used in conjunction with the procedure in Table
A57.
Procedure
Follow the steps below to calculate the time period total average travel time per vehicle.
Table A69: Steps to calculate the time period total average travel time per vehicle
Step Action
1 Use the following previously calculated values:
• free speed travel time (Table A53)
• time period additional travel time (Table A65)
• time period average delay per vehicle (Table A66)
• additional travel time due to speed changes (Table A68).
Notes:
‘Other urban roads’ only have a free speed travel time. ‘Other urban roads’ do not exhibit
reductions in travel times with increasing traffic volumes. All delays due to increasing traffic
volumes can be attributed to intersections as calculated in the procedures for traffic signals,
priority intersections and roundabouts below.
Time period additional travel time is only calculated if the VC ratio exceeds 0.7 (see Table
A61).
Bottleneck delay is only calculated if demand exceeds capacity at some time during the
time period.
2 Multiply the free speed travel time and the time period additional travel time by the road
section length.
3 Sum the values in step 2 with the bottleneck delay and additional travel time due to speed
change to get the time period total average travel time per vehicle.
Appendix 3: Traffic data and travel time estimation > Traffic signals
Traffic signals
Travel time delays associated with traffic signals are the result of a complex interaction between arrivals
on opposing phases, the response of the signal controller to detector impulses and external control
commands, and vehicle driver responses. The physical layout, location and phasing strategy also affect
operations.
Commonly available analysis procedures are based on simplifying assumptions that reduce an essentially
dynamic and stochastic process to a deterministic approximation of real events. Reliable estimates of
delay require the careful selection of values for the governing variables and a thorough understanding of
traffic operations at each site.
While the procedures of the HCM provide a useful guide, the more commonly understood methods of
Akcelik R (1981) should be followed.
This appendix uses HCM to derive a major modification to the ARR 123 methods to account for the
proximity of other signals including linking or coordination.
Capacity or saturation flow rate
The average delay to all vehicles, irrespective of the turns made, shall be the basis of the analysis. For
this reason, the methodology is approach based, not movement based.
Ideally, saturation flow rates for each approach should be determined from direct observation at the site.
Approach saturation flow rates for the relevant lane groups can be estimated as specified below.
The procedure consists of adjusting an ideal saturation flow rate of 2000 passenger cars units per hour of
green by the factors tabulated in Table A70, Table A71, Table A73 and Table A74.
Parking movements refers to the number of such movements, in and out, within a length of 50m on either
side of the intersection.
Table A70: Lane width factors
Lane width (metres) Factor

3.5 1.00
3.4 0.99
3.3 0.98
3.2 0.97
3.1 0.96
3.0 0.95
Table A71: Approach grade factors

Gradient % Factor
-4 1.02
-2 1.01
0 1.00
+2 0.99
+4 0.98
Table A72: Parking factors
Parking movements Approach lanes

(number/hour) 1 2 3
0 0.90 0.95 0.97
10 0.85 0.92 0.95
20 0.80 0.89 0.93
30 0.75 0.87 0.85
40 0.70 0.85 0.89
Table A73: Locality factors
Type of street Factor

CBD shopping 0.90
Suburban shopping 0.95
Other 1.00
Cycle times and phase splits

Appropriate cycle times and phase splits shall be determined according to the conditions that prevail
during the peak interval. In particular, the influence of minimum phase times for parallel pedestrian
facilities, actual all-red periods, and other influences on lost-time shall be included.
Peak interval average travel time
The peak interval average travel time shall be the average delay calculated by the methods of ARR 123
adjusted to account for controller type and the arrival pattern of platoons produced by nearby
intersections by applying the relevant delay adjustment factor specified below.
The arrival type is best observed in the field, but can be assessed by examining time-space diagrams for
the arterial or street on which the approach is located.
It should be noted that fully vehicle actuated controllers, remote from other signals, produce delays 15%
below that estimated by the methods of ARR 123.
Care must be exercised in applying the adjustment factors in Table A75. Arrival types 1 and 5, from Table
A74, will seldom occur unless either unfavourable or efficient linking control is imposed.
Platoons released by upstream signals will disperse according to the prevailing speed environment and
the distance between successive signal controlled intersections. Table A76 provides a broad guide to
such effects.
Table A74: Arrival type

Arrival type Condition
1 Dense platoon arriving at the commencement of red.
Dense platoon arriving near the middle of the red phase, or
2
Dispersed platoon arriving at the commencement of red.
Random arrivals or dispersed platoons arriving throughout both the green and red
3 phases. This condition applies to isolated intersections or those with cycle times
differing from nearby signal controlled intersections.
Dense platoon arriving near the middle of the green phase, or
4
Dispersed platoon arriving throughout the green phase.
5 Dense platoon arriving at the commencement of the green phase.
Table A75: Delay adjustment factor
Type of Volume to Adjustment factor

signal capacity Arrival type
ratio
1 2 3 4 5
Pre-timed ≤0.6 1.85 1.35 1.00 0.72 0.53
0.8 1.50 1.22 1.00 0.82 0.67
1.0 1.40 1.18 1.00 0.90 0.82
Actuated ≤0.6 1.54 1.08 0.85 0.62 0.40
0.8 1.25 0.98 0.85 0.71 0.50
1.0 1.16 0.94 0.85 0.78 0.61
Semi- ≤0.6 1.85 1.35 1.00 0.72 0.42
actuated on 0.8 1.50 1.22 1.00 0.82 0.53
main road
1.0 1.40 1.18 1.00 0.90 0.65
approach
Semi- ≤0.6 1.48 1.18 1.00 0.86 0.70
actuated on 0.8 1.20 1.07 1.00 0.98 0.89
side road
1.0 1.12 1.04 1.00 1.00 1.00
approach
Table A76: Platoon dispersal distances (m)

Platoon type Speed environment (km/h)
50–64 65–105
Dense <100 <300
Dispersed 150–500 350–1,000
Random >1,000 >2,000
Intersection departure delay

The HCM specifies reductions in the free speed according to the distance between signal-controlled
intersections along the route. This amounts to a nearly constant delay of six seconds (0.10 minutes) at
each intersection. The effect can be represented by adding this constant delay in addition to actual
intersection delays.
Time period total average travel time
The time period total average travel time for the intersection is approximated by the peak interval time
period delay obtained plus the intersection departure delay, as described in the previous sections of this
appendix.
Application of traffic models
Delays associated with traffic signals can be estimated by traffic models, provided:
• input parameters such as running speeds and saturation flow rates are determined in a manner
consistent with this appendix
• the delay calculated by the model is consistent with the definitions of this appendix, ie the
average delay per vehicle over the relevant approach
• the delay outputs of the model are based on the general procedure and delay equations of ARR
123 and this appendix.
A worked example of the Traffic signals procedure is provided in Appendix 8: Worked examples.
Appendix 3: Traffic data and travel time estimation > Priority intersections
Priority intersections
Priority intersections include all intersections where entry is not controlled by traffic signals. Roundabouts
are a particular class, and are separately considered in a procedure below.
Travel time delays are only incurred by movements where the priority of entry is controlled by stop signs,
give way signs, or by the general intersection driving rules. Three levels of priority are involved:
1. movements that have priority

2. movements that yield the right-of-way to the priority flows
3. movements that must give way to both the above categories.
Only priority levels 2 and 3 will experience delays.
Minimum headway in conflicting flow
The distribution of headways in the opposing traffic streams in turn depends on other variables, and is
influenced by the proximity of signal controlled intersections. When the priority intersection is remote from
traffic signals and the conflicting flows well below the capacities of their approach roadways, the
distribution of headways in the conflicting traffic flows can be assumed to be random with a minimum
headway of either 2.0 seconds (single-lane conflict) or 0.5 seconds in other cases.
Capacity
The capacity of a non-priority movement shall be determined as a function of the following variables:
• the distribution of headways, being the time between successive users of the conflict area
• the critical gap in the opposing traffic flow through which a non-priority movement vehicle will
move
• the follow-up headway being the time interval between successive vehicles which use the same
gap in the opposing traffic stream.
The capacity of the non-priority movement shall be then estimated from:
c = (3600 / Tf) × exp (-V × To / 3600)
where: c is capacity
To is Tg – Hm (Hm = 0.5 or 2.0)
Hm is minimum headway in conflicting flow
Tg is critical gap
Tf is follow-up headway
V is conflicting volume during peak interval, veh/h.
To, Tg, Hm and Tf are expressed in seconds, and c and V are expressed in vehicles per hour.
Critical gap and follow up headways
The critical gap (Tg) and follow-up headway (Tf) are related and depend on the speed of the conflicting
traffic flow, the class of control, and the movement type. In the absence of actual values determined by
observations at the site or similar sites elsewhere in New Zealand, the values in Table A77 should be
used.
Where the turning movement is required to cross more than one lane, a further 0.5 seconds shall be
added to the values of the table.
If the left turn from a minor road is provided with an acceleration lane, the critical gap of the table shall be
reduced by 1.0 seconds.
The follow-up headway is related to the critical gap, by the expression: Tf = 2.0 + 0.2 Tg
Appendix 3: Traffic data and travel time estimation > Roundabouts
Table A77: Critical gap (Tg)
Movement and control Average speed (km/h)

<60 ≥60
Right turn from
major road 4.5 5.0
Stop sign on minor road:
left turn 5.0 6.0
through 5.5 7.0
right turn 6.0 7.5
Give way on minor road:
left turn 4.5 5.0
through 5.0 6.0
right turn 5.5 6.5
Average peak interval delay

For Table A78 the following conditions apply:
• The movement VC ratio is the ratio of the average movement traffic demand for that movement
during the peak interval divided by the capacity.
• The peak interval average travel time is equivalent to the delay for each movement. This delay
depends on the VC ratio as tabulated in the table on the next page.
• The total average travel for the intersection is approximated by the peak interval time period.
Table A78: Average peak interval delay
Volume to capacity ratio Average peak interval delay (min/veh)

0.20 0.05
0.30 0.06
0.40 0.07
0.50 0.10
0.60 0.12
0.70 0.17
0.80 0.28
0.90 0.58
1.00 2.75
1.05 5.70
1.10 10.2
>1.10 12.0
Where a traffic model has been used to calculate delays at priority intersections, the provisions of the
procedure for traffic signals also apply.
Roundabouts
Roundabouts are a special case of a priority intersection. Delays at each approach can be estimated in a
manner similar to that given in the procedure for priority intersections, ie each approach can be
considered as an independent elemental intersection with one-way conflicting flows circulating round the
central island.
Procedure
The procedures and methods of Austroads (1993) Guide to traffic engineering practice part 6:
roundabouts shall be used to obtain the capacities of each approach lane.
Appendix 3: Traffic data and travel time estimation > Roundabouts
The VC ratio for each approach lane shall be estimated as the expected average flow during the peak
interval using that lane divided by the capacity.
The peak interval travel time is equivalent to the peak interval average delay for each lane. The peak
interval delay shall be estimated from Table A78 up to a maximum VC ratio of 1.05, and the average peak
period delay for the approach shall be estimated as the weighted average of the individual approach
lanes.
The performance of a roundabout becomes indeterminate for high flows, much beyond the capacity of an
approach, due to a tendency for the flows to ‘lock’ round the central island.
The time period total average travel time is the average delay during the time period, and shall be
estimated from the peak interval delay.
Where a traffic model has been used to calculate delays at roundabouts, the provisions of the procedure
for traffic signals also apply.
Appendix 4: Vehicle operating cost tables > Roundabouts
Appendix 4: Vehicle operating cost tables

Table A79: Passenger car VOC by speed and gradient (cents/km – July 2015)
Speed Gradient in percent (both directions)
(km/h) 0 1 2 3 4 5 6 7 8 9 10 11 12
10 34.0 34.1 34.2 34.2 34.3 34.4 34.5 34.6 34.8 35.0 35.2 35.5 35.9
15 30.4 30.5 30.6 30.7 30.8 30.9 31.1 31.3 31.5 31.8 32.1 32.5 32.9
20 27.7 27.8 28.0 28.1 28.2 28.4 28.6 28.8 29.0 29.4 29.7 30.2 30.7
25 25.8 25.9 26.0 26.2 26.3 26.5 26.7 27.0 27.3 27.6 28.0 28.5 29.1
30 24.4 24.5 24.7 24.8 24.9 25.1 25.4 25.6 25.9 26.3 26.8 27.3 27.9
35 23.4 23.5 23.6 23.8 23.9 24.1 24.4 24.7 25.0 25.4 25.9 26.4 27.0
40 22.7 22.8 22.9 23.1 23.2 23.4 23.7 24.0 24.3 24.7 25.2 25.8 26.4
45 22.2 22.3 22.4 22.5 22.7 22.9 23.2 23.5 23.8 24.3 24.8 25.3 26.0
50 21.8 21.9 22.1 22.2 22.4 22.6 22.8 23.2 23.5 24.0 24.5 25.1 25.7
55 21.7 21.8 21.9 22.0 22.2 22.4 22.7 23.0 23.3 23.8 24.3 24.9 25.6
60 21.6 21.7 21.8 21.9 22.1 22.3 22.6 22.9 23.3 23.7 24.3 24.9 25.6
65 21.6 21.7 21.8 21.9 22.1 22.3 22.6 22.9 23.3 23.8 24.3 24.9 25.6
70 21.7 21.8 21.9 22.0 22.2 22.4 22.7 23.0 23.4 23.9 24.4 25.0 25.7
75 21.9 22.0 22.1 22.2 22.4 22.6 22.8 23.2 23.6 24.0 24.6 25.2 25.9
80 22.1 22.2 22.3 22.4 22.6 22.8 23.0 23.4 23.8 24.2 24.8 25.4 26.2
85 22.4 22.4 22.5 22.7 22.8 23.0 23.3 23.6 24.0 24.5 25.1 25.7 26.5
90 22.7 22.7 22.8 23.0 23.1 23.3 23.6 23.9 24.3 24.8 25.4 26.0 26.8
95 23.0 23.1 23.2 23.3 23.5 23.7 23.9 24.3 24.7 25.1 25.7 26.4 27.1
100 23.4 23.4 23.5 23.6 23.8 24.0 24.3 24.6 25.0 25.5 26.1 26.7 27.5
105 23.8 23.8 23.9 24.0 24.2 24.4 24.7 25.0 25.4 25.9 26.5 27.1 27.9
110 24.2 24.3 24.3 24.5 24.6 24.8 25.1 25.4 25.8 26.3 26.9 27.6 28.3
115 24.7 24.7 24.8 24.9 25.0 25.2 25.5 25.8 26.3 26.7 27.3 28.0 28.8
120 25.1 25.2 25.2 25.3 25.5 25.7 26.0 26.3 26.7 27.2 27.8 28.4 29.2
Table A80: LCV VOC by speed and gradient (cents/km – July 2015)
(km/h) 0 1 2 3 4 5 6 7 8 9 10 11 12
10 44.2 44.2 44.3 44.3 44.5 44.7 44.9 45.2 45.6 46.1 46.7 47.3 48.1
15 39.2 39.3 39.4 39.5 39.8 40.0 40.4 40.8 41.4 42.0 42.7 43.5 44.4
20 35.5 35.6 35.7 35.9 36.2 36.6 37.0 37.5 38.1 38.8 39.6 40.5 41.5
25 32.9 33.0 33.1 33.3 33.6 34.0 34.5 35.0 35.7 36.4 37.3 38.3 39.4
30 30.9 31.0 31.2 31.4 31.8 32.2 32.7 33.2 33.9 34.7 35.7 36.7 37.8
35 29.6 29.7 29.8 30.1 30.4 30.8 31.4 32.0 32.7 33.5 34.5 35.6 36.8
40 28.6 28.7 28.9 29.2 29.5 29.9 30.5 31.1 31.8 32.7 33.7 34.8 36.0
45 28.0 28.1 28.3 28.5 28.9 29.3 29.9 30.5 31.3 32.2 33.2 34.3 35.6
50 27.6 27.7 27.9 28.2 28.5 29.0 29.5 30.2 31.0 31.9 32.9 34.1 35.4
55 27.5 27.6 27.8 28.0 28.4 28.8 29.4 30.1 30.9 31.8 32.8 34.0 35.3
60 27.5 27.6 27.8 28.0 28.4 28.9 29.4 30.1 30.9 31.9 32.9 34.1 35.5
65 27.7 27.8 27.9 28.2 28.6 29.0 29.6 30.3 31.1 32.0 33.1 34.3 35.7
70 28.0 28.0 28.2 28.5 28.8 29.3 29.9 30.6 31.4 32.3 33.4 34.7 36.0
75 28.3 28.4 28.6 28.8 29.2 29.7 30.2 30.9 31.8 32.7 33.8 35.1 36.4
80 28.8 28.9 29.0 29.3 29.6 30.1 30.7 31.4 32.2 33.2 34.3 35.5 36.9
85 29.3 29.4 29.5 29.8 30.1 30.6 31.2 31.9 32.7 33.7 34.8 36.1 37.5
90 29.9 30.0 30.1 30.4 30.7 31.2 31.8 32.5 33.3 34.3 35.4 36.7 38.1
95 30.5 30.6 30.7 31.0 31.3 31.8 32.4 33.1 34.0 34.9 36.1 37.3 38.8
100 31.2 31.3 31.4 31.7 32.0 32.5 33.1 33.8 34.6 35.6 36.7 38.0 39.5
105 31.9 32.0 32.1 32.4 32.7 33.2 33.8 34.5 35.3 36.3 37.5 38.8 40.2
110 32.7 32.7 32.9 33.1 33.4 33.9 34.5 35.2 36.1 37.1 38.2 39.5 41.0
115 33.5 33.5 33.6 33.9 34.2 34.7 35.3 36.0 36.8 37.8 39.0 40.3 41.7
120 34.3 34.3 34.4 34.7 35.0 35.5 36.1 36.8 37.6 38.6 39.8 41.1 42.5
Table A81: MCV VOC by speed and gradient (cents/km – July 2015)
(km/h) 0 1 2 3 4 5 6 7 8 9 10 11 12
10 68.1 68.5 69.6 71.3 73.5 76.0 78.7 81.6 84.6 87.4 90.1 92.6 94.6
15 63.2 63.5 64.5 66.2 68.3 70.8 73.7 76.7 79.9 83.0 86.0 88.8 91.3
20 59.7 59.9 60.8 62.4 64.5 67.1 70.0 73.1 76.4 79.7 82.9 85.9 88.7
25 57.2 57.3 58.2 59.7 61.9 64.5 67.4 70.6 74.0 77.4 80.8 84.0 86.9
30 55.6 55.6 56.4 58.0 60.1 62.7 65.7 68.9 72.4 75.9 79.4 82.7 85.8
35 54.6 54.6 55.3 56.8 58.9 61.5 64.5 67.8 71.4 75.0 78.5 82.0 85.3
40 54.1 54.0 54.7 56.1 58.2 60.8 63.9 67.2 70.8 74.5 78.1 81.7 85.1
45 53.9 53.7 54.4 55.8 57.9 60.5 63.6 67.0 70.6 74.3 78.1 81.7 85.2
50 54.0 53.8 54.4 55.8 57.9 60.5 63.6 67.0 70.7 74.4 78.3 82.0 85.5
55 54.4 54.1 54.7 56.1 58.1 60.7 63.8 67.3 71.0 74.8 78.7 82.5 86.1
60 54.9 54.6 55.1 56.5 58.5 61.2 64.3 67.7 71.4 75.3 79.2 83.1 86.8
65 55.5 55.2 55.7 57.1 59.1 61.7 64.8 68.3 72.1 76.0 79.9 83.9 87.6
70 56.3 55.9 56.5 57.8 59.8 62.4 65.5 69.0 72.8 76.7 80.8 84.7 88.6
75 57.2 56.8 57.3 58.6 60.6 63.2 66.3 69.9 73.7 77.6 81.7 85.7 89.6
80 58.2 57.8 58.2 59.5 61.5 64.1 67.2 70.8 74.6 78.6 82.7 86.8 90.7
85 59.3 58.8 59.2 60.5 62.5 65.1 68.2 71.8 75.6 79.6 83.8 87.9 91.9
90 60.4 59.9 60.3 61.5 63.5 66.1 69.3 72.8 76.7 80.7 84.9 89.1 93.1
95 61.6 61.0 61.4 62.6 64.6 67.2 70.4 73.9 77.8 81.9 86.1 90.3 94.4
100 62.8 62.2 62.6 63.8 65.8 68.4 71.5 75.1 79.0 83.1 87.3 91.5 95.6
105 64.0 63.4 63.8 65.0 66.9 69.5 72.7 76.3 80.2 84.3 88.5 92.8 97.0
110 65.3 64.7 65.0 66.2 68.1 70.7 73.9 77.5 81.4 85.5 89.8 94.1 98.3
115 66.6 65.9 66.3 67.4 69.4 72.0 75.1 78.7 82.7 86.8 91.1 95.5 99.7
120 67.9 67.2 67.5 68.7 70.6 73.2 76.4 80.0 83.9 88.1 92.5 96.8 101.1
Table A82: HCVI VOC by speed and gradient (cents/km – July 2015)
(km/h) 0 1 2 3 4 5 6 7 8 9 10 11 12
10 118.9 118.9 120.5 123.4 127.4 132.2 137.7 143.6 149.7 155.6 161.3 166.4 170.8
15 111.5 111.4 112.9 115.8 119.9 124.9 130.6 136.9 143.3 149.7 156.0 161.7 166.8
20 105.4 105.1 106.6 109.5 113.6 118.8 124.7 131.1 137.9 144.7 151.3 157.5 163.1
25 100.8 100.5 101.9 104.8 109.0 114.2 120.3 126.9 133.8 140.9 147.8 154.4 160.4
30 97.5 97.2 98.5 101.5 105.7 111.0 117.2 123.9 131.1 138.3 145.5 152.4 158.7
35 95.3 94.9 96.2 99.2 103.4 108.8 115.1 122.0 129.2 136.7 144.1 151.2 157.8
40 94.0 93.5 94.8 97.7 102.0 107.5 113.8 120.8 128.2 135.8 143.4 150.7 157.5
45 93.2 92.7 94.0 96.9 101.3 106.8 113.2 120.2 127.8 135.5 143.3 150.8 157.8
50 93.0 92.4 93.7 96.7 101.0 106.6 113.0 120.2 127.9 135.7 143.6 151.3 158.5
55 93.2 92.6 93.9 96.9 101.2 106.8 113.3 120.6 128.3 136.3 144.4 152.2 159.6
60 93.8 93.2 94.5 97.4 101.8 107.4 114.0 121.3 129.1 137.2 145.4 153.4 160.9
65 94.7 94.0 95.3 98.2 102.6 108.3 114.9 122.3 130.2 138.4 146.7 154.8 162.4
70 95.8 95.1 96.3 99.3 103.7 109.4 116.1 123.5 131.5 139.8 148.2 156.4 164.2
75 97.1 96.4 97.6 100.6 105.0 110.7 117.4 125.0 133.0 141.4 149.9 158.2 166.1
80 98.6 97.8 99.1 102.0 106.5 112.2 119.0 126.5 134.7 143.1 151.7 160.1 168.1
85 100.2 99.5 100.7 103.6 108.1 113.9 120.7 128.3 136.5 145.0 153.6 162.1 170.3
90 102.0 101.2 102.4 105.4 109.9 115.6 122.5 130.1 138.4 147.0 155.7 164.3 172.6
95 103.8 103.0 104.2 107.2 111.7 117.5 124.4 132.1 140.4 149.1 157.8 166.5 174.9
100 105.8 105.0 106.2 109.1 113.7 119.5 126.4 134.1 142.5 151.2 160.1 168.9 177.3
105 107.8 107.0 108.2 111.2 115.7 121.5 128.5 136.2 144.7 153.4 162.4 171.2 179.8
110 109.9 109.1 110.3 113.2 117.8 123.7 130.6 138.4 146.9 155.7 164.7 173.7 182.3
115 112.1 111.2 112.4 115.4 119.9 125.8 132.8 140.7 149.2 158.1 167.1 176.1 184.8
120 114.3 113.4 114.6 117.6 122.1 128.1 135.1 143.0 151.5 160.5 169.6 178.6 187.4
Table A83: HCVII VOC by speed and gradient (cents/km – July 2015)
(km/h) 0 1 2 3 4 5 6 7 8 9 10 11 12
10 172.9 178.2 186.6 197.5 210.4 224.7 239.9 255.6 271.2 286.2 300.1 312.3 322.5
15 173.1 176.8 183.9 193.7 205.8 219.6 234.7 250.5 266.5 282.1 296.9 310.4 322.1
20 169.5 172.5 178.9 188.3 200.2 214.0 229.3 245.4 262.0 278.4 294.3 309.0 322.0
25 166.0 168.5 174.6 183.9 195.8 209.8 225.3 242.0 259.2 276.4 293.2 309.0 323.3
30 163.2 165.4 171.5 180.7 192.7 207.0 222.9 240.1 257.9 275.9 293.6 310.4 325.9
35 161.3 163.4 169.4 178.7 190.9 205.5 221.8 239.5 257.9 276.7 295.2 313.0 329.5
40 160.1 162.2 168.2 177.7 190.1 205.0 221.8 239.9 259.0 278.4 297.8 316.4 333.9
45 159.7 161.7 167.8 177.5 190.2 205.4 222.5 241.2 260.9 281.0 301.0 320.5 338.9
50 159.8 161.9 168.1 178.0 190.9 206.4 224.0 243.2 263.4 284.1 304.9 325.1 344.4
55 160.5 162.6 169.0 179.0 192.2 208.1 226.1 245.7 266.4 287.8 309.2 330.2 350.2
60 161.5 163.8 170.3 180.5 194.0 210.2 228.6 248.7 269.9 291.8 313.9 335.6 356.4
65 163.0 165.3 172.0 182.4 196.2 212.7 231.5 252.0 273.7 296.2 318.8 341.2 362.7
70 164.7 167.2 174.0 184.7 198.7 215.5 234.7 256.6 277.8 300.8 324.1 347.0 369.2
75 166.8 169.3 176.3 187.2 201.5 218.6 238.1 259.5 282.2 305.7 329.5 353.0 375.9
80 169.0 171.7 178.9 190.0 204.5 222.0 241.8 263.6 286.7 310.7 335.0 359.2 382.6
85 171.5 174.3 181.6 193.0 207.8 225.5 245.7 267.8 291.4 315.8 340.7 365.4 389.5
90 174.1 177.1 184.6 196.1 211.2 229.2 249.7 272.2 296.2 321.1 346.4 371.7 396.3
95 176.9 180.0 187.7 199.4 214.7 233.1 253.9 276.8 301.1 326.5 352.3 378.1 403.3
100 179.8 183.0 190.9 202.9 218.4 237.0 258.2 281.4 306.2 331.9 358.2 384.5 410.2
105 182.8 186.2 194.2 206.4 222.2 241.1 262.6 286.1 311.3 337.4 364.2 390.9 417.2
110 186.0 189.5 197.7 210.1 226.1 245.2 267.0 290.9 316.4 343.0 370.2 397.4 424.1
115 189.2 192.8 201.2 213.8 230.0 249.5 271.5 295.8 321.6 348.6 376.2 403.8 431.1
120 192.5 196.3 204.8 217.6 234.1 253.8 276.1 300.7 326.9 354.2 382.2 410.3 438.0
Table A84: Bus VOC by speed and gradient (cents/km – July 2015)
(km/h) 0 1 2 3 4 5 6 7 8 9 10 11 12
10 79.0 80.4 82.6 85.5 88.9 92.7 96.8 101.0 105.2 109.3 113.0 116.4 119.1
15 75.4 76.5 78.4 81.1 84.5 88.4 92.6 97.1 101.6 106.1 110.4 114.4 117.9
20 72.0 72.7 74.5 77.0 80.4 84.2 88.5 93.2 97.9 102.7 107.4 111.8 115.8
25 69.5 70.0 71.6 74.0 77.3 81.2 85.5 90.2 95.2 100.1 105.1 109.8 114.2
30 67.9 68.3 69.7 72.1 75.2 79.1 83.5 88.3 93.3 98.5 103.6 108.6 113.2
35 67.2 67.4 68.7 70.9 74.1 77.9 82.3 87.1 92.3 97.6 102.8 108.0 112.9
40 67.1 67.1 68.3 70.5 73.6 77.4 81.8 86.7 91.9 97.3 102.7 108.0 113.2
45 67.6 67.4 68.5 70.6 73.6 77.4 81.8 86.7 92.0 97.5 103.0 108.5 113.8
50 68.4 68.2 69.1 71.2 74.1 77.9 82.3 87.2 92.6 98.1 103.8 109.4 114.8
55 69.6 69.3 70.1 72.1 75.0 78.7 83.2 88.1 93.5 99.1 104.8 110.6 116.1
60 71.1 70.6 71.4 73.3 76.2 79.9 84.3 89.3 94.7 100.3 106.2 112.0 117.7
65 72.8 72.2 72.9 74.8 77.6 81.3 85.7 90.7 96.1 101.8 107.7 113.6 119.5
70 74.7 74.0 74.6 76.4 79.2 82.9 87.3 92.3 97.7 103.5 109.5 115.5 121.4
75 76.8 76.0 76.5 78.3 81.0 84.6 89.0 94.0 99.5 105.3 111.4 117.5 123.5
80 79.0 78.1 78.6 80.2 82.9 86.6 90.9 96.0 101.5 107.3 113.4 119.6 125.6
85 81.3 80.3 80.7 82.3 85.0 88.6 93.0 98.0 103.5 109.4 115.5 121.8 127.9
90 83.7 82.6 83.0 84.5 87.2 90.7 95.1 100.1 105.7 111.6 117.8 124.1 130.3
95 86.1 85.1 85.3 86.8 89.4 93.0 97.3 102.4 107.9 113.9 120.1 126.4 132.7
100 88.7 87.5 87.7 89.2 91.8 95.3 99.6 104.7 110.2 116.2 122.5 128.9 135.3
105 91.3 90.1 90.2 91.6 94.2 97.7 102.0 107.0 112.6 118.6 124.9 131.4 137.8
110 93.9 92.6 92.8 94.1 96.6 100.1 104.4 109.5 115.1 121.1 127.4 133.9 140.4
115 96.6 95.3 95.3 96.7 99.1 102.6 106.9 111.9 117.5 123.6 130.0 136.5 143.0
120 99.4 97.9 97.9 99.2 101.7 105.1 109.4 114.4 120.1 126.2 132.5 139.1 145.7
Table A85: Urban arterial VOC by speed and gradient (cents/km – July 2015)
(km/h) 0 1 2 3 4 5 6 7 8 9 10 11 12
10 39.4 39.5 39.8 40.1 40.6 41.0 41.5 42.1 42.7 43.4 44.0 44.7 45.3
15 35.6 35.8 36.0 36.4 36.8 37.3 37.9 38.6 39.3 40.0 40.8 41.6 42.4
20 32.8 32.9 33.2 33.6 34.0 34.6 35.2 35.9 36.6 37.4 38.3 39.2 40.1
25 30.7 30.9 31.1 31.5 32.0 32.6 33.2 33.9 34.7 35.6 36.5 37.5 38.5
30 29.2 29.3 29.6 30.0 30.5 31.1 31.8 32.5 33.4 34.3 35.2 36.2 37.3
35 28.1 28.3 28.5 28.9 29.4 30.0 30.7 31.5 32.4 33.3 34.3 35.4 36.5
40 27.4 27.5 27.8 28.2 28.7 29.3 30.0 30.8 31.7 32.7 33.7 34.8 36.0
45 26.9 27.0 27.3 27.7 28.2 28.8 29.5 30.4 31.3 32.3 33.4 34.5 35.7
50 26.5 26.7 26.9 27.3 27.9 28.5 29.3 30.1 31.0 32.1 33.2 34.4 35.6
55 26.4 26.5 26.8 27.2 27.7 28.4 29.1 30.0 31.0 32.0 33.1 34.3 35.6
60 26.4 26.5 26.8 27.2 27.7 28.4 29.1 30.0 31.0 32.1 33.2 34.5 35.7
65 26.5 26.6 26.8 27.2 27.8 28.5 29.2 30.1 31.1 32.2 33.4 34.7 36.0
70 26.6 26.7 27.0 27.4 28.0 28.6 29.4 30.3 31.4 32.5 33.7 34.9 36.3
75 26.9 27.0 27.3 27.7 28.2 28.9 29.7 30.6 31.6 32.8 34.0 35.3 36.7
80 27.2 27.3 27.6 28.0 28.5 29.2 30.0 31.0 32.0 33.1 34.4 35.7 37.1
85 27.6 27.7 27.9 28.3 28.9 29.6 30.4 31.4 32.4 33.6 34.8 36.1 37.6
90 28.0 28.1 28.3 28.8 29.3 30.0 30.8 31.8 32.9 34.0 35.3 36.6 38.1
95 28.4 28.5 28.8 29.2 29.8 30.5 31.3 32.3 33.3 34.5 35.8 37.2 38.6
100 28.9 29.0 29.3 29.7 30.3 31.0 31.8 32.8 33.9 35.1 36.4 37.7 39.2
105 29.4 29.5 29.8 30.2 30.8 31.5 32.3 33.3 34.4 35.6 36.9 38.3 39.8
110 30.0 30.1 30.3 30.8 31.3 32.0 32.9 33.9 35.0 36.2 37.5 38.9 40.4
115 30.5 30.6 30.9 31.3 31.9 32.6 33.5 34.5 35.6 36.8 38.1 39.6 41.1
120 31.1 31.2 31.5 31.9 32.5 33.2 34.1 35.1 36.2 37.4 38.8 40.2 41.7
Table A86: Urban other VOC by speed and gradient (cents/km – July 2015)
(km/h) 0 1 2 3 4 5 6 7 8 9 10 11 12
10 39.0 39.1 39.3 39.6 39.9 40.3 40.8 41.3 41.8 42.3 42.9 43.6 44.2
15 35.1 35.3 35.5 35.8 36.2 36.6 37.1 37.7 38.3 39.0 39.7 40.4 41.2
20 32.3 32.4 32.6 33.0 33.4 33.8 34.4 35.0 35.7 36.4 37.2 38.0 38.9
25 30.2 30.3 30.6 30.9 31.3 31.8 32.4 33.1 33.8 34.6 35.4 36.3 37.2
30 28.7 28.8 29.1 29.4 29.8 30.4 31.0 31.6 32.4 33.2 34.1 35.0 36.0
35 27.7 27.8 28.0 28.3 28.8 29.3 29.9 30.6 31.4 32.3 33.2 34.2 35.2
40 26.9 27.0 27.2 27.6 28.0 28.6 29.2 29.9 30.7 31.6 32.6 33.6 34.6
45 26.4 26.5 26.7 27.1 27.5 28.1 28.7 29.5 30.3 31.2 32.2 33.2 34.3
50 26.1 26.2 26.4 26.7 27.2 27.8 28.4 29.2 30.0 30.9 31.9 33.0 34.1
55 25.9 26.0 26.2 26.6 27.0 27.6 28.3 29.0 29.9 30.8 31.9 33.0 34.1
60 25.9 26.0 26.2 26.6 27.0 27.6 28.3 29.0 29.9 30.9 31.9 33.0 34.2
65 26.0 26.1 26.3 26.6 27.1 27.7 28.4 29.1 30.0 31.0 32.1 33.2 34.4
70 26.2 26.2 26.4 26.8 27.3 27.8 28.5 29.3 30.2 31.2 32.3 33.4 34.7
75 26.4 26.5 26.7 27.0 27.5 28.1 28.8 29.6 30.5 31.5 32.6 33.7 35.0
80 26.7 26.8 27.0 27.3 27.8 28.4 29.1 29.9 30.8 31.8 32.9 34.1 35.4
85 27.1 27.1 27.3 27.7 28.2 28.7 29.5 30.3 31.2 32.2 33.3 34.5 35.8
90 27.5 27.5 27.7 28.1 28.6 29.2 29.9 30.7 31.6 32.7 33.8 35.0 36.3
95 27.9 28.0 28.2 28.5 29.0 29.6 30.3 31.1 32.1 33.1 34.3 35.5 36.8
100 28.4 28.5 28.7 29.0 29.5 30.1 30.8 31.6 32.6 33.6 34.8 36.0 37.3
105 28.9 29.0 29.2 29.5 30.0 30.6 31.3 32.1 33.1 34.2 35.3 36.6 37.9
110 29.5 29.5 29.7 30.0 30.5 31.1 31.8 32.7 33.6 34.7 35.9 37.1 38.5
115 30.0 30.1 30.2 30.6 31.1 31.7 32.4 33.2 34.2 35.3 36.4 37.7 39.1
120 30.6 30.6 30.8 31.1 31.6 32.2 33.0 33.8 34.8 35.9 37.0 38.3 39.7
Table A87: Rural strategic VOC by speed and gradient (cents/km – July 2015)
(km/h) 0 1 2 3 4 5 6 7 8 9 10 11 12
10 45.4 45.7 46.1 46.8 47.6 48.6 49.6 50.7 51.9 53.0 54.2 55.3 56.3
15 41.5 41.8 42.2 42.9 43.7 44.7 45.8 47.0 48.3 49.6 50.9 52.1 53.3
20 38.6 38.8 39.2 39.9 40.8 41.8 42.9 44.2 45.5 46.9 48.3 49.7 51.1
25 36.4 36.6 37.0 37.7 38.6 39.6 40.8 42.1 43.5 45.0 46.5 48.0 49.4
30 34.8 35.0 35.4 36.1 37.0 38.0 39.3 40.6 42.1 43.6 45.2 46.8 48.4
35 33.7 33.8 34.3 34.9 35.8 36.9 38.2 39.6 41.1 42.7 44.3 46.0 47.7
40 32.9 33.0 33.5 34.2 35.1 36.2 37.5 38.9 40.5 42.1 43.8 45.5 47.3
45 32.4 32.5 33.0 33.6 34.6 35.7 37.0 38.5 40.1 41.8 43.5 45.3 47.1
50 32.1 32.2 32.7 33.4 34.3 35.5 36.8 38.3 39.9 41.6 43.4 45.3 47.1
55 32.0 32.1 32.5 33.3 34.2 35.4 36.7 38.3 39.9 41.7 43.5 45.4 47.3
60 32.0 32.1 32.6 33.3 34.3 35.4 36.8 38.4 40.1 41.9 43.7 45.7 47.6
65 32.1 32.3 32.7 33.4 34.4 35.6 37.0 38.6 40.3 42.2 44.1 46.1 48.1
70 32.4 32.5 33.0 33.7 34.7 35.9 37.3 38.9 40.7 42.5 44.5 46.5 48.6
75 32.7 32.9 33.3 34.0 35.0 36.3 37.7 39.3 41.1 43.0 45.0 47.0 49.1
80 33.1 33.3 33.7 34.5 35.5 36.7 38.2 39.8 41.6 43.5 45.5 47.6 49.8
85 33.6 33.7 34.2 34.9 36.0 37.2 38.7 40.3 42.2 44.1 46.2 48.3 50.4
90 34.1 34.3 34.7 35.5 36.5 37.8 39.2 40.9 42.8 44.7 46.8 49.0 51.2
95 34.7 34.8 35.3 36.0 37.1 38.4 39.9 41.5 43.4 45.4 47.5 49.7 51.9
100 35.3 35.4 35.9 36.6 37.7 39.0 40.5 42.2 44.1 46.1 48.2 50.5 52.7
105 35.9 36.1 36.5 37.3 38.3 39.6 41.2 42.9 44.8 46.8 49.0 51.2 53.5
110 36.6 36.7 37.2 38.0 39.0 40.3 41.9 43.6 45.5 47.6 49.8 52.0 54.4
115 37.3 37.4 37.9 38.7 39.7 41.1 42.6 44.4 46.3 48.4 50.6 52.9 55.2
120 38.0 38.1 38.6 39.4 40.5 41.8 43.4 45.1 47.1 49.2 51.4 53.7 56.1
Table A88: Rural other VOC by speed and gradient (cents/km – July 2015)
(km/h) 0 1 2 3 4 5 6 7 8 9 10 11 12
10 43.5 43.8 44.2 44.7 45.4 46.2 47.0 48.0 48.9 49.9 50.9 51.8 52.7
15 39.7 39.9 40.3 40.8 41.5 42.4 43.3 44.3 45.3 46.4 47.6 48.7 49.8
20 36.8 36.9 37.3 37.9 38.6 39.4 40.4 41.5 42.6 43.8 45.0 46.2 47.5
25 34.6 34.8 35.1 35.7 36.4 37.3 38.3 39.4 40.6 41.9 43.2 44.5 45.8
30 33.0 33.2 33.6 34.1 34.9 35.8 36.8 38.0 39.2 40.5 41.9 43.3 44.7
35 31.9 32.1 32.4 33.0 33.8 34.7 35.8 36.9 38.2 39.6 41.0 42.5 43.9
40 31.1 31.3 31.6 32.2 33.0 33.9 35.0 36.3 37.6 39.0 40.5 42.0 43.5
45 30.6 30.7 31.1 31.7 32.5 33.5 34.6 35.8 37.2 38.6 40.1 41.7 43.3
50 30.3 30.4 30.8 31.4 32.2 33.2 34.3 35.6 37.0 38.5 40.0 41.6 43.3
55 30.2 30.3 30.7 31.3 32.1 33.1 34.2 35.5 36.9 38.5 40.1 41.7 43.4
60 30.2 30.3 30.7 31.3 32.1 33.1 34.2 35.5 37.0 38.6 40.2 41.9 43.6
65 30.3 30.4 30.8 31.4 32.2 33.3 34.4 35.8 37.3 38.8 40.5 42.2 44.0
70 30.6 30.7 31.0 31.7 32.5 33.5 34.7 36.1 37.6 39.2 40.8 42.6 44.4
75 30.9 31.0 31.3 32.0 32.8 33.8 35.1 36.4 37.9 39.6 41.3 43.1 44.9
80 31.2 31.4 31.7 32.3 33.2 34.2 35.5 36.9 38.4 40.0 41.8 43.6 45.5
85 31.7 31.8 32.2 32.8 33.6 34.7 35.9 37.3 38.9 40.5 42.3 44.2 46.1
90 32.2 32.3 32.6 33.3 34.1 35.2 36.4 37.9 39.4 41.1 42.9 44.8 46.7
95 32.7 32.8 33.2 33.8 34.7 35.7 37.0 38.4 40.0 41.7 43.5 45.4 47.4
100 33.3 33.4 33.7 34.4 35.2 36.3 37.6 39.0 40.6 42.4 44.2 46.1 48.1
105 33.9 34.0 34.3 35.0 35.8 36.9 38.2 39.7 41.3 43.0 44.9 46.8 48.8
110 34.5 34.6 35.0 35.6 36.5 37.6 38.9 40.3 42.0 43.7 45.6 47.6 49.6
115 35.1 35.2 35.6 36.3 37.1 38.2 39.6 41.0 42.7 44.5 46.3 48.3 50.4
120 35.8 35.9 36.3 36.9 37.8 38.9 40.3 41.7 43.4 45.2 47.1 49.1 51.2
Back to 3.4 Impact on greenhouse gas emissions: Assessment of CO2 emissions >>
Back to 3.6 Impact on network productivity and utilisation: Vehicle operating cost >>
Table A89: Urban additional VOC due to roughness by vehicle class (cents/km – July 2015)
Roughness Additional VOC in cents/km by vehicle class
IRI (m/km) NAASRA Passenger LCV MCV HCVI HCVII Bus
(count/km) car
0–2.5 0–66 0.0 0.0 0.0 0.0. 0.0 0.0
3.0 79 0.3 0.3 0.7 0.9 1.4 0.8
3.5 92 0.8 0.7 2.3 3.1 4.4 2.8
4.0 106 1.6 1.4 4.7 6.3 8.5 5.7
4.5 119 2.7 2.6 7.6 10.2 13.6 9.1
5.0 132 4.0 4.2 10.9 14.7 19.3 12.9
5.5 145 5.5 6.0 14.5 19.6 25.4 16.9
6.0 158 7.1 8.1 18.3 24.7 31.9 21.2
6.5 172 8.8 10.3 22.2 30.0 38.6 25.5
7.0 185 10.5 12.6 26.1 35.4 45.3 29.8
7.5 198 12.3 14.9 30.1 40.8 52.1 34.1
8.0 211 14.1 17.2 34.0 46.2 58.8 38.5
8.5 224 15.8 19.5 37.9 51.6 65.5 42.7
9.0 238 17.5 21.7 41.8 56.8 72.2 46.9
9.5 251 19.2 23.9 45.6 62.0 78.7 51.1
10.0 264 20.8 26.0 49.2 67.1 85.0 55.2
10.5 277 22.4 28.0 52.8 72.0 91.3 59.2
11.0 290 23.3 29.0 55.0 74.9 95.4 61.7
11.5 304 24.1 29.8 57.1 77.5 99.3 64.2
12.0 317 24.9 30.6 59.1 80.1 103.1 66.6
12.5 330 25.6 31.5 61.1 82.7 107.0 69.0
13.0 343 26.4 32.3 63.1 85.3 110.8 71.4
13.5 356 27.2 33.1 65.2 87.9 114.6 73.8
14.0 370 28.0 33.9 67.2 90.6 118.5 76.2
14.5 383 28.7 34.7 69.2 93.2 122.3 78.6
15.0 396 29.5 35.5 71.2 95.8 126.1 81.0
Table A90: Rural additional VOC due to roughness by vehicle class (cents/km – July 2015)
Roughness Additional VOC in cents/km by vehicle class
IRI (m/km) NAASRA Passenger LCV MCV HCVI HCVII Bus
(count/km) car
0–2.5 0–66 0.0 0.0 0.0 0.0 0.0 0.0
3.0 79 0.2 0.2 0.5 0.7 1.1 0.6
3.5 92 0.9 0.9 2.5 3.5 4.8 3.1
4.0 106 2.4 2.7 6.4 8.8 11.1 7.4
4.5 119 4.4 5.3 11.2 15.5 18.9 12.7
5.0 132 6.6 8.2 16.3 22.7 27.3 18.2
5.5 145 8.9 11.1 21.3 29.8 35.6 23.7
6.0 158 11.0 14.0 26.2 36.6 43.7 29.0
6.5 172 13.2 16.9 31.0 43.2 51.7 34.2
7.0 185 15.3 19.8 35.7 49.8 59.7 39.4
7.5 198 17.6 22.8 40.6 56.6 67.9 44.7
8.0 211 18.5 23.8 42.8 59.2 72.0 47.2
8.5 224 19.3 24.6 44.8 61.8 75.9 49.6
9.0 238 20.0 25.5 46.8 64.4 79.7 52.0
9.5 251 20.8 26.3 48.9 67.0 83.6 54.4
10.0 264 21.6 27.2 50.9 69.7 87.5 56.8
10.5 277 22.4 28.0 52.9 72.3 91.4 59.3
11.0 290 23.2 28.9 55.0 74.9 95.3 61.7
11.5 304 24.0 29.7 57.0 77.5 99.1 64.1
12.0 317 24.8 30.6 59.1 80.1 103.0 66.5
12.5 330 25.6 31.4 61.1 82.7 106.9 69.0
13.0 343 26.4 32.3 63.1 85.3 110.8 71.4
13.5 356 27.2 33.1 65.2 87.9 114.7 73.8
14.0 370 28.0 34.0 67.2 90.6 118.5 76.2
14.5 383 28.8 34.8 69.2 93.2 122.4 78.6
15.0 396 29.6 35.7 71.3 95.8 126.3 81.1
Table A91: Additional VOC due to roughness by road category (cents/km – July 2015)
Roughness Additional VOC in cents/km by road category
IRI (m/km) NAASRA Urban Rural strategic Rural others
(count/km)
0–2.5 0–66 0.0 0.0 0.0
3.0 79 0.3 0.3 0.3
3.5 92 0.9 1.2 1.2
4.0 106 1.8 3.2 3.1
4.5 119 3.0 5.8 5.6
5.0 132 4.6 8.6 8.3
5.5 145 6.3 11.5 11.1
6.0 158 8.1 14.3 13.8
6.5 172 10.0 17.0 16.4
7.0 185 12.0 19.7 19.1
7.5 198 14.0 22.6 21.8
8.0 211 16.0 23.7 22.9
8.5 224 18.0 24.8 23.9
9.0 238 19.9 25.8 24.9
9.5 251 21.8 26.9 25.9
10.0 264 23.7 27.9 26.9
10.5 277 25.4 29.0 27.9
11.0 290 26.5 30.0 28.9
11.5 304 27.3 31.1 29.9
12.0 317 28.2 32.1 30.9
12.5 330 29.1 33.1 31.9
13.0 343 30.0 34.2 32.9
13.5 356 30.9 35.2 33.9
14.0 370 31.8 36.3 34.9
14.5 383 32.7 37.3 35.9
15.0 396 33.5 38.4 36.9
Table A92: Urban arterial and urban other – additional VOC due to congestion by vehicle class
(cents/km – July 2015)
VC ratio Additional VOC in cents/km by vehicle class
Passenger LCV MCV HCVI HCVII Bus
car
0.00 0.0 0.0 0.0 0.0 0.0 0.0
0.05 0.0 0.0 0.0 0.2 0.5 0.1
0.10 0.0 0.0 0.1 0.3 1.0 0.2
0.15 0.0 0.1 0.2 0.6 1.7 0.3
0.20 0.0 0.1 0.3 0.8 2.5 0.4
0.25 0.1 0.2 0.4 1.2 3.5 0.6
0.30 0.1 0.2 0.5 1.6 4.7 0.9
0.35 0.1 0.3 0.7 2.1 6.1 1.1
0.40 0.2 0.4 0.9 2.7 7.8 1.5
0.45 0.2 0.5 1.1 3.4 9.9 1.9
0.50 0.3 0.7 1.4 4.2 12.3 2.4
0.55 0.4 0.9 1.7 5.3 15.3 3.0
0.60 0.5 1.2 2.2 6.5 18.8 3.8
0.65 0.7 1.5 2.7 8.0 23.0 4.7
0.70 1.0 1.9 3.3 9.9 28.1 5.8
0.75 1.3 2.5 4.1 12.1 34.2 7.3
0.80 1.7 3.2 5.0 14.8 41.5 9.0
0.85 2.2 4.1 6.2 18.0 50.3 11.1
0.90 2.9 5.2 7.6 21.9 60.8 13.7
0.95 3.8 6.6 9.0 26.0 69.5 16.8
1.00 4.4 7.1 9.0 26.0 69.5 16.8
1.05 4.4 7.1 9.0 26.0 69.5 16.8
1.10 4.4 7.1 9.0 26.0 69.5 16.8
1.15 4.4 7.1 9.0 26.0 69.5 16.8
1.20 4.4 7.1 9.0 26.0 69.5 16.8
Table A93: Rural strategic and rural other – additional VOC due to congestion by vehicle class
(cents/km – July 2015)
car
0.00 0.0 0.0 0.0 0.0 0.0 0.0
0.05 0.0 0.0 0.1 0.3 1.0 0.1
0.10 0.0 0.0 0.2 0.5 2.2 0.1
0.15 0.0 0.0 0.3 0.9 3.5 0.2
0.20 0.0 0.0 0.4 1.2 4.9 0.3
0.25 0.0 0.0 0.5 1.7 6.6 0.5
0.30 0.0 0.0 0.7 2.2 8.4 0.6
0.35 0.0 0.0 0.9 2.7 10.5 0.8
0.40 0.0 0.0 1.1 3.4 12.9 1.0
0.45 0.0 0.0 1.3 4.1 15.5 1.3
0.50 0.0 0.0 1.6 4.9 18.5 1.7
0.55 0.0 0.0 1.9 5.9 21.9 2.1
0.60 0.1 0.0 2.2 7.0 25.7 2.6
0.65 0.1 0.0 2.6 8.2 30.0 3.2
0.70 0.2 0.1 3.1 9.6 34.8 3.9
0.75 0.3 0.2 3.6 11.2 40.2 4.8
0.80 0.5 0.4 4.2 13.1 46.4 5.8
0.85 0.8 0.7 4.9 15.2 53.3 7.1
0.90 1.3 1.5 5.7 17.6 61.1 8.6
0.95 2.2 3.0 6.7 20.4 69.8 10.5
1.00 3.7 6.0 7.5 23.5 70.6 12.7
1.05 3.9 6.0 7.5 24.2 70.6 13.1
1.10 3.9 6.0 7.5 24.2 70.6 13.1
1.15 3.9 6.0 7.5 24.2 70.6 13.1
1.20 3.9 6.0 7.5 24.2 70.6 13.1
Table A94: Motorway – additional VOC due to congestion by vehicle class (cents/km – July 2015)
car
0.00 0.0 0.0 0.0 0.0 0.0 0.0
0.05 0.0 0.0 0.0 0.0 0.1 0.0
0.10 0.0 0.0 0.0 0.0 0.2 0.0
0.15 0.0 0.0 0.0 0.0 0.3 0.0
0.20 0.0 0.0 0.0 0.1 0.4 0.0
0.25 0.0 0.0 0.0 0.1 0.6 0.0
0.30 0.0 0.0 0.1 0.2 0.9 0.0
0.35 0.0 0.0 0.1 0.2 1.4 0.0
0.40 0.0 0.0 0.1 0.3 1.9 0.0
0.45 0.0 0.0 0.2 0.5 2.7 0.1
0.50 0.0 0.0 0.2 0.7 3.7 0.1
0.55 0.0 0.0 0.3 1.0 5.1 0.2
0.60 0.0 0.0 0.5 1.4 7.1 0.3
0.65 0.0 0.0 0.7 2.1 9.7 0.5
0.70 0.0 0.0 1.0 3.0 13.3 0.8
0.75 0.0 0.0 1.4 4.2 18.2 1.2
0.80 0.0 0.0 2.1 6.0 24.9 2.0
0.85 0.1 0.1 2.9 8.6 34.0 3.1
0.90 0.3 0.3 4.2 12.3 46.4 4.9
0.95 0.9 1.1 6.0 17.6 63.4 7.7
1.00 3.1 4.8 7.2 23.1 70.0 12.1
1.05 3.3 5.0 7.2 23.1 70.0 12.1
1.10 3.3 5.0 7.2 23.1 70.0 12.1
1.15 3.3 5.0 7.2 23.1 70.0 12.1
1.20 3.3 5.0 7.2 23.1 70.0 12.1
Table A95: Additional VOC due to congestion by road category (cents/km – July 2015)
VC ratio Additional cost in cents/km
Urban Rural two-lane Rural two-lane Motorway
strategic other
0.00 0.0 0.0 0.0 0.0
0.05 0.0 0.1 0.0 0.0
0.10 0.1 0.1 0.1 0.0
0.15 0.1 0.2 0.1 0.0
0.20 0.2 0.3 0.2 0.0
0.25 0.3 0.4 0.3 0.0
0.30 0.4 0.5 0.4 0.0
0.35 0.5 0.6 0.5 0.0
0.40 0.7 0.8 0.6 0.1
0.45 0.9 0.9 0.8 0.1
0.50 1.1 1.2 0.9 0.1
0.55 1.5 1.4 1.2 0.2
0.60 1.8 1.7 1.4 0.3
0.65 2.3 2.1 1.7 0.4
0.70 2.9 2.5 2.1 0.7
0.75 3.7 2.9 2.5 1.0
0.80 4.6 3.5 3.0 1.4
0.85 5.7 4.2 3.6 2.1
0.90 7.2 4.9 4.3 3.2
0.95 8.9 5.8 5.1 4.7
1.00 9.2 6.9 6.1 6.9
1.05 9.2 7.7 7.0 7.1
1.10 9.2 7.7 7.0 7.1
1.15 9.2 7.7 7.0 7.1
1.20 9.2 7.7 7.0 7.1
Table A96: Passenger car additional travel time due to speed change cycles (seconds/speed cycle)
Initial Additional travel time in seconds/speed cycle by final speed
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 2.2
10 4.1 1.1
15 5.8 2.8 0.8
20 7.4 4.4 2.1 0.6
25 8.9 6.0 3.6 1.7 0.5
30 10.4 7.5 5.1 3.0 1.5 0.4
35 11.8 9.0 6.5 4.4 2.6 1.3 0.4
40 13.1 10.4 8.0 5.8 3.9 2.3 1.1 0.3
45 13.7 11.4 9.2 7.2 5.2 3.5 2.1 1.0 0.3
50 14.3 12.1 10.0 8.1 6.3 4.7 3.2 1.9 0.9 0.3
55 14.9 12.8 10.8 8.9 7.2 5.6 4.2 2.9 1.8 0.9 0.2
60 15.4 13.4 11.5 9.7 8.1 6.5 5.1 3.8 2.6 1.7 0.8 0.2
65 15.9 14.0 12.2 10.5 8.9 7.4 5.9 4.6 3.5 2.4 1.5 0.8 0.2
70 16.4 14.6 12.9 11.2 9.6 8.2 6.8 5.5 4.3 3.2 2.2 1.4 0.7 0.2
75 16.9 15.2 13.5 11.9 10.4 8.9 7.5 6.2 5.0 3.9 2.9 2.0 1.3 0.7 0.2
80 17.4 15.7 14.1 12.5 11.1 9.6 8.3 7.0 5.8 4.7 3.7 2.7 1.9 1.2 0.6 0.2
85 17.8 16.2 14.7 13.2 11.7 10.3 9.0 7.7 6.6 5.4 4.4 3.4 2.5 1.8 1.1 0.6 0.2
90 18.3 16.7 15.2 13.8 12.4 11.0 9.7 8.5 7.3 6.2 5.1 4.1 3.2 2.4 1.7 1.0 0.5 0.2
95 18.8 17.2 15.8 14.4 13.0 11.7 10.4 9.1 8.0 6.9 5.8 4.8 3.9 3.0 2.3 1.6 1.0 0.5 0.2
100 19.2 17.7 16.3 14.9 13.6 12.3 11.0 9.8 8.7 7.5 6.5 5.5 4.6 3.7 2.9 2.1 1.5 0.9 0.5 0.2
105 19.6 18.2 16.8 15.5 14.2 12.9 11.7 10.5 9.3 8.2 7.2 6.2 5.2 4.3 3.5 2.7 2.0 1.4 0.9 0.5 0.1
110 20.1 18.7 17.3 16.0 14.7 13.5 12.3 11.1 10.0 8.9 7.8 6.8 5.9 5.0 4.1 3.3 2.6 1.9 1.3 0.8 0.4 0.1
115 20.5 19.1 17.8 16.5 15.3 14.0 12.9 11.7 10.6 9.5 8.5 7.5 6.5 5.6 4.7 3.9 3.2 2.5 1.8 1.3 0.8 0.4 0.1
120 20.9 19.6 18.3 17.0 15.8 14.6 13.4 12.3 11.2 10.1 9.1 8.1 7.1 6.2 5.4 4.5 3.8 3.0 2.4 1.8 1.2 0.8 0.4 0.1
Table A97: Passenger car additional VOC due to speed change cycles (cents/speed cycle – July 2015)
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 0.1
10 0.1 0.0
15 0.2 0.1 0.0
20 0.3 0.2 0.1 0.0
25 0.4 0.3 0.2 0.1 0.1
30 0.5 0.4 0.3 0.2 0.1 0.1
35 0.6 0.5 0.4 0.3 0.2 0.1 0.1
40 0.7 0.6 0.6 0.5 0.4 0.2 0.1 0.1
45 0.9 0.8 0.7 0.6 0.5 0.4 0.2 0.1 0.1
50 1.0 0.9 0.9 0.8 0.7 0.5 0.4 0.3 0.1 0.1
55 1.2 1.1 1.0 0.9 0.8 0.7 0.5 0.4 0.3 0.1 0.1
60 1.3 1.3 1.2 1.1 1.0 0.8 0.7 0.5 0.4 0.3 0.1 0.1
65 1.5 1.5 1.4 1.3 1.2 1.0 0.9 0.7 0.6 0.4 0.3 0.1 0.1
70 1.7 1.6 1.6 1.5 1.4 1.2 1.0 0.9 0.7 0.6 0.4 0.3 0.1 0.1
75 1.9 1.8 1.8 1.7 1.6 1.4 1.2 1.1 0.9 0.7 0.6 0.4 0.3 0.1 0.1
80 2.1 2.0 2.0 1.9 1.7 1.6 1.4 1.3 1.1 0.9 0.8 0.6 0.4 0.3 0.1 0.1
85 2.3 2.3 2.2 2.1 1.9 1.8 1.6 1.5 1.3 1.1 0.9 0.8 0.6 0.5 0.3 0.1 0.0
90 2.5 2.5 2.4 2.3 2.2 2.0 1.8 1.6 1.5 1.3 1.1 1.0 0.8 0.6 0.5 0.3 0.1 0.0
95 2.8 2.7 2.6 2.5 2.4 2.2 2.0 1.8 1.7 1.5 1.3 1.2 1.0 0.8 0.6 0.5 0.3 0.1 0.0
100 3.0 2.9 2.8 2.7 2.6 2.4 2.2 2.0 1.9 1.7 1.5 1.3 1.2 1.0 0.8 0.6 0.5 0.3 0.1 0.0
105 3.2 3.1 3.0 2.9 2.8 2.6 2.4 2.2 2.1 1.9 1.7 1.5 1.3 1.2 1.0 0.8 0.6 0.5 0.3 0.1 0.0
110 3.5 3.4 3.3 3.1 3.0 2.8 2.6 2.4 2.3 2.1 1.9 1.7 1.5 1.3 1.2 1.0 0.8 0.6 0.4 0.3 0.1 0.0
115 3.7 3.6 3.5 3.3 3.2 3.0 2.8 2.6 2.4 2.3 2.1 1.9 1.7 1.5 1.3 1.2 1.0 0.8 0.6 0.4 0.3 0.1 0.0
120 3.9 3.8 3.7 3.6 3.4 3.2 3.0 2.8 2.6 2.4 2.2 2.0 1.9 1.7 1.5 1.3 1.1 1.0 0.8 0.6 0.4 0.3 0.1 0.0
Table A98: LCV additional travel time due to speed change cycles (seconds/speed cycle)
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 2.4
10 4.4 1.2
15 6.2 3.0 0.8
20 8.0 4.8 2.3 0.6
25 9.6 6.5 3.9 1.8 0.5
30 11.1 8.1 5.4 3.3 1.6 0.4
35 12.6 9.7 7.0 4.7 2.8 1.4 0.4
40 14.1 11.2 8.6 6.2 4.2 2.5 1.2 0.3
45 14.8 12.2 9.9 7.7 5.6 3.8 2.3 1.1 0.3
50 15.4 13.0 10.8 8.7 6.8 5.1 3.5 2.1 1.0 0.3
55 16.0 13.8 11.6 9.6 7.8 6.1 4.5 3.1 1.9 0.9 0.3
60 16.6 14.5 12.5 10.5 8.7 7.0 5.5 4.1 2.8 1.8 0.9 0.2
65 17.2 15.2 13.2 11.4 9.6 8.0 6.4 5.0 3.7 2.6 1.6 0.8 0.2
70 17.8 15.9 14.0 12.2 10.5 8.8 7.3 5.9 4.6 3.4 2.4 1.5 0.8 0.2
75 18.4 16.5 14.7 12.9 11.3 9.7 8.2 6.8 5.5 4.3 3.2 2.2 1.4 0.7 0.2
80 18.9 17.1 15.4 13.7 12.0 10.5 9.0 7.6 6.3 5.1 4.0 3.0 2.1 1.3 0.7 0.2
85 19.5 17.7 16.0 14.4 12.8 11.3 9.8 8.4 7.1 5.9 4.8 3.7 2.8 1.9 1.2 0.6 0.2
90 20.0 18.3 16.7 15.1 13.5 12.0 10.6 9.2 7.9 6.7 5.6 4.5 3.5 2.6 1.8 1.1 0.6 0.2
95 20.5 18.9 17.3 15.7 14.2 12.7 11.3 10.0 8.7 7.5 6.3 5.2 4.2 3.3 2.5 1.7 1.1 0.5 0.2
100 21.0 19.4 17.9 16.3 14.9 13.4 12.1 10.7 9.5 8.3 7.1 6.0 5.0 4.0 3.1 2.3 1.6 1.0 0.5 0.2
105 21.5 20.0 18.4 17.0 15.5 14.1 12.8 11.5 10.2 9.0 7.8 6.7 5.7 4.7 3.8 3.0 2.2 1.5 1.0 0.5 0.2
110 22.0 20.5 19.0 17.6 16.2 14.8 13.5 12.2 10.9 9.7 8.6 7.5 6.4 5.4 4.5 3.6 2.8 2.1 1.5 0.9 0.5 0.1
115 22.5 21.0 19.6 18.2 16.8 15.4 14.1 12.9 11.6 10.4 9.3 8.2 7.1 6.1 5.2 4.3 3.5 2.7 2.0 1.4 0.9 0.4 0.1
120 23.0 21.5 20.1 18.7 17.4 16.1 14.8 13.5 12.3 11.1 10.0 8.9 7.9 6.8 5.9 5.0 4.1 3.3 2.6 1.9 1.3 0.8 0.4 0.1
Table A99: LCV additional VOC due to speed change cycles (cents/speed cycle – July 2015)
Initial Additional VOC in cents/speed cycle by final speed
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 0.1
10 0.2 0.1
15 0.3 0.2 0.1
20 0.4 0.3 0.1 0.1
25 0.6 0.5 0.3 0.2 0.1
30 0.8 0.6 0.5 0.4 0.2 0.1
35 1.0 0.9 0.7 0.6 0.4 0.2 0.1
40 1.2 1.1 0.9 0.8 0.6 0.4 0.3 0.1
45 1.4 1.3 1.2 1.0 0.8 0.6 0.4 0.3 0.1
50 1.7 1.6 1.4 1.3 1.1 0.9 0.6 0.5 0.3 0.1
55 1.9 1.8 1.7 1.6 1.4 1.1 0.9 0.7 0.5 0.3 0.1
60 2.2 2.1 2.0 1.8 1.7 1.4 1.2 0.9 0.7 0.5 0.3 0.1
65 2.5 2.4 2.3 2.1 1.9 1.7 1.5 1.2 1.0 0.7 0.5 0.3 0.1
70 2.8 2.7 2.6 2.4 2.3 2.0 1.8 1.5 1.3 1.0 0.8 0.5 0.3 0.1
75 3.2 3.1 2.9 2.8 2.6 2.3 2.1 1.8 1.6 1.3 1.0 0.8 0.5 0.3 0.1
80 3.5 3.4 3.3 3.1 2.9 2.6 2.4 2.1 1.9 1.6 1.3 1.1 0.8 0.5 0.3 0.1
85 3.8 3.7 3.6 3.4 3.2 3.0 2.7 2.4 2.2 1.9 1.6 1.4 1.1 0.8 0.5 0.3 0.1
90 4.2 4.1 3.9 3.7 3.5 3.3 3.0 2.7 2.5 2.2 1.9 1.7 1.4 1.1 0.8 0.6 0.3 0.1
95 4.6 4.4 4.3 4.1 3.9 3.6 3.3 3.0 2.8 2.5 2.2 1.9 1.7 1.4 1.1 0.8 0.6 0.3 0.1
100 4.9 4.8 4.6 4.4 4.2 3.9 3.6 3.4 3.1 2.8 2.5 2.2 2.0 1.7 1.4 1.1 0.8 0.5 0.3 0.1
105 5.3 5.1 4.9 4.7 4.5 4.2 3.9 3.6 3.4 3.1 2.8 2.5 2.2 1.9 1.7 1.4 1.1 0.8 0.5 0.3 0.1
110 5.6 5.5 5.3 5.1 4.8 4.5 4.2 3.9 3.6 3.3 3.1 2.8 2.5 2.2 1.9 1.6 1.4 1.1 0.8 0.5 0.3 0.1
115 6.0 5.8 5.6 5.4 5.1 4.8 4.5 4.2 3.9 3.6 3.3 3.0 2.7 2.4 2.2 1.9 1.6 1.3 1.1 0.8 0.5 0.2 0.1
120 6.3 6.1 5.9 5.7 5.4 5.1 4.7 4.4 4.1 3.8 3.5 3.2 2.9 2.7 2.4 2.1 1.8 1.6 1.3 1.0 0.7 0.5 0.2 0.1
Table A100: MCV additional travel time due to speed change cycles (seconds/speed cycle)
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 2.5
10 4.6 1.3
15 6.5 3.1 0.9
20 8.3 5.0 2.4 0.7
25 10.0 6.8 4.0 1.9 0.5
30 11.6 8.5 5.7 3.4 1.6 0.5
35 13.2 10.1 7.3 5.0 3.0 1.4 0.4
40 14.7 11.7 9.0 6.5 4.4 2.6 1.3 0.4
45 15.4 12.8 10.3 8.1 5.9 4.0 2.4 1.1 0.3
50 16.1 13.6 11.3 9.1 7.1 5.3 3.6 2.2 1.0 0.3
55 16.8 14.4 12.2 10.1 8.1 6.4 4.7 3.3 2.0 1.0 0.3
60 17.4 15.2 13.0 11.0 9.1 7.4 5.7 4.3 3.0 1.9 0.9 0.3
65 18.0 15.9 13.8 11.9 10.1 8.3 6.7 5.2 3.9 2.7 1.7 0.8 0.2
70 18.6 16.6 14.6 12.7 10.9 9.2 7.7 6.2 4.8 3.6 2.5 1.6 0.8 0.2
75 19.2 17.3 15.4 13.5 11.8 10.1 8.6 7.1 5.7 4.5 3.3 2.3 1.4 0.7 0.2
80 19.8 17.9 16.1 14.3 12.6 11.0 9.4 8.0 6.6 5.3 4.2 3.1 2.1 1.3 0.7 0.2
85 20.4 18.5 16.7 15.0 13.4 11.8 10.3 8.8 7.5 6.2 5.0 3.9 2.9 2.0 1.3 0.6 0.2
90 20.9 19.1 17.4 15.7 14.1 12.6 11.1 9.7 8.3 7.0 5.8 4.7 3.7 2.7 1.9 1.2 0.6 0.2
95 21.5 19.7 18.1 16.4 14.9 13.3 11.9 10.5 9.1 7.8 6.6 5.5 4.4 3.5 2.6 1.8 1.1 0.6 0.2
100 22.0 20.3 18.7 17.1 15.6 14.1 12.6 11.2 9.9 8.6 7.4 6.3 5.2 4.2 3.3 2.4 1.7 1.1 0.5 0.2
105 22.5 20.9 19.3 17.7 16.2 14.8 13.4 12.0 10.7 9.4 8.2 7.1 6.0 4.9 4.0 3.1 2.3 1.6 1.0 0.5 0.2
110 23.0 21.4 19.9 18.4 16.9 15.5 14.1 12.7 11.4 10.2 9.0 7.8 6.7 5.7 4.7 3.8 3.0 2.2 1.5 1.0 0.5 0.2
115 23.5 22.0 20.5 19.0 17.5 16.1 14.8 13.4 12.2 10.9 9.7 8.6 7.5 6.4 5.4 4.5 3.6 2.8 2.1 1.5 0.9 0.5 0.1
120 24.0 22.5 21.0 19.6 18.2 16.8 15.4 14.1 12.9 11.6 10.5 9.3 8.2 7.2 6.2 5.2 4.3 3.5 2.7 2.0 1.4 0.9 0.4 0.1
Table A101: MCV additional VOC due to speed change cycles (cents/speed cycle – July 2015)
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 0.2
10 0.4 0.2
15 0.7 0.4 0.2
20 1.0 0.8 0.5 0.3
25 1.5 1.3 1.0 0.7 0.3
30 2.0 1.8 1.5 1.2 0.8 0.4
35 2.7 2.4 2.2 1.8 1.4 1.0 0.5
40 3.4 3.2 2.9 2.5 2.1 1.6 1.1 0.5
45 4.2 4.0 3.7 3.3 2.9 2.3 1.8 1.2 0.6
50 5.2 4.9 4.6 4.2 3.8 3.2 2.6 2.0 1.3 0.7
55 6.2 6.0 5.7 5.3 4.8 4.2 3.5 2.8 2.1 1.4 0.7
60 7.4 7.1 6.8 6.4 5.9 5.3 4.6 3.9 3.1 2.3 1.5 0.7
65 8.6 8.4 8.0 7.6 7.1 6.5 5.8 5.1 4.3 3.4 2.5 1.6 0.8
70 9.9 9.7 9.4 8.9 8.4 7.8 7.1 6.3 5.5 4.6 3.7 2.7 1.7 0.8
75 11.4 11.1 10.8 10.3 9.8 9.2 8.5 7.7 6.9 6.0 5.0 4.0 2.9 1.8 0.8
80 12.9 12.7 12.3 11.8 11.3 10.6 9.9 9.1 8.3 7.4 6.4 5.3 4.2 3.1 1.9 0.8
85 14.6 14.3 13.9 13.4 12.9 12.2 11.5 10.7 9.8 8.9 7.9 6.8 5.7 4.5 3.3 2.0 0.8
90 16.3 16.0 15.6 15.1 14.5 13.9 13.1 12.3 11.4 10.5 9.5 8.4 7.2 6.0 4.8 3.4 2.1 0.8
95 18.2 17.8 17.4 16.9 16.3 15.6 14.8 14.0 13.1 12.1 11.1 10.0 8.9 7.7 6.4 5.0 3.6 2.2 0.8
100 20.1 19.7 19.3 18.8 18.1 17.4 16.6 15.8 14.9 13.9 12.9 11.8 10.6 9.3 8.0 6.7 5.3 3.8 2.3 0.9
105 22.2 21.8 21.3 20.7 20.1 19.3 18.5 17.6 16.7 15.7 14.7 13.5 12.4 11.1 9.8 8.4 7.0 5.5 3.9 2.4 0.9
110 24.3 23.9 23.4 22.8 22.1 21.3 20.5 19.6 18.6 17.6 16.6 15.4 14.2 13.0 11.6 10.3 8.8 7.3 5.7 4.1 2.5 0.9
115 26.5 26.1 25.5 24.9 24.2 23.4 22.5 21.6 20.6 19.6 18.5 17.4 16.1 14.9 13.5 12.2 10.7 9.2 7.6 5.9 4.2 2.5 0.9
120 28.9 28.4 27.8 27.1 26.4 25.5 24.6 23.7 22.7 21.6 20.5 19.4 18.1 16.8 15.5 14.1 12.6 11.1 9.5 7.9 6.1 4.4 2.6 0.9
Table A102: HCVI additional travel time due to speed change cycles (seconds/speed cycle)
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 2.9
10 5.3 1.5
15 7.6 3.6 1.0
20 9.6 5.7 2.8 0.8
25 11.6 7.8 4.7 2.2 0.6
30 13.5 9.8 6.6 3.9 1.9 0.5
35 15.3 11.7 8.5 5.7 3.4 1.6 0.5
40 17.0 13.5 10.3 7.5 5.1 3.0 1.5 0.4
45 17.8 14.8 11.9 9.3 6.8 4.6 2.7 1.3 0.4
50 18.6 15.8 13.0 10.5 8.2 6.1 4.2 2.5 1.2 0.3
55 19.4 16.7 14.1 11.7 9.4 7.3 5.5 3.8 2.3 1.1 0.3
60 20.1 17.5 15.1 12.7 10.6 8.5 6.6 4.9 3.4 2.1 1.0 0.3
65 20.9 18.4 16.0 13.8 11.6 9.6 7.8 6.1 4.5 3.1 2.0 1.0 0.3
70 21.6 19.2 16.9 14.7 12.7 10.7 8.9 7.1 5.6 4.1 2.9 1.8 0.9 0.3
75 22.3 20.0 17.8 15.7 13.6 11.7 9.9 8.2 6.6 5.2 3.8 2.7 1.7 0.9 0.2
80 22.9 20.7 18.6 16.5 14.6 12.7 10.9 9.2 7.6 6.2 4.8 3.6 2.5 1.5 0.8 0.2
85 23.6 21.5 19.4 17.4 15.5 13.6 11.9 10.2 8.6 7.2 5.8 4.5 3.3 2.3 1.4 0.7 0.2
90 24.2 22.2 20.2 18.2 16.4 14.6 12.8 11.2 9.6 8.1 6.7 5.4 4.2 3.1 2.2 1.4 0.7 0.2
95 24.9 22.9 20.9 19.0 17.2 15.4 13.7 12.1 10.6 9.1 7.7 6.4 5.1 4.0 3.0 2.1 1.3 0.7 0.2
100 25.5 23.5 21.6 19.8 18.0 16.3 14.6 13.0 11.5 10.0 8.6 7.3 6.0 4.9 3.8 2.8 2.0 1.2 0.6 0.2
105 26.1 24.2 22.4 20.6 18.8 17.1 15.5 13.9 12.4 10.9 9.5 8.2 6.9 5.7 4.6 3.6 2.7 1.9 1.2 0.6 0.2
110 26.7 24.8 23.0 21.3 19.6 17.9 16.3 14.7 13.2 11.8 10.4 9.1 7.8 6.6 5.5 4.4 3.4 2.6 1.8 1.1 0.6 0.2
115 27.3 25.5 23.7 22.0 20.3 18.7 17.1 15.6 14.1 12.7 11.3 9.9 8.7 7.4 6.3 5.2 4.2 3.3 2.4 1.7 1.1 0.5 0.2
120 27.8 26.1 24.4 22.7 21.1 19.5 17.9 16.4 14.9 13.5 12.1 10.8 9.5 8.3 7.1 6.0 5.0 4.0 3.2 2.3 1.6 1.0 0.5 0.2
Table A103: HCVI additional VOC due to speed change cycles (cents/speed cycle – July 2015)
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 0.3
10 0.6 0.3
15 1.1 0.8 0.4
20 1.7 1.4 0.9 0.5
25 2.7 2.3 1.8 1.3 0.6
30 3.8 3.4 3.0 2.3 1.6 0.8
35 5.1 4.7 4.2 3.6 2.8 1.9 0.9
40 6.5 6.2 5.7 5.0 4.2 3.2 2.1 1.0
45 8.3 7.9 7.3 6.6 5.7 4.7 3.5 2.4 1.2
50 10.2 9.8 9.3 8.5 7.5 6.3 5.1 3.9 2.6 1.3
55 12.4 12.0 11.4 10.6 9.7 8.4 7.1 5.6 4.2 2.8 1.4
60 14.7 14.3 13.8 13.0 12.0 10.7 9.3 7.8 6.2 4.5 3.0 1.4
65 17.3 16.9 16.3 15.5 14.5 13.2 11.8 10.2 8.6 6.8 4.9 3.1 1.5
70 20.1 19.6 19.0 18.2 17.2 15.8 14.4 12.8 11.1 9.3 7.4 5.4 3.3 1.6
75 23.0 22.6 21.9 21.1 20.0 18.7 17.2 15.6 13.9 12.0 10.1 8.0 5.8 3.5 1.6
80 26.2 25.7 25.0 24.2 23.1 21.7 20.2 18.6 16.8 14.9 12.9 10.8 8.5 6.2 3.7 1.6
85 29.5 29.0 28.3 27.4 26.3 24.9 23.4 21.7 20.0 18.0 16.0 13.8 11.5 9.1 6.5 4.0 1.6
90 33.1 32.6 31.8 30.9 29.7 28.3 26.7 25.1 23.2 21.3 19.2 17.0 14.7 12.2 9.6 6.9 4.2 1.6
95 36.9 36.3 35.5 34.5 33.3 31.9 30.3 28.5 26.7 24.7 22.6 20.4 18.0 15.5 12.9 10.1 7.3 4.4 1.6
100 40.8 40.2 39.4 38.3 37.1 35.6 34.0 32.2 30.3 28.3 26.2 23.9 21.5 19.0 16.3 13.5 10.6 7.6 4.6 1.7
105 45.0 44.3 43.4 42.3 41.0 39.5 37.8 36.0 34.1 32.0 29.9 27.6 25.2 22.6 19.9 17.1 14.2 11.1 8.0 4.8 1.7
110 49.4 48.6 47.6 46.5 45.2 43.6 41.8 40.0 38.0 35.9 33.7 31.4 29.0 26.4 23.7 20.9 17.9 14.8 11.6 8.3 4.9 1.8
115 53.9 53.1 52.1 50.9 49.5 47.8 46.0 44.1 42.1 40.0 37.7 35.4 32.9 30.3 27.6 24.7 21.7 18.6 15.4 12.0 8.6 5.1 1.8
120 58.7 57.8 56.7 55.4 53.9 52.2 50.4 48.4 46.3 44.2 41.9 39.5 37.0 34.4 31.6 28.7 25.7 22.6 19.4 16.0 12.5 8.9 5.3 1.9
Table A104: HCVII additional travel time due to speed change cycles (seconds/speed cycle)
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 3.2
10 6.0 1.6
15 8.4 4.0 1.1
20 10.7 6.4 3.1 0.9
25 12.9 8.7 5.2 2.5 0.7
30 15.0 10.9 7.3 4.4 2.1 0.6
35 17.0 13.0 9.4 6.4 3.8 1.8 0.5
40 18.9 15.0 11.5 8.4 5.7 3.4 1.6 0.5
45 19.9 16.5 13.3 10.4 7.5 5.1 3.1 1.5 0.4
50 20.8 17.6 14.6 11.7 9.2 6.8 4.6 2.8 1.3 0.4
55 21.7 18.6 15.7 13.0 10.5 8.2 6.1 4.2 2.6 1.2 0.3
60 22.5 19.6 16.9 14.2 11.8 9.5 7.4 5.5 3.8 2.4 1.2 0.3
65 23.4 20.6 17.9 15.4 13.0 10.8 8.7 6.8 5.0 3.5 2.2 1.1 0.3
70 24.2 21.5 18.9 16.5 14.2 12.0 9.9 8.0 6.2 4.6 3.2 2.0 1.0 0.3
75 24.9 22.4 19.9 17.5 15.3 13.1 11.1 9.2 7.4 5.8 4.3 3.0 1.9 0.9 0.3
80 25.7 23.2 20.8 18.5 16.3 14.2 12.2 10.3 8.6 6.9 5.4 4.0 2.8 1.7 0.9 0.3
85 26.5 24.1 21.8 19.5 17.4 15.3 13.3 11.5 9.7 8.0 6.5 5.0 3.7 2.6 1.6 0.8 0.3
90 27.2 24.9 22.6 20.5 18.4 16.3 14.4 12.5 10.8 9.1 7.5 6.1 4.7 3.5 2.4 1.5 0.8 0.2
95 27.9 25.7 23.5 21.4 19.3 17.3 15.4 13.6 11.8 10.2 8.6 7.1 5.7 4.5 3.3 2.3 1.4 0.7 0.2
100 28.6 26.4 24.3 22.2 20.2 18.3 16.4 14.6 12.9 11.2 9.6 8.2 6.8 5.4 4.2 3.2 2.2 1.4 0.7 0.2
105 29.3 27.2 25.1 23.1 21.1 19.2 17.4 15.6 13.9 12.2 10.7 9.2 7.8 6.4 5.2 4.0 3.0 2.1 1.3 0.7 0.2
110 30.0 27.9 25.9 23.9 22.0 20.1 18.3 16.6 14.9 13.2 11.7 10.2 8.8 7.4 6.1 4.9 3.9 2.9 2.0 1.2 0.6 0.2
115 30.7 28.7 26.7 24.7 22.9 21.0 19.2 17.5 15.8 14.2 12.7 11.2 9.7 8.4 7.1 5.9 4.7 3.7 2.7 1.9 1.2 0.6 0.2
120 31.3 29.4 27.4 25.5 23.7 21.9 20.1 18.4 16.8 15.2 13.6 12.1 10.7 9.3 8.0 6.8 5.6 4.5 3.5 2.6 1.8 1.1 0.6 0.2
Table A105: HCVII additional VOC due to speed change cycles (cents/speed cycle – July 2015)
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 0.4
10 1.1 0.6
15 2.2 1.7 0.9
20 3.7 3.2 2.3 1.2
25 5.8 5.3 4.4 3.2 1.7
30 8.4 7.8 6.9 5.6 4.0 2.1
35 11.4 10.8 9.9 8.6 6.9 4.8 2.4
40 14.9 14.3 13.3 12.0 10.2 8.0 5.5 2.8
45 19.3 18.6 17.4 15.9 14.1 11.8 9.2 6.3 3.2
50 24.3 23.6 22.4 20.8 18.7 16.1 13.4 10.4 7.1 3.6
55 29.9 29.2 28.0 26.3 24.2 21.6 18.5 15.1 11.5 7.8 3.9
60 36.2 35.5 34.2 32.5 30.3 27.6 24.5 20.9 17.0 12.7 8.5 4.2
65 43.2 42.4 41.1 39.4 37.1 34.3 31.1 27.5 23.4 19.0 14.1 9.2 4.6
70 50.9 50.0 48.7 46.9 44.6 41.7 38.4 34.7 30.6 26.0 21.0 15.6 9.9 4.8
75 59.3 58.4 57.0 55.1 52.7 49.8 46.4 42.6 38.4 33.7 28.6 23.0 17.1 10.9 5.1
80 68.5 67.5 66.0 64.1 61.6 58.6 55.2 51.3 46.9 42.1 36.9 31.2 25.1 18.6 11.8 5.4
85 78.5 77.4 75.8 73.8 71.3 68.2 64.6 60.6 56.2 51.3 45.9 40.1 33.9 27.2 20.1 12.7 5.6
90 89.3 88.1 86.4 84.3 81.6 78.5 74.8 70.7 66.2 61.2 55.8 49.8 43.4 36.6 29.3 21.7 13.7 5.8
95 100.9 99.6 97.8 95.6 92.8 89.5 85.8 81.6 77.0 71.9 66.3 60.3 53.8 46.8 39.4 31.5 23.2 14.7 6.1
100 113.4 112.0 110.1 107.7 104.9 101.4 97.6 93.3 88.6 83.4 77.7 71.5 64.9 57.8 50.2 42.2 33.7 24.8 15.6 6.5
105 126.8 125.2 123.2 120.7 117.7 114.2 110.2 105.8 100.9 95.6 89.8 83.6 76.8 69.6 61.9 53.7 45.0 35.9 26.4 16.6 6.9
110 141.1 139.4 137.2 134.6 131.5 127.8 123.7 119.1 114.1 108.7 102.8 96.4 89.5 82.2 74.4 66.0 57.2 47.9 38.2 28.0 17.6 7.3
115 156.4 154.5 152.2 149.4 146.1 142.3 138.0 133.3 128.2 122.6 116.6 110.1 103.1 95.6 87.6 79.2 70.2 60.8 50.9 40.5 29.7 18.6 7.7
120 172.7 170.6 168.1 165.2 161.7 157.7 153.3 148.4 143.2 137.4 131.2 124.6 117.5 109.9 101.8 93.2 84.1 74.5 64.4 53.8 42.8 31.3 19.6 8.1
Table A106: Bus additional travel time due to speed change cycles (seconds/speed cycle)
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 2.5
10 4.6 1.3
15 6.5 3.1 0.9
20 8.3 5.0 2.4 0.7
25 10.0 6.8 4.0 1.9 0.5
30 11.6 8.5 5.7 3.4 1.6 0.5
35 13.2 10.1 7.3 5.0 3.0 1.4 0.4
40 14.7 11.7 9.0 6.5 4.4 2.6 1.3 0.4
45 15.4 12.8 10.3 8.1 5.9 4.0 2.4 1.1 0.3
50 16.1 13.6 11.3 9.1 7.1 5.3 3.6 2.2 1.0 0.3
55 16.8 14.4 12.2 10.1 8.1 6.4 4.7 3.3 2.0 1.0 0.3
60 17.4 15.2 13.0 11.0 9.1 7.4 5.7 4.3 3.0 1.9 0.9 0.3
65 18.0 15.9 13.8 11.9 10.1 8.3 6.7 5.2 3.9 2.7 1.7 0.8 0.2
70 18.6 16.6 14.6 12.7 10.9 9.2 7.7 6.2 4.8 3.6 2.5 1.6 0.8 0.2
75 19.2 17.3 15.4 13.5 11.8 10.1 8.6 7.1 5.7 4.5 3.3 2.3 1.4 0.7 0.2
80 19.8 17.9 16.1 14.3 12.6 11.0 9.4 8.0 6.6 5.3 4.2 3.1 2.1 1.3 0.7 0.2
85 20.4 18.5 16.7 15.0 13.4 11.8 10.3 8.8 7.5 6.2 5.0 3.9 2.9 2.0 1.3 0.6 0.2
90 20.9 19.1 17.4 15.7 14.1 12.6 11.1 9.7 8.3 7.0 5.8 4.7 3.7 2.7 1.9 1.2 0.6 0.2
95 21.5 19.7 18.1 16.4 14.9 13.3 11.9 10.5 9.1 7.8 6.6 5.5 4.4 3.5 2.6 1.8 1.1 0.6 0.2
100 22.0 20.3 18.7 17.1 15.6 14.1 12.6 11.2 9.9 8.6 7.4 6.3 5.2 4.2 3.3 2.4 1.7 1.1 0.5 0.2
105 22.5 20.9 19.3 17.7 16.2 14.8 13.4 12.0 10.7 9.4 8.2 7.1 6.0 4.9 4.0 3.1 2.3 1.6 1.0 0.5 0.2
110 23.0 21.4 19.9 18.4 16.9 15.5 14.1 12.7 11.4 10.2 9.0 7.8 6.7 5.7 4.7 3.8 3.0 2.2 1.5 1.0 0.5 0.2
115 23.5 22.0 20.5 19.0 17.5 16.1 14.8 13.4 12.2 10.9 9.7 8.6 7.5 6.4 5.4 4.5 3.6 2.8 2.1 1.5 0.9 0.5 0.1
120 24.0 22.5 21.0 19.6 18.2 16.8 15.4 14.1 12.9 11.6 10.5 9.3 8.2 7.2 6.2 5.2 4.3 3.5 2.7 2.0 1.4 0.9 0.4 0.1
Table A107: Bus additional VOC due to speed change cycles (cents/speed cycle – July 2015)
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 0.2
10 0.5 0.2
15 0.8 0.6 0.3
20 1.3 1.1 0.8 0.4
25 2.0 1.7 1.4 1.0 0.5
30 2.7 2.5 2.1 1.7 1.2 0.6
35 3.6 3.3 3.0 2.5 2.0 1.4 0.7
40 4.5 4.3 3.9 3.5 2.9 2.3 1.5 0.8
45 5.7 5.4 5.0 4.5 3.9 3.3 2.5 1.7 0.9
50 7.0 6.7 6.3 5.8 5.1 4.4 3.6 2.7 1.9 0.9
55 8.4 8.1 7.7 7.2 6.5 5.7 4.9 3.9 2.9 2.0 1.0
60 10.0 9.7 9.3 8.7 8.0 7.2 6.3 5.3 4.3 3.1 2.1 1.0
65 11.7 11.4 10.9 10.3 9.6 8.8 7.9 6.9 5.8 4.6 3.4 2.2 1.1
70 13.5 13.2 12.7 12.1 11.4 10.5 9.6 8.6 7.5 6.3 5.0 3.6 2.2 1.1
75 15.5 15.1 14.6 14.0 13.2 12.4 11.4 10.4 9.2 8.0 6.7 5.3 3.9 2.3 1.1
80 17.6 17.1 16.6 15.9 15.2 14.3 13.3 12.3 11.1 9.9 8.6 7.1 5.6 4.1 2.5 1.0
85 19.7 19.3 18.7 18.0 17.2 16.3 15.3 14.3 13.1 11.8 10.5 9.1 7.5 6.0 4.3 2.6 1.0
90 22.1 21.6 20.9 20.2 19.4 18.5 17.4 16.3 15.1 13.9 12.5 11.1 9.5 7.9 6.2 4.5 2.7 1.0
95 24.5 23.9 23.3 22.5 21.7 20.7 19.6 18.5 17.3 16.0 14.6 13.2 11.6 10.0 8.3 6.5 4.7 2.8 1.0
100 27.0 26.4 25.7 24.9 24.0 23.0 21.9 20.7 19.5 18.2 16.8 15.3 13.8 12.1 10.4 8.6 6.8 4.8 2.8 1.0
105 29.7 29.0 28.2 27.4 26.4 25.4 24.2 23.0 21.8 20.4 19.0 17.5 16.0 14.3 12.6 10.8 8.9 7.0 5.0 2.9 1.0
110 32.4 31.7 30.9 30.0 28.9 27.8 26.7 25.4 24.1 22.7 21.3 19.8 18.2 16.6 14.8 13.1 11.2 9.2 7.2 5.1 3.0 1.0
115 35.3 34.5 33.6 32.6 31.5 30.4 29.1 27.8 26.5 25.1 23.6 22.1 20.5 18.8 17.1 15.3 13.4 11.5 9.5 7.4 5.2 3.0 1.0
120 38.2 37.3 36.3 35.3 34.2 32.9 31.6 30.3 28.9 27.5 26.0 24.4 22.8 21.1 19.4 17.6 15.7 13.8 11.8 9.7 7.5 5.3 3.0 1.0
Table A108: Urban arterial additional travel time due to speed change cycles (seconds/speed cycle)
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 2.3
10 4.2 1.2
15 6.0 2.8 0.8
20 7.6 4.5 2.2 0.6
25 9.1 6.1 3.7 1.8 0.5
30 10.6 7.7 5.2 3.1 1.5 0.4
35 12.0 9.2 6.7 4.5 2.7 1.3 0.4
40 13.4 10.7 8.2 5.9 4.0 2.4 1.2 0.3
45 14.0 11.6 9.4 7.3 5.3 3.6 2.2 1.0 0.3
50 14.6 12.4 10.2 8.3 6.5 4.8 3.3 2.0 1.0 0.3
55 15.2 13.1 11.0 9.2 7.4 5.8 4.3 3.0 1.8 0.9 0.2
60 15.7 13.7 11.8 10.0 8.3 6.7 5.2 3.9 2.7 1.7 0.8 0.2
65 16.3 14.3 12.5 10.7 9.1 7.5 6.1 4.7 3.5 2.5 1.5 0.8 0.2
70 16.8 14.9 13.2 11.5 9.9 8.3 6.9 5.6 4.4 3.2 2.3 1.4 0.7 0.2
75 17.3 15.5 13.8 12.2 10.6 9.1 7.7 6.4 5.2 4.0 3.0 2.1 1.3 0.7 0.2
80 17.8 16.1 14.4 12.9 11.3 9.9 8.5 7.2 5.9 4.8 3.7 2.8 1.9 1.2 0.6 0.2
85 18.3 16.6 15.0 13.5 12.0 10.6 9.2 7.9 6.7 5.6 4.5 3.5 2.6 1.8 1.1 0.6 0.2
90 18.8 17.2 15.6 14.1 12.7 11.3 9.9 8.7 7.5 6.3 5.2 4.2 3.3 2.4 1.7 1.1 0.5 0.2
95 19.2 17.7 16.2 14.7 13.3 12.0 10.6 9.4 8.2 7.0 5.9 4.9 4.0 3.1 2.3 1.6 1.0 0.5 0.2
100 19.7 18.2 16.7 15.3 13.9 12.6 11.3 10.1 8.9 7.7 6.7 5.6 4.7 3.8 2.9 2.2 1.5 1.0 0.5 0.2
105 20.1 18.7 17.3 15.9 14.5 13.2 12.0 10.7 9.6 8.4 7.3 6.3 5.3 4.4 3.6 2.8 2.1 1.4 0.9 0.5 0.1
110 20.6 19.2 17.8 16.4 15.1 13.8 12.6 11.4 10.2 9.1 8.0 7.0 6.0 5.1 4.2 3.4 2.7 2.0 1.4 0.9 0.4 0.1
115 21.0 19.6 18.3 17.0 15.7 14.4 13.2 12.0 10.9 9.8 8.7 7.7 6.7 5.7 4.9 4.0 3.3 2.5 1.9 1.3 0.8 0.4 0.1
120 21.4 20.1 18.8 17.5 16.2 15.0 13.8 12.6 11.5 10.4 9.3 8.3 7.3 6.4 5.5 4.7 3.9 3.1 2.4 1.8 1.3 0.8 0.4 0.1
Table A109: Urban arterial additional VOC due to speed change cycles (cents/speed cycle – July 2015)
speed 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 0.1
10 0.2 0.1
15 0.3 0.2 0.1
20 0.4 0.3 0.2 0.1
25 0.5 0.4 0.3 0.2 0.1
30 0.7 0.6 0.5 0.4 0.3 0.1
35 0.9 0.8 0.7 0.6 0.4 0.3 0.1
40 1.2 1.1 1.0 0.8 0.7 0.5 0.3 0.1
45 1.4 1.3 1.2 1.1 0.9 0.7 0.5 0.3 0.2
50 1.7 1.6 1.5 1.4 1.2 1.0 0.7 0.5 0.3 0.2
55 2.0 1.9 1.8 1.7 1.5 1.3 1.0 0.8 0.6 0.4 0.2
60 2.4 2.3 2.2 2.0 1.9 1.6 1.4 1.1 0.9 0.6 0.4 0.2
65 2.8 2.7 2.6 2.4 2.2 2.0 1.7 1.5 1.2 0.9 0.7 0.4 0.2
70 3.2 3.1 2.9 2.8 2.6 2.4 2.1 1.9 1.6 1.3 1.0 0.7 0.4 0.2
75 3.6 3.5 3.4 3.2 3.0 2.8 2.5 2.3 2.0 1.7 1.4 1.1 0.7 0.4 0.2
80 4.0 3.9 3.8 3.7 3.5 3.2 3.0 2.7 2.4 2.1 1.8 1.5 1.1 0.8 0.5 0.2
85 4.5 4.4 4.3 4.1 3.9 3.7 3.4 3.1 2.8 2.5 2.2 1.9 1.5 1.2 0.8 0.5 0.2
90 5.0 4.9 4.8 4.6 4.4 4.1 3.9 3.6 3.3 3.0 2.7 2.3 2.0 1.6 1.2 0.9 0.5 0.2
95 5.6 5.4 5.3 5.1 4.9 4.6 4.4 4.1 3.8 3.5 3.1 2.8 2.4 2.1 1.7 1.3 0.9 0.5 0.2
100 6.1 6.0 5.8 5.7 5.4 5.2 4.9 4.6 4.3 3.9 3.6 3.3 2.9 2.5 2.2 1.8 1.4 1.0 0.6 0.2
105 6.7 6.6 6.4 6.2 6.0 5.7 5.4 5.1 4.8 4.5 4.1 3.8 3.4 3.0 2.6 2.2 1.8 1.4 1.0 0.6 0.2
110 7.3 7.2 7.0 6.8 6.5 6.3 5.9 5.6 5.3 5.0 4.6 4.3 3.9 3.5 3.1 2.7 2.3 1.9 1.5 1.0 0.6 0.2
115 7.9 7.8 7.6 7.4 7.1 6.8 6.5 6.2 5.9 5.5 5.2 4.8 4.4 4.0 3.6 3.2 2.8 2.4 1.9 1.5 1.0 0.6 0.2
120 8.6 8.4 8.2 8.0 7.7 7.4 7.1 6.8 6.4 6.1 5.7 5.3 5.0 4.6 4.2 3.8 3.3 2.9 2.5 2.0 1.5 1.1 0.6 0.2
Table A110: Urban other additional travel time due to speed change cycles (seconds/speed cycle)
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 2.3
10 4.2 1.2
15 5.9 2.8 0.8
20 7.6 4.5 2.2 0.6
25 9.1 6.1 3.7 1.8 0.5
30 10.6 7.7 5.2 3.1 1.5 0.4
35 12.0 9.2 6.7 4.5 2.7 1.3 0.4
40 13.4 10.6 8.1 5.9 4.0 2.4 1.2 0.3
45 14.0 11.6 9.4 7.3 5.3 3.6 2.2 1.0 0.3
50 14.6 12.3 10.2 8.3 6.5 4.8 3.3 2.0 1.0 0.3
55 15.2 13.0 11.0 9.1 7.4 5.8 4.3 3.0 1.8 0.9 0.2
60 15.7 13.7 11.8 10.0 8.2 6.7 5.2 3.9 2.7 1.7 0.8 0.2
65 16.2 14.3 12.5 10.7 9.1 7.5 6.1 4.7 3.5 2.5 1.5 0.8 0.2
70 16.8 14.9 13.2 11.5 9.9 8.3 6.9 5.6 4.3 3.2 2.3 1.4 0.7 0.2
75 17.3 15.5 13.8 12.2 10.6 9.1 7.7 6.4 5.2 4.0 3.0 2.1 1.3 0.7 0.2
80 17.8 16.1 14.4 12.8 11.3 9.9 8.5 7.2 5.9 4.8 3.7 2.8 1.9 1.2 0.6 0.2
85 18.3 16.6 15.0 13.5 12.0 10.6 9.2 7.9 6.7 5.5 4.5 3.5 2.6 1.8 1.1 0.6 0.2
90 18.7 17.1 15.6 14.1 12.7 11.3 9.9 8.7 7.4 6.3 5.2 4.2 3.3 2.4 1.7 1.1 0.5 0.2
95 19.2 17.6 16.1 14.7 13.3 11.9 10.6 9.4 8.2 7.0 5.9 4.9 4.0 3.1 2.3 1.6 1.0 0.5 0.2
100 19.6 18.1 16.7 15.3 13.9 12.6 11.3 10.0 8.9 7.7 6.6 5.6 4.7 3.8 2.9 2.2 1.5 0.9 0.5 0.2
105 20.1 18.6 17.2 15.8 14.5 13.2 11.9 10.7 9.5 8.4 7.3 6.3 5.3 4.4 3.6 2.8 2.1 1.4 0.9 0.5 0.1
110 20.5 19.1 17.7 16.4 15.1 13.8 12.6 11.4 10.2 9.1 8.0 7.0 6.0 5.1 4.2 3.4 2.7 2.0 1.4 0.9 0.4 0.1
115 21.0 19.6 18.2 16.9 15.6 14.4 13.2 12.0 10.8 9.7 8.7 7.6 6.7 5.7 4.9 4.0 3.3 2.5 1.9 1.3 0.8 0.4 0.1
120 21.4 20.0 18.7 17.4 16.2 15.0 13.8 12.6 11.5 10.4 9.3 8.3 7.3 6.4 5.5 4.6 3.9 3.1 2.4 1.8 1.3 0.8 0.4 0.1
Table A111: Urban other additional VOC due to speed change cycles (cents/speed cycle – July 2015)
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 0.1
10 0.2 0.1
15 0.3 0.1 0.1
20 0.4 0.2 0.1 0.1
25 0.5 0.4 0.3 0.2 0.1
30 0.7 0.6 0.5 0.4 0.2 0.1
35 0.9 0.8 0.7 0.6 0.4 0.3 0.1
40 1.1 1.0 0.9 0.8 0.6 0.4 0.3 0.1
45 1.3 1.2 1.1 1.0 0.8 0.7 0.5 0.3 0.1
50 1.6 1.5 1.4 1.3 1.1 0.9 0.7 0.5 0.3 0.1
55 1.9 1.8 1.7 1.6 1.4 1.2 0.9 0.7 0.5 0.3 0.2
60 2.2 2.1 2.0 1.9 1.7 1.5 1.3 1.0 0.8 0.5 0.3 0.2
65 2.5 2.5 2.4 2.2 2.0 1.8 1.6 1.3 1.1 0.8 0.6 0.4 0.2
70 2.9 2.8 2.7 2.6 2.4 2.2 1.9 1.7 1.4 1.2 0.9 0.6 0.4 0.2
75 3.3 3.2 3.1 2.9 2.8 2.5 2.3 2.0 1.8 1.5 1.2 0.9 0.7 0.4 0.2
80 3.7 3.6 3.5 3.3 3.2 2.9 2.7 2.4 2.1 1.9 1.6 1.3 1.0 0.7 0.4 0.2
85 4.1 4.0 3.9 3.7 3.6 3.3 3.1 2.8 2.5 2.3 2.0 1.7 1.4 1.0 0.7 0.4 0.2
90 4.6 4.5 4.3 4.2 4.0 3.7 3.5 3.2 2.9 2.7 2.4 2.1 1.7 1.4 1.1 0.8 0.4 0.2
95 5.0 4.9 4.8 4.6 4.4 4.2 3.9 3.6 3.4 3.1 2.8 2.5 2.1 1.8 1.5 1.1 0.8 0.4 0.2
100 5.5 5.4 5.3 5.1 4.9 4.6 4.4 4.1 3.8 3.5 3.2 2.9 2.6 2.2 1.9 1.5 1.2 0.8 0.5 0.2
105 6.0 5.9 5.7 5.6 5.4 5.1 4.8 4.5 4.2 3.9 3.6 3.3 3.0 2.6 2.3 1.9 1.6 1.2 0.8 0.5 0.2
110 6.5 6.4 6.2 6.1 5.8 5.6 5.3 5.0 4.7 4.4 4.1 3.7 3.4 3.1 2.7 2.4 2.0 1.6 1.2 0.9 0.5 0.2
115 7.1 6.9 6.8 6.6 6.3 6.0 5.8 5.5 5.1 4.8 4.5 4.2 3.9 3.5 3.2 2.8 2.4 2.0 1.7 1.3 0.9 0.5 0.2
120 7.6 7.5 7.3 7.1 6.8 6.5 6.2 5.9 5.6 5.3 5.0 4.6 4.3 3.9 3.6 3.2 2.9 2.5 2.1 1.7 1.3 0.9 0.5 0.2
Table A112: Rural strategic additional travel time due to speed change cycles (seconds/speed cycle)
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 2.3
10 4.3 1.2
15 6.1 2.9 0.8
20 7.7 4.6 2.2 0.6
25 9.3 6.3 3.7 1.8 0.5
30 10.8 7.9 5.3 3.2 1.5 0.4
35 12.3 9.4 6.8 4.6 2.8 1.3 0.4
40 13.7 10.9 8.3 6.0 4.1 2.4 1.2 0.3
45 14.3 11.9 9.6 7.5 5.4 3.7 2.2 1.1 0.3
50 14.9 12.6 10.5 8.4 6.6 4.9 3.4 2.0 1.0 0.3
55 15.5 13.3 11.3 9.3 7.5 5.9 4.4 3.1 1.9 0.9 0.3
60 16.1 14.0 12.0 10.2 8.4 6.8 5.3 4.0 2.8 1.7 0.8 0.2
65 16.6 14.6 12.8 11.0 9.3 7.7 6.2 4.8 3.6 2.5 1.6 0.8 0.2
70 17.2 15.3 13.5 11.7 10.1 8.5 7.1 5.7 4.4 3.3 2.3 1.4 0.7 0.2
75 17.7 15.9 14.1 12.4 10.8 9.3 7.9 6.5 5.3 4.1 3.1 2.1 1.3 0.7 0.2
80 18.2 16.4 14.8 13.1 11.6 10.1 8.7 7.3 6.1 4.9 3.8 2.8 2.0 1.2 0.6 0.2
85 18.7 17.0 15.4 13.8 12.3 10.8 9.4 8.1 6.9 5.7 4.6 3.6 2.7 1.9 1.2 0.6 0.2
90 19.2 17.5 16.0 14.4 13.0 11.5 10.2 8.9 7.6 6.4 5.3 4.3 3.4 2.5 1.7 1.1 0.6 0.2
95 19.7 18.1 16.5 15.0 13.6 12.2 10.9 9.6 8.4 7.2 6.1 5.0 4.1 3.2 2.4 1.6 1.0 0.5 0.2
100 20.1 18.6 17.1 15.6 14.2 12.9 11.6 10.3 9.1 7.9 6.8 5.8 4.8 3.8 3.0 2.2 1.6 1.0 0.5 0.2
105 20.6 19.1 17.6 16.2 14.8 13.5 12.2 11.0 9.8 8.6 7.5 6.5 5.5 4.5 3.7 2.9 2.1 1.5 0.9 0.5 0.1
110 21.0 19.6 18.2 16.8 15.4 14.1 12.9 11.6 10.4 9.3 8.2 7.2 6.2 5.2 4.3 3.5 2.7 2.0 1.4 0.9 0.5 0.1
115 21.5 20.1 18.7 17.3 16.0 14.7 13.5 12.3 11.1 10.0 8.9 7.8 6.8 5.9 5.0 4.1 3.3 2.6 1.9 1.3 0.8 0.4 0.1
120 21.9 20.5 19.2 17.9 16.6 15.3 14.1 12.9 11.7 10.6 9.5 8.5 7.5 6.5 5.6 4.8 3.9 3.2 2.5 1.9 1.3 0.8 0.4 0.1
Table A113: Rural strategic additional VOC due to speed change cycles (cents/speed cycle – July 2015)
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 0.1
10 0.2 0.1
15 0.3 0.2 0.1
20 0.5 0.4 0.2 0.1
25 0.7 0.6 0.5 0.3 0.2
30 1.0 0.9 0.7 0.6 0.4 0.2
35 1.3 1.2 1.1 0.9 0.7 0.4 0.2
40 1.7 1.6 1.4 1.2 1.0 0.7 0.5 0.2
45 2.1 2.0 1.8 1.6 1.4 1.1 0.8 0.5 0.3
50 2.5 2.4 2.3 2.1 1.8 1.5 1.2 0.9 0.6 0.3
55 3.0 2.9 2.8 2.6 2.3 2.0 1.7 1.3 1.0 0.6 0.3
60 3.6 3.5 3.3 3.1 2.9 2.5 2.2 1.8 1.4 1.0 0.7 0.3
65 4.2 4.1 3.9 3.7 3.5 3.1 2.8 2.4 2.0 1.6 1.1 0.7 0.3
70 4.9 4.7 4.6 4.4 4.1 3.8 3.4 3.0 2.6 2.1 1.7 1.2 0.7 0.3
75 5.6 5.4 5.3 5.0 4.8 4.4 4.1 3.7 3.2 2.8 2.3 1.8 1.3 0.8 0.4
80 6.3 6.2 6.0 5.8 5.5 5.1 4.8 4.4 3.9 3.5 3.0 2.5 1.9 1.4 0.8 0.4
85 7.1 7.0 6.8 6.5 6.3 5.9 5.5 5.1 4.7 4.2 3.7 3.2 2.6 2.1 1.5 0.9 0.4
90 8.0 7.8 7.6 7.4 7.1 6.7 6.3 5.9 5.4 5.0 4.5 3.9 3.4 2.8 2.2 1.6 0.9 0.4
95 8.9 8.7 8.5 8.2 7.9 7.5 7.1 6.7 6.3 5.8 5.3 4.7 4.2 3.6 2.9 2.3 1.7 1.0 0.4
100 9.8 9.6 9.4 9.1 8.8 8.4 8.0 7.6 7.1 6.6 6.1 5.6 5.0 4.4 3.8 3.1 2.4 1.7 1.0 0.4
105 10.8 10.6 10.4 10.1 9.8 9.4 8.9 8.5 8.0 7.5 7.0 6.4 5.8 5.2 4.6 3.9 3.3 2.5 1.8 1.1 0.4
110 11.9 11.6 11.4 11.1 10.8 10.3 9.9 9.4 9.0 8.4 7.9 7.3 6.7 6.1 5.5 4.8 4.1 3.4 2.7 1.9 1.1 0.4
115 13.0 12.7 12.5 12.2 11.8 11.4 10.9 10.4 9.9 9.4 8.9 8.3 7.7 7.1 6.4 5.7 5.0 4.3 3.5 2.8 2.0 1.2 0.5
120 14.1 13.9 13.6 13.2 12.9 12.4 11.9 11.5 10.9 10.4 9.8 9.3 8.6 8.0 7.4 6.7 6.0 5.2 4.5 3.7 2.9 2.0 1.2 0.5
Table A114: Rural other additional travel time due to speed change cycles (seconds/speed cycle)
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 2.3
10 4.3 1.2
15 6.0 2.9 0.8
20 7.7 4.6 2.2 0.6
25 9.2 6.2 3.7 1.8 0.5
30 10.7 7.8 5.3 3.1 1.5 0.4
35 12.2 9.3 6.8 4.6 2.7 1.3 0.4
40 13.6 10.8 8.3 6.0 4.1 2.4 1.2 0.3
45 14.2 11.8 9.5 7.4 5.4 3.7 2.2 1.1 0.3
50 14.8 12.5 10.4 8.4 6.6 4.9 3.3 2.0 1.0 0.3
55 15.4 13.2 11.2 9.3 7.5 5.8 4.4 3.0 1.8 0.9 0.3
60 16.0 13.9 12.0 10.1 8.4 6.8 5.3 3.9 2.7 1.7 0.8 0.2
65 16.5 14.5 12.7 10.9 9.2 7.6 6.2 4.8 3.6 2.5 1.6 0.8 0.2
70 17.0 15.2 13.4 11.6 10.0 8.5 7.0 5.7 4.4 3.3 2.3 1.4 0.7 0.2
75 17.6 15.8 14.0 12.4 10.8 9.3 7.8 6.5 5.2 4.1 3.0 2.1 1.3 0.7 0.2
80 18.1 16.3 14.7 13.0 11.5 10.0 8.6 7.3 6.0 4.9 3.8 2.8 2.0 1.2 0.6 0.2
85 18.6 16.9 15.3 13.7 12.2 10.7 9.4 8.1 6.8 5.6 4.6 3.6 2.6 1.8 1.1 0.6 0.2
90 19.0 17.4 15.8 14.3 12.9 11.4 10.1 8.8 7.6 6.4 5.3 4.3 3.3 2.5 1.7 1.1 0.6 0.2
95 19.5 17.9 16.4 14.9 13.5 12.1 10.8 9.5 8.3 7.1 6.0 5.0 4.0 3.1 2.3 1.6 1.0 0.5 0.2
100 20.0 18.5 17.0 15.5 14.1 12.8 11.5 10.2 9.0 7.9 6.8 5.7 4.7 3.8 3.0 2.2 1.5 1.0 0.5 0.2
105 20.4 19.0 17.5 16.1 14.7 13.4 12.1 10.9 9.7 8.6 7.5 6.4 5.4 4.5 3.6 2.8 2.1 1.5 0.9 0.5 0.1
110 20.9 19.4 18.0 16.7 15.3 14.0 12.8 11.6 10.4 9.2 8.1 7.1 6.1 5.2 4.3 3.5 2.7 2.0 1.4 0.9 0.4 0.1
115 21.3 19.9 18.6 17.2 15.9 14.6 13.4 12.2 11.0 9.9 8.8 7.8 6.8 5.8 4.9 4.1 3.3 2.6 1.9 1.3 0.8 0.4 0.1
120 21.8 20.4 19.1 17.7 16.5 15.2 14.0 12.8 11.7 10.6 9.5 8.4 7.4 6.5 5.6 4.7 3.9 3.2 2.5 1.8 1.3 0.8 0.4 0.1
Table A115: Rural other additional VOC due to speed change cycles (cents/speed cycle – July 2008)
speed
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115
(km/h)
5 0.1
10 0.2 0.1
15 0.3 0.2 0.1
20 0.5 0.3 0.2 0.1
25 0.7 0.6 0.4 0.3 0.1
30 0.9 0.8 0.7 0.5 0.3 0.2
35 1.2 1.1 0.9 0.8 0.6 0.4 0.2
40 1.5 1.4 1.3 1.1 0.9 0.7 0.4 0.2
45 1.9 1.7 1.6 1.4 1.2 1.0 0.7 0.5 0.2
50 2.3 2.2 2.0 1.8 1.6 1.3 1.0 0.8 0.5 0.2
55 2.7 2.6 2.5 2.3 2.0 1.7 1.4 1.1 0.8 0.5 0.3
60 3.2 3.1 2.9 2.7 2.5 2.2 1.9 1.6 1.2 0.9 0.6 0.3
65 3.7 3.6 3.5 3.3 3.0 2.7 2.4 2.1 1.7 1.3 1.0 0.6 0.3
70 4.3 4.2 4.0 3.8 3.6 3.3 2.9 2.6 2.2 1.8 1.4 1.0 0.6 0.3
75 4.9 4.8 4.6 4.4 4.2 3.9 3.5 3.2 2.8 2.4 2.0 1.5 1.1 0.7 0.3
80 5.5 5.4 5.2 5.0 4.8 4.5 4.1 3.8 3.4 3.0 2.6 2.1 1.7 1.2 0.7 0.3
85 6.2 6.1 5.9 5.7 5.4 5.1 4.8 4.4 4.0 3.6 3.2 2.7 2.2 1.8 1.2 0.7 0.3
90 6.9 6.8 6.6 6.4 6.1 5.8 5.5 5.1 4.7 4.3 3.8 3.4 2.9 2.4 1.9 1.3 0.8 0.3
95 7.7 7.6 7.4 7.1 6.9 6.5 6.2 5.8 5.4 5.0 4.5 4.0 3.5 3.0 2.5 1.9 1.4 0.8 0.3
100 8.5 8.3 8.2 7.9 7.6 7.3 6.9 6.5 6.1 5.7 5.2 4.7 4.2 3.7 3.2 2.6 2.0 1.5 0.9 0.3
105 9.4 9.2 9.0 8.7 8.4 8.1 7.7 7.3 6.9 6.4 6.0 5.5 5.0 4.4 3.9 3.3 2.7 2.1 1.5 0.9 0.3
110 10.2 10.1 9.8 9.6 9.3 8.9 8.5 8.1 7.7 7.2 6.7 6.2 5.7 5.2 4.6 4.1 3.5 2.9 2.2 1.6 0.9 0.4
115 11.2 11.0 10.7 10.5 10.1 9.7 9.3 8.9 8.5 8.0 7.5 7.0 6.5 6.0 5.4 4.8 4.2 3.6 3.0 2.3 1.6 1.0 0.4
120 12.1 11.9 11.7 11.4 11.0 10.6 10.2 9.8 9.3 8.8 8.4 7.8 7.3 6.8 6.2 5.6 5.0 4.4 3.7 3.1 2.4 1.7 1.0 0.4
Appendix 5: Passing lanes > Passing lane evaluation procedures
Appendix 5: Passing lanes

This appendix contains procedures to evaluate the benefits of providing passing lanes, typically through
the provision of passing lanes, climbing lanes, slow vehicle bays and increases in the natural passing
opportunities from improved alignments.
A wide range of vehicle types travel on New Zealand highways each day and inevitably some slower
vehicles impede other faster vehicles. In order to overtake these slower vehicles on two-lane highways,
drivers must use the opposing traffic lane. However, this is not always possible or safe. Suitable gaps in
the opposing traffic may be limited and the road alignment may restrict the forward sight distance. The
result is increased travel times as well as increases in driver frustration. Research suggests that the latter
may lead to an increase in unsafe passing manoeuvres and crashes (Thrush 1996).
Passing lanes (and climbing lanes) provide a relatively safe environment for vehicles to overtake other
vehicles, allowing them to travel at their desired speed until such time as the platoons reform. As a
consequence, the benefits of passing lanes generally extend much further than the physical length of the
passing lane section itself, as shown in Figure A8 below.
Figure A8: Benefit length of installing passing lanes
Physical extent of passing lane

Level of platooning
Extent of passing lane benefits
Running distance
Passing lanes free impeded vehicles from slow moving platoons, and in doing so they improve levels of
service, reduce travel times and driver frustration. These benefits will be greatest at locations where road
and traffic conditions result in significant passing demand.
In hilly and mountainous terrain, passing lanes (and climbing lanes) may not be viable, particularly on
lower volume roads. In these situations, other improvement options, such as slow vehicle bays and
shoulder widening, should be considered. The benefit of full-length passing lanes in less severe terrain
can also be low, when traffic volumes are low. Improving sight lines through clearance of vegetation and
vertical or horizontal realignment may increase the available passing opportunities and generate other
safety benefits.
Passing lane evaluation procedures

There are three procedures in this appendix:
1. Passing lane strategy for determining the location of individual passing lanes
2. Assessment of individual passing lanes identified as feasible from a passing lane strategy
3. Detailed analysis of passing lane projects using rural traffic simulation software, such as TRARR.
Appendix 5: Passing lanes > Passing lane evaluation procedures
Figure A9 should be used to determine the appropriate procedure.
Figure A9: Selection of passing lane analysis procedure
Start analysis
No
Are passing opportunities Consider alternative options such as:
the only option to address
safety and efficiency • realignment
deficiencies? • road widening
• removal of roadside hazards
Yes
Has a passing lane

Determine a passing No strategy been developed Yes Is passing lane
lane strategy using and location of future the preferred
procedure 1 passing opportunities option?
determined?
No
Yes
Evaluate other options
using MBCM procedures
Does the passing lane:

1) lie on a prolonged gradient
2) have a high construction cost
3) have a high % of slow No
vehicles
Yes
Yes Use procedure 2 –

Is the analysis for a assessment of individual
prefeasibility passing opportunities
analysis?
Yes
Is the passing lane Yes Use procedure 3 – analysis

expected to be the by rural road simulation
main source of package (eg TRARR)
benefits?
Appendix 5: Passing lanes > Background
Background
Travel time and driver frustration savings
Travel time and driver frustration benefits are generated when passing lanes reduce the amount of time
drivers spend travelling in platoons. The demand for passing and consequently the benefits, are a
function of a number of parameters including:
• traffic variables:
o traffic volume
• road variables:
o terrain/alignment
o grades
The downstream distance over which road user benefits accrue reduces as traffic volumes, the proportion
of slower vehicles (HCVs), and the speed differential between fast and slow vehicles increase. Features
that re-platoon the traffic stream, such as urban areas and major intersections, may limit the available
benefits. While passing lanes also have an impact on the passing opportunities available to traffic
travelling in the opposite direction (where passing is not prohibited), these impacts are typically quite
small and are ignored.
These procedures provide graphs of travel time and driver frustration benefits, which are used or
incorporated into graphs of BCR for different input parameters. These graphs were developed from a
simulation model, which simulates two traffic streams (fast and slow vehicles) travelling along sections of
highway. The simulations are used to determine the demand for passing lanes. The travel time benefits of
passing lanes are then assessed using the ‘unified passing model’ developed by Werner and Morrall
(1984). The changes observed in the level of platooning determine the driver frustration benefits, while
the reduction in travel time is a benefit in its own right. It is also used to determine the change in mean
travel speed and the subsequent change in vehicle operating costs.
Crash rates
A crash rate analysis has been undertaken to produce the crash reduction benefit graphs shown in Figure
A16 to Figure A19. The typical crash rate by terrain type is taken from Table A36. The crash rate at the
passing lane and downstream of the passing lane is less than the typical rate and varies depending on
proximity to the passing lane. The maximum reduction is along the passing lane where the reduction in
the typical rate is 25%. The reduction in the crash rate reduces linearly to zero from the end of the
passing lane to either the location where vehicle platooning returns to normal (generally 5–10km
downstream), or where another passing lane begins.
Table A116 shows the crash rate before the installation of a passing site. The typical crash rates for hilly
terrain have been interpolated as mid-way between the crash rates for rolling and mountainous terrain.
If the passing lane forms part of a rural realignment or there are five or more injury crashes or two or
more serious and fatal crashes in any 1km section (up to 10km downstream of the passing lane) then
crash-by crash analysis may be suitable. To determine if such an analysis is appropriate refer to Figure
A1.
In the majority of cases crash benefits should only be claimed up to 5km downstream of a passing lane,
unless a rural simulation analysis indicates that vehicle platooning will not return to normal until more than
5km downstream. No upstream crash benefits can be included unless international or local research is
produced to justify such benefits.
Table A116: Crash rates for rural mid-block locations (/108 veh/km)
Terrain type Typical crash rate – no passing lane

Flat 16
Rolling 20
Hilly 24 (interpolated from rolling and mountainous crash rates)
Mountainous 28
Passing lane length

A standard passing lane length of 1km is assumed in these procedures. When evaluating passing lanes
with a length greater or shorter than 1km, the appropriate factors in Table A125 should be applied to the
road user benefits.
Proportion of heavy traffic
Two traffic streams, ‘cars’ (passenger cars and light commercial vehicles) and ‘trucks’ (medium/heavy
commercial vehicles and buses) are assumed. The relative proportions are based on the ‘all periods’
composition for a rural strategic road, which is 88% light vehicles and 12% heavy vehicles (refer Table
A45). This assumption impacts on both the level of travel time benefits and on the value of these benefits.
The adjustment in equation 1 (Table A126) can be applied when the percentage of heavy vehicles is
above or below 12%.
Traffic flow profile
The benefits of passing lanes are a function of the traffic using the road during a particular period
(vehicles/hour). To express the benefits of passing lanes as a function of annual average daily traffic
(AADT), it is necessary to assume a traffic flow profile and the number of hours per year that this
particular level of traffic flow (percentage of AADT) occurs. The traffic flow profile assumed for these
procedures is based on that recorded for rural state highways that do not carry high volumes of seasonal
holiday or recreational traffic.
Although it may be expected that additional benefits will accrue to passing lanes on roads that carry high
volumes of recreational traffic, the differences have been found to be insignificant. The exceptional peaks
of the roads with high volumes of recreational traffic are offset by a reduction in the proportion of time the
road operates at around 7% of AADT (refer Table A117).
The relationship between the benefits and the flow profile is relatively robust. In situations where the
traffic flow profile differs significantly from the above, the simplified procedure may not be applicable, and
more detailed analysis using rural simulation (eg TRARR) may be required.
Table A117: Traffic flow profiles
Hourly flow as Roads with low volumes of recreational Roads with high volumes of recreational
% of AADT traffic traffic
Hours/year % hours % AADT Hours/year % hours % AADT
0.9 3,979 45.42 9.7% 3,797 43.35 9.3%
3.5 933 10.65 8.9% 2,062 23.54 19.8%
7.0 3,210 36.64 61.6% 1,819 20.76 34.9%
10.5 541 6.18 15.6% 822 9.38 23.6%
14.0 97 1.11 3.7% 96 1.10 3.7%
17.5 10 0.11 0.5% 120 1.37 5.8%
21.0 – – – 6 0.07 0.4%
25.0 – – – 38 0.43 2.6%
Total 8,760 100% 100% 8,760 100% 100%
Traffic growth
The procedures have been developed using a traffic growth of 2%. Adjustment factors are produced to
modify benefit graphs when the traffic growth is 0%, 1%, 3% and 4%. Where the traffic growth does not
correspond to these values an appropriate adjustment factor can be calculated using interpolation or
extrapolation.
Speed
The variation in traffic speed of individual vehicles within each traffic stream is expressed in terms of the
coefficient of variation (standard deviation divided by the mean) of all vehicle speeds. The procedure
assumes the coefficient of variation (COV) to be 13.5% for both traffic streams.
In situations where road geometry or terrain type has a significant impact on the speeds of particular
vehicle types, it is likely that the COV will increase. In such cases the simplified model will underpredict
the benefits of releasing faster vehicles from platoons. Similarly, on long flat straights where there is likely
to be less variation in speed the model can be expected to overpredict the travel time benefits. The
adjustment in equation 2 (Table A126) can be applied when the COV is above or below 13.5%.
Construction costs
The construction costs presented here, and used in the analysis for determining the appropriate passing
lane strategy, are based on the average costs of constructing a 1km passing lane in each of the terrain
categories. These average costs are generally weighted to the lower end of the reported range, as in
most instances passing lanes are located to avoid costly items, such as bridges.
Average construction and maintenance costs have been calculated for each of the terrain types, using
real costs from a number of projects and from data collected for passing lane research. The construction
costs per linear metre from these projects determined the cost categories shown in Table A118. Table
A119 relates each of the four terrain types to the cost categories, together with the unit and total
construction costs used in the analysis. All costs include the end tapers.
Table A118: Classification of passing lane costs

Category Cost/m Typically had some or all of the following features: Assumed cost/m
($2005) ($2005)
Easy $120 to Flat, straight road and terrain $170
$250 Very good ground conditions
Two or three passing lanes projects in one contract
Existing road 10m seal width, new passing lanes on both
sides of road
No expensive special features
Average $250 to Flat or gently rolling terrain $320
$500 Straight or curved alignment
Good or average ground conditions (soft material
encountered on some projects)
Typically one passing lane per contract
Some special features on some projects
Difficult ≥$500 Poor ground requiring removal and replacement $800
Curved or straight alignment (Estimates in this
Awkward or hilly terrain category were as
Short length of passing lane in one contract high as $1,700 per
High traffic count and control costs linear metre)
Often expensive special features such as rehabilitation and
intersection improvements
Appendix 5: Passing lanes > Passing lane strategies
Table A119: Passing lane average costs ($ 2005)
Terrain type Cost category Unit cost (per m) Total cost (for 1km)
Flat Easy/average $250 $250,000
Rolling Average $320 $320,000
Hilly Average/difficult $500 $500,000
Mountainous Difficult $800 $800,000
Note, however, that cost estimates vary widely depending on site-specific and therefore standardised and
average costs should be used with caution.
Where the estimated cost of construction differs significantly from that assumed in Table A119, an
adjustment to the BCR could be made using equation 3 (Table A126).
Be aware that the analysis of data from selected passing lane sites indicated:
• passing lanes generally cost between $120 and $800 per linear metre, but can cost up to $1700
in some cases. Specific cost estimates should be prepared for each site under consideration
• significant savings in both design and construction costs are possible if two or three projects are
combined into one contract.
Special features can be very expensive and should be avoided where possible, and local knowledge is
important to achieving accurate estimates. Special features include:
• swamps/soft ground
• significant earthworks quantities
• large culvert and/or drain extensions
• intersection improvements
• expensive service relocations.
Construction period
The procedures outlined in this appendix assume that the construction of the passing lane is completed
within the first year.
Update factors
Update factors for user benefits and constructions costs should be used with these procedures. These
can be found on the Waka Kotahi website. When applying an update factor to the combined travel time
and vehicle operating costs, the adjustment factor for travel time costs should be used.
Passing lane strategies

This section provides a procedure for assessing passing lane strategies and is divided into two sections.
The first gives a coarse analysis for identifying passing lane spacing strategies and when increased
passing lane frequency may become economic. The second section is used for determining actual
locations for passing lanes and approximate BCRs of individual projects. More detailed guidance on
individual passing lanes can be found in Assessment of individual passing lanes.
The assumptions made in this procedure are affected by local conditions (refer to Background).
Strategy identification procedure
This procedure is required as an initial step to evaluate strategies. It can also be used in isolation as a
coarse analysis to identify the approximate BCR for each passing lane within a particular strategy.
This procedure can be used to determine the most appropriate passing lane spacing strategy for sections
of strategic rural roads and by doing so identify when increased passing lane frequency may be required.
Table A120: Steps to determine passing lane spacing strategy
Step Action
1 Break the network into sections, as specified in the Waka Kotahi state highway performance
indicators and targets guidelines (or similar for local authority roads). Further classify these
traffic sections into sub-sections with consistent traffic volume and terrain type. Sub-sections
should start or finish at main urban areas.
Sub-sections should not be shorter than:
• 10km for passing lanes at 5km spacing
• 20km for passing lanes at 10km spacing.
When terrain and traffic volumes change frequently, then smaller sections should be combined
and the average traffic volume used in the analysis. The predominant terrain type should also
be used in the analysis. Where this procedure does not seem appropriate, such as on a steep
grade on a route that has typically a rolling or flat alignment, analysts should use a simulation
model such as TRARR to calculate the benefits.
2 Classify the terrain. This can be done vertically by generalised gradient (sum of the absolute
value of rises and falls expressed as m/km) and horizontally by generalised curvature
(degrees/km). Combined classifications of vertical and horizontal terrain are shown in Table
A121, and are a result of analysis of 500m lengths using a 1500m moving average of these
parameters. The curvature, or degrees per kilometre specified in Table A121, is estimated by
summing the deviation angles of the horizontal curves from plans or aerial photography and
dividing by the road length. Rise and fall can be obtained from profile drawings or highway
information sheets. Alternatively, this profile and curve data can be obtained from surveyed
road geometry data.
3 Determine percentage of road with passing sight distance (% PSD) for each sub-section.
The % PSD is the proportion of the section that has visibility greater than 450m. This can be
calculated using surveyed gradient and horizontal curvature data.
In the absence of survey data, each sub-section can be classified according to terrain type,
based on average gradient and curvature. Terrain type sectioning can then be converted to
percentage passing sight distance using Table A122. Note that this method is not as accurate
and may not be sufficient in situations where the benefits are sensitive to % PSD, especially
where traffic volumes are higher.
In Table A122 PSD has been calculated as a moving average over 15km, with the PSD
ascribed to the centre 5km. This is the basis of the BCR graphs and should be observed when
applying the method. The curvature can be estimated as in step 2.
4 Use the analysis year AADT, and % PSD to calculate a BCR, using Figure A10 to Figure
A13.
If traffic growth is not 2% per year, multiply the BCR by the correction factors in Table A123. If
the traffic growth is not in Table A123, extrapolate or interpolate to obtain a correction factor.
The analysis is carried out in both directions, generally with a stagger between opposing
passing lanes where the terrain and available width allows.
5 Repeat step 4 using the predicted AADT for future years in increments of five years from the
analysis year, to identify when it may be worthwhile to adopt a strategy that involves more
frequent passing lanes.
Table A121: Combined terrain classification

Horizontal terrain (degrees/km)
Vertical terrain Straight Curved Winding Tortuous
(rise and fall, m/km) (0–50) (50–150) (150–300) (>300)
Flat (0–20) Flat Rolling Hilly Mountainous
Rolling (20–45)
Hilly (45–60) Rolling Hilly
Mountainous (>60) Mountainous
Table A122: Terrain relationship to passing sight distance
Vertical terrain
Measure
Straight Curved Windy Tortuous
Curvature, degrees per km 0–50 50–150 150–300 >300
Number of curves per km <1.0 1.0–3.0 3.0–6.0 >6.0
Average % passing sight distance 35 15 10 5
Percentage of road length with:
less than 25% sight distance 45 85 95 98
25 to 50% sight distance 30 15 5 2
50 to 75% sight distance 15 – – –
over 75% sight distance – – – –
Table A123: Traffic growth correction factors for BCR graphs
AADT Traffic growth

0% 1% 2% 3% 4%
2,000 0.80 0.90 1.00 1.10 1.21
3,000 0.82 0.91 1.00 1.09 1.18
4,000 0.84 0.92 1.00 1.08 1.16
6,000 0.84 0.92 1.00 1.08 1.16
8,000 0.84 0.92 1.00 1.08 1.15
10,000 0.86 0.93 1.00 1.07 1.15
Refinement of strategy
The following steps determine the location of passing lanes before evaluating individual passing lanes
(Table A126).
Table A124: Steps to refine passing lane spacing strategy
Step Action
6 Identify existing and planned passing lanes for each section where passing lanes can be
justified.
If existing passing lanes spacing ≤ calculated, then
no new passing lanes required
If existing passing lanes spacing > calculated, then
identify potential new sites at the calculated interval
Older sites are unlikely to be at set intervals (as part of a strategy) and judgement is required
in determining whether new sites are justified. Where relevant, identify possible sites for
future years.
7 Identify suitable sites. Sites should be within 1km of either side of the calculated spacing.
Construction cost, land availability and forward visibility at the exit merge are important
factors for site selection. Site spacing or length may be adjusted to balance passing demand
and opportunities. For wider spacing it will be necessary to combine each of the sub-sections
identified in step 1.
Where the strategy results in similar site spacing for each sub-section, this spacing must be
maintained over sub-section boundaries. If the optimal spacing for each sub-section results
in different desired site spacing for each sub-section, the overall strategy should be based on
the largest spacing, ie where the spacing changes from 5km in sub-section one to 10km in
sub-section two, then the spacing should be increased to the higher values (10km) over the
boundary.
Step Action
Any inbound sites in the vicinity of towns should commence at least 5km from the urban
speed limit, unless reasons for a closer facility can be justified. This normally requires
modelling using TRARR.
Use the following guidance to maximise passing lane benefits:
• Select locations where large numbers of vehicles are observed travelling in slow
moving platoons.
• Select locations where there is the greatest speed differential between slow and fast
vehicles (for example, on steep grades).
• Locate sites leading away from congestion (such as urban areas).
• Where possible locate sites on sections with existing no-overtaking lines to maximise
the increase in net passing opportunities.
• Avoid significant intersections (particularly right-turn bays).
• Consider site lengths of between 800m and 1500m in most rural areas – shorter
lengths are unlikely to release all platooned vehicles and little benefit is gained from
excessively long lengths.
• Do not locate the merge area at the end of the sites where there is limited forward
sight distance or where there is a sudden reduction in the desired speed, eg at a
tight horizontal curve.
• The termination of sites in opposing directions should not be adjacent to each other.
• Ensure that sufficient shoulder width and merge space are provided, otherwise an
increase in lost-control and merging crashes could occur.
• Avoid costly physical restraints such as narrow bridges and culverts that require
widening.
Refer to Austroads (2003) Rural road design for further information.
8 Sections of prolonged gradient should be identified, as possible opportunities for
climbing lanes (or slow vehicle bays) using Table A125 below, which is adapted from
Austroads (2003) and considers the length of sustained gradient necessary to reduce the
speed of a heavy commercial vehicle to 40km/h. To assess the benefits of such sites a more
detailed analysis is required using rural simulation software (refer to Rural simulation for
assessing passing lanes).
Table A125: Limiting lengths m for consideration of climbing lanes
Gradient % Approach speed (km/h)

60 80 100
10 100 200 450
9 100 250 550
8 100 300 650
7 150 300 800
6 150 350 1,000
5 200 450
4 300 650
Figure A10: Graphs of strategy BCR for flat terrain
– 5km
Flat terrainFlat Terrain, – 2%
spacing 5 kmtraffic growth 2%
Spacing, Traffic Growth
40.0
35.0
B/C Ratio (July 2005)
30.0
25.0
20.0
15.0
10.0
5.0
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
% Passing Sight Distance
2000 AADT 3000 AADT 4000 AADT 6000 AADT 8000 AADT 10000 AADT
Flat terrain – 10
Flat Terrain, 10 –km
km spacing 2% Spacing,
traffic growth
2% Traffic Growth
40.0
35.0
30.0
25.0
20.0
15.0
10.0
5.0
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Flat terrain – 20km

Flat Terrain, 20– km
spacing 2% traffic growth2%
Spacing, Traffic Growth
40.0
35.0
30.0
25.0
20.0
15.0
10.0
5.0
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Figure A11: Graphs of strategy BCR for rolling terrain
Rolling terrain – 5km spacing – 2% traffic growth

Rolling Terrain, 5 km Spacing, 2% Traffic Growth
35.0
30.0
25.0
20.0
15.0
10.0
5.0
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
terrain – 10km
Rolling Rolling Terrain, 10 –km
spacing 2%Spacing,
traffic growth
2% Traffic Growth
35.0
30.0
25.0
20.0
15.0
10.0
5.0
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
terrain – 20km
RollingRolling Terrain, 20 –km
spacing 2%Spacing,
traffic growth
2% Traffic Growth
35.0
30.0
25.0
20.0
15.0
10.0
5.0
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Figure A12: Graphs of strategy BCR for hilly terrain
Hilly terrain – 5km spacing – 2% traffic growth

Hilly Terrain, 5 km Spacing, 2% Traffic Growth
25.0
20.0
15.0
10.0
5.0
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Hilly– 10km
Hilly terrain spacing
Terrain, – 2%Spacing,
10 km traffic growth
2% Traffic Growth
25.0
20.0
15.0
10.0
5.0
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

25.0
20.0
15.0
10.0
5.0
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Figure A13: Graphs of strategy BCR for mountainous terrain
Hilly terrain – 5km

Hilly spacing 5– km
Terrain, 2% traffic growth2% Traffic Growth
Spacing,
25.0
20.0
15.0
10.0
5.0
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Hilly terrain – 10km

Hilly Terrain, 10–km
spacing 2% Spacing,
traffic growth
2% Traffic Growth
25.0
20.0
15.0
10.0
5.0
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

25.0
20.0
15.0
10.0
5.0
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Appendix 5: Passing lanes > Assessment of individual passing lanes
Assessment of individual passing lanes

This procedure is suitable for establishing the benefits of individual passing lane projects. This method is
not suitable for:
• slow vehicle bays and crawling lanes at the indicative business case stage
• locations where there is a large proportion of slow vehicles such as campervans, coaches, or
slow heavily loaded commercial vehicles
• passing lanes with significant construction costs or significant construction and preconstruction
periods.
For locations where one or more of the above factors apply, a rural traffic simulation model is required to
assess the benefits (refer to Rural simulation for assessing passing lanes).
It is assumed that before using this procedure an appropriate passing lane strategy has been developed
using the method in Table A120 and individual passing lanes are being investigated. This procedure is
used to calculate the benefits of passing lanes in one direction only. For dual passing lanes (passing
lanes in both directions), the procedure needs to be undertaken for both directions separately.
To use the procedure in this section, the BCR graphs in Figure A10 to Figure A13 are not to be used.
Instead, separate graphs for each category of road user benefits are used (Figure A14 to Figure A19),
and these can be adjusted where necessary to account for local conditions.
Procedure for individual passing lanes
Table A126: Steps for assessment of individual passing lanes
Step Action
1 Calculate the travel time and vehicle operating savings, using graphs in Figure A14. If
necessary multiply by the traffic growth correction factor in Table A127 and the travel time
update factor from the most recent update factors, available on the Waka Kotahi website. The
inputs to the graphs are:
• passing lane spacing (either 5, 10 or 20km – for isolated passing lanes use 20km
spacing)
• analysis year AADT
• % PSD (to calculate see Table A120).
2 Calculate the driver frustration savings, using graphs in Figure A15. If necessary, multiply
by the traffic growth correction factor in Table A128 and the driver frustration update factor from
the most recent update factors, available on the Waka Kotahi website.
3 Sum the road user benefits from steps 1 and 2. These are the road user benefits that need
to be adjusted to account for the site-specific characteristics such as passing lane length,
speed distribution and proportion of heavy traffic.
4 Adjustment for the passing lane length. The benefits calculated in the previous steps are
based on passing lanes of 1km in length. Where individual passing lanes are less than 1km in
length, the benefits are reduced because a lesser number of platooned vehicles will be
released. Where the proposed passing lane is longer than 1km, additional benefits may result.
The formation of platoons depends on the spacing between passing lanes, therefore an
adjustment to the benefits is calculated based on the combined effect of passing lane length
and spacing, as provided in Table A129 and Table A130 below (intermediate values may be
interpolated).
5 Adjustment for the proportion of heavy traffic, by comparing the medium plus heavy vehicle
component of the traffic flow at the site with the component for rural strategic roads identified
from Table A46. For every percentage above the assumed 12% proportion of heavy vehicles
(rural strategic), increase the road user benefits by 1%. Similarly, for every percentage point
below the assumed 12% of heavy vehicles decrease the road user benefits by 1%.
Equation 1: Road user benefits (adjusted)
Step Action
= 𝑅𝑜𝑎𝑑 𝑢𝑠𝑒𝑟 𝑏𝑒𝑛𝑒𝑓𝑖𝑡𝑠 (𝑢𝑛𝑎𝑑𝑗𝑢𝑠𝑡𝑒𝑑) × (1 + [𝑝𝑟𝑜𝑝 ℎ𝑒𝑎𝑣𝑦 𝑣𝑒ℎ𝑖𝑐𝑙𝑒𝑠 − 0.12])
6 Adjustment for differences in the speed distribution. This adjustment of road user benefits
(from step 5) is performed if the speed distribution at the site varies from the assumed 13.5%.
A current sample of vehicle speeds over the road sections being analysed is required.
The adjustment is to increase the road user benefits by 2.5% for each percentage point above
the assumed COV of speed of 13.5%. Similarly reduce the road user benefits for a lower COV.
Equation 2: Road user benefits (adjusted)
= 𝑅𝑜𝑎𝑑 𝑢𝑠𝑒𝑟 𝑏𝑒𝑛𝑒𝑓𝑖𝑡𝑠 (𝑢𝑛𝑎𝑑𝑗𝑢𝑠𝑡𝑒𝑑) × (1 + [𝐶𝑂𝑉 − 0.135] × 2.5)
7 Calculate crash costs savings, using graphs in Figure A16 to Figure A19 (interpolate or
extrapolate if necessary) and multiply with the appropriate traffic growth correction factors in
Table A131.
If the passing lane forms part of a rural realignment, or there are either five or more injury
crashes, or two or more serious and fatal crashes in any 1km section (up to 10km downstream
of the passing lane), then crash-by-crash analysis can be used. To determine if such an
analysis is appropriate, refer to Figure A1.
8 Calculate the BCR for the individual passing lanes using the cost estimates for the site and the
benefits calculated in the preceding steps. The BCR can be recalculated using the following
formula (if the unit costs are taken from Table A119).
Equation 3:
𝐵𝐶𝑅 (𝑐𝑎𝑙𝑐𝑢𝑙𝑎𝑡𝑒𝑑 𝑎𝑏𝑜𝑣𝑒) × 𝑇𝑎𝑏𝑙𝑒 𝐴7.4 𝑢𝑛𝑖𝑡 𝑐𝑜𝑠𝑡
𝐵𝐶𝑅 (𝑎𝑑𝑗𝑢𝑠𝑡𝑒𝑑) =
𝐿𝑜𝑐𝑎𝑙 𝑢𝑛𝑖𝑡 𝑐𝑜𝑠𝑡 (𝑝𝑒𝑟 𝑚)
Table A127: Traffic growth correction factors for travel time and VOC graphs
AADT Traffic growth

0% 1% 2% 3% 4%
2,000 0.66 0.83 1.00 1.18 1.39
3,000 0.70 0.85 1.00 1.17 1.34
4,000 0.72 0.86 1.00 1.14 1.27
6,000 0.80 0.90 1.00 1.10 1.20
8,000 0.82 0.91 1.00 1.09 1.18
10,000 0.82 0.91 1.00 1.09 1.17
Table A128: Traffic growth correction factors for driver frustration graphs
AADT Traffic growth
0% 1% 2% 3% 4%
2,000 0.64 0.82 1.00 1.19 1.40
3,000 0.70 0.85 1.00 1.15 1.30
4,000 0.76 0.88 1.00 1.11 1.22
6,000 0.84 0.92 1.00 1.08 1.15
8,000 0.86 0.93 1.00 1.07 1.15
10,000 0.86 0.93 1.00 1.07 1.15
Table A129: Passing lane length factors for travel time delays and vehicle operating cost savings
AADT Passing lane length (m, excl tapers)

(veh/day) 400 600 800 1,000 1,200 1,400 1,600 1,800 2,000
2,000 0.39 0.65 0.91 1.00 1.17 1.15 1.13 1.16 1.18
4,000 0.30 0.60 0.86 1.00 1.19 1.30 1.40 1.48 1.55
6,000 0.08 0.35 0.80 1.00 1.21 1.38 1.54 1.65 1.76
8,000 0.04 0.18 0.60 1.00 1.22 1.43 1.63 1.76 1.88
10,000 0.02 0.11 0.38 0.82 1.24 1.47 1.69 1.83 1.96
12,000 0.02 0.08 0.27 0.57 1.06 1.49 1.73 1.88 2.03
14,000 0.01 0.06 0.20 0.43 0.80 1.32 1.76 1.93 2.09
16,000 0.01 0.05 0.16 0.34 0.63 1.04 1.59 1.97 2.14
18,000 0.01 0.04 0.13 0.28 0.51 0.85 1.30 1.81 2.19
20,000 0.01 0.03 0.11 0.23 0.43 0.71 1.09 1.51 2.03
22,000 0.01 0.03 0.09 0.20 0.37 0.60 0.93 1.29 1.73
24,000 0.01 0.02 0.08 0.17 0.32 0.52 0.80 1.11 1.50
26,000 0.00 0.02 0.07 0.15 0.28 0.46 0.70 0.98 1.31
Notes:
1. Shaded values show either excluded values 1.6–2km passing lane with 2000–4000 vpd or drop-off
in efficiency.
2. The values are for passing lanes on flattish gradient with 110km/h overtaking speed.
3. Refer to Waka Kotahi for passing lanes that lie outside of the above range of values.
4. These factors do not apply to passing lanes in 2+1 layouts (continuous alternating passing lanes).
5. One-way hourly flows were converted to AADT, using a 45%/55% directional split and a peak hourly
flow of 7.6% AADT.
Table A130: Passing lane length factors for frustration cost savings
AADT Passing lane length (m, excl tapers)

(veh/day) 400 600 800 1000 1200 1400 1600 1800 2000
2,000 0.17 0.52 0.87 1.00 1.13 1.33 1.52 1.62 1.71
4,000 0.13 0.48 0.82 1.00 1.18 1.30 1.41 1.50 1.59
6,000 0.03 0.29 0.80 1.00 1.20 1.29 1.37 1.47 1.56
8,000 0.02 0.15 0.60 1.00 1.21 1.30 1.38 1.48 1.58
10,000 0.01 0.09 0.38 0.82 1.21 1.31 1.40 1.51 1.61
12,000 0.01 0.07 0.27 0.57 1.03 1.32 1.43 1.55 1.66
14,000 0.01 0.05 0.20 0.43 0.78 1.17 1.47 1.59 1.71
16,000 0.00 0.04 0.16 0.34 0.61 0.92 1.32 1.61 1.73
18,000 0.00 0.03 0.13 0.28 0.50 0.75 1.08 1.47 1.75
20,000 0.00 0.03 0.11 0.23 0.42 0.63 0.90 1.23 1.62
22,000 0.00 0.02 0.09 0.20 0.36 0.53 0.77 1.05 1.38
24,000 0.00 0.02 0.08 0.17 0.31 0.46 0.66 0.91 1.19
26,000 0.00 0.02 0.07 0.15 0.27 0.41 0.58 0.80 1.05
Notes:
1. Shaded values show either excluded values 1.6–2km passing lane with 2,000–4,000 vpd or drop-off
in efficiency.
2. The values are for passing lanes on flattish gradient with 110km/h overtaking speed.
3. Refer to Waka Kotahi for passing lanes that lie outside of the above range of values.
4. These factors do not apply to passing lanes in 2+1 layouts (continuous alternating passing lanes).
5. One-way hourly flows were converted to AADT, using a 45%/55% directional split and a peak hourly
flow of 7.6% AADT.
Table A131: Traffic growth correction factors for crash savings graphs
AADT Traffic growth

0% 1% 2% 3% 4%
2,000 0.84 0.92 1.00 1.08 1.15
3,000 0.88 0.94 1.00 1.04 1.07
4,000 0.88 0.94 1.00 1.02 1.05
6,000 0.88 0.94 1.00 1.06 1.12
8,000 0.88 0.94 1.00 1.06 1.12
10,000 0.88 0.94 1.00 1.06 1.12
Figure A14: Graphs of vehicle operating cost and delay savings for all terrain
5km spacing – 2% traffic growth

5 km Spacing, 2% Traffic Growth
Time and VOC Savings, NPV/passing
6,000,000
5,000,000
lane ($July 2002)
4,000,000
3,000,000
2,000,000
1,000,000
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
10km spacing – 2%
10traffic growth
km Spacing, 2% Traffic Growth
7,000,000
6,000,000
lane ($July 2002)
5,000,000
4,000,000
3,000,000
2,000,000
1,000,000
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

7,000,000
6,000,000
lane ($July 2002)
5,000,000
4,000,000
3,000,000
2,000,000
1,000,000
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Figure A15: Graphs of driver frustration benefits for all terrain

200,000
NPV/passing lane ($July 2002)
180,000
Driver Frustration Benefits,
160,000
140,000
120,000
100,000
80,000
60,000
40,000
20,000
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
10km spacing – 2%
10 traffic growth
100,000
90,000
80,000
70,000
60,000
50,000
40,000
30,000
20,000
10,000
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
20km spacing – 2%
20traffic growth
80,000
70,000
60,000
50,000
40,000
30,000
20,000
10,000
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Figure A16: Graphs of crash savings for flat terrain
Flat terrain – 5km spacing – 2% traffic growth

Flat Terrain, 5 km Spacing, 2% Traffic Growth
700,000
Accident Reduction Benefits,
600,000
500,000
400,000
300,000
200,000
100,000
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Flat terrain – 10km spacing – 2% traffic growth

Flat Terrain, 10 km Spacing, 2% Traffic Growth
1,200,000
1,000,000
800,000
600,000
400,000
200,000
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Flat terrain – 20km

Flat Terrain, 20 –km
spacing 2%Spacing,
traffic growth
2% Traffic Growth
2,500,000
2,000,000
1,500,000
1,000,000
500,000
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Figure A17: Graphs of crash savings for rolling terrain

1,000,000
900,000
800,000
700,000
600,000
500,000
400,000
300,000
200,000
100,000
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

1,800,000
1,600,000
1,400,000
1,200,000
1,000,000
800,000
600,000
400,000
200,000
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
terrain – Terrain,
RollingRolling 20km spacing – 2%
20 km traffic growth
Spacing, 2% Traffic Growth
3,500,000
3,000,000
2,500,000
2,000,000
1,500,000
1,000,000
500,000
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Figure A18: Graphs of crash savings for hilly terrain

1,200,000
1,000,000
800,000
600,000
400,000
200,000
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

2,500,000
2,000,000
1,500,000
1,000,000
500,000
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

4,000,000
3,500,000
3,000,000
2,500,000
2,000,000
1,500,000
1,000,000
500,000
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Figure A19: Graphs of crash savings for mountainous terrain
Mountainous terrain – 5km spacing – 2% traffic growth

Mountainous Terrain, 5 km Spacing, 2% Traffic Growth
1,400,000
1,200,000
1,000,000
800,000
600,000
400,000
200,000
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Mountainous terrain – 10km spacing – 2% traffic growth

Mountainous Terrain, 10 km Spacing, 2% Traffic Growth
2,500,000
2,000,000
1,500,000
1,000,000
500,000
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Mountainous
Mountainous – 20km spacing
terrainTerrain, 20 km– Spacing,
2% traffic growth
2% Traffic Growth
5,000,000
4,500,000
4,000,000
3,500,000
3,000,000
2,500,000
2,000,000
1,500,000
1,000,000
500,000
-
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Appendix 5: Passing lanes > Rural simulation for assessing passing lanes
Rural simulation for assessing passing lanes

Due to the complex nature of vehicle interactions on two-lane rural roads, traffic simulation programmes
such as TRARR (or TWOPASS) should be used where a more detailed analysis is required or the costs
of a passing lane project are very high. Rural road simulation should be used for:
• slow vehicle bays and climbing lanes at the scheme assessment stage
• locations where there is a large proportion of slow vehicles such as campervans, coaches or slow
moving heavy vehicles.
Rural simulation can be used to obtain a more precise calculation of travel time and vehicle operating
cost benefits resulting from passing lanes, particularly when the sites are constructed as part of road
realignments. For strategic assessment of road links, rural simulation can also be used to evaluate the
relative benefits of passing lanes at various spacing or where local circumstances suggest that these
procedures may not be appropriate, or the assumptions have been violated.
TRARR has traditionally been the rural simulation package used for evaluating passing lanes; however,
other packages are also available and can be used. Koorey (2003) discusses some of the advantages
and disadvantages of TRARR and other packages. The following sub-sections describe analysis by
TRARR as well as model calibration and validation.
Analysis using TRARR
TRARR requires particular care to accurately model traffic flows for both existing and proposed road
layouts. The following notes are provided as a guide. Refer to Hoban et al (1991) for further details about
the TRARR model.
• The modelled road section should include 2km of road upstream of the actual passing site(s).
The modelled road section shall, where appropriate, start and end at points where significant
changes in the nature of the traffic stream occur, such as restricted speed zones (as in urban
areas) and major intersections. The length of the road modelled downstream of the project end
point shall be sufficient to ensure that traffic platooning differences between the do-minimum and
the passing lane option will have tapered out over this length. Depending on the traffic volume,
terrain and passing lanes downstream of the project section, this may be up to 10km.
• A sufficient range of traffic volumes should be modelled to adequately represent all existing and
predicted traffic flows. The proportion of trucks to be modelled should be checked from traffic
data, as it may vary with time of day or volume. For traffic flows of fewer than 50 veh/h the
benefits can be assumed to be negligible and not included if desired.
• Select a sufficient settling-down period to enable traffic (including the slowest vehicles) to fully
traverse the modelled section.
• A New Zealand-based set of vehicle classes and parameters (as specified in VEHS and TRAF
files) should be used for accurate representation of the traffic stream. Refer to Tate (1995) for
examples.
• Suitable intermediate observation points should be specified to enable an accurate assessment
of vehicle operating costs. The same points should be used for all options (except where
realignments preclude this).
• Driver frustration benefits are derived from the ‘Time spent following’ information (given in the
TRARR OUT file). Research by Koorey et al (1999) established a willingness-to-pay value for the
provision of passing lanes of 3.5 cents per vehicle per kilometre of constructed passing lane (this
is in addition to other benefits such as travel time savings). This benefit is applied to all vehicles
that are freed from a platoon at the passing lane over the length they remain free from a platoon.
The value of 3.5 cents/veh/km shall only apply to vehicles travelling in the direction of the passing
site. The veh/km to apply the willingness-to-pay factor to shall be determined by multiplying the
traffic volume by the analysis length and the change in time spent following.
Appendix 5: Passing lanes > Definitions
Example: TRARR is used to analyse 12km of road.
• For a traffic volume of 200veh/h, the do-minimum option gives 50% of time spent following.
• A passing lane option gives 35% of time spent following. The resulting veh/km to apply the
willingness to pay value to, is: 200 × 12 × (50% - 35%) = 360 veh/km/h.
• Crash benefits should be considered up to 10km downstream of the passing lane, depending on
where the traffic platooning differences between the do-minimum and the option have tapered
out.
Calibration/validation of TRARR
TRARR modelling requires care to ensure that it accurately models the actual flows. Although Tate (1995)
found that the relative changes were typically not as sensitive as the absolute values, it is desirable to
match the two where possible. To this end, sufficient field data must be obtained to verify the models.
• The same random traffic generation shall be used for both the do-minimum and project options.
Likewise, for each traffic volume, an equal number of vehicles (at least 1000) shall be simulated
for each option.
• Field data must be collected on typical travel times along the modelled section, including
intermediate points, for both cars and trucks in each direction. These should be used to calibrate
the do-minimum model, adjusting the TRARR desired vehicle speeds to replicate the observed
travel time under the given volume. Overall modelled travel times should match to within 5%,
while intermediate times should be within 10%.
• The proportion of bunching at the start and end of the modelled section should be collected,
along with any desired intermediate points. This data should be calibrated against the do-
minimum model for the particular traffic volume by adjusting the TRARR initial bunching
parameters and intermediate passing lanes. Modelled bunching values should be within 5%
(absolute value) of the field data.
• Once calibrated the models may then be validated by assessing their performance against
outputs measured under different traffic conditions. So if, for example, calibration data was
collected when the average traffic flow was 100 veh/h, the models may be validated by
comparing the model outputs against field measurements taken when traffic volumes were 200
veh/h.
Refer to section 4.3 for further information on checking traffic models.
Definitions
Bunching The proportion of vehicles travelling behind others in platoons. Calculated as the ratio
of following vehicles over total vehicles.
Climbing lane An additional lane provided on steep grades, where large and heavy vehicles travel at
reduced speeds.
Desired speed The speed that drivers would like to travel when not constrained by other traffic. This
is largely dependent on the road alignment. Also known as free speed or unimpeded
speed.
Following Vehicles that are sufficiently close to the vehicle in front to be affected by the speed of
vehicles the front vehicle. Vehicles with headways of less than six seconds are usually
considered to be following.
Free vehicles Vehicles able to travel at their desired speed. This includes vehicles on their own, ie
not part of a multi-vehicle platoon, and leading vehicles. Vehicles with headways of
more than six seconds are usually considered to be free.
Appendix 5: Passing lanes > Definitions
Headway The amount of space between successive vehicles. Can be measured either by
distance or time. Usually measured from the front of one vehicle to the front of the
next.
Leading The vehicle at the head of a multi-vehicle platoon. Leading vehicles are able to travel
vehicles at their desired speed.
Merge area The zone at the end of the passing lane where the two lanes taper into one.
Overtaking An equivalent term for passing.
Passing lane An additional lane, providing two lanes in one direction. A common form of passing
lane. Typically, 400m to 2km in length. Also known as auxiliary lanes or climbing lanes
(on grades). For the purposes of analysis, the length of the passing lane does not
include the end tapers.
Passing Any measure designed to improve the likelihood of vehicles passing safely. These
opportunity include passing lanes, slow vehicle bays, shoulder widening, and improved passing
sight distance (eg realignments).
Platoon A group of vehicles clustered together (ie, small headways) and all travelling at
approximately the same speed as the leading vehicle. Also known as queues or
bunches. The size of the platoon is defined by the number of vehicles. A vehicle on its
own is considered a platoon of size one.
Sight distance The road distance ahead of the driver that is visible. This enables the driver to assess
whether it is safe to pass. Refer to Austroads (2003) for further information, especially
with regard to object and eye heights.
Slow vehicle A short section of shoulder marked as a lane for slow vehicles to move over and let
bay other vehicles pass. Typically up to 400m in length. Slow vehicles have to give way to
the main traffic flow at the end of the bay.
TRARR A rural road simulation package from ARRB transport research in Australia – the latest
version is TRARR 4 (Shepherd 1994). The name ‘TRARR’ is a contraction of ‘TRAffic
on Rural Roads’. TRARR uses various vehicle performance models together with
terrain data to establish, in detail, the speeds of vehicles at each location along the
road. This establishes the demand for passing and determines whether or not passing
manoeuvres may be executed. The outputs, mean travel times and journey speeds,
are used to calculate the benefits of various project options.
Back to 3.8 Impact on user experience … : Driver frustration related to passing lanes >>
Appendix 6: Discount factors > Definitions
Appendix 6: Discount factors

Table A132: Quarterly single payment present worth factors
Time (years from time 4% discount rate SPPWF 3% discount rate 6% discount rate
zero in quarters from SPPWF SPPWF
1 July to 30 June) (sensitivity test) (sensitivity test)
0 1.0000 1.0000 1.0000
0.25 0.9902 0.9926 0.9855
0.50 0.9806 0.9853 0.9713
0.75 0.9710 0.9781 0.9572
1.00 0.9615 0.9709 0.9433
1.25 0.9522 0.9637 0.9298
1.50 0.9429 0.9566 0.9163
1.75 0.9337 0.9496 0.9031
2.00 0.9246 0.9426 0.8900
2.25 0.9155 0.9357 0.8771
2.50 0.9066 0.9288 0.8644
2.75 0.8978 0.9219 0.8519
3.00 0.8890 0.9151 0.8396
3.25 0.8803 0.9084 0.8275
3.50 0.8717 0.9017 0.8155
3.75 0.8632 0.8951 0.8037
4.00 0.8548 0.8885 0.7921
4.25 0.8465 0.8819 0.7806
4.50 0.8382 0.8755 0.7693
4.75 0.8300 0.8690 0.7582
5.00 0.8219 0.8626 0.7473
5.25 0.8139 0.8563 0.7365
5.50 0.8060 0.8500 0.7258
5.75 0.7981 0.8437 0.7153
6.00 0.7903 0.8375 0.7050
6.25 0.7826 0.8313 0.6948
6.50 0.7750 0.8252 0.6847
6.75 0.7674 0.8191 0.6748
7.00 0.7599 0.8131 0.6651
7.25 0.7525 0.8071 0.6554
7.50 0.7452 0.8012 0.6460
7.75 0.7379 0.7953 0.6366
8.00 0.7307 0.7894 0.6274
Table A133: Annual uniform series present worth factors
Time 4% discount rate 3% discount rate USPWF 6% discount rate USPWF

(years from time zero) USPWF (sensitivity test) (sensitivity test)
0 0.0000 0.0000 0.0000
1 0.9806 0.9854 0.9714
2 1.9236 1.9420 1.8879
3 2.8302 2.8708 2.7524
4 3.7020 3.7726 3.5680
5 4.5403 4.6481 4.3375

6 5.3463 5.4980 5.0634
7 6.1213 6.3233 5.7482
8 6.8665 7.1245 6.3943
9 7.5831 7.9023 7.0038
10 8.2721 8.6575 7.5787
11 8.9345 9.3907 8.1212
12 9.5715 10.1026 8.6329
13 10.1841 10.7937 9.1157
14 10.7730 11.4647 9.5711
15 11.3393 12.1161 10.0008
16 11.8838 12.7486 10.4061
17 12.4074 13.3626 10.7885
18 12.9108 13.9588 11.1493
19 13.3949 14.5376 11.4896
20 13.8604 15.0995 11.8107
21 14.3079 15.6451 12.1136
22 14.7382 16.1748 12.3993
23 15.1520 16.6890 12.6689
24 15.5499 17.1883 12.9232
25 15.9325 17.6731 13.1631
26 16.3003 18.1437 13.3895
27 16.6540 18.6006 13.6030
28 16.9941 19.0442 13.8044
29 17.3212 19.4749 13.9945
30 17.6356 19.8930 14.1738
31 17.9380 20.2990 14.3429
32 18.2287 20.6931 14.5025
33 18.5082 21.0757 14.6530
34 18.7770 21.4473 14.7950
35 19.0355 21.8079 14.9290
36 19.2840 22.1581 15.0554
37 19.5229 22.4981 15.1746
38 19.7527 22.8282 15.2871
39 19.9736 23.1487 15.3932
40 20.1860 23.4598 15.4933
41 20.3903 23.7619 15.5877
42 20.5867 24.0551 15.6768
43 20.7755 24.3399 15.7609
44 20.9571 24.6163 15.8402
45 21.1317 24.8847 15.9150
46 21.2996 25.1453 15.9856
47 21.4610 25.3982 16.0522
48 21.6163 25.6438 16.1150
49 21.7655 25.8823 16.1742
50 21.9090 26.1138 16.2301
51 22.0470 26.3386 16.2829
52 22.1797 26.5568 16.3326

53 22.3073 26.7687 16.3795
54 22.4299 26.9744 16.4238
55 22.5479 27.1741 16.4656
56 22.6613 27.3680 16.5050
57 22.7704 27.5562 16.5422
58 22.8752 27.7390 16.5773
59 22.9761 27.9164 16.6103
60 23.0730 28.0887 16.6416
Table A134: Annual arithmetic growth present worth factors

Time 4% discount rate 3% discount rate 6% discount rate AGPWF
(years from time zero) AGPWF AGPWF (sensitivity test)
(sensitivity test)
0 0.0000 0.0000 0.0000
1 0.4871 0.4903 0.4810
2 1.8984 1.9229 1.8512
3 4.1621 4.2426 4.0084
4 7.2105 7.3965 6.8591
5 10.9799 11.3340 10.3180
6 15.4104 16.0069 14.3069
7 20.4455 21.3688 18.7549
8 26.0321 27.3758 23.5971
9 32.1204 33.9856 28.7748
10 38.6635 41.1582 34.2343
11 45.6175 48.8550 39.9273
12 52.9410 57.0395 45.8098
13 60.5953 65.6768 51.8420
14 68.5442 74.7334 57.9883
15 76.7537 84.1777 64.2163
16 85.1919 93.9794 70.4971
17 93.8292 104.1097 76.8048
18 102.6376 114.5411 83.1162
19 111.5914 125.2474 89.4107
20 120.6663 136.2038 95.6699
21 129.8396 147.3867 101.8778
22 139.0905 158.7736 108.0200
23 148.3994 170.3431 114.0842
24 157.7481 182.0748 120.0593
25 167.1198 193.9496 125.9362
26 176.4990 205.9492 131.7068
27 185.8711 218.0561 137.3643
28 195.2228 230.2541 142.9030
29 204.5419 242.5274 148.3182
30 213.8170 254.8614 153.6061
31 223.0377 267.2421 158.7639
32 232.1944 279.6563 163.7893
Time 4% discount rate 3% discount rate 6% discount rate AGPWF

(years from time zero) AGPWF AGPWF (sensitivity test)
(sensitivity test)
33 241.2786 292.0916 168.6808
34 250.2821 304.5362 173.4374
35 259.1978 316.9791 178.0587
36 268.0191 329.4097 182.5449
37 276.7401 341.8182 186.8963
38 285.3554 354.1954 191.1139
39 293.8603 366.5326 195.1989
40 302.2505 378.8216 199.1528
41 310.5222 391.0547 202.9773
42 318.6723 403.2248 206.6744
43 326.6977 415.3251 210.2463
44 334.5960 427.3494 213.6954
45 342.3651 439.2919 217.0240
46 350.0033 451.1472 220.2348
47 357.5091 462.9101 223.3304
48 364.8815 474.5760 226.3136
49 372.1196 486.1406 229.1872
50 379.2228 497.5998 231.9540
51 386.1908 508.9501 234.6170
52 393.0235 520.1880 237.1789
53 399.7209 531.3105 239.6428
54 406.2835 542.3146 242.0116
55 412.7115 553.1980 244.2880
56 419.0058 563.9583 246.4749
57 425.1670 574.5934 248.5753
58 431.1962 585.1015 250.5918
59 437.0942 595.4810 252.5273
60 442.8624 605.7305 254.3844
Back to 1.9 Discounting: Present value >>
Back to 5. Discounting >>
Appendix 7: Risk analysis worksheets > Risk analysis worksheet 1
Appendix 7: Risk analysis worksheets

Risk analysis worksheet 1
In this worksheet nine overall categories of risk are defined, within each of which a number of risk sub-
categories have been identified as being potentially material. For each item in the worksheet, the analyst
should assess the risk according to the risk matrix and assign a risk rating. In cases of doubt, specific
sensitivity tests are proposed, but these may be amended if, in the analyst’s judgement, there are more
appropriate tests. Space is allowed for identifying other material risks in the worksheet.
Each identified risk is to be assigned a risk owner who shall be a named individual or team. A risk owner
can be defined as: ‘the person best placed to manage the risk, suitably qualified and experienced to do
so’.
Responsibilities of the risk owner include:
• managing owner risks – definition, analysis and evaluation

• managing risk treatment – definition, effectiveness, programme requirements and conduct
• ensuring owned risk and treatment data is robust and well maintained
• participating in reviews/workshops as appropriate.
Although it will generally be appropriate to report on the risks for the detailed sub-categories, in those
circumstances where only broad risk information is available, such as in early project stages, it would be
acceptable to report on the risks for each category as a whole, and the worksheet is structured to permit
this.
Table A135: Summary of benefit risks
Risk description (risk source) Risk Risk events and Likelihood Risk rating
owner consequences
Benefit risks
1 Base travel demand
1.1 Age of data source
1.2 Data scope
1.3 Data quantity and statistical
reliability
1.4 Diversion assumption
1.5 Travel demand validation
1.6 Traffic composition
1.7 Other
2 Growth forecasts
2.1 High city population growth
2.2 Development-related traffic as
proportion of scheme traffic
2.3 Time series projection
2.4 Other
3 Assignment
3.1 Changes in user safety – observed
crash sample size
3.2 Changes in user safety –
judgemental crash reduction risk
3.3 Changes in human health – walking
and cycling tourists
3.4 Changes in human health –
environment
Risk description (risk source) Risk Risk events and Likelihood Risk rating
owner consequences
3.5 Changes in transport costs – travel
time valuation
3.6 Wider economic impact – tourism
benefit
3.7 Changes in climate – CO2 valuation
3.8 Changes in access to social and
economic opportunities – user
experience
3.9 Other
4 Benefit realisation
4.1 Tourism
4.2 Dependency on overall economy
4.3 Dependency on future projects or
technology
4.4 Force majeure
4.5 Other future projects
4.6 Diversion from private vehicle
4.7 Supply relationships
4.8 Routing parameters
4.9 Other
Table A136: Summary of cost risks
Cost risks Risk Risk events and Likelihood Risk

owner consequences rating
5 Environmental and planning
5.1 Tangata whenua
5.2 Emissions
5.3 Landscape and visual
5.4 Ecological effects
5.5 Archaeological and historic sites
5.6 Social networks and severance
5.7 Economic/amenity impacts on land users
5.8 Natural hazards
5.9 Other
6 Land and property
6.1 Property acquisition
6.2 Property economic value
6.3 Other
7 Earthworks
7.1 Knowledge of ground conditions
7.2 Complex/unpredictable conditions
7.3 Road design form
7.4 Extent of topographical data
7.5 Source and disposal of material
7.6 Other
8 Other engineering costs
8.1 Engineering complexity
8.2 Signalling and communications
8.3 Transport service operating surplus/deficit

8.4 Other
9 Services
9.1 Existence, location and condition
9.2 Site flexibility
9.3 Cooperation of utilities
9.4 Other
Table A137: Summary of other risks

Other risks Risk Risk events and Likelihood Risk rating
owner consequences
10
10.1
10.2
10.3
11
11.1
11.2
11.3
11.4
12
12.1
12.2
12.3
12.4
13
13.1
13.2
13.3
13.4
14
14.1
14.2
14.3
14.4

In this worksheet, additional information should be supplied on the nature of the high or critical risks
identified in each of the main risk categories and their implications for project decisions. Where possible
and appropriate, courses of action for treating the risks should also be proposed and the costs of these
actions estimated; a brief discussion of courses of action is given in the risk treatment section.
In respect of high or critical risks identified in Table A135, Table A136 and Table A137, additional
information should be supplied under the following headings.
Estimated impacts on benefits/cost (as appropriate):
The analyst’s judgement as to the potential size of the risks, in terms of the percentage impact on either
benefits or costs, should be provided where feasible. It is, however, accepted that it is the nature of some
risks that reliable estimation of their potential impacts is impossible.
Description of implications for option selection and/or project timing:

Risks may impact on decisions on either option selection (where the risks are not common to all options)
or project timing (where, for example, the risks of a non-qualifying BCR may be so high as to suggest a
delay in project implementation).
Recommended actions and estimated costs of those actions (where relevant):
Waka Kotahi will wish to consider the appropriate treatment for each risk (the generic options are: accept,
avoid or transfer risks, reduce likelihood or reduce consequences of risks), and recommendations are
sought on specific actions and their potential costs.
Table A138: Identified high or critical risks

Risk Description and Estimated impacts on Implications Recommended
category nature of the risk benefits and costs actions
(a) (b) (c) (d) (e)
The worksheet should be completed for the identified high risks. The risk categories are labelled R1 to
R17. Leave a risk category blank if it is not high risk. If it is high risk, but the impact cannot be quantified,
simply tick the relevant box. Where the risk impact can be broadly quantified, insert the expected
percentage impact on benefits, costs or the anticipated programme delay in the relevant box.
The worksheet also provides a means of combining the identified and quantified high benefit and cost
risks to give an indication of the impact of these high risks on the overall level of project risk relative to
what might normally be expected for a typical project at a late stage in project development.

This worksheet allows for the calculation of relative risk indicators for a project’s benefits, costs and BCR.
In order to compute the overall project risk, it is necessary to account for the typical risks to be expected
in the other risk categories (the ‘medium’ or ‘baseline’ risks). Therefore, for the purpose of this worksheet,
a broad judgement has been made on the expected levels of benefit, cost and BCR risks associated with
a typical medium risk project in the later stages of development.
These measures of risk have been called ‘relative risk indicators’: there are three, RB, RC and RBCR, for
benefit, cost and BCR risks respectively. They combine the particularly high risks identified in the table
with the expected medium risk levels in other categories to give an overall indication of the impact on
project risk. The relative risk indicators measure the project risk relative to the baseline overall risk of a
typical project.
Figure A20 illustrates the concept. If, for example, we estimate the baseline cost risk of a typical project to
be $1M ±12% (95% confidence limits) and the risk for a specific project is higher at ±18%, then the
relative risk indicator is 1.5, the ratio of the two values. Thus the ‘high’ risks identified for this project
increase the overall risk by 50% over what would normally be expected.
Figure A20: The relative risk indicator for project costs
As the calculation takes no account of identified ‘low’ risk categories, the risk indicator is not a
comprehensive measure of the overall project risk – it is partly for this reason that it is termed an
‘indicator’. Until knowledge is gained of the performance of this indicator as a measure of risk and the
degree to which it varies from project to project, it will not be a factor in funding decisions.
The relative risk indicators labelled RC and RB should be computed using the formulae:
RB = [1 + (1/0.03) * ∑i (Vi – 0.0056)] 0.5
where Vi = (Ri/100)2 and the summation is only for Ri values in the table.
RC = [1 + (1/0.015) * ∑i (Vi – 0.0025)]0.5
where Vi = (Ri/100)2 and the summation is only for Ri values in the table. That is, the benefit risk is
computed from values R1 to R4 and R11 provided in the table and the cost risk from R5 to R10, where the
risks are converted from percentage, eg, 30%, to a fraction, eg, 0.3.
The relative risk indicators RB and RC thus calculated are combined to give the overall BCR relative risk
indicator RBCR as follows.
RBCR = [0.35 * RC2 + 0.65* RB2] 0.5
After the applying risk treatments and contingency, any residual risks shall be reported and quantified to
produce risk adjusted BCR. Use this worksheet along with the BCR risk tool (BCR optimism bias testing).
Table A139: Risk adjusted BCR
Risk category Residual cost risk Residual benefit risk Residual programming risk
Overall relative risk indicators (RC =) (RB =) (RBCR =)

Adjusted BCR =
A worked example for Risk analysis worksheet 3: relative risk indicator calculation is provided in Appendix
8: Worked examples.
Back to 7.4 Risk analysis procedures: Risk assessment >>
Appendix 8: Worked examples > Consumer surplus and the rule of half
Appendix 8: Worked examples

This appendix contains worked examples of benefit quantification and monetisation procedures,
discounting and incremental cost–benefit appraisal, and application of the risk procedures.
Consumer surplus and the rule of half

The basic technique for evaluating consumer impacts of price changes is to use the incremental cost to
consumers who do not change their travel, plus half the change in price times the number of trips that
increase or decrease. This is known as the ‘rule of half’, which represents the midpoint between the old
price and the new price.
For example, if a $1 highway toll increase causes annual vehicle trips to decline from three million to two
million, the reduction in consumer surplus (the total net cost to consumers) is $2.5 million ($1 × 2 million
for existing trips, plus $1 × 1 million × ½ for vehicle trips foregone). Similarly, if a 50c per trip public
transport fare reduction results in an increase from 10 million to 12 million annual public transport trips,
this can be considered to provide $5.5 million in consumer surplus benefits (50c × 10 million for existing
trips, plus 50c × two million × ½ for added trips).
Back to 1.7 Benefits: Rule of half >>
Crash cost procedure

This worked example uses methods B and C from Appendix 2: Crash analysis.
Do-minimum crash costs
A straight and flat 3.3km section of (near rural) state highway in a 100km/h area is identified as having a
high incident of loss of control crashes. This section of road has two 3.5 metre lanes and no sealed
shoulder. The road is a primary collector with an AADT of 2800 and a traffic growth rate of 4%. Nine injury
crashes were recorded in CAS for the previous five years. Two of these were serious injury crashes.
The option is to widen the seal to 9 metres in total: two 3.5 metre lanes and 1 metre wide sealed
shoulders. Time zero is 2015.
The weighted crash procedure is used, as there are less than three injury crashes, or one serious or fatal
crash, per kilometre in the previous five years. Crash rates are available for the do-minimum and
option(s). These are provided in the Crash estimation compendium (CEC).
The proposed improvement (seal widening) is not considered a fundamental change and hence the crash
history is still relevant in calculating the site-specific crash rate (refer to Appendix 2 Definitions) and costs.
Site specific crash rate AS:
AS = nine injury crashes/five years for the site history × 1.10
where: 1.10 is the crash trend adjustment factor from Table A21.
AS = 9 / 5 × 1.10 = 1.98 crashes per year
Typical crash rate (see Appendix 2 Definitions) AT:
AT = (b0 × CMF) × X (from CEC table 5, rural two-lane roads)
where: coefficient b0 = 18 from CEC table 2, for a mean seal width of 8.2 metres, for a
primary collector on level terrain.
Exposure (X) = 3.3km × 2,800 AADT × 365 / 108
= 0.034
CMF = 1.21 (from CEC table 5). This adjusts b0 upward, because the current seal width
of 7 metres is narrower than the mean seal width of 8.2 metres for a rural
primary collector.
AT,dm = 18 × 0.034 × 1.21 = 0.74 crashes per year.
Appendix 8: Worked examples > Crash cost procedure
No adjustment is required for time zero as year zero is 2015.

Weighted crash rate (refer to Appendix 2 Definitions) AW for the do-minimum
The weighted crash rate equation from Weighted crash rate for the do-minimum is:
AW,dm = w × AT + (1 - w) × AS
w = k
k × AT x Y
where: k = 3.0 (from CEC table 2(b)) and Y=5 (years)
Because k is per kilometre, AT needs to be divided by the site length (3.3km),
therefore AT = 0.74 / 3.3
= 0.224.
w = 3.0
3.0 + 0.224 x 5
= 0.73
Therefore, the weighted crash rate is:
AW,dm = 0.73 × 0.74 + (1 – 0.73) × 1.98
= 1.07 crashes per year
Do-minimum crash costs:
= 1.07 × $585,000 (from Table A36)
= $628,758 per year
Option (a) crash costs: no significant changes at site
Typical crash rate AT:
AT,opt = b0 × exposure × CMF (cross-section)
= 18 × 0.034 × 0.69
= 0.42 crashes per year
where: the CMF (cross-section) from CEC table 5 adjusts b0 downwards as the proposed
seal width of 9 metres is wider than the mean seal width of 8.2 metres (for a
primary collector).
Weighted crash rate AW for the option:
AW,opt = AT,opt × AW,dm / AT,dm (from Weighted crash rate for
project option)
= 0.42 × 1.07 / 0.74 = 0.61 crashes per year
Option (a) crash costs = 0.61 × $585,000 = $355,269 per year
Option (a) crash benefits = $628,758 - $355,269 = $273,489 per year
Option (b) crash costs: site significantly changed
If the proposed improvement is considered a fundamental change, in this case due to other works such
as the protection of steep drop-offs or removal of obstacles in the roadside clear zone, then the site-
specific crash history used in the weighted crash procedure (method C) is not relevant in the calculation
of the option crash rate and costs. When there is a fundamental change, the crash costs for the option
are calculated using method B.
Typical crash rate AT for option:
Appendix 8: Worked examples > Cycling benefits
AT,opt = b0 × exposure × CMF (cross-section)

= 18 × 0.034 × 0.69 = 0.42 crashes per year
Option (b) crash costs = 0.42 × $585,000 = $247,034 per year
Option (b) crash benefits = $628,758 - $247,034 = $381,058 per year
Back to 3.1 Impact on social cost of deaths and serious injuries >>
Cycling benefits
For this example calculation, assume:
• an average (two-way) distance cycled per user: 10km

• cycling specific number of days per week for 52 weeks per year
• $4.90/km health benefit per cyclist (no e-bikes) with $6200/user/year health benefit cap
• cycleway has over 100 users per day
• health benefits capping calculations based on frequency of use as shown below.
Days cycled/week Annual health benefits Health benefits cap Annual health benefits after
(uncapped) capping
1 $2,548 No $2,548
2 $5,096 No $5,096
3 $7,644 Yes $6,200
4 $10,192 Yes $6,200
5 $12,740 Yes $6,200
In this example, the health benefits are capped for new individual users cycling three or more days per
week.
Back to 3.2 Impact of mode on physical and mental health >>
Vehicle emissions procedure

For a 1km road with 1000 vehicles travelling along it with a calculated travel time of 2.371 min/veh and a
vehicle flow composition of 95% light and 5% heavy.
Speed = 1 x 60 / 2.371 = 25.3km/h
Light PM10 = 0.02 g/vkt
Heavy PM10 = 0.22 g/vkt
Weighted PM10 emission rate = 95% x 0.02 + 5% × 0.22
= 0.03 g/vkt
PM10 emissions = weighted PM10 emission rate × vkt
= 0.03 × (1km × 1,000 vehicles) = 57g
Back to 3.3 Impact of air emissions on health >>
Traffic congestion procedure

Two worked examples are provided for different road categories as defined in Table A46.
Rural highway realignment
An activity involves the realignment of a busy 2 kilometre section of rural highway, which improves sight
distances, providing more overtaking opportunities for following traffic. The road is classified as rolling
terrain.
Appendix 8: Worked examples > Traffic congestion procedure
From calculations in Appendix 3: Traffic data and travel time estimation, the road section carries 12,500
veh/day, with a peak interval intensity of 1000 veh/h, 60/40 directional split and 12% heavy truck
component. In the do-minimum, the alignment offers no passing opportunities (0% overtaking sight
distance), and after realignment there is no restriction on overtaking sight distance (100% overtaking sight
distance). The hourly capacity of the road in the do-minimum is calculated as:
2800 × ft × fd = 2,800 × 0.675 × 0.94
= 1,775 veh/h
where: 2,800 is the ideal capacity of the road section;
ft and fd are adjustment factors for directional distribution and
the proportion of trucks (see Table A60).
The peak interval traffic intensity (1000 veh/h) divided by capacity gives a VC ratio of 56%.
From Figure 3 the PTD in the do-minimum is 79%, and 71.5% after realignment. The CRV for rural
strategic roads is $4.23 per veh/h (from Table 16).
The incremental values for congestion for the do-minimum and project option are calculated as follows:
Do-minimum = 4.23 × 79/90
= $3.71 per veh/h
Activity option = 4.23 × 71.5/90
= $3.36 per veh/h
The time period total average travel time for the road section is calculated using the procedures in Table
A69 (based on component values calculated in other sections of Appendix 3: Traffic data and travel time
estimation). For this example, the average travel times per vehicle have been calculated as 1.70 and 1.30
min/veh for the do-minimum and realignment options, respectively.
The congestion cost savings are calculated by multiplying the peak interval traffic intensity by the
incremental value for congestion and the time period average travel time divided by 60. For example:
Do-minimum = 1,000 × 3.71 × 1.70/60
= $105.1/h
Project option = 1,000 × 3.36 × 1.30/60
= $72.8/h
Congestion cost saving = $105.1 - $72.8
= $32.3/h over the peak period.
Urban intersection improvement
A project proposal will reduce delay and improve safety at a priority-controlled T-intersection through the
installation of a roundabout. Traffic volumes on the three approaches to the intersection are evenly
balanced, there is a high proportion of turning traffic and the configuration of the site is such that a
roundabout can be constructed without additional land take.
Bottleneck delay to side road traffic during the peak interval of the morning peak period has been
observed to average 35 s/veh for the 500 veh/h on the side road approach, and 5 s/veh for the 300 veh/h
turning off the main road. With the roundabout, traffic volume and bottleneck delay for the three
approaches has been modelled at: 500 veh/h and 7 s/veh; 700 veh/h and 5.5 s/veh; and 600 veh/h and 6
s/veh.
Total bottleneck delay is calculated as:
Do-minimum = (500 × 35 + 300 × 5) / 3,600 = 5.28 veh/h
Roundabout option = (500 × 7 + 700 × 5.5 + 600 × 6) / 3,600 = 3.04 veh/h
Appendix 8: Worked examples > Trip reliability procedure
Reduction in bottleneck delay = 5.28 - 3.04 = 2.24 veh/h

Congestion cost saving = 2.24 × CRV = 2.24 × $3.88 = $8.68/h over time period.
Back to 3.6 Impact on network productivity and utilisation: Traffic congestion values >>
Trip reliability procedure

An activity provides a township (urban arterial) bypass from A to E to remove through traffic from the town
centre. The existing through-traffic between A and E is 2,400 veh/h with 1,200 vehicles in each direction.
It is expected that the traffic volumes between A and E will remain the same once the bypass is built, but
1,400 vehicles will use the new bypass each hour (700 in each direction).
Figure A21: Township bypass overview
Arterial Retail Arterial

C 1,800 C 1,500 C 1,800
v1 1,200 v1 1,200 v1 1,200
v2 500 v2 500 v2 500
A B Signalised E
intersection
N
Township bypass
Arterial D
C 1,800
v1 700
Traffic volumes and VC ratios at signalised intersection I are summarised in Table A140, Table A141,
Table A142 and Table A143.
Table A140: Do-minimum VC ratios

Approach Lane no. Movement Traffic volume VC ratio
(veh/h)
South (B) 1 LT 1,121 0.840
2 R 82 0.595
East (D) 1 L 249 0.706
2 TR 62 0.442
North (E) 1 L 252 0.271
2 T 947 0.774
3 R 9 0.072
West (C) 1 LTR 35 0.290
Table A141: Option VC ratios
Approach Lane no. Movement Traffic volume VC ratio

(veh/h)
South (B) 1 LT 421 0.664
2 R 82 0.330
East (D) 1 L 249 0.286
2 TR 62 0.246
North (E) 1 L 252 0.237
2 T 247 0.433
3 R 9 0.040
West (C) 1 LTR 35 0.161
Table A142: Do-minimum flow matrices

To To To To To E To E Sum
A B C D via town via
bypass
From A 0 0 1 82 1,120 0 1,203
B 0 0 0 0 0 0 0
C 4 0 0 11 20 0 35
D 249 0 2 0 60 0 311
E via town 947 0 9 252 0 0 1,208
E via bypass 0 0 0 0 0 0 0
Sum 1,200 0 12 345 1,200 0 2,757
Table A143: Option flow matrices
A B C D via town via bypass
From A 0 0 1 82 420 700 12,03
B 0 0 0 0 0 0 0
C 4 0 0 11 20 0 35
D 249 0 2 0 60 0 311
E via town 247 0 9 252 0 0 508
E via bypass 700 0 0 0 0 0 700
Sum 1,200 0 12 345 500 700 2,757
For road section, standard deviations of travel times in minutes are calculated by:
SD(TT) = S0 + (S - S0) / 1 + eb*(VC ratio - a))
For urban arterial: S = 0.89, b = -28, a = 1, S0 =0.117 (Table 69)
For urban retail road: S = 0.87, b = -16, a = 1, S0 =0.150 (Table 69)
Table A144: Standard deviations of travel time (minutes)
From To Do-minimum Activity option

A B 0.117 0.117
B I 0.178 0.150
I E 0.117 0.117
A E - 0.117
For intersection C, standard deviations of delays in minutes for each movement are calculated by:
SD(TT) = S0 + (S - S0) / (1 + eb*(VC ratio - a))

For signalised intersection: S =1.25, b = -32, a = 1, S0 =0.120 (Table 69).
Table A145: Standard deviations of intersection travel times

From To Do-minimum Activity option
B C 0.127 0.120
B E 0.127 0.120
B D 0.120 0.120
D B 0.120 0.120
D C 0.120 0.120
D E 0.120 0.120
E D 0.120 0.120
E B 0.121 0.120
E C 0.120 0.120
C E 0.120 0.120
C D 0.120 0.120
C B 0.120 0.120
The total variability is the square root of the sum of individual link/intersection variability. For instance,
from origin A to destination C, the total variability for ‘do-minimum’ and ‘activity option’ are calculated by:
Variability A-C do-minimum = 2 2 2
(SDLink(AB)) + (SDLink(BI)) + (SDIntersection(BC))
= 0.117 2 + 0.178 2 + 0.127 2

= 0.248 min
2
Variability A-C activity option = 0.117 + 0.150 2 + 0.120 2
= 0.225 min
Table A146: Do-minimum matrices of standard deviations of travel times
To A To B To C To D To E via To E via
town bypass
From A 0 0 0.248 0.244 0.274 0
B 0 0 0 0 0 0
C 0.244 0 0 0.120 0.168 0
D 0.244 0 0.120 0 0.168 0
E via town 0.271 0 0.168 0.168 0 0
E via bypass 0 0 0 0 0 0
Table A147: Option matrices of standard deviations of travel times

To A To B To C To D To E via To E via
town bypass
From A 0 0 0.225 0.225 0.254 0.117
B 0 0 0 0 0 0
C 0.225 0 0 0.120 0.168 0
D 0.225 0 0.120 0 0.168 0
E via town 0.254 0 0.168 0.168 0 0
E via bypass 0.117 0 0 0 0 0
Appendix 8: Worked examples > Risk analysis procedure for resilience
Multiply the element in the flow matrix (Table A142 and Table A143) with the corresponding element in
the standard deviation matrix (Table A146 and Table A147) to derive the variability for each movement.
Sum each line to get the total for the approach. Add the final column together to derive the network-wide
variability.
Table A148: Do-minimum matrices of flow × standard deviation of travel time
bypass
From A 0.000 0 0.248 20.008 306.880 0 327.136
B 0.000 0 0.000 0.000 0.000 0 0.000
C 0.976 0 0.000 1.320 3.360 0 5.656
D 60.756 0 0.240 0.000 10.080 0 71.076
E via town 256.637 0 1.512 42.336 0.000 0 300.485
E via bypass 0.000 0 0.000 0.000 0.000 0 0.000
Sum 318.369 0 2.000 63.664 320.320 0 704.353
Table A149: Option matrices of flow × standard deviation of travel time
bypass
From A 0.000 0 0.225 18.450 106.680 81.900 207.255
B 0.000 0 0.000 0.000 0.000 0.000 0.000
C 0.900 0 0.000 1.320 3.360 0.000 5.580
D 56.025 0 0.240 0.000 10.080 0.000 66.345
E via town 62.738 0 1.512 42.336 0.000 0.000 106.586
E via bypass 81.900 0 0.000 0.000 0.000 0.000 81.900
Sum 201.563 0 1.977 62.106 120.120 81.900 467.666
The total variability for 'do-minimum' is 704.353 veh/min and for 'activity option' is 467.666 veh/min.
Variability benefits per peak hour are calculated as:
0.9 × $15.13 × (704.353 - 467.666) / 60 × 30 % = $16.11/h
where: $15.13 is the value of travel time for morning commuter peak hour for urban arterial
(Table 16)
0.9 is the variability travel time factor
30% is the adjustment factor as there is only one major source of variability.
Back to full procedures for road improvement activities: Stage 4g. Impact on system reliability >>
Risk analysis procedure for resilience

A minor bridge structure has been assessed to have a limited residual life and has been tentatively
programmed for replacement after five years. However, the design of the bridge pre-dates modern
earthquake design codes and the bridge would be damaged to an extent requiring replacement in an
earthquake of return period of 200 years or more.
Appendix 8: Worked examples > Risk analysis procedure for resilience
Calculating probability of risk

The annual probability of the bridge being destroyed by earthquake in any one year, denoted as p, is
1/200 = 0.005. The probability of the bridge surviving for five years and then being replaced as
programmed, is calculated as follows:
• The probability of an earthquake in the first year = p = 1/200 = 0.005.

• The probability of the bridge surviving for one year is therefore (1 - p) = 0.995.
• The probability of the bridge being destroyed in year two is the probability of it surviving through
year one multiplied by the probability of an earthquake in year two = p(1 - p) = 0.005 × 0.995 =
0.004975 and so on for five years.
In the general case, the probabilities of the bridge being destroyed in each year are:
• year 1: p
• year 2: p (1 - p)
• year 3: p (1 - p)2
• year n: p (1 - p) n - 1
and the probability of the bridge surviving to n years and then being replaced is therefore:
1 - p - p (1 - p) - p (1 - p)2 - … - p (1 - p)(n – 1) = (1 - p)n
The probability of survival to the end of year five is therefore:
(1 - 0.005)5 = 0.97525
In the event of earthquake damage, a temporary Bailey bridge would have to be erected while a new
permanent structure was being built. This would impose an additional cost on the road controlling
authority, which would not occur in the case of a planned replacement. There would also be disruption to
traffic at the time of the earthquake.
Calculating costs if risk occurs
Assume that the bridge replacement cost is $2.5 million over two years. Making the assumption that an
earthquake, if it occurred, would on average occur mid-year, it is then assumed that these costs are
distributed $1.5 million in the first year and $1.0 million in the next year.
Assume that the cost of erecting a temporary Bailey bridge is $0.2 million spread over six months, the
disruption cost during planned replacement of the bridge is zero (the old bridge remains open), and the
disruption cost of unplanned delays while the Bailey is being constructed is $0.5 million and disruption
during Bailey use (during the two years it takes to construct the new bridge) is $0.2 million per year.
If the bridge is destroyed before planned replacement, then the costs at the start of the year in which the
earthquake occurs are:
Roading costs $million
Bailey bridge $0.1 × 0.9713 (SPPWF yr 0.5)
$0.1 × 0.9433 (SPPWF yr 1.0)
Permanent replacement bridge $1.5 × 0.9433 (SPPWF yr 1.0)
$1.0 × 0.8900 (SPPWF yr 2.0)
total $2.496 million
Road user costs:
Initial disruption costs $0.5 × 0.9713 (SPPWF yr 0.5)
$0.2 × 0.5 × 0.9433 (SPPWF yr 1.0)
Ongoing disruption costs $0.2 × 0.9163 (SPPWF yr 1.5)
$0.2 × 0.5 × 0.8900 (SPPWF yr 2.0)
Appendix 8: Worked examples > Funding gap analysis
total $0.663 million
Where: SPPWF is the single payment present worth factor (Table A150).
Calculating expected values
The probability of the bridge being destroyed by an earthquake in each of years one, two three and four
are then multiplied by the above costs and benefits to give expected values in each year. The same is
done in year five for the costs of planned replacement of the bridge. The expected values of costs and
benefits in each year are then as follows.
Table A150: Example expected value calculations

Year Probability Costs Benefits Expected value Expected value
(costs) (benefits)
1 0.005000 $2,496,000 -$663,000 $12,480 -$3,315
2 0.004975 $2,496,000 -$663,000 $12,418 -$3,298
3 0.004950 $2,496,000 -$663,000 $12,355 -$3,282
4 0.004925 $2,496,000 -$663,000 $12,293 -$3,265
5 0.004901 $2,496,000 -$663,000 $12,233 -$3,249
Year 5 0.975250 $2,305,000 $2,248,000
replacement
Remaining calculations
The above costs and benefits are effectively discounted to the start of each year and each must be
further discounted by the SPPWF factor for (year - 1).
The example does not take account of any benefits that may arise from bridge replacement, such as a
reduction in annual maintenance costs, road user benefits from improved alignment or reduction in bridge
loading restrictions. These should be dealt with in a similar way, by discounting future costs and benefits
to the start of each year one to five and then multiplying by the probability of loss of earthquake
occurrence to give expected values, which should then be further discounted to time zero.
Back to full procedures for road improvement activities: Stage 4j. Other significant impacts …>>
Back to 7.3 Risk analysis overview: Risk reduction benefit >>
Funding gap analysis

In this example of improvement(s) to an existing service, a 12% service provider’s required rate of return
is used. Different activities may justify lower or higher rates of return.
The period of analysis for this particular activity is 40 years. The revenue flow is the increase or change in
revenue from the base case (pre-existing service levels). The revenue for a new service would be
equivalent to the number of users multiplied by the proposed user charge.
The funding gap is included in the table as a payment spread over year’s two to nine of the proposal.
Different values were inserted for the funding gap until the sum of the last column equalled zero.
As the funding gap is positive, the activity is not commercial and funding assistance is required to make it
viable. The value of the funding gap is $1,064,809 per year spread over years two to nine. The present
value of the funding gap is $4,722,845, which does not change irrespective of how the funding gap is
defined. However, this present value is at the service provider’s desired rate of return, not the discount
rate used in economic evaluation.
The cumulative amount of the funding gap is $8,518,471. This depends on how the funding gap is
defined. It is smallest when all funding for the gap is provided at the start of the proposal, eg $5,924,337 if
all paid in year two.
Appendix 8: Worked examples > Discounting
Table A151: Example funding gap calculation
Year Capital cost O&M cost Revenue Funding gap Annual total SPPWF Net present
value
1 -$2,500,000 -$2,500,000 0.8929 -$2,232,143
2 -$2,500,000 -$484,600 $346,000 $1,064,809 -$1,573,791 0.7972 -$1,254,617
3 -$484,600 $356,380 $1,064,809 $936,589 0.7118 $666,645
4 -$484,600 $367,071 $1,064,809 $947,280 0.6355 $602,014
5 -$484,600 $378,084 $1,064,809 $958,292 0.5674 $543,761
6 -$484,600 $389,426 $1,064,809 $969,635 0.5066 $491,247
7 -$484,600 $401,109 $1,064,809 $981,318 0.4523 $443,898
8 -$484,600 $413,142 $1,064,809 $993,351 0.4039 $401,198
9 -$484,600 $425,536 $1,064,809 $1,005,745 0.3606 $362,682
10 -$484,600 $438,302 -$46,298 0.3220 -$14,907
11 -$484,600 $451,452 -$33,148 0.2875 -$9,529
12 -$484,600 $464,995 -$19,605 0.2567 -$5,032
13 -$484,600 $478,945 -$5,655 0.2292 -$1,296
14 -$484,600 $493,313 $8,713 0.2046 $1,783
15 -$484,600 $508,113 $23,513 0.1827 $4,296
16 -$484,600 $523,356 $38,756 0.1631 -
17 -$484,600 $539,057 $54,457 0.1456 -
18 -$484,600 $555,228 $70,628 0.1300 -
19 -$484,600 $571,885 $87,285 0.1161 -
20 -$484,600 $589,042 $104,442 0.1037 -
21 -$484,600 $606,713 $122,113 0.0926 -
22 -$484,600 $624,914 $140,314 0.0826 -
Present value = $4,722,845 Sum of net present value = $0
Back to full procedures for public transport activities: Stage 7d. Calculate service provider’s funding gap >>
Discounting
Single payment present worth factor for a single period
For a section of road resealed 15 years after time zero at a cost of $50,000, the present value of the
reseal cost using a discount rate of 6% is:
Present value = $50,000 x SPPWF615 (Table 82)
= $50,000 x 0.4173
= $20,865
Single payment present worth factor for multiple periods
A project costing $2 million with a implementation period of 15 months starting in the 8th month after time
zero, has the following cash flow for expenditure:
Table A152: Example costs for the 2nd half of year 1

Month 7 8 9 10 11 12 Total
$ (000s) 0 50 50 50 100 150 400
Appendix 8: Worked examples > Discounting
Table A153: Example costs for the 1st half of year 2
Month 13 14 15 16 17 18 Total
$ (000s) 200 200 300 300 200 100 1,300
Table A154: Example costs for the 2nd half of year 2

Month 19 20 21 22 23 24 Total
$ (000s) 50 50 100 100 0 0 300
The present value of the implementation expenditure is:

Using annual SPPWF from Table 82
Present value = ($400,000 + $1,300,000) × SPPWF61 + $300,000 × SPPWF62
= $1,700,000 × 0.9434 + $300,000 × 0.8900
= $1,870,780
A more accurate calculation using quarterly SPPWF from Table A150.
Present value = $150,000 × SPPWF60.75 + $450,000 × SPPWF61.00
+ $800,000 × SPPWF61.25 + $350,000 × SPPWF61.50
+ $250,000 × SPPWF61.75
= $150,000 × 0.9572 + $450,000 × 0.9433
+ $800,000 × 0.9298 + $350,000 × 0.9163
+ $250,000 × 0.9031
= $1,858,385
Figure A22: Example cashflows
time zero
+1.25 years
900
Implementation expenditure
time zero $800k

+1.00 years
time zero
+1.50 years time zero
600 time zero +1.75 years
+0.75 years $450k
$350k
$300k$300k
300 $250k
$200k$200k $200k
$150k $150k
$100k $100k $100k$100k
$50k $50k $50k $50k $50k
0
8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Months after time zero
Quarterly cashflow Monthly cashflow
Uniform series present worth

If maintenance costs for the do-minimum are $30,000 a year over a 42-year analysis period (40 years
plus two years to the start of construction), from Table A133 the present value of the maintenance costs
is:
Present value = $30,000 × (USPWF642 - USPWF60)
= $30,000 × (15.677 - 0)
Appendix 8: Worked examples > BCRG procedure (using simplified numbers)
= $470,310
Arithmetic growth present worth factor
If vehicle operating costs are $70,000 with traffic growth of 1% at time zero, and construction finishes two
years from time zero, from Table A134 the present value of the vehicle operating costs on the new
construction is:
Present value = $70,000 × [(USPWF642 - USPWF62) + 0.01 × (AGPWF642 - AGPWF62)]
= $70,000 × [(15.677 - 1.888) + 0.01 × (206.674 - 1.851)]
= $1,108,606
Back to 5. Discounting >>
BCRG procedure (using simplified numbers)

Toll road example
BCRN
PV benefits = $600m
PV costs = $300m (including cost of toll facilities and toll collection)
BCRN = 600 / 300 = 2.0
BCRG
PV benefits = $600m
PV tolls = $150 m (gross toll collections)
PV costs = $300 m (including cost of toll facilities and toll collection)
BCRG = (600 - 150) / (300 - 150) = 3.0
PT example
BCRN
PV benefits = $900m
PV costs = $300m
BCRN = 900 / 300 = 3.0
BCRG
PV benefits = $900m
PV farebox revenue = $150m (PV of gross revenue)
PV costs = $300 m
BCRG = (900 - 150) / (300 - 150) = 5.0
Private sector contribution example
BCRN
PV benefits = $300m
PV costs = $200m
BCRN = 300/200 = 1.5
Appendix 8: Worked examples > Incremental BCRs
BCRG
PV benefits = $300m
PV private sector contribution = $100m
PV costs = $200m
BCRG = (300 - 100) / (200 - 100) = 2.0
Back to 6.2 Government benefit–cost ratio >>
Incremental BCRs
The concept of incremental cost–benefit analysis is illustrated in the figure below, which considers two
options – A and B.
The BCR for option B is 4.0 (4,000/1,000). Such a value would usually result in the project receiving a
High rating for the economic efficiency criteria considered under the Waka Kotahi funding allocation
process. The less-costly option A, with a BCR of 7.5 (3,000/400), would receive the same High rating.
However, incremental cost–benefit analysis demonstrates that the incremental benefits gained by
supporting option B ahead of option A represent only a small return on the additional cost, as the
incremental BCR is 1.7 ((4,000–3,000)/(1,000–400))
Figure A23: Incremental BCR between two options
Costs = option B - do-minimum
Option B
4000
Incremental benefits
= option B - option A
Option A
3000
Incremental costs
Present value benefits
Benefits = option B - option A

= option B
- do-minimum
2000
Benefits
= option A
1000
Costs
= option A
0
0 200 400 600 800 1000 1200
Do-minimum Present value costs
Appendix 8: Worked examples > Calculating bottleneck delay
Applying incremental CBA to mutually exclusive options

To analyse five mutually exclusive project options against a target incremental BCR of 4.0, first rank the
options in order of increasing cost as in Table A155:
Table A155: Mutually exclusive options ranked by cost
Option Benefits Costs BCR

A 110 15 7.3
B 140 30 4.7
C 260 45 5.8
D 345 65 5.3
E 420 100 4.2
Next, calculate the incremental BCR of each higher cost option, discarding those below the target
incremental BCR as in Table A156.
Table A156: Calculating the incremental BCR of mutually exclusive options

Current Next- Calculation Incremental Above/below New preferred option
preferred higher BCR the target
option cost incremental
option BCR
A B (140 - 110)/(30 - 15) 2.0 Below A (No change)
A C (260 - 110)/(45 - 15) 5.0 Above C
C D (345 - 260)/(65 - 45) 4.3 Above D
D E (420 - 345)/(100 - 65) 2.1 Below D (No change)
Finally select the option that has the highest cost and an incremental BCR greater than the target
incremental BCR, which in this example is option D.
Back to 6.3 Incremental cost–benefit analysis: Procedure for calculating the incremental BCR>>
Calculating bottleneck delay

An example of the bottleneck delay calculation using the data from step 4 of Table A66 and a road
capacity of 500 vehicles.
Table A157: Example data for bottleneck delay calculation
Start Demand Cumulative Vehicles Cumulative Queue at Queue at Average

time (veh) demand discharged discharge end of start of delay
(veh) (veh) (veh) interval interval (veh/min)
Step 4 5 6 7 8 9 10
7:00 264 264 264 264 0 0 0.0
7:15 475 739 475 739 0 0 0.0
7:30 591 1,330 500 1,239 91 0 682.5
7:45 600 1,930 500 1,739 191 91 2,115.0
8:00 591 2,521 500 2,239 282 191 3,547.5
8:15 475 2,996 500 2,739 257 282 4,042.5
8:30 264 3,260 500 3,239 21 257 2,085.0
8:45 250 3,510 271 3,510 0 21 157.5
9:00 234 3,744 234 3,744 0 0 0.0
Appendix 8: Worked examples > Traffic signals
Step 11: Time period total delay

= 682.5 + 2,115 + 3,547.5 + 4,042.5 + 2,085 + 157.5
= 12,630 veh-mins
Step 12: Time period average delay per vehicle
= 12,630 / 3744
= 3.37 mins/veh
Calculating the time period total average travel time
Section length = 1km
Free speed travel time = 0.636 min/km
Time period additional travel time = 0.232 min/km
Speed change additional travel time = 0.003 min
Bottleneck delay per vehicle = 1.5 min/veh
Time period total average travel time
= (TTFS + TTATT) × length + bottleneck delay + speed change
= (0.636 + 0.232) × 1.00 + 1.5 + 0.003
= 2.371 min/veh
Back to Appendix 3. Traffic data and travel time estimation: Calculating bottleneck delay>>
Traffic signals
Table A158: Example traffic signal data
Basic data
Lane width 3.3m
Number of lanes 2
Approach grade +2%
Parking movements/h 20
Locality CBD
Arrival type Random
Signal type Actuated
Lane width factor (from Table A70) = 0.98

Approach grade factor (from Table A71) = 0.99
Parking factor (from Table A72) = 0.89
Locality factor (from Table A73) = 0.90
Saturation flow rate = 2000 × 0.98 × 0.99 × 0.89 × 0.90
= 1554 pcu/h
Arrival type (from Table A74) = 3
Delay adjustment factor (from Table A75) = 0.85
In using a traffic model to analyse this example intersection, a saturation flow rate of 1554 pcu/h shall be
used, and the resulting delays multiplied by 0.85.
Back to Appendix 3. Traffic data and travel time estimation: Traffic signals >>
Appendix 8: Worked examples > Risk analysis worksheet 3: relative risk indicator calculation
Risk analysis worksheet 3: relative risk indicator calculation

Estimated 95% confidence limits on quantifiable risk category (expressed as a % of the impact on TOTAL
costs or TOTAL benefits).
Table A159: Example data for relative risk indicator calculation
Risk category Benefit risk Cost risk Programming risk

1
2 (R2 =) 10%
3
4
5 (R5 =) 15%
6
7 (R14=) 6 months
8 (R8 =) 25%
9
10 (R10 =) ✓
Overall relative risk (RB =) 1.07 (RC =) 2.52 (RBCR=) 1.72
indicators
The notes below illustrate the calculation of the relative risk indicators, using data from Table A159 and
the methodology from Risk analysis worksheet 3.
Relative benefit risk indicator:
RB = [1 + (1/0.03) × (R22 - 0.0056)] 0.5 = 1.07
That is, the estimated benefit confidence limit (95%) risk is 7% larger than the nominal value.
Relative cost risk indicator:
RC = {1 + (1/0.015) × [(R52 - 0.0025) + (R82 - 0.0025)]} 0.5 = 2.52
That is, the estimated cost confidence limit (95%) risk is 152% larger than the nominal value.
Relative BCR risk indicator:
RBCR = [0.35 × RC2 + 0.65 × RB2 ] 0.5 = 1.72
That is, the estimated BCR confidence limit (95%) risk is 72% larger than the nominal value.

Monetised Benefits and Costs Manual

Uploaded by

Copyright:

Available Formats

Monetised Benefits and Costs Manual

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Monetised Benefits and Costs Manual

Uploaded by

Copyright:

Available Formats

f

ISBN 978-1-98-856173-8 (online)

June 2023 │ Version 1.6.1

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 4

3.7 Impact on system reliability .............................................................................................79

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 5

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 6

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 7

Table 100: Sensitivity tests ................................................................................................................. 201

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 8

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 9

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 10

Table A142: Do-minimum flow matrices ............................................................................................. 418

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 11

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 12

Introduction > Purpose of this manual

Who this manual is for

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 13

1. Concepts > 1.1 Social cost–benefit analysis

1.2 Equity or distributional effects of land transport initiatives

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 14

1. Concepts > 1.3 Steps in BCR calculation

1.3 Steps in BCR calculation

• Investigation and design

• Determine the benefit–cost ratios (BCR) of the preferred option

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 15

1. Concepts > 1.4 Counterfactuals and the do-minimum

1.4 Counterfactuals and the do-minimum

Do-minimum for safety activities

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 16

1. Concepts > 1.5 Alternatives and options

1.5 Alternatives and options

1.6 Period of analysis

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 17

1. Concepts > 1.7 Benefits

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 18

1. Concepts > 1.7 Benefits

Wider economic impacts

Level of data collection and analysis

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 19

1. Concepts > 1.8 Costs

Types of monetised benefits

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 20

1. Concepts > 1.8 Costs

Avoided costs and cost savings

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 21

1. Concepts > 1.9 Discounting

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 22

1. Concepts > 1.10 Benefit–cost ratios and other appraisal tools

where: n is time in years after time zero; and

1.10 Benefit–cost ratios and other appraisal tools

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 23

1. Concepts > 1.11 Sensitivity analysis

Net present value

First year rate of return

1.11 Sensitivity analysis

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 24

1. Concepts > 1.11 Sensitivity analysis

Risk and uncertainty

Monetised benefits and costs manual │ version 1.6.1, June 2023 // 25