Dynamic Response and Life-Cycle Analysis of Floating Production Storage and Offloading Systems

School of Civil and Mechanical Engineering
Dynamic Response and Life-Cycle Analysis of Floating

Production Storage and Offloading Systems
Rini Nishanth
This thesis is presented for the Degree of

Doctor of Philosophy
of
Curtin University
October 2018
DECLARATION OF THESIS
To the best of my knowledge and belief this thesis contains no material previously
published by any other person except where due acknowledgment has been made.
This thesis contains no material which has been accepted for the award of any other
degree or diploma in any university.
Signature: ………………………………………….
11 Oct 2018
Date: ……………
DEDICATION
To my husband and Son
To my mother and sister To my grand parents
To my family members To my friends
v
ACKNOWLEDGEMENTS
First and Foremost, I thank the GOD Almighty for directing me towards this research
opportunity, giving me the knowledge and ability to complete this research and for
giving strength and patience to face the trials and challenges in life, I faced over this
period. Without his provisions and blessings, this achievement would not have been
possible.
I am immensely grateful to Dr. Andrew Whyte, Associate Professor and Head of

the Department, Civil Engineering, at Curtin University, for supervising me and
encouraging me to find my way through the life-cycle aspects of FPSO. I am indebted
to him for the confidence he placed in me and for all the provisions he made possible
at Curtin University. Apart from keeping a constant check on my research progress, I
am immensely grateful for the time and effort he spent on helping me complete this
thesis, without his help, the timely submission of this thesis would not have been
possible. I am grateful for the moral support provided by him while completing the
thesis corrections and I can surely say this thesis wouldn’t have been possible without
his guidance and support.
I would like to express my deep and sincere gratitude to my research supervisors,

Prof Ir Dr. Mohd Shahir Liew and Dr. Kurian Velluruzhathil John, Universiti Teknologi
PETRONAS, for supporting me immensely through this journey, giving valuable
suggestions, encouragement and support to pursue this joint PhD programme. I am
deeply grateful for choosing me as the candidate for this Collaborative PhD programme
and for giving me support and answers to my relentless queries consistently, while I
was at Universiti Teknologi Petronas and at Curtin University.
I am thankful to my supervisor Prof Ir Dr. Mohd Shahir Liew for his kind help and
support in the final submission of this thesis and for the moral support while completing
vi
the thesis corrections. I am extremely grateful for his guidance while completing the
thesis corrections and the confidence he placed in me.
I am thankful to my previous supervisor Dr Kurian Velluruzhathil John for not only

being a good teacher and mentor, but also for being an inspiration. He made sure I
develop not only as a researcher, but also as a professional Engineer in my work. He
gave me all the freedom, while silently ensuring that I stay on course and complete
work on time. Without his guidance, this thesis would not have been possible and I am
forever grateful for his assistance.
I am grateful to Curtin University and Universiti Teknologi PETRONAS staff,

particularly School of Civil and Mechanical Engineering Staff, offshore laboratory
technicians and the Centre of Graduate Studies staff, this smooth voyage would not
have been possible without them. Their time and effort is gratefully acknowledged. I
am deeply grateful to the Det Norske Veritas team for supplying me with the SESAM
license throughout this research period and immensely grateful to their software support
team for clearing my doubts and answering to my queries. The financial support for
this study was provided by Universiti Teknologi PETRONAS, Malaysia and Curtin
University, Perth. I thank my family members and friends, particularly my mother for
giving me constant strength and moral support to complete this thesis. I am eternally
grateful to everyone who has contributed to make this journey a success.
Last, but not the least, I am forever grateful to my husband and son for all the
sacrifices they made during this PhD journey, for being understanding and supporting
in all the ways possible and for placing their confidence in me during the hardest times.
vii
ABSTRACT
Floating Production Storage and Offloading (FPSO) systems are efficient means of
natural resource management in the deep-water fields of South East Asia and Western
Australia. For short-term projects of 10 years, most energy-companies use FPSOs with
refurbished hulls, stabilised by mooring-line connections subjected to unpredictability
of waves and non-linearities from mooring lines along with risk associated from
fluctuating market rates. The suitability of both converted tankers and newly built
floating platforms for projects under site specific metocean conditions can be assessed
by the computation of respective motion responses and operational downtime, with life
cycle cost analyses providing a means to compare alternative vessel specifications. In
this study, uncoupled and coupled software simulation models for both spread moored
and turret moored FPSOs are developed using SESAM software. The uncoupled
simulation model is validated using the model testing results performed at Universiti
Teknologi PETRONAS (UTP) Offshore Laboratory in the presence of unidirectional
random waves and coupled simulation model using published results from tests
performed at Offshore Technology Research Center wave basin at Texas A&M
university, by investigating the six degree of freedom responses of FPSOs under the
action of wind, waves and current. The results correspond well for wave period ranges
of 5 s – 25 s. Life-cycle cost analysis methodology is developed and used to compute
the whole life cost of the FPSO system, identifying economic FPSO options for 10-year
and 25-year design periods. Riser turret moored FPSOs are found to be the costliest.
The parametric study results covering the influence of metocean conditions, water
depth, hull loading conditions, hull length to beam ratio, mooring line azimuth angle,
mooring line length, and position of mooring line fairleads can be used along with the
life-cycle costs and net present values when choosing the economic and efficient option
in the initial design phase as well as relocation of FPSO. The response amplitude
operators, relative motions as well as sensitivity analysis chart reflecting the varying
market rates are also developed for the Malaysian and Australian waters.
viii
In compliance with the terms of the Copyright Act 1987 and the IP Policy of the
university, the copyright of this thesis has been reassigned by the author to the legal
entities of the universities,
Institute of Technology PETRONAS Sdn Bhd. / Curtin University
Due acknowledgement shall always be made of the use of any material contained
in, or derived from, this thesis.
© RINI NISHANTH, 2018
Institute of Technology PETRONAS Sdn Bhd
Curtin University
All rights reserved.
ix
TABLE OF CONTENT
ABSTRACT ................................................................................................................viii
LIST OF FIGURES ..................................................................................................... xv
LIST OF TABLES ....................................................................................................... xx
LIST OF ABBREVIATIONS ....................................................................................xxii
LIST OF SYMBOLS ................................................................................................ xxiv
CHAPTER 1 INTRODUCTION ................................................................................... 1
1.1 Chapter Overview .............................................................................................. 1
1.2 Background of FPSO ......................................................................................... 1
1.2.1 History of FPSO .................................................................................... 1
1.2.2 Structure of FPSO ................................................................................. 3
1.2.3 Advantages of FPSOs............................................................................ 4
1.3 Relevance of Cost and Motion Analysis of FPSO ............................................ 5
1.4 Relevance of Dynamic Analysis of FPSO ......................................................... 6
1.5 Relevance of Life-cycle Cost Analysis for FPSO ............................................. 8
1.6 Problem Statement ............................................................................................. 8
1.7 Objectives of the Study.................................................................................... 10
1.8 Scope of the Study ........................................................................................... 11
1.9 Significance of the Project ............................................................................... 13
1.10 Overview of the Thesis .................................................................................. 13
CHAPTER 2 LITERATURE REVIEW ...................................................................... 15
2.1 Chapter Overview ............................................................................................ 15
2.2 Reported Studies .............................................................................................. 15
2.2.1 Dynamic Analysis of SPM and Ships ................................................. 16
2.2.1.1 Model testing on SPM and ships ............................................. 16
2.2.1.2 Numerical investigation of SPM and ship motions ................. 17
2.2.2 Dynamic Analysis of FPSO ................................................................ 19
2.2.3 FPSO Motion Response Using Uncoupled and Coupled Analysis ..... 22
2.2.4 FPSO Motion Response using Model Testing .................................... 24
2.2.4.1 Modelling of FPSO .................................................................. 24
x
2.2.4.2 Modelling of mooring lines ..................................................... 26
2.2.5 Hydrodynamics of FPSO .................................................................... 27
2.2.5.1 Representation of wave ............................................................ 27
2.2.5.2 Wave force on FPSO ............................................................... 29
2.2.5.3 Wave force on mooring lines ................................................... 32
2.2.6 Operability Conditions for FPSO and Downtime Cost ....................... 32
2.2.6.1 Parametric studies for environmental loads and water depth
on FPSO motions ..................................................................... 34
2.2.6.2 Parametric studies for mooring line and hull parameters on
FPSO motions .......................................................................... 35
2.2.6.3 Metocean conditions and FPSO motions ................................. 36
2.2.6.4 FPSO motions and green water effects .................................... 37
2.2.6.5 Operability analysis ................................................................. 38
2.2.7 Life-cycle Cost Analysis ..................................................................... 39
2.2.7.1 Definition of LCCA ................................................................. 39
2.2.7.2 Benefits of LCCA .................................................................... 40
2.2.7.3 Previous studies using LCCA .................................................. 40
2.2.7.4 LCCA Procedure ...................................................................... 42
2.2.7.5 Limitations of LCCA ............................................................... 46
2.2.8 Life-cycle costs for FPSO ................................................................... 47
2.3 Critical Literature Review ............................................................................... 49
2.3.1 FPSO Motion Performance and Cost .................................................. 50
2.3.2 Dynamic Responses of FPSO ............................................................. 51
2.3.3 FPSO Operability ................................................................................ 53
2.3.4 Life-cycle cost of FPSO ...................................................................... 55
2.4 Chapter Summary ............................................................................................ 58
CHAPTER 3 METHODOLOGY ................................................................................ 59
3.1 Chapter Overview ............................................................................................ 59
3.2 Overall Research Methodology ....................................................................... 59
3.3 Experimental Tests .......................................................................................... 62
3.3.1 Test Facility and Instrumentation ........................................................ 62
3.3.1.1 Wave maker system ................................................................. 64
xi
3.3.1.2 Qualysis motion tracking system ............................................. 66
3.3.1.3 Wave probes............................................................................. 67
3.3.1.4 Load Cells and Data Loggers ................................................... 67
3.3.2 Choice of the Scale and Physical Modelling Law ............................... 68
3.3.3 FPSO Model ........................................................................................ 69
3.3.3.1 Model description .................................................................... 69
3.3.3.2 Mooring system ....................................................................... 72
3.3.4 Laboratory Tests .................................................................................. 73
3.3.4.1 General ..................................................................................... 73
3.3.4.2 Wave calibration ...................................................................... 73
3.3.4.3 Calibration of model ................................................................ 75
3.3.4.4 Experimental setup................................................................... 77
3.3.4.5 Static offset test ........................................................................ 79
3.3.4.6 Free decay test.......................................................................... 79
3.3.4.7 Seakeeping tests ....................................................................... 79
3.4 Dynamic Analysis of FPSO ............................................................................. 83
3.4.1 FPSO Modelling .................................................................................. 84
3.4.2 Frequency Domain Analysis ............................................................... 87
3.4.2.1 Assumptions and theories ........................................................ 87
3.4.2.2 Analysis Procedure .................................................................. 88
3.4.3 Time Domain Analysis........................................................................ 89
3.4.3.1 Assumptions and theories ........................................................ 89
3.4.3.2 Analysis Procedure .................................................................. 89
3.5 Operability Analysis ........................................................................................ 91
3.5.1 Green water on FPSO and Downtime cost .......................................... 91
3.5.2 FPSOs for operability analysis and site specific metocean data ......... 95
3.6 Life - Cycle Cost Analysis of FPSO .............................................................. 102
3.6.1 Identification of design alternatives to be compared......................... 102
3.6.2 Establishment of basic assumptions and determination of exact
LCC procedure to be adopted. ....................................................... 103
3.6.2.1 Life expectancy of FPSO ....................................................... 104
3.6.2.2 Cash flow timings .................................................................. 104
xii
3.6.2.3 Discount rate .......................................................................... 104
3.6.2.4 Source and reliability of data ................................................. 105
3.6.3 Data Collection .................................................................................. 105
3.6.4 Cost Data for FPSOs ......................................................................... 106
3.6.5 Life-cycle Cost Calculation for FPSO .............................................. 107
3.6.6 NPV Calculation for FPSO ............................................................... 107
3.6.7 Risk/Uncertainty Assessment ............................................................ 108
3.6.8 Selection of Design Alternative ........................................................ 109
3.7 Chapter Summary .......................................................................................... 109
CHAPTER 4 RESULTS AND DISCUSSION .......................................................... 110
4.1 Chapter Overview .......................................................................................... 110
4.2 Experimental Results ..................................................................................... 110
4.2.1 Static Offset Test ............................................................................... 111
4.2.2 Free Decay Test ................................................................................. 111
4.3 Validation of Uncoupled Frequency Domain Analysis ................................. 114
4.4 Validation of Coupled Time Domain Analysis ............................................. 118
4.4.1 FPSO and Metocean Conditions ....................................................... 118
4.4.2 Natural Periods & Damping Ratios ................................................... 121
4.4.3 Response Spectra............................................................................... 124
4.5 Parametric Study............................................................................................ 128
4.5.1 Influence of Metocean Data on FPSO Motions ................................ 128
4.5.2 Influence of Water Depth and Wave Periods on FPSO Motions ...... 136
4.5.3 Influence of Mooring Line Azimuth Angle on FPSO Motions ........ 138
4.5.4 Influence of Mooring Line Length on FPSO Motions ...................... 142
4.5.5 Influence of Spread Mooring Fairlead Location on Hull on FPSO
Motions .......................................................................................... 145
4.5.6 Influence of Hull Length to Beam Ratio on FPSO Motions ............. 148
4.5.7 Influence of Hull Loading Condition on FPSO Motions .................. 152
4.6 Motion Response of FPSOs with Similar Dimensions as Operating FPSOs
in Malaysia and Australia ........................................................................... 155
4.7 Green Water on FPSOs .................................................................................. 159
xiii
4.7.1 Relative motion and freeboard exceedance for Malaysian and
Australian FPSOs ........................................................................... 159
4.7.2 Downtime cost due to green water .................................................... 165
4.8 Life-Cycle Cost of FPSOs in Malaysia and Australia ................................... 166
4.9 Net Present Value of FPSOs in Malaysia and Australia ............................... 172
4.10 Present Value of FPSOs in Malaysia and Australia .................................... 174
4.11 Sensitivity Analysis ..................................................................................... 177
4.12 Cost Proportions .......................................................................................... 180
4.12.1 Cendor II and Glas Dowr FPSO ...................................................... 180
4.12.2 Nganhurra and Stybarrow FPSO ..................................................... 181
4.12.3 Okha and Pyrenees Venture FPSO.................................................. 182
4.12.4 Perisai Kamelia and Glas Dowr FPSO ............................................ 183
4.13 Cost and Motion Performance of FPSOs..................................................... 184
4.14 Chapter Summary ........................................................................................ 186
CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS .............................. 187
5.1 Conclusions ................................................................................................... 187
5.1.1 Adequacy of simulation model and simulation procedure adopted to
predict wave frequency motion responses ..................................... 188
5.1.2 Suitability of spread moored and turret moored FPSOs in extreme
weather conditions ......................................................................... 189
5.1.3 Effect of water depth, mooring line and hull parameters on FPSO
motions........................................................................................... 190
5.1.4 Downtime cost due to green water effects for site specific
conditions of Malaysia and Australia ............................................ 191
5.1.5 Cost effective FPSO configurations for 10-year and 25-year use in
Malaysia and Australia .................................................................. 192
5.2 Recommendation for Future Work ................................................................ 193
REFERENCES………………………………………………………………….......195
PUBLICATIONS LIST…………………………………………………………......210
xiv
LIST OF FIGURES
Figure 1.1: Worldwide Distribution of FPSO 2017 ....................................................... 2

Figure 1.2: Various Parts of FPSO ................................................................................ 4
Figure 1.3: Six Degrees of Freedom of FPSO ............................................................. 11
Figure 2.1: Schematic Diagram for a Progressive wave in x direction ....................... 28
Figure 2.2: Definition of Boundary Conditions for the Linear Diffraction Problem ... 30
Figure 2.3: LCC Framework ........................................................................................ 43
Figure 2.4: Life-cycle of FPSO .................................................................................... 47
Figure 3.1: Overview of Research Methodology ......................................................... 61
Figure 3.2: UTP Offshore Lab ..................................................................................... 63
Figure 3.3: Wave Basin at UTP Offshore Laboratory ................................................. 63
Figure 3.4: H R Wallingford Wave Maker .................................................................. 64
Figure 3.5: Performance of the Wave Generator at 1 m and 0.8 m water Depth [85] . 66
Figure 3.6: Berantai FPSO Model ............................................................................... 69
Figure 3.7: Drawing of FPSO Model ........................................................................... 70
Figure 3.8: Mooring Line Arrangement for Model Tests ............................................ 72
Figure 3.9: Wave Probe Arrangement for Wave Calibration ...................................... 74
Figure 3.10: Wave probes in tank during wave calibration ......................................... 74
Figure 3.11: Three Point Mass System ........................................................................ 76
Figure 3.12: Experimental Setup ................................................................................. 78
Figure 3.13: FPSO Model in Wave Tank .................................................................... 78
Figure 3.14: SESAM Communication ......................................................................... 84
Figure 3.15: FPSO Concept Model .............................................................................. 85
Figure 3.16: Panel Model Assigned with Wet Surface ................................................ 86
Figure 3.17: Finite Element Mesh Generated .............................................................. 86
Figure 3.18: Spread Moored Model in HydroD V4.5-08 ............................................ 88
Figure 3.19: FPSO Model in SESAM DeepC Interface .............................................. 91
Figure 3.20: Points along the vessel hull at which wave elevation RAOs are calculated
in SESAM HydroD V4.5-08 ........................................................................................ 92
Figure 3.21: Representation of Relative Wave Motion ............................................... 94
xv
Figure 3.22: Operation location for selected Malaysian FPSOs for Operability
analysis and LCCA ...................................................................................................... 97
Figure 3.23: Operation location for selected Australian FPSOs for Operability
analysis and LCCA ...................................................................................................... 98
Figure 3.24: LCCA Procedure for FPSO ................................................................... 103
Figure 4.1: Force- excursion relationship of FPSO with horizontal mooring system
(test result in prototype Scale) ................................................................................... 111
Figure 4.2: Free decay time series – Surge (test result in prototype scale) ............... 112
Figure 4.3: Free decay time series – Sway (test result in prototype scale) ................ 112
Figure 4.4: Free decay time series – Yaw (test result in prototype scale) ................. 113
Figure 4.5: Free decay time series – Heave (test result in prototype scale)............... 113
Figure 4.6: Free decay time series – Roll (test result in prototype scale) .................. 113
Figure 4.7: Free decay time series – Pitch (test result in prototype scale)................. 114
Figure 4.8: Surge RAO (Comparison of HydroD and experiment results) ............... 115
Figure 4.9: Heave RAO (Comparison of HydroD and experiment results) .............. 115
Figure 4.10: Pitch RAO (Comparison of HydroD and experiment results) .............. 116
Figure 4.11: Sway RAO (Comparison of HydroD and experiment results) .............. 116
Figure 4.12: Roll RAO (Comparison of HydroD and experiment results) ................ 116
Figure 4.13: Yaw RAO (Comparison of HydroD and experiment results) ............... 117
Figure 4.14: Wind Spectrum Comparison ................................................................. 121
Figure 4.15: Wave Spectrum Comparison ................................................................. 121
Figure 4.16: Free Decay Time Series – Surge (SESAM DeepC) .............................. 122
Figure 4.17: Free Decay Time Series – Heave (SESAM DeepC) ............................. 123
Figure 4.18: Free Decay Time Series – Roll (SESAM DeepC) ................................ 123
Figure 4.19: Free Decay Time Series – Pitch (SESAM DeepC) ............................... 123
Figure 4.20: Surge Motion Spectrum Comparison (SESAM DeepC and Kim et al
results [42]) ................................................................................................................ 124
Figure 4.21: Sway Motion Spectrum Comparison (SESAM DeepC and Kim et al
results [42]) ................................................................................................................ 125
Figure 4.22: Heave Motion Spectrum Comparison (SESAM DeepC and Kim et al
results [42]) ................................................................................................................ 125
xvi
Figure 4.23: Roll Motion Spectrum Comparison (SESAM DeepC and Kim et al
results [42]) ................................................................................................................ 126
Figure 4.24: Pitch Motion Spectrum Comparison (SESAM DeepC and Kim et al
results [42]) ................................................................................................................ 126
Figure 4.25: Yaw Motion Spectrum Comparison (SESAM DeepC and Kim et al
results [42]) ................................................................................................................ 127
Figure 4.26: Motion response of turret moored OTRC FPSO subjected to wind, wave
and current (SESAM DeepC results) ......................................................................... 130
Figure 4.27: Motion response of turret moored OTRC FPSO subjected to only wave
(SESAM DeepC results) ............................................................................................ 130
Figure 4.28: Motion response of spread moored OTRC FPSO subjected to wind, wave
and current (SESAM DeepC results) ......................................................................... 131
Figure 4.29: Motion response of spread moored OTRC FPSO subjected to only wave
Figure 4.30: Motion response of turret moored Berantai FPSO subjected to wind,
wave and current (SESAM DeepC results) ............................................................... 132
Figure 4.31: Motion response of spread moored Berantai FPSO subjected to wind,
wave and current (SESAM DeepC results) ............................................................... 132
Figure 4.32: Spread moored Berantai FPSO subjected to only wave (SESAM DeepC
results) ........................................................................................................................ 133
Figure 4.33: Variation of Surge, Heave and Pitch Motion with Water Depth
(Experimental results in prototype scale) .................................................................. 137
Figure 4.34: Variation of Surge, Heave and Pitch Motion with Mooring Line Azimuth
Angle under Long Crested Waves (Experimental results in prototype scale) ........... 139
Angle under Short Crested Waves (Experimental results in prototype scale) ........... 141
Figure 4.36: Variation of FPSO Translational Motions with Mooring Line Length
Figure 4.37: Variation of FPSO Rotational Motions with Mooring Line Length
Figure 4.38: Variation in Translational FPSO Motions with Location of Spread
Mooring Fairleads on Hull (SESAM DeepC results) ................................................ 146
xvii
Figure 4.39: Variation in Rotational FPSO Motions with Location of Spread Mooring
Fairleads on Hull (SESAM DeepC results) ............................................................... 147
Figure 4.40: Variation in Translational FPSO Motions with Length to Beam Ratio of
Hull (SESAM HydroD results) .................................................................................. 150
Figure 4.41: Variation in Rotational FPSO Motions with Length to Beam Ratio of
Hull (SESAM HydroD results) .................................................................................. 151
Figure 4.42: Variation in Translational FPSO Motions with Hull Loading Condition
(SESAM HydroD results) .......................................................................................... 153
Figure 4.43: Variation in Rotational FPSO Motions with Hull Loading Condition
(SESAM HydroD results) .......................................................................................... 154
Figure 4.44: RAOs of Operating Malaysian FPSOs (SESAM HydroD results) ....... 157
Figure 4.45: RAOs of Operating Australian FPSOs (SESAM HydroD results) ....... 158
Figure 4.46: Capital Cost of FPSOs in Malaysia and Australia under study ............. 168
Figure 4.47: Total Life-Cycle cost of FPSOs in Malaysia and Australia for 10-year
and 25-year life-cycle period ..................................................................................... 169
Figure 4.48: Cumulative Real Cost of FPSOs in Malaysia and Australia for 10-year
Life-cycle Period ........................................................................................................ 170
Figure 4.49: Cumulative Real Cost of FPSOs in Malaysia and Australia for 25-year
Life-cycle Period ........................................................................................................ 171
Figure 4.50: NPV of FPSOs plotted against their capital cost (NPV calculated with
only expenses) ............................................................................................................ 173
expenses and revenue from oil production) ............................................................... 174
Figure 4.52: Present Value of Australian FPSOs chosen for study for 10-year life-
cycle period ................................................................................................................ 175
cycle period ................................................................................................................ 175
Figure 4.54: Present Value of Malaysian FPSOs chosen for study for 10-year life-
cycle period ................................................................................................................ 176
cycle period ................................................................................................................ 176
Figure 4.56: NPV Profile for Spread Moored FPSOs................................................ 178
xviii
Figure 4.57: NPV Profile for External Turret Moored FPSOs ................................. 178
Figure 4.58: NPV Profile for Internal Turret Moored FPSOs ................................... 179
Figure 4.59: NPV Profile for Riser Turret Moored FPSOs ....................................... 179
Figure 4.60: Cost Proportions of Cendor II and Glas Dowr ...................................... 180
Figure 4.61: Cost Proportions of Nganhurra and Stybarrow Venture ....................... 181
Figure 4.62: Cost Proportions of Okha and Pyrenees Venture .................................. 182
Figure 4.63: Cost Proportions of Perisai Kamelia and Glas Dowr ............................ 184
xix
LIST OF TABLES
Table 1.1: Typical Differences Between Spread Moored and Turret Moored FPSO .... 3
Table 2.1: Scale Factors as per Froude’s Law of Similitude [62] ............................... 25
Table 2.2: Critical Literature Review .......................................................................... 56
Table 3.1: Specifications of Wave Maker System ....................................................... 65
Table 3.2: FPSO Model Details ................................................................................... 71
Table 3.3: Regular Wave Series................................................................................... 81
Table 3.4: Long-crested Random Wave Series............................................................ 82
Table 3.5: Short-crested Random Wave Series ........................................................... 83
Table 3.6: FPSOs for operability analysis and LCCA Study [2] ................................. 96
Table 3.7: Annual wave scatter table for Zone A to be used for operability analysis of
Perisai Kamelia FPSO.................................................................................................. 99
Table 3.8: Annual wave scatter table for Zone D to be used for operability analysis of
Berantai, Bunga Kertas and Cendor II FPSO .............................................................. 99
Table 3.9: Annual wave scatter table for Zone I to be used for operability analysis of
Kikeh FPSO ................................................................................................................. 99
Table 3.10: Annual wave scatter table at location 21°S 114°E to be used for
operability analysis of Nganhurra, Pyrenees Venture, Stybarrow Venture and
Ningaloo Vision FPSO .............................................................................................. 100
Table 3.11: Annual wave scatter table at location 19°S 116°E to be used for
operability analysis of Okha and Modec Venture II .................................................. 100
Table 3.12: Annual wave scatter table at location 10.5°S 125°E to be used for
operability analysis of Glas Dowr FPSO ................................................................... 101
Table 3.13: Representative Extreme cyclonic condition in NWA [61] ..................... 101
Table 4.1: Measured Natural Periods for FPSO [56] ................................................. 112
Table 4.2: Main Particulars of OTRC FPSO [42]...................................................... 119
Table 4.3: Mooring Line Particulars for OTRC FPSO [42] ...................................... 119
Table 4.4: Hydrodynamic coefficients of the chain and polyester rope [42]............. 120
Table 4.5: Metocean Data used by Kim et al [42] ..................................................... 120
Table 4.6: Comparison of Natural Periods & Damping Ratios ................................. 122
xx
Table 4.7: Mooring Line Details for Berantai FPSO Model ..................................... 129
Table 4.8: Vessel dimensions for hull length to beam ratio parametric study ........... 148
Table 4.9: Operating FPSOs in Malaysia 2016 [2] .................................................... 155
Table 4.10: Extreme Metocean Conditions used for the Analysis............................. 155
Table 4.11: Operating FPSOs in Australia 2016 [2] .................................................. 156
Table 4.12: Annual MPM relative motion for Perisai Kamelia FPSO in m .............. 160
Table 4.13: Annual MPM relative motion for Berantai FPSO in m .......................... 160
Table 4.14: Annual MPM relative motion for Bunga Kertas FPSO in m.................. 160
Table 4.15: Annual MPM relative motion for Cendor II FPSO in m ........................ 161
Table 4.16: Annual MPM relative motion for Kikeh FPSO in m .............................. 161
Table 4.17: Annual MPM relative motion for Nganhurra FPSO in m ...................... 162
Table 4.18: Annual MPM relative motion for Pyrenees venture FPSO in m ............ 162
Table 4.19: Annual MPM relative motion for Ningaloo Vision FPSO in m ............. 162
Table 4.20: Annual MPM relative motion for Okha FPSO in m ............................... 163
Table 4.21: Annual MPM relative motion for Modec Venture II FPSO in m ........... 163
Table 4.22: Annual MPM relative motion for Glas Dowr FPSO in m ...................... 163
Table 4.23: Annual maximum freeboard exceedance of FPSOs for extreme cyclonic
condition in NWA ...................................................................................................... 164
Table 4.24: Annual maximum freeboard exceedance of FPSOs for extreme metocean
condition in Kikeh field ............................................................................................. 165
Table 4.25: Cost Data of selected FPSOs [2] [141]–[153] ........................................ 167
Table 4.26: Cost, Dimensions and DWT of Cendor II and Glas Dowr ..................... 181
Table 4.27: Cost, Dimensions and DWT of Nganhurra and Stybarrow Venture ...... 182
Table 4.28: Cost, Dimensions and DWT of Okha and Pyrenees Venture ................. 183
Table 4.29: Cost, Dimensions and DWT of Perisai Kamelia and Glas Dowr ........... 184
xxi
LIST OF ABBREVIATIONS
2D Two Dimensional
3D Three Dimensional
CALM Catenary Anchor Leg Mooring
CAPEX Capital Expenditure
CF Compact Flash
CG Center of Gravity
DOF Degree of Freedom
DWT Dead weight tonnage
ET External Turret
FB Freeboard
FEED Front End Engineering Design
FP Forward Perpendicular
FPSO Floating Production Storage and Offloading
FSO Floating Storage and Offloading
IT Internal Turret
LBP Length between Perpendicular
LCCA Life-cycle Cost Analysis
LCG Longitudinal Center of Gravity
LCR Long-crested Random
LF Low Frequency
LNG Liquefied Natural Gas
LOA Overall Length
LPG Liquefied Petroleum Gas
LWT Light Ship Weight
MBOPD Thousand Barrels Oil Per Day
MWL Mean Water Level
MPM Most Probable Maximum
NPV Net Present Value
NWA North Western Australia
xxii
OPEX Operating Expenditure
OTRC Offshore Technology Research center
PTF Paddle Transfer Function
RAO Response Amplitude Operator
RCIS Royal Institution of Chartered Surveyors
RTM Riser Turret Mooring
SCR Short-crested Random
SLR Single-Lens Reflex
SPM Single Point Mooring
STP Submerged Turret Production
UTP Universiti Teknologi PETRONAS
USD US Dollars
VCG Vertical Center of Gravity
WF Wave Frequency
xxiii
LIST OF SYMBOLS
𝐵 Beam of vessel
B8 Revenue generated
B9 Salvage value/Scrap value
𝐶𝐴 , 𝐶𝑑 , 𝐶𝑚 Added mass, drag force and inertia force coefficients respectively
C1 Capital costs
C2 Installation costs
C3 Operation costs
C4 Maintenance costs
C5 Refurbishment/Replacement costs
C6 Downtime costs
C7 Decommissioning/Disposal cost
D Diameter of cylinder
𝐹𝐷 Drag Force
𝐹𝐼 Inertia Force
𝐹𝐵 Freeboard
H Wave height
𝐻𝑠 Significant wave height
L Wave Length
N life expectancy of FPSO
𝑃𝐸 Short term probability of exceedance
𝑃𝑉𝑛 Discounted present value
𝑅𝑚 Relative motion of FPSO
Rn Real costs incurred per year
𝑆𝑅 (𝜔) Response Spectrum
𝑆(𝜔) Wave Spectrum
T Wave Period
𝑇𝑝 Peak Period
𝑈𝑟𝑒𝑙 Relative velocity between structure and wave
xxiv
𝑋̈ Structural acceleration
d Water depth
g Acceleration due to gravity
𝑘 Wave number
𝑚 Mass of the vessel
𝑚0𝑅 Zeroth order moment of area under the relative motion spectrum
t Instantaneous time
u, v Fluid particle velocity in x, y direction respectively
𝑢̇ , 𝑣̇ The fluid particle acceleration in x, y direction respectively
(x, y, z) Cartesian co-ordinate system used for the definition of wave
kinematics
γ Spectrum peak enhancement factor
∆ Weight displacement of FPSO
Δt Time interval
Displaced volume of vessel
𝜂, 𝜂𝑎 Wave amplitude
𝜂𝑡 Wave Profile
Roll motion
Pitch motion
Yaw motion
λ Linear scale factor
Density of water
τ Spectral shape parameter
Potential Function
Incident wave potential
Scattered wave potential
Circular frequency of the incident wave
Peak circular wave frequency
xxv
CHAPTER 1
INTRODUCTION
1.1 Chapter Overview
In this chapter, the history and advantages of Floating Production Storage and
Offloading (FPSO) systems are explained together with the need for conducting the
dynamic analysis of FPSO, how cost and motion are interrelated while choosing FPSO
configurations and the relevance of conducting Life Cycle Cost Analysis (LCCA) for
the same. Further, the objectives of the present study are stated and, the scope of the
study is furnished. Finally, the significance of the project is addressed.
1.2 Background of FPSO
The history, advantages and structure of FPSO are detailed in the below sections.
1.2.1 History of FPSO
Over the past 40 years, ship – shaped offshore units have proven to be reasonably
reliable solutions for deep water offshore fields. These include FPSOs and FSOs
operating in harsh environmental conditions and waters of more than 1500 m depth.
Even though oil storage and shuttle tanker – mooring facilities using converted trading
tankers existed in late 1960s, the entry of ship – shaped units to the offshore industry is
not known accurately. The first such vessels were connected by hawsers to catenary
anchor leg mooring (CALM) systems. These then evolved into the familiar systems
employing single – point mooring. The first FPSO was Arco in the Ardjuna field in the
Java Sea offshore Indonesia in 1976. This was a concrete barge with steel tanks, used
1
to store refrigerated liquefied petroleum gas (LPG) moored to a buoy using a rigid arm
system in 42.7 m water depth. The first tanker-based single-point moored FPSO facility
for oil is said to be the Castellon for Shell offshore Spain in 1976. This facility was
meant to produce oil from a subsea completed well, some 65 km offshore Tarragona. It
began operations in 1977, and was designed for a 10-year field life. Compared to these
early days, floating production systems have now evolved to a mature technology that
potentially opens the development of offshore oil and gas resources that would be
otherwise impossible or uneconomical to tap. The technology now enables production
far beyond the water-depth constraints of fixed-type offshore platforms and provides a
flexible solution for developing short-lived fields with marginal reserves and fields in
remote locations where installation of a fixed facility would be difficult [1]. Many
FPSOs have been installed and operated worldwide and many new FPSOs will be
installed in the coming years. Out of a total of 178 FPSOs operating now globally, 9
are in Western Australia offshore, 8 are in Malaysia offshore, 2 are in Thailand offshore
and 9 are in Indonesia offshore regions. Another 5 FPSOs are ready for redeployment
in Australia and 2 in Malaysia. Figure 1.1 shows the distribution of FPSO worldwide
[2].
Figure 1.1: Worldwide Distribution of FPSO 2017

(source: Offshore Magazine, 2017)
2
1.2.2 Structure of FPSO
FPSOs are usually ship shaped floating structures, even though other variety shapes of
FPSOs are coming into industry like the cylindrical FPSO. An FPSO is equipped with
provisions for production, storage and offloading hydrocarbons. A typical FPSO
consists of mooring area, process area, storage and offloading systems, hull structure,
utilities and marine systems, dynamic positioning system or station keeping system and
means of escape and evacuation. Mooring systems can be spread mooring or turret
mooring. There are different types of turret moored FPSOs, they can be external turret
moored, internal turret moored, riser turret moored or with submerged turret production
systems. For spread moored FPSOs, riser hang-off points are attached at the side of the
vessel. For turret moored FPSOs, turrets should be located at or near the bow so that
the FPSO can weathervane passively without thruster assistance. Table 1.1 summarises
typical differences between spread moored and turret moored FPSO.
Table 1.1: Typical Differences Between Spread Moored and Turret Moored FPSO
Description Spread Moored Turret Moored
Can be used in mild to Can be used in mild to
Environment
moderate environment extreme environment
Vessel
Fixed orientation 360° weathervaning
Orientation
Adapts to various riser Location of turret requires
Riser Systems
systems robust design
Station keeping Variable offsets (due to low
Minimized offsets
Performance frequency wave)
Dependent on relative vessel- Weathervaning capability
Vessel Motions
environment directionality reduces motion
Offloading Dependent on vessel- Aligned with mean
Performance environment orientation environment
FPSOs can have double skin hull or single skin hull. While most of the newly built
hulls have double hulls, the old converted tankers have only single hulls [2]. The
process area consists of the gas separation and compression systems and metering
systems. Storage is provided in the center tanks, with water ballast in the double bottom
3
(if fitted) and side tanks. A common way of exporting crude oil from a FPSO is by
shuttle tanker transport to a shore terminal. The export may take place by direct transfer
from the FPSO to a shuttle tanker by a hose, or by transfer from FPSO via separate
offloading system (hose-riser pipeline- riser). The former method is mainly used when
the production/storage unit is a ship or barge shaped floating production, storage and
offloading unit (FPSO). Both alongside transfer and tandem transfer methods can be
used, depending on operational criteria [3]. Figure 1.2 shows the various parts of an
FPSO.
Figure 1.2: Various Parts of FPSO

(source: Marine Insight, 2011)
1.2.3 Advantages of FPSOs
Advantages of FPSOs are
• Early Production and huge storage capacity.
• Easy to remove and reuse.
• Reduced upfront investment.
• Can be used in any water depth.
• Abandonment costs are less than for fixed platforms.
4
• Retained values; because they can be relocated to other fields.
• Earlier cash flows because they are faster to develop than fixed platforms.
• Ample deck space reducing risk of oil spilling.
1.3 Relevance of Cost and Motion Analysis of FPSO
Extreme FPSO motion can lead to significant operational downtime, disrupting

processing and production of oil on board FPSO and offloading to tankers. Excessive
heave and pitch motion of FPSO can result in green water on board FPSO damaging
production equipment and fatalities in some case. Roll motion of FPSOs should be
within ±5° to enable crew habitability. Excessive roll motion of FPSO can also lead to
green water impacts on deck. Extreme surge and yaw motion of FPSO can disrupt
offloading operation during tandem and side-by side offloading configuration by
driving off the oil tanker or by imposing risk of collision with them. In turret moored
FPSOs, excessive sway and yaw motion can result in fishtailing phenomenon, again
disrupting offloading operations [4]. Hence all six FPSO motions need to be minimised
to ensure safe, profitable and efficient oil production using FPSOs. Hence an extensive
parametric study covering hull, mooring and metocean parameters are conducted in this
research and the results are presented, identifying the range of parameters over which
FPSO motions are minimum.
Spread moored and turret moored FPSOs are two main types of FPSO
configuration. While spread moored FPSOs have a fixed orientation, turret moored
FPSOs can weathervane reducing the loads on mooring lines. The FPSO motion
behaviour in varying wave heights are studied with and without wind and current in the
presence of wave to understand the range of FPSO motion amplitudes in the six degrees
of freedom, trend in FPSO motion variation with wave height and to study the influence
of current and wind in FPSO motions. Especially, excessive heave and pitch motion in
extreme wave heights can result in green water on FPSO deck. If the relative motion of
FPSO (heave) and wave is higher than 3m, medium to high risk level green water
impacts can occur, resulting in operational downtime. This in turn can influence the
5
life-cycle cost of the system. Hence a downtime analysis is done to compare the loss in
production when spread moored and turret moored FPSOs are used. Relative motions
are found by performing dynamic analysis and probability distribution for freeboard
exceedance are generated. If the free board exceedance is above 3 m, this can result in
downtime cost. Downtime cost computed by measuring motion responses in the event
of green water is one of the cost factors affecting the total life-cycle cost of the FPSO
and associated mooring line system.
FPSO hull and mooring system should be designed such that the six motions of
FPSO are minimum as well as its cost. Cost of FPSO hull is only 10-15% of the cost of
mooring system. Hence proper care should be taken while using converted hulls in
FPSO, as it is not specifically designed for the metocean conditions it is installed and
can sometimes lead to frequent mooring line damages. Hence cost and motion are two
important aspects while choosing an FPSO configuration to maximize operational time
and profit. The study was done to compare the different FPSO configuration in terms
of motion performance and cost and can significantly aid in the decision-making
process in the FFED phase of FPSO projects. This research is a unique attempt to
identify the factors affecting the choice of FPSO in terms of its motion performance
and cost.
1.4 Relevance of Dynamic Analysis of FPSO
Floating Production Storage and Offloading System (FPSO) has proven to be one of
the most effective means to carry out oil drilling and processing especially in deep
waters. Malaysian and West-Australian waters are prominent regions in supplying the
energy resource for the global needs with over 15 FPSOs operating in this region from
water depth ranging from 50 m to 3000 m. Even though Malaysian and West-
Australian waters are calm when compared to the North Sea, proper consideration
should be given to the various criteria like water depth and metocean data. FPSO
motions are minimised by employing proper mooring and hull configurations which
has sufficient storage facilities, deck space, required natural period and designed to
project specific metocean criteria. Often, this is done through an iterative procedure,
6
where FPSO hull sizing is determined through a reverse iteration to achieve target
natural periods and minimum FPSO motions. A sensitivity study covering the
metocean, hull and mooring line parameters can aid in minimising the FPSO motion by
identifying the critical parameters, their range of effectiveness and trend of FPSO
motion to those varying parameters. Hence model tests and extensive simulations have
been performed using SESAM suit of programs – SESAM HydroD and SESAM DeepC
under first order wave loads, current and wind to study FPSO motion responses in wave
frequency ranges through uncoupled and coupled analysis approach to identify the
parameters and their range of application to ensure minimum FPSO motion.
Since FPSOs under study have ship- like forms with one plane of symmetry and
with the longitudinal dimension much larger than the transverse one with large aspect
ratio (L/B ratio) in the range of 5 to 6, it makes them particularly sensitive to the
direction of incoming waves. Sometimes the waves, winds and current can be quite
non- parallel, and subject the vessel to quartering or beam seas that can significantly
influence the response of a ship – shaped vessel. To determine the stress distribution
on such a structure the motions of the structure should be known in addition to the wave
forces on it.
Spread mooring system helps to maintain a fixed orientation of FPSO in global

coordinates [5] under calm weather conditions while turret mooring system helps in
timely deployment of the mooring system under hostile weather conditions, preventing
further damage to the mooring lines. If the mooring system is not specifically designed
for the location, the repair and maintenance of the same prove costly. It is thus
necessary to have the knowledge of the FPSO motion responses in the preliminary
design phase, where appropriate choice of the mooring system can be made. Also, for
short projects of 10 years, most companies use converted tankers. The suitability of
these converted tankers or even newly built tankers in these projects under site specific
metocean condition can be assessed by the computation of motion responses by
conducting a dynamic analysis of the floater. An accurate prediction of hydrodynamic
response involves the use of model testing and numerical simulations in a balanced
way. Hence dynamic analysis of FPSOs to compute vessel response and wave loads
acting on it are extremely important in the initial design phase of any FPSO project.
7
1.5 Relevance of Life-cycle Cost Analysis for FPSO
Every offshore structure has a ‘life’, starting with the planning, design or development
of the structure, followed by resource extraction, production, use or consumption and
finally end of life activities including decommissioning or conversion of structure while
the oil field is nearly exploited. The life cycle costing method is an economic evaluation
technique which is well suited to compare alternative designs with different cost
expenditures over the project life. Generally, life cycle cost is defined as sum of all
costs over the full life span of a system, which includes purchase price, installation cost,
operating costs, maintenance and upgrade costs, and remaining value at the end of
ownership or its useful life. The changes made in one phase of life cycle of an offshore
structure can have a drastic effect in the succeeding phases. For example: reduction in
initial cost of the hull by using converted old tankers may result in higher maintenance
cost or vice versa; production costs of FPSOs can be higher when compared to the cost
of oil available from the field; and, choosing turret mooring instead of spread mooring
in calm waters. Hence LCCA should be performed early in the design process while
there is still a chance to refine the design to ensure a reduction in life cycle costs.
Designing of each phase needs to be done carefully and well planned as the safety of
structure is very important in this case. If decisions are taken based on effects in only
one part of the life cycle, it may do more harm than good. LCCA assists in choosing
the conceptual design which is best in both performance and safety. Hence a thorough
life cycle cost analysis of the FPSO is required to have practicably profitable, safe and
successful oil production and consumption.
1.6 Problem Statement
The number of floaters operating in Malaysian and West-Australian waters is mounting

and FPSOs are prominent structures contributing to the exploitation of oil and gas
resources in these regions. From a detailed review of the existing FPSOs off Australia
and Malaysia, it was found that most of the operating FPSOs are converted tankers. In
2016, out of the 10 FPSOs operating in Australia, 7 of them are converted tankers
modified to serve the purpose and all the operating FPSOs in Malaysia (5 FPSOs) are
8
converted tankers [2]. The conversion of the FPSO includes replacing certain parts or
sometimes all of them except the hull [6]. Also, converted FPSOs are frequently
removed to replace the mooring system and usually supplied with turret mooring to be
more resistant to environmental loads. Sometimes this process will be costlier when
compared to building new purpose-built vessels if the mooring system must be replaced
more often than planned replacements. Since the CAPEX of the mooring system is
very high when compared to that of the hull, care should be taken while designing a
newly built hull or choosing a converted one.
Hull motions should be minimised to achieve increased operational time and

maximum productivity. Extreme surge, sway and yaw motion can disrupt offloading
operations while excessive roll motions could disable crew habitability [4]. Excessive
heave, roll and pitch motions can cause green water on deck and can cause downtime
[7]. The relative motion due to combined heave, roll and pitch should not exceed 3 m
as it can damage equipments on the deck [8]. Hence all the six FPSO motions should
be minimum to ensure longer operational time. FPSO motions can be controlled by
choosing hull and mooring system with minimum motions when subjected to site
specific metocean conditions. Often, the hull sizing is done through a reverse iteration
to achieve target natural periods and minimum motion responses. These iterative
processes can be time consuming and cost incurring due to the use of special softwares
and personnel with specific skills. Hence parametric studies covering hull dimensions,
loading conditions and mooring line parameters are conducted to find an effective
initial solution to be used in this iterative procedure to reduce this design process time.
Along with that the motion response of turret moored and spread moored FPSO in
varying wave heights are studied to compare their motion behaviour in increased wave
heights. This is important as the West Australian region is often prone to extreme
cyclonic conditions [4] and can result in high wave heights which can result in damage
to FPSO deck due to green water impacts.
Often, industries choose turret mooring in hostile weather conditions and spread
mooring in calm weather conditions to avoid these iterative procedures. These decisions
should be backed up by proper research data, cost calculations and motion response
study and that is achieved through this research. The cost of the FPSO configuration
9
should also be minimum when compared to its motion, for that to be an effective
solution. Life-cycle cost study for the existing FPSOs in Malaysia and Australia are
never reported and a cost study would be ideal to assess the long-term worth of these
assets. Since FPSOs are rarely demolished during decommissioning and has high resale
values and its reusable in other locations, an initial construction cost can mask its long-
term asset value. A life-cycle cost analysis reporting the initial cost, down time cost due
to green water, net present value for short-term and long-term use would be ideal as this
will strengthen the decision made during FEED phase and can avoid huge expenses
later in the life of FPSO.
The research addresses the issues mentioned above through dynamic analysis of
FPSOs using state of the art hydrodynamic analysis software SESAM and life-cycle
costing from first principles.
1.7 Objectives of the Study
To solve the issues mentioned above, the objectives of this research are as follows:
1. Validation of software simulation model and simulation procedures by

comparing with model test results to assess the capability of simulation model
and analysis procedures adopted to predict wave frequency motion responses.
2. To compare spread moored and turret moored FPSO behaviour in varying wave
heights to assess the suitability of these options in extreme weather conditions.
3. To investigate the effect of mooring line azimuth angle, mooring line length,
spread mooring fairlead location, hull length to beam ratio and hull loading
condition and water depth on FPSO motions and identify range of parameters
over which FPSO motion will be minimum and to produce parametric charts.
4. To calculate the down time cost due to green water by computing the relative
motion and freeboard exceedance for the FPSOs operating in Malaysia and
Australia when subjected to wind generated sea conditions and in turn assess
10
the operability under site specific metocean condition where the FPSOs are
installed.
5. To identify cost effective FPSO configurations for 10-year and 25-year use by
comparing life-cycle cost, downtime cost and Net Present value of chosen
FPSOs in Malaysia and Australia through data collection and life-cycle cost
analysis.
1.8 Scope of the Study
The scope of the study is as follows:
 The FPSO is considered free to move in all six degrees of freedom, i.e. surge,
sway, heave, roll, pitch and yaw as shown in figure 1.3 and is subjected to
regular, unidirectional and multidirectional random waves under the action of
wind and current.
 The FPSO is anchored to sea bed using horizontal spread mooring system
modelled using soft springs with negligible mass and damping in model tests
and only horizontal excursion of the mooring line is considered for the
experimental study.
Figure 1.3: Six Degrees of Freedom of FPSO
11
 First order 3D Diffraction potential theory is used to calculate the wave load on
FPSO and Modified Morison equation is used to calculate the mooring load in
coupled analysis[9] using commercial software. Second order forces and
responses are not included in the study [10]. Uncoupled analysis using
commercial software calculates motion by distributing free surface source
potentials across the hull surface as the Green’s functions and using Green’s
theorem in frequency domain [11].
 The research parameters are metocean conditions, water depth, hull loading
conditions, hull length to beam ratio, mooring line length, mooring line azimuth
angle, and position of spread mooring fairleads.
 The metocean and market conditions are limited to Malaysian and Australian
waters to generate the FPSO motion response and life cycle cost data of FPSOs
in this region.
 Market fluctuations in capital, operation and maintenance, lease rate and income
from oil are not considered while calculating life-cycle cost in section 4.8 – 4.10
and section 4.12 for the ease of comparison of different FPSO configurations.
Hence a sensitivity analysis has been performed to incorporate the market
fluctuations in section 4.11 and the variation in NPV and life-cycle cost in a
good and bad market condition is studied.
 Operability analysis is conducted only for green water incidents and downtime
cost is calculated only if downtime is reported for chosen FPSOs under the
action of wind generated waves. Site specific annual wave scatter table based
analysis was performed.
 FPSO is assumed to be producing oil to its maximum capacity and life-cycle

cost is calculated based on capital cost, operation and maintenance cost, revenue
from oil production and downtime cost due to green water events when
subjected to wind generated sea condition only.
 Accidents, shut downs due to factors other than green water, mooring line
breakages and profit from oil production are not considered in LCCA.
12
1.9 Significance of the Project
FPSO and the associated mooring system is chosen for a project mainly based on its
performance and the profit from the project. An initial cost calculation of the project
can help in choosing the best possible FPSO system and the mooring type based on its
life-cycle cost. But the cost alone is not sufficient to determine the mooring system to
be used. FPSO motions are greatly dependent on the site specific metocean conditions
and water depth, apart from the hull condition and mooring system data. Hence the
results generated by conducting the parametric study can be used in obtaining an idea
of the motion performance of similar configured FPSOs during conversion, based on
the dimension of FPSO hull and mooring line length and configurations, in addition to
the metocean data and water depth details, which covers the major environmental
conditions and structural details for the FPSOs in Malaysia and Australia and thus can
be used as a reference while planning for a converted tanker or newly built hull. If any
of the converted tankers are not suitable for the designated oil field, the RAOs generated
for the location can be used to generate a new purpose-built vessel, while relative
motions generated for FPSOs in Malaysia and Australia gives the trend and magnitude
of subsequent freeboard exceedance to be expected. Hence precaution can be taken
while choosing the FPSO dimension to avoid green water phenomenon and further
downtime cost. Also, the life cycle cost data generated for the FPSOs in Malaysia and
Australia gives information about their net asset value and cost details during their
construction, which can be used as a reference for upcoming projects in this region. In
short, this project brings the technical and cost aspects of FPSOs under one umbrella,
enabling the design of FPSOs with maximized productivity and field life, leading to
maximum financial benefits from the project.
1.10 Overview of the Thesis
Chapter 1 introduces the study conducted. Chapter 2 reviews the literature pertaining
to the subject of this thesis. The reported researches are classified in eight categories
and a general description of each category is given. After reporting the past researches
in this area, a critical discussion is presented along with a table of gaps in the studies.
13
Chapter 3 details the methodology adopted to conduct the dynamic analysis of
FPSOs using software simulation, model tests and numerical code. The modelling of
FPSO using SESAM Genie V5.3-10 is detailed along with the analysis procedures used
for frequency domain analysis in SESAM HydroD V4.5-08 to generate RAOs, relative
motions and motion response time series in SESAM DeepC V5.0-06. The physical
model testing setup, facilities, model specification and laboratory tests conducted are
also detailed. The life-cycle costing methodology used to compute the life-cycle cost
of FPSOs are detailed towards the last sections of this chapter. Methodology to evaluate
the significant cost for different FPSO options, adding the expenditures and subtracting
the revenues and choosing the final design with the minimum life cycle cost is detailed
including data collection, sensitivity analysis and calculation of Net Present Value
(NPV).
Chapter 4 presents the results of the study. The validation of the numerical models,
graphs generated by conducting the parametric study, RAOs, relative motions,
downtime cost and cost data for FPSOs in Malaysia and Australia are presented. All
results are accompanied by critically detailed discussions.
Chapter 5 concludes the study conducted. An overview of the research work carried
out is given based on the problem discussed throughout the thesis, addressing each
objective of the study. Finally, recommendations for further improvements and future
works related to the research work carried out are proposed.
14
CHAPTER 2
LITERATURE REVIEW
In this chapter, the research studies on the dynamic response and the life cycle cost
analysis of the FPSO are reviewed. Major focus is given to the theoretical background
for the FPSO responses, experimental and numerical works conducted in the past
decades and the life cycle cost analysis procedures adopted. The reported studies are
classified into eight major sections and are presented here. Finally, a critical review of
the conducted literature survey is furnished along with the gap study.
2.2 Reported Studies
In this literature survey, the reported researches are grouped into eight categories based
on the research direction. The studies related to the dynamic analysis of single point
mooring systems and ships using model testing and numerical investigations, dynamic
analysis of FPSO using model testing and numerical investigations, coupled and
uncoupled analysis of FPSO, hydrodynamics of FPSO, operability studies on FPSO
including parametric studies and green water impacts and finally, life-cycle cost
analysis studies and LCCA for FPSO are reviewed in detail in the following sections.
A significant amount of studies were done to investigate the dynamic behavior of ship
shaped vessels and single point mooring systems as done by Pinkster et al [10] by
conducting model tests on single point mooring attached to a tanker. Also, it should be
noted that considerable research work has been reported on numerical and experimental
investigation of dynamic response of FPSO in the presence of wind, wave and current.
For example, Wichers [11] initiated a comprehensive study for numerical simulations
of a turret moored FPSO in irregular waves with winds and Garett [12] developed a
numerical model to allow accurate and efficient fully coupled global analysis of
Floating Production systems including the vessel, mooring system and the riser system.
However, very few studies have reported the comparison of motion behaviour of spread
moored and turret moored FPSOs under varying environmental loads and their
susceptibility to green water and subsequent downtime in their life-cycle. Also, it is
noted that very few studies are conducted on life-cycle costing methodology for FPSO,
while no studies have reported cost comparison for ET, IT, RTM and SM FPSOs and
impact of choosing converted hulls on their life-cycle cost, albeit that life cycle costing
analysis for ships have been reported as investigated by Gratsos et al [13]. Detailed
knowledge gaps from previous studies are discussed later in the chapter in section 2.3.
2.2.1 Dynamic Analysis of SPM and Ships
The preceding developments in the field of dynamic analysis of ship shaped vessels
have given ample insight on the dynamic behavior of FPSOs. Researches were
conducted to study the dynamic response of SPM and ships using model testing and
numerical investigations.
2.2.1.1 Model testing on SPM and ships
Pinkster et al [10] studied the role of model tests in the design of single point mooring
(SPM) terminals attached to a tanker. They gave necessary information regarding how
to set up test programs, the scope of tests, the characteristics simulated, and
measurements carried out, possible sources of error and analysis of results. They
concluded that the results from the model tests should be used without applying any
correction to rectify error due to scale effect because of uncertainties exist concerning
the drag coefficient of prototypes. Since Froude scale of models are used, viscous force
will be overestimated, but the significance of inertia and wave effect on the structure
16
can be effectively presented. Mansard et al [14] conducted physical model tests to show
the effect of wave grouping on moored vessel response. It was shown that the wave
grouping present in irregular waves is an important parameter in assessing vessel
response and a correct reproduction of bounded long wave component is required in
model studies to get a realistic response of vessels. Chakrabarti [15] gave emphasis on
the sea keeping and towing tests of an offshore structure on station or in transit and
gave an insight in to the difficulties in conducting tests and the remedial measures used
in the setup. Fournier et al [16] conducted a physical model test to study the ship
response to establish critical wave conditions that cause excessive vessel motions for
safe/ efficient cargo transfer.
2.2.1.2 Numerical investigation of SPM and ship motions
The preliminary step in the calculation of vessel responses are the computation of added
mass, damping and exciting force which is then incorporated to the equation of motion
along with mass and restoring matrix. Korvin-Kroukovsky [17] developed the first
motion theory in this field which was used for computation of vessel responses. It was
later found that, there were inconsistencies in this theory in the mathematics and
Salvesen et al [18] modified it. Newman [19] developed unified slender body theory,
which was later extended to a diffraction problem by Sclavounos [20].
Unlike Strip theory and unified slender body theory, 3D methods like Green
function method and Rankine Panel method can give detailed force distribution over
hull surface of large structures. Chang et al [21] and Inglis et al [22][23] proposed the
Green function method which is later extended by Wu et al [24] and Chen et al [25].
While Green function method uses a time harmonic function with forward speed on the
free surface, Rankine Panel Method which was initiated by Dawson [26] uses Rankine
sources on body surface as well as free surface, allowing more general free surface
conditions to be used. These 3D methods have proved to give better agreement with
17
the experimental data [27], at the same time, strip theory requires only less
computational effort and gives reliable reasonable results for engineering applications.
Also, the 3D method requires the user to input the full 3D vessel co-ordinates while
Strip theory requires only the cross-sectional water line breadths, draft and area, if the
conformal mapping technique is used.
Vugts [28] conducted experimental and theoretical investigations to determine the

hydrodynamic coefficients and exciting forces in swaying, heaving and rolling for 2D
cylinders with various cross-sections in beam sea condition. The results can be used to
predict the hydrodynamic coefficients and exciting forces of matching Lewis forms or
actual section fits. The sway and heave results showed good agreement. The simpler
force calculation derived by Newman [29] fits the experimental results for exciting
forces by Vugts [28], but it does not predict the phase angles. His studies also prove
that the force and moment calculation using Froude-Krylov hypothesis underestimates
the actual wave force and moment. Salvesen et al [18] presented the equations of
motion for ships which consists of two sets of independent linear coupled differential
equations. Since the floating structure under consideration has lateral symmetry, the
surge, heave and pitch are not hydrodynamically coupled with sway, roll and yaw. He
also derived the equation for the calculation of exciting forces in these degrees of
freedom.
Journee [30] used two parameter Lewis conformal mapping method to develop a
quick strip theory calculation. His approach helps in avoiding the major human error
in giving inputs of ship offsets. For this method, only the cross-sectional water line
breadths, draft and area is needed. Das et al [31] investigated the coupled sway, roll
and yaw responses of a floating body with hydrodynamic coefficients derived from
Frank’s close-fit curve. His results showed that yaw motions exist for a floating body
under the action of regular waves in beam sea condition if the center of gravity of the
floating body does not coincide with assumed position of co-ordinate origin and that
the magnitude of yaw rotation decreases as the wave period decreases. Hem Lata et al
18
[32] compared the hydrodynamic coefficients obtained using conformal mapping and
state of the art analysis software AQUA and SESAM. The recurrent form of
Bieberbach Method was used for conformal mapping and the results were agreeing with
a maximum error less than 10%. Fan et al [27] computed ship motions based on
methods provided by Salvesen et al [18] in time domain and seconds in applying a
correction factor to the roll damping coefficients to achieve accurate results. Momoki
et al [33] gives calculation methods for the pressure acting on the hull for analyzing the
ship structural response in waves. Ship motion was calculated using a nonlinear strip
method. Using this as input to the CFD program, pressures acting on the hull were
found. This pressure distribution was used to analyze the ship’s structural response
using Finite Element Method (FEM).
2.2.2 Dynamic Analysis of FPSO
The foregoing developments in the field of FPSOs have given generous insight on the
dynamic behavior of ship shaped FPSO in unidirectional random waves and irregular
waves. Researches were conducted to study the dynamic response of FPSO using both
model testing and numerical experiments utilising coupled and uncoupled approach.
Wichers [11] initiated a comprehensive study for numerical simulations of a turret
moored FPSO in irregular waves with winds and currents, neglecting the inertia and
damping effect of mooring lines. A procedure to obtain practical values of added mass
and damping to calculate the nature of the stability and natural frequencies of the system
was also given by him. The uncoupled analysis technique was later found to give
smaller values for the motion response after verification through a multitude of
experiments and analysis techniques by several scholars.
Low frequency and wave frequency motions of FPSOs due to environmental loads
were studied by Jiang et al [34], putting forward the possibility of large amplitude slow
drift oscillations in the horizontal degree of freedom due to low frequency wave
19
components. However, heave, roll and pitch motions are significantly affected by the
presence of first order wave excitations [35]. Heurtier et al [36] compared the coupled
and uncoupled analysis for a moored FPSO in harsh environments and suggested that
the uncoupled analysis results are efficient to be used in the early design phase of the
mooring system. There was relatively good agreement between the uncoupled and
coupled analysis values even though the maximum values were different; while
Wichers et al [37], [38] established the need for including coupling effects between
FPSO hull and mooring lines and the effect of viscous damping. These studies showed
that, uncoupled analysis will give large errors in the case of FPSOs due to the mooring
line interactions; fully coupled time domain analysis is required in obtaining accurate
results, was their final respective conclusion.
Lou et al [39] studied the FPSO motions using both coupled and uncoupled time
domain analysis methods and suggested coupled analysis should be the preferred
method for investigating FPSO responses; Lou also concludes that model testing should
be combined with numerical analysis for accurate prediction of system responses as
model testing alone is not sufficient. Low et al [40] developed a computer program to
calculate the coupled motion response of floating structures. The program used both
frequency domain and time domain approaches to estimate the response. The results
obtained after the simulation of a spread moored FPSO in 2000 m water depth under
the action of wave with 100 m significant wave height, matched very well for the two
approaches. The frequency domain method gave good results where the geometric non-
linearity is not prevalent.
Tahar et al [41] developed a hull/mooring/riser coupled analysis program for a

turret moored FPSO in the presence of wind, wave and current to understand the motion
characteristics, coupling effects and the role of various hydrodynamic contributions.
The results obtained were compared with MARINS wave tank test results. The
numerical time domain program developed by Kim et al [42] to calculate vessel and
line dynamics were validated using OTRC wave basin model testing results, and are
20
matching very well except for roll motions. This discrepancy can be attributed to the
use of truncated mooring system which underestimates the dynamic mooring tension.
Some of the reported studies on FPSO were carried out by providing additional
attachments on FPSO and by varying the usual ship shape of FPSO. Priyanto et al [43]
examined the wave exciting surge forces on FPSO when provided with a submerged
plate on lee side of FPSO using numerical method based on diffraction theory and the
results were verified by conducting experimental tests. He concluded that, at low
frequencies, the surge forces are effectively reduced due to the presence of submerged
plate. Siow et al [44] provided preparatory procedures for round shaped FPSO model
testing and details on mooring design and model set up. The vertical motion of FPSO
experienced only wave frequency motions. Siow et al [45] also conducted model tests
to study the effect of different mooring system on FPSO motions. He concluded that
the mooring system do not have significant effect on FPSO motions in wave frequency
range. He also showed that in wave frequency ranges, absence of mooring lines does
not produce any difference in results and is matching with the experimental results.
Long term FPSO responses are found to be critical when compared to other sea
states by Vázquez-Hernández et al [46] while Rho et al [47] has studied the FPSO
motion responses in most conservative environmental condition with 100-year return
period. Fontaine et al [48] reassessed the reliability of mooring system of an existing
FPSO in West Africa using field metocean conditions and compared the different
design approaches for FPSO. The effectiveness of response based design is emphasized,
noting the main drawback as computing time. Ma et al [49] conducted numerical
experiments using fully nonlinear potential theory and experimental investigation to
study the interactions between a simplified FPSO and focusing waves. The incoming
waves produced during experimental tests were reproduced in numerical wave tank by
in cooperating a self-correction time domain technique which produced agreeable
results with experimental outcome. Feng et al [35] and Chen et al [50] have
21
demonstrated the efficacy of commercial softwares like ANSYS and SESAM in the
modelling and meshing of the FPSO vessel.
Recently, Ji et al [51] has studied the influence of middle water arch in FPSO
motion response and its capability of suppressing FPSO motions except heave. Kang
et al [52] conducted fatigue analysis on mooring lines of a spread moored FPSO and
observed that it is highly impacted by the wave frequency motions of FPSO. Lopez et
al [53] conducted experimental investigations using hybrid passive truncated FPSO and
mooring model to assess hydrodynamic performance of a proposed FPSO in Gulf of
Mexico at a water depth of 1000 to 2000 m. The surge motion of FPSO was found to
be twice in non-collinear environmental loads when compared to be under the influence
of collinear loads and the mooring lines are more sensitive to dynamic response in non-
collinear condition. Hong et al [54] investigated the effect of impact load by steep
waves on FPSO bow using model tests. The impact loads were found to be increasing
with wave steepness and so a recommendation was given to include steep waves in
addition to the representative wave condition of significant wave height and pitch
forcing period while applying structural load during design of FPSOs in North Sea.
2.2.3 FPSO Motion Response Using Uncoupled and Coupled Analysis
In the traditional methods of vessel response calculation, an uncoupled approach to

calculate the same was utilized by considering the load effects from moorings modelled
as linear restoring forces applied at nodes of the finite element model of the FPSO.
Once the vessel response is calculated, these motions were applied as terminal
excitation at the top end of the mooring to calculate the line dynamics. But this
approach has many drawbacks such as:
• The current induced mean loads on mooring lines are not considered;
• The damping from mooring lines on LF vessel motion must be simplified;
22
• Mooring line dynamics is not considered while calculating vessel response.
As the water depth increases, the effect of these drawbacks will increase the
inaccuracy of the results [55]. In shallow waters, floater motions are triggered to a large
extent by the fluid forces on the floater itself. As the water depth increases, the length
of the mooring line increases as does the coupling effects between mooring and FPSO
[56], [57]. Where the non-linearities are not prevalent, uncoupled frequency domain
analysis gives good results [40], [58]. So, to study the floater responses due to the
change in hull dimensions, hull loading conditions and for water depth parametric
studies below 100 m, uncoupled analysis seems to be a good choice and can be time
saving. First order motion responses could be studied using an uncoupled approach
where non-linearities from mooring lines are not the primary concern of the study,
although mooring lines affects the mean position of FPSO in low frequency regime
[59], [39]. However, change in these mean offsets are not the primary concern of this
research. Wave frequency regimes fall between wave frequencies 0.2 rad/s – 2 rad/s
with low frequency regimes for wave frequency around 0.02 rad/s [40]. The first
harmonic wave energy is contained in the wave period range of 5 s – 25 s [56]. The
effect of spread mooring system on the linear wave induced motion is generally quite
small. In special cases, like in higher wave periods (greater than 25 s is a rare
occurrence), the mooring system will have an influence [60]. This allows the usage of
uncoupled frequency domain analysis tools like wadam wizard in SESAM HydroD to
be used for finding motion RAOs.
However, in deep waters, the effect of geometric non- linearities and cable
dimensions affect the system response in two ways. Firstly, the restoring forces of the
vessel due to the mooring lines are affected. Secondly, large changes to the line
configuration affect the dynamic response characteristics and damping levels provided
to the vessel [58]. The presence of mooring and risers introduces sources of damping
that are not included in the classical roll damping problem used in uncoupled analysis
[61]. Coupling effects are contributed by static restoring forces, current loading,
23
mooring line damping effect, hull/mooring contact and additional inertia forces other
than that of a hull [55]. These effects are considered in a fully time domain coupled
analysis. Hence when the effect of mooring line dimensions and metocean conditions
on FPSO motions are studied, coupled analysis is a must to obtain accurate results in
deep waters. SESAM DeepC is a fully coupled time domain program that can be used
for dynamic analysis of deep water floating bodies [58], [55]. It utilises an implicit
time stepping scheme and the dynamic equation of motion is solved by equilibrium
iterations at every time step. Material non-linearity, geometric non-linearity, explicit
loads and hydrodynamic loads can be effectively treated using DeepC [55].
2.2.4 FPSO Motion Response using Model Testing
Literatures pertaining to the modelling of FPSO hull and mooring system are reported
below.
2.2.4.1 Modelling of FPSO
The choice of FPSO model scale depends on water depth of the basin, accuracy of
results (the smaller the model, less accurate results) and capability of generating
required wave height and period at a particular scale in the basin. There are mainly two
ways to relate the prototype and model. One is by matching the non-dimensional terms
developed by inspection analysis of the mathematical description of the physical system
under investigation. In this method, the equality of the corresponding non-dimensional
parameters in model and prototype govern the scaling laws. The non-dimensional form
of differential equations derived from the physical system dynamics is ensured to be
duplicated by the simulated physical system. The non-dimensional quantities in the
differential equations must be equal for both model and the prototype; albeit that this
method can be adopted only when the governing equation of the prototype and the
24
model is explicitly known. The second method is based on Buckingham Pi theorem by
relating the model properties to the prototype properties. In this method, the important
variables influencing the dynamics of the system are identified first along with their
dimension. Then, from these variables, an independent and convenient set of non-
dimensional parameters is constructed. The similitude requirements are yielded from
the equality of the pi terms for the model and the prototype. The model and the
prototype structural systems are similar if the corresponding pi terms are equal [62].
Table 2.1: Scale Factors as per Froude’s Law of Similitude [62]
Variable Quantity Scale factor
Length L λ
Area L2 λ2
Angle none 1
Mass M λ3
Time T λ1/2
Acceleration LT-2 1
Velocity LT-1 λ1/2
Variable Quantity Scale factor
Spring constant MT-2 λ2
Force MLT-2 λ3
Wave height L λ
Wave period T λ1/2
Wave length L λ
Density ML-3 1
Where the action of waves and the inertia of the body is predominant, the law of
similitude between prototype and model is formulated using Froude’s law [10]. If λ is
the linear scale factor, application of Froude’s law of similitude results in the scaling
shown in Table 2.1 to be adopted for model testing [62]. Using these scale factors,
adjustments for water depth, centre of gravity of model and calibrations for wind,
25
current and wave can be done prior to the actual model tests [62]. These adjustments
and calibrations are done before keeping the model in the basin [10]. The spectral
energies of the generated wave are compared with the numerical one and adjusted
through an iterative procedure until the required accuracy is obtained [62].
Wood is used to construct the models of FPSO hulls. The principle of physical
pendulum is used to adjust the longitudinal weight distribution and the transverse
stability is adjusted by means of inclining tests [10]. While ballasting the model to
adjust the CG, Moment of Inertia and draft, it is better to use weight than to use water
to avoid sloshing and alteration of loads acting on it. Once the mooring lines are
attached, the natural periods of the system can be found by conducting free decay tests
[62].
2.2.4.2 Modelling of mooring lines
Horizontal mooring lines are reported to be used to restraint FPSO during model tests
under the action of unidirectional waves, regular waves and current [63] [64].
Horizontal soft moorings are mostly used due to the limitation in maximum water depth
in the wave basins. To model the mooring lines and the associated viscous effects,
Reynold’s scaling should be adopted and the model should be comparatively big [62].
This is not possible in wave tanks with depths of 1 m. Also soft mooring lines are
preferred over stiffer ones in model testing, so that the wave induced vessel motion will
not be affected while enough restoring forces and moments sufficient to prevent large
drift motions are given [59]. In such circumstances, truncated mooring systems can be
used along with a numerical tool to extrapolate them to full depth (Hybrid verification
method) [55]. But, it underestimates the dynamic mooring load and the dynamic
similitude is very hard to achieve even if clumps/buoys and springs are used to match
the surge stiffness in model testing by Kim et al [42]. Also, the effect of mooring lines
on FPSO motions are quite small in shallow waters at WF [59], [60]. Hence soft
26
horizontal mooring with negligible mass and damping will be ideal to be used under
such circumstances.
2.2.5 Hydrodynamics of FPSO
2.2.5.1 Representation of wave
Theoretical simulation of water waves and sea motion, in general involves rigorous
mathematical analysis. The basic hydrodynamic equations that govern the wave
kinematics are the equation of continuity (Laplace’s equation) and the equation of the
conservation of the momentum (Bernoulli’s equation). The form and solution of these
equations vary depending on the intended application of the wave kinematics.
However, in general, all solutions assume incompressible, inviscid and irrotational fluid
particles. The solution of the boundary value problem can be solved in different ways
using the existing wave theories [65]. The simplest solution of the hydrodynamic
equations involves further assumption, that the waves are of small amplitude compared
to the water depth and the wave length. This solution was introduced by Airy (1845)
and became known as the linear Airy wave theory or sinusoidal wave theory [66], [65].
This assumption allows the free surface boundary condition to be linearized dropping
the wave height terms beyond first order. It was shown to provide a good solution in
deep water when water depth to wave length ratio is greater than 0.5 [67]. For the range
of water depths, wave periods and wave heights used for the first order analysis using
regular waves, the linear wave theory was used since it is simple and reliable over a
large segment of whole wave regime and sufficient to obtain the kinematics of waves
to be used in the analysis of FPSO in deep water [65]. A schematic diagram of an
elementary, sinusoidal progressive wave is presented in Figure 2.1.
27
Figure 2.1: Schematic Diagram for a Progressive wave in x direction
In representing the random sea state, mathematical spectrums are widely used,
which are based on significant wave height, wave period or shape factors [65], [67],
[42]. Two of the most commonly used spectrums are Pierson-Moskowitz (P-M)
spectrum with single parameter (based on either significant wave height or wind speed)
and JONSWAP five parameter spectrum; usually three parameters held constant, which
describes fully developed and fetch limited seas respectively [68], [65].
The JONSWAP spectrum, which was derived from fetch-limited measurements

made in the North Sea has a mean shape represented with a peak enhancement factor,
γ, equal to 3.3, which in engineering application is often adopted on the assumption that
this spectral shape is valid for all locations with the North Sea known for the most
hostile weather conditions [68]. The JONSWAP spectrum model is given by Eq. 2.1
[65].
(𝜔−𝜔𝑜 )2
0.0081𝑔2 −5 𝜔 −4 𝑒𝑥𝑝[−
(2𝜏2 𝜔𝑜 2 )
]
𝑆(𝜔) = 𝜔 exp⁡[−1.25 ( ) ] 𝛾 (2.1)
2𝜋 4 𝜔𝑜
28
where
0.161𝑔
𝜔𝑜 = (2.2)
𝐻𝑠
Also P-M spectrum was widely used by the engineers as it is one of the most
representative spectrum for many areas over the world [67]. The P-M spectrum model
is mathematically represented as shown in Eq. 2.3 [65].
0.0081𝑔2 𝜔 −4
𝑆(𝜔) = 𝜔−5 exp⁡[−1.25 (𝜔 ) ] (2.3)
2𝜋 4 𝑜
2.2.5.2 Wave force on FPSO
Since FPSO is a very large structure compared to the wave length of the incident wave,
the incident wave reaching the structure experiences scattering from the surface of the
structure in the form of reflected wave that is of the order of the magnitude of the
incident wave. In this case the diffraction of the waves from the surface of the structure
should be considered in the wave-force calculations [27] . Under diffraction theory, the
basic flow is assumed to be oscillatory, incompressible and irrotational so that the fluid
velocity may be represented as the gradient of a scalar potential, 𝜙. In diffraction
potential theory, the total velocity potential representing the flow around the hull is
obtained as a sum of the incident (𝜙0 ) and scattered potential (𝜙𝑠 ) [65].
𝜙 = 𝜙0 + 𝜙𝑠 ⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡ (2.4)
It satisfies the Laplace equation given in a rectangular Cartesian coordinate system

OXYZ as shown in Eq. 2.5 within the fluid region where 𝜙 = 𝜙(𝑥, 𝑦, 𝑧) and (𝑥, 𝑦, 𝑧)
are the coordinates of a point in the fluid at which the potential 𝜙 is calculated at time
t [65]. The definition of the boundary conditions for the linear diffraction problem is
given in Figure 2.2.
29
𝜕2 𝜙 𝜕2 𝜙 𝜕2 𝜙
∇2 𝜙 = + 𝜕𝑦 2 + 𝜕𝑧 2 = 0⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡ (2.5)
𝜕𝑥 2
Figure 2.2: Definition of Boundary Conditions for the Linear Diffraction Problem
The free surface boundary conditions are [65]:
1. Dynamic Boundary Condition
𝜕𝜙 1 𝜕𝜙 2 𝜕𝜙 2 𝜕𝜙 2
= 𝑔𝜂 + 2 [( 𝜕𝑥 ) + ( 𝜕𝑦 ) + ( 𝜕𝑧 ) ] = 0⁡on⁡𝑦 = 𝜂⁡ (2.6)
𝜕𝑡
2. Kinematic Boundary Condition
𝜕𝜂 𝜕𝜙 𝜕𝜂 𝜕𝜙 𝜕𝜂 𝜕𝜙
+ 𝜕𝑥 𝜕𝑥 + − 𝜕𝑦 = 0⁡on⁡𝑦 = 𝜂⁡ (2.7)
𝜕𝑡 𝜕𝑧 𝜕𝑧
3. Bottom Boundary Condition
𝜕𝜙
= 0⁡on⁡𝑦 = −𝑑⁡⁡ (2.8)
𝜕𝑦
30
4. Body surface Boundary Condition
𝜕𝜙
= 0⁡on − 𝑑 ≤ 𝑦 ≤ 𝜂⁡⁡⁡ (2.9)
𝜕𝜂
The problem is to solve for the velocity potential𝜙, where 𝜙 is the sum of incident
potential, 𝜙0 and scattered potential, 𝜙𝑠 . The incident potential satisfies the boundary
value problem mentioned above in the absence of the structure with a change in body
surface condition as shown in Eq. 2.10 [65].
𝜕𝜙0 𝜕𝜙𝑠
𝜕𝜂
= 𝜕𝜂
⁡on − 𝑑 ≤ 𝑦 ≤ 𝜂⁡⁡⁡⁡⁡⁡⁡⁡⁡ (2.10)
The additional boundary condition for the scattered potential is the Sommerfeld
radiation condition which is stated as below where 𝜑 is Eigen values [65].
𝜕
lim √𝑅 (𝜕𝑅 ± 𝑖𝜑) 𝜙𝑠 = 0 (2.11)
𝑅→∝
The complete boundary value problem is highly nonlinear, especially because of the
free surface boundary conditions. Once 𝜙 is solved for boundary value problem, the
pressure on the surface, 𝑝 of the body and water particle velocities can be calculated as
[65]
𝜕𝜙
𝑢= ⁡⁡ (2.12)
𝜕𝑥
𝜕𝜙
𝑣= ⁡⁡⁡⁡⁡ (2.13)
𝜕𝑦
𝜕𝜙
𝑤= ⁡⁡ (2.14)
𝜕𝑧
𝜕𝜙 1
𝑝 = 𝜌 𝜕𝑡 + 2 𝜌(∇𝜙)2 ⁡⁡⁡⁡ (2.15)
Once pressure is known, the force in specific direction is obtained from the integration
of the component of the pressure in that direction over the submerged surface [65]. This
31
method is used in the 3D diffraction analysis of floating structures [55]. Instead of
solving the total velocity potential function 𝜙 ,the diffraction problem can be solved
using other methods like strip theory as well using approximations to calculate the
hydrodynamic coefficients [29], [18] and subsequently the responses.
2.2.5.3 Wave force on mooring lines
Since mooring lines are slender members compared to the wave length, Modified
Morison equation is used to calculate the wave load acting on them. The original
version of was proposed by Morison [69] for the evaluation of the exciting wave force
on vertical pile, which composed of two inertia and drag components. This equation is
considered semi-empirical equation and was proved reliable for evaluating forces on
slender rigid cylinders. Later, for compliant structures the original force equation can
be modified to account for relative velocity and acceleration between the structure and
the fluid particles. The drag FD and inertia FI forces on an element of a unit length of
the cylinder are given by Eq. 2.16 and Eq. 2.17 respectively. This formula of the force
equation was used for evaluation of wave frequency forces.
𝐷
𝐹𝐷 = 𝜌𝐶𝑑 2 𝑈𝑟𝑒𝑙 |𝑈𝑟𝑒𝑙 |⁡⁡⁡⁡⁡⁡⁡⁡ (2.16)
𝜌𝜋𝐷 2
𝐹𝐼 = (𝐶𝑚 𝑈̇ − 𝐶𝐴 𝑋̈)⁡⁡⁡⁡⁡⁡⁡ (2.17)
4
Where 𝑈𝑟𝑒𝑙 is relative velocity between structure and wave [65], [55], [70].
2.2.6 Operability Conditions for FPSO and Downtime Cost
FPSO operability is determined based on ability of FPSO to perform under extreme

weather conditions without interrupting the offloading operation and processing on
deck. Factors affecting the operability of FPSO are identified as follows:
32
1. Offloading operability: FPSO should avoid fishtailing motion (i.e. large sway and
yaw motion) mainly applicable to turret moored FPSOs while high amplitude low
frequency horizontal FPSO motions (Surge, sway and yaw) should be avoided as well
for both turret moored and spread moored FPSOs [4].
2. Green water impacts: When relative motion of waves and FPSO heave motion (deck
clearance against green water) exceeds freeboard, impact loads are placed on deck due
to green water. These events can even result in fatalities if accommodation modules
are affected by the impact. In the event of impact to processing plants, loss of
containment may occur. These events can cause operational downtime and subsequent
loss in oil production [7]. Green water height on FPSO deck should be less than 3 m to
have low level of risk associated with it.
3. Excessive vertical motion of FPSO: Excessive heave, roll and pitch motion can affect
processing on board FPSO and crew habitability. Extreme motions in heave, roll and
pitch occurs in WF ranges [71]. Roll motion of FPSOs should be within ±5° to enable
crew habitability [61].
Green water on FPSO deck should be minimised to achieve safe operation period and
avoid damage to equipments on board FPSO. Related to FPSO station keeping
capabilities, accepted risk level for loss of production by Whitman [72] is exceeded if
the probability of occurrence of green water exceeds 0.01 and if FPSO motions exceeds
acceptable limit with a probability of occurrence higher than 0.001. Studies should be
aimed at reducing such risks so that both loss of life and loss in production could be
avoided.
FPSO parameters conductive to occurrence of phenomena disrupting operational

time are large vertical motions of ship because of length of ship, continuously changing
FPSO draft and heavy weather conditions [73]. In actual, all the six FPSO motions
should be minimised to ensure safe processing and offloading operation. The following
section details the previous efforts taken to conduct parametric studies to optimise
33
FPSO motions, influence of environmental loads on FPSO motions, green water effects
due to vertical motions of FPSO and operability analysis undertaken to calculate FPSO
downtime due to these vertical motions and green water impacts.
2.2.6.1 Parametric studies for environmental loads and water depth on FPSO motions
Li et al [74] investigated the motion performance of a fully loaded single point moored
FPSO in heave, roll and pitch in water depth varying from 21 m to 26 m for 100-year
environment condition. He used water depth to draft ratio ranging from 1.3 to 1.1 and
the results shows that as the water depth decreases, the WF motions of the FPSO
decreases in shallow water. Wang et al [75] studied the surge, heave and pitch motions
of a FPSO with soft yoke mooring system using ANSYS AQUA, comparing Newman’s
approximation and Pinksters method for water depth varying from 20 m to 33 m. Result
using Newman’s approximation differed from the results from later and concluded that
it is due to the inclusion of second order forces in Newman’s approximation. Hence a
model test to study the variation of FPSO surge, heave and pitch motion in water depth
up to 100 m is of interest to arrive at conclusion regarding the motion behaviour of
FPSO since the shallowest oil field in Malaysia is at 55 m in Berantai oil field and in
Australia is at 78 m in Cossack-Wanaea-Lambert-Hermes.
Soares et al [76] concluded in his study that surge, heave and pitch motions of a
turret moored FPSO varied linearly with significant wave height and reported that the
surge motion for turret moored FPSO decreases while heave and pitch increases. The
motions are reported to be varying linear with wave height [77], however a detailed
investigation comparing all the 6 motions of FPSO for spread mooring and turret
mooring configuration has not reported before and the trend in variation of vertical and
horizontal motions of FPSO while different mooring are used. This is very much in
need as the vertical motions of FPSO highly determines the mooring system employed
and there by associated costs in the life-cycle of FPSO.
34
Caire et al [57] studied the effect of wave directionality on FPSO riser top tension
responses and concluded that the responses in heave and roll decreases as spreading
parameter decreases. Munipalli et al [77] studied the effect of wave steepness on yaw
motions of a weathervaning FPSO and observed large yaw rotations for low wave
steepness and large wave lengths.
2.2.6.2 Parametric studies for mooring line and hull parameters on FPSO motions
Kannah et al [78] did experimental study on an externally turret moored FPSO of 1:100
scale. The study was conducted for different loading conditions and hawser lengths
while the water depth was limited to 1m. Their study reported that the surge RAO
increases with an increase in DWT and an increase in hawser length to ship length ratio.
They have identified the limitations of their results that, it cannot be applied to FPSOs
in deep water. Due to the limitation in water depth modelling in wave tank, a numerical
modelling to investigate the effect of mooring line length to ship length ratio is of
interest and will aid in filling the knowledge gap.
Kannah et al [79] also studied the effect of turret position on FPSO motions and found
that keeping turret in forward position reduces surge, heave and pitch motion when
compared to keeping it in midship. Yadav et al [80] conducted parametric study on a
weathervaning FPSO studying the effect of turret position and hull length on FPSO yaw
motion. It was observed that yaw motion is more influenced by ship length to wave
length ratio than natural roll period and the horizontal offset increased as turret moved
close to mid-ship. However, the effect of spread mooring fairlead on FPSO motion has
not been reported before.
Baghernezhad et al [81] compared the effect of FPSO shape in the overall

performance. He studied both ship shaped and cylindrical shaped FPSO in full loading
condition at a water depth of 100 m. They concluded that cylindrical FPSO
35
performance is better in terms of stability, sea keeping, mooring and riser tension as it
absorbs less energy from waves due to its geometry.
Montasir et al [82] studied the effect of mooring line azimuth angle on a turret
moored FPSO’s heave, pitch and surge motions. The heave motion was observed to be
highly sensitive to increase in azimuth angle from 30º to 60°.
2.2.6.3 Metocean conditions and FPSO motions
The presence of current and wind can significantly influence the wave viscosity.
Viscous damping was found to be increasing linearly with the tanker surge velocity
[62], also mentioning that when current is introduced, the amount of viscous damping
in wave increases. As per Ewans et al [83] wind has a greater part in determining the
heading of the turret moored FPSO. Hassan et al [84] studied the effect of current in
the damping ratio of the system for a catenary and horizontal mooring system. For both
the systems, the damping ratio of the system increased after introducing current. The
damping of the catenary mooring system was higher when compared to that of the
horizontal mooring system due to the interaction of mooring lines with wave and
current.
The current loading on the mooring system may dominate the total steady force
while slowly varying wind loads may give rise to LF motions in horizontal directions.
Also, wind and current can sometimes induce fishtailing effects in the FPSO motions
by inducing unstable coupled sway and yaw motions [56]. Stansberg et al [85] deduced
that the wave-current interaction effect on FPSO and semisubmersible motions can be
much larger when compared to the effects from current and wind alone. Teles et al [86]
conducted model testing and sensitivity study to investigate the wave current interaction
effects and found that the mean horizontal velocities near the free surface are
significantly affected by the introduction of current in the presence of waves.
36
The presence of wave-current –wind interaction is shown to affect the FPSO
motions significantly from the previous studies. The choice of the mooring system is
based mainly on the floater motions [9]. Hence a thorough study is required to
determine the variation of floater motion in the presence of wind, wave and current.
2.2.6.4 FPSO motions and green water effects
Green water is defined as unbroken waves overtopping FPSO deck [87]. Buchner [88]
studied the impact of green water on FPSOs using model tests. He considered the
relative motion of FPSO with respect to wave height, water flowing on to the deck and
water hitting FPSO. He concluded that green water effects are sensitive to wave height,
wave period and current velocity. He [8] also studied the impact of green water through
the sides of a weathervaning FPSO and linear diffraction theory may be used to assess
the green water effects in the preliminary design phase [71] [89]. Nielson studied green
water loads on ships having forward velocity using numerical methods [90]. Buchner
et al suggests the use of 3D diffraction theory to predict green water incidents [91].
Kleefsman et al performed green water simulation for FPSO using a domain

decomposition method. The method of decomposing far field and near filed of FPSO
gave good results when linear diffraction theory was used to represent far field. Local
flow around deck was simulated using Navier-Stokes solver [92]. Lu et al [93] studied
the green water effect on moored FPSO, freely floating FPSO and a side-by-side
moored FPSO/LNG in extreme waves and he was successfully able to model the highly
nonlinear interactions and mooring effects. Tao et al [94] detailed the simulation of
green water using numerical and experimental methods. Akandu et al [95] developed
a numerical program called ProGreen to optimise the principal dimensions of FPSO
based on freeboard exceedance analysis. Veer et al [96] emphasized the importance of
model testing in green water flow analysis and gave detailed description of three typical
37
flooding events by experimentally studying the phenomenon using deck mounted wave
probes and on-board video.
Recently, Werter [97] investigated the short term and long-term probability of green
water on FPSO and used ANSYS Aqwa to generate vessel motion RAOs. It was seen
that ANSYS Aqua generated motion RAOs higher than DNV software. Zhang et al
[98] investigated overtopping through a CFD tool and proposed to use relative
overtopping duration to combine the coupled effects of overtopping duration and
freeboard exceedance to estimate damage on deck.
Green water on FPSO is now recognized as an important aspect during the design
of FPSO [71] and deck and topside design should be considered as an ultimate limit
state rather than accidental limit state criteria [7]. Green water impacts can have
dangerous effects on processing facilities on FPSO deck [71]. This event is likely to
occur in low wave height and period and in conditions lower than design criteria where
the wave height and period could be close to 1-year return period conditions than the
design 100-year period. The largest relative wave elevation occurs when wave length
is equal to ship length [99]. Also, freeboards may be insufficient to prevent this from
happening at high loading condition of FPSO [87]. Freeboard exceedance is
categorised into low, medium and high levels. Low level of freeboard exceedance is
when the water height is less than 3 m, medium level when it is between 3 m and 6 m
and high level of susceptibility if it is above 6 m of water height. Buchner assumed
freeboard exceedance limit of 2.8 m to be acceptable [99].
2.2.6.5 Operability analysis
Workability or the operability of an offshore structure is whether it is capable to operate

in wind, wave and current where it is installed in a safe and reliable functioning
condition. Wal et al outlines the scatter diagram and scenario based method to calculate
38
workability of offshore structures. He developed a tool called Dredsim 2000 to calculate
the workability of dredging tools used in oil and gas industry [100].
Djatmiko et al studied the operability of FPSO based on green water and slamming
effects and observed that most persistent green water loads are at the fore upper deck at
15% LBP from FP and downtime is increased when FPSO draft was changed to 9.92
m from 8.05 m [101]. Ewans et al studied the heave, roll and pitch motions of a FPSO
to determine operability conditions for locations at offshore Namibia and at west coast
of New Zealand where the swells act perpendicular to wind sea [83]. Correa et al
investigated offloading downtime of a spread moored FPSO in tandem with a
dynamically positioned shuttle tanker. He observed that by incrementing angle that
defines the area where the shuttle tanker is allowed to weathervane, offloading
downtime was reduced [102].
Downtime due to green water in location under consideration is of great interest as

no reported studies have previously assessed the same. Also, the results generated will
aid in the choice of mooring system in these locations based on both cost and motion
criteria.
2.2.7 Life-cycle Cost Analysis
2.2.7.1 Definition of LCCA
LCCA is defined as the process of economic analysis to assess the total cost of
acquisition and ownership of a product over its life-cycle or a portion thereof [103].
The main objective of LCCA is to quantify the total cost of ownership of a product or
a project throughout its full life-cycle, which includes research and development,
construction, operation and maintenance, and disposal or reuse. Life-cycle costing is a
concept used for making decisions between alternative options, optimizing design,
39
scheduling maintenance and revamping project planning. The option identified with
the highest net present value is the most economical or least cost option/ approach
[104]. The two major factors that influence such options are scalability and
customizability and thus such new concepts need powerful life-cycle cost models that
can cope with the influence of scale and customer requirements on the whole life cycle
[105].
2.2.7.2 Benefits of LCCA
The benefits of doing LCCA are as follows [106], [107]:
 It results in earlier actions to generate revenue or to lower costs than

otherwise might be considered.
 It ensures better decision from a more accurate and realistic assessment of

revenues and costs, at-least within a particular life-cycle stage.
 It promotes long-term rewarding.
 It provides an overall framework for considering total incremental costs

over the life span of the product.
2.2.7.3 Previous studies using LCCA
One of the earliest recognizable LCCA application in civil engineering is the World
Bank Highway Cost model in 1969. Life-cycle costing application in offshore
structures is a relatively new research area and some of the previous studies using
LCCA for offshore structures are reported below.
40
Nam et al [108] developed a new life-cycle cost methodology with the risk
expenditure taken in to account for comparative evaluation of offshore process options
at their conceptual design stage. The risk expenditure consisted of the failure risk
expenditure and the accident risk expenditure. The former accounted for the production
loss and the maintenance expense due to equipment failures while the latter reflected
the asset damage and the fatality worth caused by disastrous accidents such as fire and
explosion. It was demonstrated that the adopted LCCA methodology can play the role
of a process selection basis in choosing the best of the liquefaction process options
including the power generation systems for a floating LNG (Liquefied natural gas)
production facility.
Thalji et al [105] conducted a case study on innovative vertical axis wind turbine
concept to generate a scalable and customer oriented life-cycle costing model for the
same. The cost analysis of the wind turbine concept covers the whole life processes,
manufacturing, installation, operating and maintenance. Santos et al [109] developed
a theoretical methodology to study the life-cycle cost of floating offshore wind farms.
Six life-cycle phases needed to install a floating offshore wind farm was defined:
conception and definition, design and development, manufacturing, installation,
exploitation and dismantling. They suggested that the proposed methodology could be
used to calculate the real cost of constructing the floating offshore wind farms.
Gratsos et al [13] investigates through a cost/benefit analysis, how the average

annual cost of ship transport varies with the corrosion additions elected at the design
stage. The results of this study clearly indicated that ships built with corrosion
allowances dictated by experience, adequate for the ship’s design life, when all factors
have been considered, have a lower life-cycle cost per annum for the maintenance of
the integrity of their structure.
Howell et al [110] discusses the various factors affecting the CAPEX and OPEX of
turret moored and spread moored FPSOs and the technical issues related to the design
41
of mooring system. He also computed the NPV for a spread moored FPSO and a turret
moored FPSO in Brazil with 10.5% discount rate. However, the detailed cost estimate
was not given. He affirmed that in addition to the CAPEX of both systems, they are
different in terms of their motion performance and offloading performance as well.
Dina et al [111] performed LCCA to compare maintenance cost of an oil and gas
production facility in the sensitive environment of Arctic, while implementing different
technical solutions. Out of the whole life-cycle, only maintenance phase was considered
in this study.
Recently, Kurniawati et al [112] evaluated the long-term charter rate in volatile or

uncertain condition of FPSO by capital budgeting principal where NPV was one of the
evaluation criteria. Miranda et al [113] derived a target reliability index for FPSO for
ultimate limit state design of turret moored FPSO’s mooring lines for hypothetical
tanker dimensions. A life-cycle cost model was used to optimise the disconnection
criteria by counting failure instances due to green water, hull and mooring and to obtain
design criteria in reliability format. However, the results cannot be generalised as there
will be cuts to life-cycle expenditures if the optimisation criteria are implemented.
Other than this, there are very limited detailed studies reporting LCCA of FPSOs. Also,
there are no previous studies reporting NPV variation when different types of turret
mooring are used.
2.2.7.4 LCCA Procedure
The success of LCCA largely depends on the level of accuracy of the cost data in use,
variable selection and ensuring that the correct economic criteria are followed [114]. It
is a method to evaluate the significant cost for different design options, adding the
expenditures and subtracting the revenues and resale values and before choosing the
final design with the minimum life-cycle cost, non-economic considerations are given
42
to accommodate benefits of the project. The primary step in LCCA includes the
identification of cost variables [103] [114]. The major steps in conducting an LCCA
study are detailed below compiling and summarizing the technique followed by Ferry
et al [115] and Kirk et al [116] and the same is given in Figure 2.3 [116].
Step1: Identification of design alternatives to be compared
Life-cycle costing assessment stems from the need to evaluate the true cost of a
construction project/asset over its entire life-cycle period. Based on the need and
after a brainstorming session, various design alternatives are identified. These
alternatives are subjected to an initial screening based on design constraints,
benefits measurable in monetary terms, ease of implementation, ability to perform
the function and magnitude of savings in the initial design phase. The remaining
alternatives are developed to obtain enough data for whole life cost computation
[116].
Identification of design alternatives
Establishment of basic assumption and

determination of exact LCC procedure
Data Collection
NPV calculation for design alternatives
Risk/uncertainty assessment
Figure 2.3:
Selection LCCalternatives
of design Framework
Figure 2.3: LCC Framework
43
Step2: Establishment of basic assumptions and determination of exact LCCA
procedure to be adopted.
Assumptions on life expectancy of asset, period of study, cash flow timings,

resale/residual values, inflation, discount rate, source and reliability of data,
maintenance policies and comprehensiveness of life-cycle costing is identified as
the next step of the procedure. A Discount-rate is required to assist with an
understanding of the time value of money such that a dollar today is worth more
than a dollar in the future. The discount rate can be calculated from the method
specified by Royal Institution of Chartered Surveyors (RCIS) [103].
Step3: Data Collection
Data collection is the most difficult step in the entire phase of LCCA unless a design
team can supply the data from their experience and brief the level of cost data
accuracy. In its absence, data collection fundamentally depends on networking with
expert practitioners related to the specific asset under analysis, literature reviews or
modification of available data to suit the study. Because of this, LCCA is done for
projects with potential benefits only. Most of the researchers start the data
collection by ‘estimates’ of elemental costs, manufacturer’s and supplier’s
quotations. Monetary costs include, Capital costs (C1), installation costs (C2),
operating costs (C3), maintenance costs (C4), refurbishment/replacement costs
(C5), downtime costs (C6) and decommissioning/disposal cost (C7). The monetary
benefits such as revenue generated (B8) and salvage values (B9) should be
subtracted while computing the Net Present Value (NPV) [103]. As per Al- Hajj’s
study, the absence of sufficient data is seen to be the major barrier in doing a life-
cycle cost assessment (LCCA) [117]. Ferry et al [115] observes that the unreal
variables can put the findings generated through LCCA in doubt category.
Ashworth suggests proper care should be taken to reduce the uncertainty in results
[118]. That being said, experienced practitioners’ can/do provide suitable datasets.
44
Step 4. NPV calculation for design alternatives
Life-cycle cost for each life-cycle phase mentioned in step 3 are calculated by
applying discount rate and finally summarized to achieve the system LCCA. Once
the system LCCA is calculated as per Eq. 2.25, system NPV is calculated as shown
in Eq. 2.26 [103].
𝑆𝑦𝑠𝑡𝑒𝑚⁡𝐿𝐶𝐶 = ⁡𝐶1 + 𝐶2 + 𝐶3 + 𝐶4 + 𝐶5 + 𝐶6 + 𝐶7 − 𝐵8 − 𝐵9 (2.18)
𝑆𝑦𝑠𝑡𝑒𝑚⁡𝐿𝐶𝐶
𝑁𝑃𝑉⁡𝑠𝑦𝑠𝑡𝑒𝑚 = ⁡ (1+𝑑𝑖𝑠𝑐𝑜𝑢𝑡⁡𝑟𝑎𝑡𝑒)×𝑝𝑒𝑟𝑖𝑜𝑑⁡𝑜𝑓⁡𝑠𝑡𝑢𝑑𝑦 (2.19)
Step 5. Risk/uncertainty assessment
LCCA involves uncertainty in its very nature and the degree of uncertainty
determines the degree of accuracy of results. LCCA results are credible only when
the uncertainties are considered and sensitivity analysis is performed to do single
variant/ multi-variant analysis to study the variation in one parameter by varying a
second parameter on which it depends [116]. The output parameter in a sensitivity
analysis is always the life-cycle cost of the least cost alternative and the input
parameter is always the input cost element. The analysis helps in studying the
variation of life-cycle cost for an economic design alternative under varying
circumstances where its life-cycle cost can be high and help in finding the
breakeven point where the alternative will no longer be cheaper when compared to
the next lowest alternative design. Uncertainties can be classified into two:
alternative-independent uncertainties (resulting from assumptions concerning all
the alternatives to some degree) and alternative dependent uncertainties (due to
specific alternatives) [116]. The simplest method in weighing the alternative-
independent uncertainties is by using the discount rate in the analysis which is
greater than the one in the absence of uncertainty as used by Whyte [103] in
calculating NPV of best and worst systems. Alternative dependent uncertainties
45
are those related to differential escalation rates, obsolescence, cost-estimate
accuracy, useful life and physical failure [116].
Step 6. Selection of design alternative
If ‘benefits’ are included in the LCCA study, then NPV or annualized equivalent
value of the alternatives are compared while deciding the design alternative. The
alternative with negative NPV means the project is going to yield a return lower
than its capital cost. If NPV is positive, then the project will bring profit through
the implementation of that design alternative. If life-cycle cost of two or more
alternatives are found to be equal or within 10% difference, then the nonmonetary
benefits like environmental sustainability, aesthetics, safety, expansion potential
and obsolescence avoidance are considered. The technique of weighted evaluation
is used when nonmonetary benefits are considered [116][103].
2.2.7.5 Limitations of LCCA
The limitations in LCCA study are normal restrictions in every engineering tool.
Surpassing these limitations, LCCA has passed the test of time by engineers who
combine proper judgement using their experience and knowledge in minimising these
limitations. The limitations of LCCA study are [119]:
 LCCA is able to indicate reasonable or unreasonable specification comparisons
 LCCA expertise relies upon a range of subjective experiences of whole-cost.
 Accuracy of LCCA depends on the cost data inputs.
 Errors in LCCA accuracy are addressed chiefly through sensitivity analysis.
 Cost data can be limited, and given its longitudinal nature difficult to obtain.
46
 LCC models require volumes of data (such as the building specific BCIS) and
often non-onshore-construction data is somewhat indicative in nature.
Notwithstanding the limitations above, LCCA is deemed a tried and tested means
to better understand design options and minimise the life-cycle cost of projects
[119].
2.2.8 Life-cycle costs for FPSO
Every FPSO has a “whole-life” , starting with the conception and definition of FPSO,
design/development of the FPSO for serviceability, producibility and safety, followed
by extracting the resources and delivering the FPSO by conducting fabrication,
installation of FPSO to the offshore field, maintenance, inspection, repair, support and
modification of the FPSO or equipment throughout its operational life and finally after
the design period of usually 30 years, the FPSO is removed from the offshore field or
decommissioned or converted for other purpose [110], [120]. Figure 2.4 shows the life-
cycle of FPSO.
Figure 2.4: Life-cycle of FPSO
47
The cost variables considered for each life cycle phase of FPSO is summarized as
follows:
a) Capital cost (C1): The capital expenditure for an FPSO includes cost of
materials for a newly built hull or purchase fee for second hand hull and conversion
cost for second hand hull. The conversion costs also include cost of fabrication to install
mooring system as well, since it is not designed for oil drilling purposes, while a
purpose-built tank is built with the facility to install mooring system based on the area
it is going to be installed. Also, cost of materials for topside, cost for labour charges in
design, development and construction of FPSO or cost of labour charges in planning
and carrying out modification for converted tanker (professional design fee,
construction supervision fee and labour charge for workers) should also be considered.
The costs for equipment hired for construction should also be considered [110], [120].
b) Installation cost (C2): Installation cost for the FPSO includes the cost for
transport of FPSO from dock to the oil field. Sometimes only the fuel charges and ship
personnel charges needs to be considered if the FPSO sails to the field of location. If
the FPSO is towed to the location, then the cost for towing arrangement, emergency
anchor and bunkering arrangements should be considered in addition to the labour
charges of riding crew and warranty surveyor. The other costs arise from installing the
mooring and riser system. The related labour costs and equipments costs will also fall
under installation costs [110], [120].
c) Operation cost (C3): The operation costs for the FPSO mainly consists of the
cost of fuels and electricity in running the plant, labour charges for FPSO crew and
technicians and cost of rented equipment. This may vary for old and new tankers based
on efficiency. This is the phase were oil production revenue is addressed (B8) [120].
d) Maintenance cost (C4): Maintenance costs mainly consists of labour charges for
inspection personnel, technicians and cleaners and the like. The cost of dry dock hiring
for planned maintenance schedules should also be considered [110].
48
e) Refurbishment/ Replacement cost (C5): Cost of materials for planned
replacements and cost of equipment for carrying out the same are calculated in this
phase. Material, equipment and labour costs for covering irregularities in hull integrity
and storage compartments should also be considered [110], [120].
f) Downtime cost (C6): Downtime costs arises from shutdown of oil drilling due
to unfavorable weather or accidents, green water events and mooring line damages.
The cost is calculated in terms of transporting the crew back to onshore facilities (costs
of hiring helicopter) and loss in terms of time value of money (unable to drill and
produce oil). Especially in the case of converted tankers, frequent mooring line
damages are reported. The replacement costs for moorings are higher and to be
calculated based on average number of damages reported. Labour charges for
inspecting accidental damages and break downs should also be counted for [110], [120].
g) Decommissioning/Disposal cost (C7): Once an oil field is fully exploited,

FPSOs are either transported back to onshore facility or to another oil field. FPSOs are
seldom dismantled or discarded like fixed offshore structures. They are converted for
other purposes or may be reused in another oil field. Hence it will always have a
residual or salvage value or resale value (B9). The decommissioning cost of FPSOs
includes the cost of fuel for riding back to another location/field or cost of towing
arrangements. Dismantling cost of mooring/riser system and the labour charges for the
crew should also be counted for [110].
2.3 Critical Literature Review
In this literature review the focus is mainly given to literatures about FPSO motion
response, their life-cycle cost and operability conditions. The critical literature review
conducted is presented in Table 2.2, where the crucial gaps in the literature has been
49
identified via key references. The following sections analyses these gaps in detail and
other critical aspects which are found out in the literature.
2.3.1 FPSO Motion Performance and Cost
The cost and motion performance of a tanker are the two important factors while
choosing the FPSO hull and the associated mooring system, especially when converted
tankers are used for oil explorations. Malaysia and Australia own the maximum
number of FPSOs in the Indian ocean region. Also, the number of FPSOs with
converted hulls are more when compared to that of with newly-built ones in Malaysia
and Australia [2]. No studies have reported the impact of choosing a converted hull on
the capital and life time cost of the FPSO system and their life-cycle cost when ET, IT,
RTM, STP and SM is used. An initial life-cycle cost calculation of the FPSO system
is desirable since they are high investment projects and large structures with difficulty
in its construction and installation. It is better to conduct LCCA in the planning stage
to choose the best possible FPSO hull and mooring type based on cost. The preliminary
step in conducting an LCCA is data collection; the major sources of data must come
from industry practitioners, albeit that a reluctance exists to share cost data due to the
competent nature of oil industry. In life-cycle analysis, cost data is often indicative in
nature [117], but in this research care is taken to ensure the quality of cost data by
collecting them from reliable industry practitioners and published cost reports.
Conducting an LCCA for FPSOs in Malaysian and Australian waters remains important
as no previous detailed studies have been carried out to address the whole-cost aspects
of FPSOs /converted FPSOs with different mooring configurations.
As discussed above, cost alone is not sufficient to determine a FPSO system, rather
respective motion performance must complement (specification) choice [110]. Whole
cost data requires to reflect different metocean conditions, water depths and, different
FPSO sizes and loading conditions. This calls for a detailed parametric study covering
50
the effect of wave, wind, current, hull dimensions and loading conditions, mooring line
configurations and dimensions on FPSO motions.
If the FPSO system is not designed properly, it will result in frequent mooring line
breakages, hull damages due to green water and result in shut down. To avoid
operational downtime due to such circumstances, motion performance of FPSOs for
various design parameters should be thoroughly studied before choosing them for an
oil field [110]. Also, downtime cost due to green water phenomena is studied to assess
the performance of FPSO in Malaysia and Australia under wind generated sea state
using location specific wave scatter table approach.
A comprehensive study to assess the motion performance, life-cycle cost and

downtime for spread moored and turret moored FPSO has never been reported before
and is of great interest as the need for cost and design optimisation is becoming more
and more prevalent in the FEED phase of project now- a- days. Results generated could
be of great impact, especially for the FPSO operators in Malaysia and Australia while
choosing FPSO configurations as there are many FPSO projects in the FEED phase by
the start of 2018 and oil and gas market is showing remarkable recovery when compared
to the last few years.
2.3.2 Dynamic Responses of FPSO
The size of the FPSO is comparable with the wave length and hence results in a
disturbance of the wave field causing diffraction of incident waves. Hence diffraction
theory is used to calculate the wave load acting on FPSO [65], whereas Modified
Morison equation is used to calculate the wave load on mooring lines as they are slender
members. Also, Linear Airy Wave theory is used to represent sinusoidal waves as the
wave heights in deep waters of Malaysia are small compared to the water depth.
Unidirectional wave spectrums are the conservative wave assumption used in
51
hydrodynamic studies [57]. P-M and JONSWAP spectrum are found to be more
suitable to represent the long-crested waves in locations of study. Limited study thus
far has examined motion performance of operating FPSOs in Malaysia and Australia.
Simplified analysis like using strip theory requires only less computational effort
and gives reliable reasonable results on the conservative side for engineering
applications, but the 3D diffraction methods agrees better with the experimental results
[27]. Hence 3D diffraction analysis should be used to study the FPSO motion
performance and hence used for the location under interest in this study.
Uncoupled frequency domain analysis is reported to have many drawbacks [55].

But, where the non-linearities are not dominant, uncoupled frequency domain analysis
gives good results [40], [58]. So, to study the floater responses due to the change in
hull dimensions, hull loading conditions and for water depth parametric studies below
100 m, uncoupled analysis seems to be a good choice and can be time saving. Also,
the effect of mooring lines and risers may not be significant for determining RAOs in
wave frequency regime, although it affects the mean position in low frequency regime
[59], [39] and the change in these mean offsets are not the primary concern of a dynamic
analysis. Since the mooring system has less effect on linear wave induced motions in
wave period range of 5 s – 25 s [61], [56], the usage of uncoupled frequency domain
analysis tools like wadam wizard in SESAM HydroD can be used for finding motion
RAOs.
But in deep waters, the effect of geometric non- linearities and cable dimensions
affect the system response considerably. When the effect of mooring line dimensions
and metocean conditions [83], [85] on FPSO motions are studied, coupled analysis is a
must to obtain accurate results in deep waters. SESAM DeepC is a fully coupled time
domain program that can be used for dynamic analysis of deep water floating bodies
[55], [58]. It utilises an implicit time stepping scheme and the dynamic equation of
motion is solved by equilibrium iterations at every time step. Material non-linearity,
52
geometric non-linearity, explicit loads and hydrodynamic loads can be effectively
treated using DeepC [55].
It is argued here that numerical experiments however can never fully replace wave tank
experiments, because many physical uncertainties will still prevail in a numerical model
and hence the numerical needs to be verified before further application of it [62]. An
efficient structural design involves complementing numerical and physical experiments
to properly guide the engineers [62]. Horizontal mooring using soft linear springs is
reported to be used for representing mooring lines of the distorted physical model due
to water depth limitation in wave tank [59], [60], [63]. Also, the effect of mooring lines
on FPSO motions are quite small in shallow waters at WF [59], [61]. Hence soft
horizontal mooring with negligible mass and damping is ideally used under such
circumstances. Since multi-directional waves represent the real sea state more closely,
a comparison of FPSO motion response in long-crested and short-crested waves is best
studied using model testing.
2.3.3 FPSO Operability
As mentioned in section 2.2.6, all the six FPSO motions should be minimised to ensure
safe operating conditions for FPSO. Hence a parametric study covering metocean
parameters, mooring line parameters and hull parameters are carried out to identify
optimum configurations were FPSO motions will be minimum.
Parametric studies including water depth in shallow water so far has been conducted
only for a maximum of 33 m water depth [74] [75]. In this study, model test is
conducted to study the variation of FPSO motion in water depth up to 100 m since the
shallowest oil field in Malaysia is at 55 m in Berantai oil field and in Australia is at 78
m in Cossack-Wanaea-Lambert-Hermes. The motions are reported to be varying linear
with wave height [77], however a detailed investigation comparing all the 6 motions of
53
FPSO for spread mooring and turret mooring configuration has not reported before and
the trend in variation of vertical and horizontal motions of FPSO while different
mooring are used. Hence influence of wave height on FPSO motions is studied with
and without the presence of wind and current for spread and turret moored FPSOs.
Effect of FPSO loading condition and mooring length was studied previously for a
water depth of 1 m and they have identified the limitations of their results that it cannot
be applied for deep waters [78]. Hence parametric studies are conducted for loading
condition and mooring line length in the present study for deep waters, so that the results
are applicable for deep waters in Malaysia and Australia. Studies have reported the
influence of turret position on FPSO motion [79] [80], however no studies have
reported the influence of spread mooring fairleads on FPSO motion. Hence the
influence of spread mooring fairleads is investigated in the present study. Also, mooring
line azimuth angle was varied from 30° to 60° for turret moored FPSO to study its
influence on motions. However, effect of mooring line azimuth angle on spread moored
FPSO motion is yet to be studied and experimental study could lead to realistic results
in optimizing the mooring configurations of spread moored FPSOs. Hence model tests
are conducted to study for mooring line azimuth angles 15°, 30°, 45° and 55° for spread
moored FPSO. In addition to these, influence of hull length to beam on FPSO motions
is also studied.
For efficient design of FPSO and associated mooring system, the numerical model
should be able to consider six degrees of freedom motion and depends on factors
including FPSO size, water depth, environmental condition and mooring line
parameters [81]. Hence parametric studies to investigate the six motions of FPSO
varying hull parameters and mooring line parameters are conducted in the present study
to identify factors reducing FPSO motions and thereby enabling increased operational
time.
54
Motion response from parametric studies for varying wave height in the presence
of wind and current, FPSO loading condition and hull length to beam ratio in the heave,
pitch and roll can also be used in minimising green water impacts. Also, downtime
analysis is carried out to calculate downtime cost of FPSOs for long-term and short-
term wave statistics for Malaysia and Australia.
2.3.4 Life-cycle cost of FPSO
As reported in the section 2.2.7.4 the parts of the technique followed by Ferry et al
[115] and Kirk et al [116] are combined to conduct this (new unique) LCCA study of
FPSOs. The associated risks and independent uncertainties are addressed by applying
calculated discount rates and sensitivity analysis as used by Whyte [103] in calculating
NPV of best practicable options from a range of available systems. Other than Howell
[110], no other studies have previously reported the life cycle cost of FPSO mooring
options. Also, no studies have been previously carried out to determine the life-cycle
cost of FPSOs operating in Malaysia and Australia comparing the options of mooring
types and hull condition (newly-built/converted).
The following table summarises research thus far and the gaps being addressed by this
work.
55
Table 2.2: Critical Literature Review
Important References and Identified Gap in the
Topic of Interest
Points addressed Literature
Howell et al [110] emphasized The current study computes
importance of studying motion the motion performance
Main research aspect-
performance along with cost of under various parameters as
Cost and motion of
FPSO. However no previous well as cost of FPSOs
FPSO
studies are conducted in this comparing mooring options
aspect. and hull conditions.
Cost and Motion Cost and Motion performance
No study has previously reported
response of Operating of operating FPSOs in
the operating FPSOs cost and
FPSOs in Malaysia and Malaysia and Australia is
Motion performance.
Australia studied here.
Physical model testing should be Numerical and physical
combined with numerical experiments are conducted
FPSO motion response
modelling to achieve accurate here to find FPSO motion
model [62]. responses.
Limitation of strip theory was Diffraction analysis has been
Full 3D diffraction given by Fan et al and the need for carried out here using
analysis of FPSO conducting 3D diffraction analysis frequency domain and time
was emphasized [27]. domain approach.
Coupled analysis is
performed for parametric
Coupled analysis gives accurate
studies involving metocean
results where non-linearities are
parameters and mooring line
Coupled and uncoupled predominant [55]. Uncoupled
parameters. Uncoupled
dynamic analysis analysis gives good results where
analysis is used for hull
non-linearities are not
parametric studies involving
predominant [40] [58]
loading condition and
dimensions.
56
Topic of Interest
Parametric study is conducted

Water depth parametric studies for water depth up to 100 m as
Parametric studies on
are performed in shallow waters shallowest oil field in
FPSO – Water depth
of up to 33 m [74] [75]. Malaysia and Australia starts
from 55 m using model tests.
Present study compared
No studies have compared the
FPSO motion for different
FPSO motion behaviour in
Parametric studies on mooring configuration for
varying wave heights when spread
FPSO – Wave height wave height 4 m to 8 m with
mooring and turret mooring is
and without the presence of
used
wind and current.
Influence of loading condition and
hull length to beam ratio is not
Influence of loading
reported to be studied for deep
Parametric Studies – condition and hull length to
water locations and the results
Hull parameters beam ratio is studied for deep
generated for different loading
water locations.
conditions by Kannah et al points
out that to be a limitation [78] .
Influence of spread mooring
fairlead location on FPSO motion Influence of spread mooring
is not studied previously and effect fairleads and mooring line
of mooring line length is studied length is studied using fully
Parametric studies - only for shallow waters, again coupled analysis for deep
Mooring line parameters Kannah et al pointing it out to be a water locations. Effect of
limitation of results generated mooring line azimuth angle is
[78]. Effect of mooring line studied for spread moored
azimuth angle is studied only for FPSOs.
turret moored FPSOs.
57
Topic of Interest
Downtime cost is calculated

based on vertical relative
Downtime cost due to green water motions of FPSO in heave,
incidents in Malaysian and roll and pitch contributing to
Downtime cost analysis
Australian seas have not green water on FPSOs using
previously been reported. Malaysian and Australian
long term and short term
metocean statistics.
This study is carried out for
Howell et al [110] computed life-
LCCA of FPSO FPSOs in Malaysia and
cycle cost for FPSO in Brazil.
Australia.
Howell et al [110] compared Comparison is made here
LCCA comparing
spread moored and internal turret between spread moored, ET,
mooring type
moored FPSOs. IT, RTM and STP.
Life-cycle cost for newly
LCCA comparing newly
built hull and converted hull
built hull and converted No studies are reported.
has been compared in this
hull
work.
2.4 Chapter Summary
The research studies handling the FPSO motion responses, operability analysis and life
cycle costing approaches in the past decade were reviewed above and categorised into
eight general research motivations. The previous studies and developments in each
category were reported. Finally, a critical review of researches pertaining to the study
has been carried out to identify the theories and methodologies to be adopted; the gaps
in literature have been identified and tabulated above with the extent to which this work
shall fill knowledge gaps has been made explicit.
58
CHAPTER 3
METHODOLOGY
This chapter discusses the methodology adopted to obtain the research objectives
mentioned in Chapter 1. The method of investigating the FPSO motion responses are
detailed using the model testing procedures followed and software simulation
procedures carried out. The frequency domain and time domain approaches are
discussed in detail using the theories and assumptions used during the analysis.
Followed by that, the calculation of freeboard exceedance to identify green water event
is detailed. Finally, the life-cycle costing procedure adopted for FPSO is detailed in
Malaysia and Australian context. The adopted methodology presented here is in the
same order as they were performed in the study; the scientific basis and the critical
review of the methods adopted here having been already delineated in the previous
chapter.
3.2 Overall Research Methodology
The previous chapter emphasized the role of both performance and cost in an efficient
cost-effective design of FPSO and associated mooring system. Two main points were
highlighted in the study, namely the dynamic motion response of FPSO and the life-
cycle cost of the FPSO. Firstly, the FPSO motion responses were computed using
numerical simulations and model testing, as both are equally significant in obtaining
accurate results [62]. Operability analysis was then performed to identify downtime
due to green water events which can lead to downtime cost in the life-cycle of FPSO.
Then, the life-cycle cost analysis of FPSO was carried out to calculate the NPV and
life-cycle costs of chosen FPSOs in Malaysia and Australia.
A complete 3D diffraction analysis was performed using SESAM HydroD in
frequency domain analysis to obtain the 6 DOF Response Amplitude Operators and the
results were compared with the physical model testing results conducted at the UTP
Offshore Laboratory. Later, a fully coupled analysis of FPSO was carried out using
SESAM DeepC and the analysis procedure and the results were verified against the
published experimental results from the FPSO tests conducted at the OTRC Wave
basin, Texas A&M. The calibrated models were then used for the further parametric
studies and operability analysis based on green water using the verified procedure in
SESAM HydroD, where SESAM DeepC was used while mooring lines plays a
significant role in motion response.
Finally, the LCCA study of FPSO was carried out using the techniques mentioned
in the previous chapter. The FPSO cost data was collected from PETRONAS Carigali
Sdn Bhd, Chevron Australia, Wood Mackenzie Asset Reports and related sources. The
different FPSO systems are compared in terms of their whole life-cycle cost to identify
the economic option of mooring system and whether to build a newly built hull or use
a converted tanker for oil drilling and processing purposes in the Malaysian and
Australian waters. Since the cost data are for FPSOs from different metocean
conditions, water depths and of different FPSO sizes and loading condition, appropriate
parametric studies have been carried out to evaluate the motion performance and cost
of these FPSOs.
The following figure 3.1 represents the overview of the research methodology
performed:
60
DYNAMIC RESPONSE AND LCCA OF FPSO
SELECTION OF FPSO AND TANKER

Based on:
 Existing Literature
 Location Oriented Study
DYNAMIC RESPONSES OF SYSTEM
SOFTWARE
MODEL TESTING
SIMULATION
MODEL TESTS IN THE WAVE

TANK SOFTWARE SIMULATION USING
Environment loads Instruments SESAM SUIT OF PROGRAMS

Environment loads Tools
 Regular waves  Load cells
 Unidirectional  Qualysis motion  Regular waves  SESAM Genie
waves capture  Random waves  SESAM HydroD
 Multidirectional  Wave probes  Wind  SESAM DeepC
waves  Current meters  Current
 Current  Data loggers
COMPARISON AND VALIDATION

 Surge
 Sway
 Heave Magnitude of
 Pitch RAO
 Roll
 Yaw
PARAMETRIC STUDIES, FLOATER RESPONSES &

OPERABILITY ANALYSIS FOR MALAYSIAN AND AUSTRALIAN WATERS
LIFE CYCLE COST ANALYSIS OF FPSO
NPV& PRESENT WORTH CHARTS FOR CHOSEN MALAYSIAN AND AUSTRALIAN FPSOs
Figure 3.1: Overview of Research Methodology
61
3.3 Experimental Tests
As mentioned earlier in this chapter, the uncoupled frequency domain numerical model
with mooring modelled as linear spring is validated by conducting an experimental
study at the UTP offshore laboratory in the presence of long-crested waves and current.
Apart from that, parametric study was carried out through several test runs at different
water depths and different mooring line azimuth angles. So, the first phase of model
test was aimed at providing data for the validation of the numerical model and the
second phase included parametric studies. In this section, the physical model of the
FPSO, mooring system used and the environmental conditions are described along with
the laboratory tests conducted. Moreover, the instrumentations and the data acquisition
systems for the tests are described.
3.3.1 Test Facility and Instrumentation
The experimental investigation was carried out in a 22 m long, 10 m wide, and 1.5m
deep wave tank in the offshore laboratory, Universiti Teknologi PETRONAS, Malaysia
which is shown in Figure 3.2. The detailed drawing of the wave tank is shown in Figure
3.3 with the basin plan and the east-west section. The wave tank is fitted with multi-
element HR Wallingford wave maker containing 16 paddles and wave dissipator. The
wave absorber at the other end of the wave tank consist of foam filled plate fixed to a
rigid framework. The lab is equipped with a current generator capable of generating a
maximum current of 0.2 m/s for a water depth of 1m. Qualysis Oqus 500+p 4 high
speed motion capture system with SLR optics is mounted on the walls with the
coordinates calibrated by choosing the centre of wave tank as origin. The wave
elevations were measured by twin wire wave probes. The detailed description of the
facilities and equipment is as follows:
62
Figure 3.2: UTP Offshore Lab
Figure 3.3: Wave Basin at UTP Offshore Laboratory
63
3.3.1.1 Wave maker system
The wave maker system at UTP offshore laboratory consists of wave maker, signal
generation computer, remote control unit and dynamic wave absorption beach. The
wave generator shown in Figure 3.4 has two modules with each having 8 individual
paddles that can move independently to one another. The paddles can move back and
forth to create waves in the wave basin.
Figure 3.4: H R Wallingford Wave Maker
The wave maker can generate waves of up to 0.3 m wave height and wave period
as short as 0.5 s (model scale) as per the performance graph plotted in Figure 3.5 in
water depth of 0.8 m and 1 m. The specifications of the wave maker system are given
in Table 3.1. The progressive mesh beach system at the other side of the wave tank
helps in minimising the interference from reflected waves during test runs. It is
designed to absorb the waves which are reflected from the model. It consists of foam
filled plate which is fixed to a rigid framework. The efficiency of the beach was found
to decrease slightly with bigger waves, dropping from 98.1 % to 97.4 % as the wave
height was increased from 0.05 m to 0.3 m [121]. Hence the wave height used for
model tests were limited to 0.05 m.
64
Table 3.1: Specifications of Wave Maker System
Description Value
Wave Maker Specification
Paddle Width (m) 0.62
Paddle Height (m) 1.3
Paddle Stroke (m) 1.08
Paddle Velocity (m/s) 0.87
Paddle Force (kN) 1.5
No. of Modules 2
Module Width (m) 4.98
Maximum Water Depth (m) 1
Spectra Available
JONSWAP
Bretschneider
P-M
ISSC
ITTC
BTTP
Derbyshire Coastal
Derbyshire Ocean
Neumann
Top hat
Sea State can be defined by
Wave height
Wave frequency
Fetch
Wind speed
Spectral density
65
Figure 3.5: Performance of the Wave Generator at 1 m and 0.8 m water Depth [85]
3.3.1.2 Qualysis motion tracking system
Qualysis Oqus 500+p 4 high speed motion capture system with SLR optics is used to
measure the motion response of the FPSO model. Principle of triangulation is the basis
for the measurement technique. Four infrared sensitive camera are set to view the area
where the model moves. Five infrared passive reflecting markers are fitted on the top
of the model, such that their positions relative to each other remains constant and care
should be taken to avoid the overlapping of marker reflections when the model is
moving [121]. Hence it is better to keep the markers at a minimum distance of 10 cm
– 15 cm and it is important to keep them in patterns which helps in identifying all the 5
markers at all the time.
66
The Qualysis tracking system was set to measure with 100 Hz real time frequency
while capturing and made sure that the wave elevation and load cell data is also being
recorded with the same frequency of data inputs. The 6DOF motion output from the
Qualysis motion tracking system is in the TSV file formats which comprises the number
of frames, number of markers, number of cameras, frequency of measurement, time and
motion data [122].
3.3.1.3 Wave probes
Twin wire wave probes were used to measure the instantaneous wave elevations. Wave
elevations were measured mainly for 1) calibration purposes, and 2) as a means of
measuring the wave – platform interaction effects by measuring the instantaneous wave
elevations during test runs. It consists of a head which is fixed to the calibration stem
and a mounting block, that allows the calibration stem to be fixed to any vertical
surface. The wave probes were attached to the tripod with the probe diameter 6.0 mm
and length 900 mm. Wave probes were connected to computer system to monitor and
record the change of water level during each test. Each probe was calibrated regularly
to ensure the accuracy of recording by measuring the change in output voltage when
the probe is raised or lowered by a known amount in still water. This operation is
enabled by means of a calibration stem which is attached to the wave probe and which
has a succession of precisely spaced holes drilled along the length of the stem.
3.3.1.4 Load Cells and Data Loggers
TML’s submersible tension/compression load cells with low capacity (250 N) are
cylindrical shaped (80 mm diameter and 42 mm height) and light weight (0.45 kg).
They can be used to measure the mooring load with high precision as these load cell’s
internal structure has both ends fixation beam for the strain sensing element. These
67
sensors are equipped with 4-core shielded chloroprene cable which is 60m long and 6
mm diameter in size, and can produce an output rate of 3000×10-6 strain and it can be
operated in the temperature ranging from -20 oC to +70 oC.
The TML’s smart dynamic strain recorder is a compact, flash recording type 4-
channel, of dimension 15.7 cm x 8.4 cm x 4.2 cm and weight 0.5 kg. It can be used to
measures strain, DC voltage and thermocouples. Measured data is automatically stored
on a compact flash card up to 2GB. The 4-channel unit can be connected in parallel up
to 8 units (total 32 channels). It consists of a built-in un-interrupted power supply (UPS)
to function when power supply is suddenly interrupted. The highest sampling speed is
5 μs with one channel and the measured data are recorded on a specified CF memory
card at the same speed.
3.3.2 Choice of the Scale and Physical Modelling Law
As mentioned in Chapter 2, Froude’s law of scaling is found to be the most suitable

scaling law to represent the action of waves and the inertia of the body due to the
limitation in the size of the wave tank and the size of the model it can occupy. The
scaling factors shown previously in Table 2.1 was used to follow the law of similitude.
The FPSO model was chosen such that it is easy to handle as well as it is economic to
construct. To obtain reliable results, the common scales used are [122]:
1. for coastal structures – 1:150 to 1:20 in towing tank
2. in 3D wave tank – 1: 150 to 1: 80 and
3. for free and moored floating platforms – 1: 100 to 1: 10.
In this study, the FPSO model is constructed using wood and the scale used is 1: 100,
so that it is easy to handle.
68
3.3.3 FPSO Model
3.3.3.1 Model description
Berantai FPSO dimensions were used to construct the FPSO model using wood and
1:100 scale factor was used. Choosing the 1:100 scale allows easy handling of the
models as FPSO’s are normally having length in the range of 200 m – 300 m. The
fabrication was done at the Marine Teknology Lab of UTM Skudai, as they have much
experience in fabricating ship and platform models. Figure 3.6 shows the Berantai
FPSO model and the Figure 3.7 gives the detailed drawing of the FPSO model. Table
3.2 gives the FPSO model dimensions and structural data.
Figure 3.6: Berantai FPSO Model
69
Figure 3.7: Drawing of FPSO Model
70
Table 3.2: FPSO Model Details
Measurement Model (1:100) Unit
LOA 2.074 m
LBP 1.987 m
Beam 0.322 m
Depth of hull 0.17 m
Max cross sect area 0.04 m2
Waterplane area 0.575 m2
Empty Hull Weight (without top cover) 15.8 kg
Mass of model at 50% loading condition 30.9 kg
Draft at 50% loading condition 0.063 m
L.C.G (from aft) 1.09 m
V.C.G (from keel) 0.075 m
𝑘𝑥𝑥 14.5 cm
Radius of Gyration 𝑘𝑦𝑦 51.75 cm
𝑘𝑧𝑧 51.75 cm
71
3.3.3.2 Mooring system
Modeling of FPSO system involves modeling both the floating structure and the
mooring system. Due to the limitations of the wave basin mentioned earlier in this
chapter, it is common to model the mooring lines as linear springs [123]. Soft springs
were used to minimise their influence on FPSO motions and to prevent the FPSO from
drifting away. Soft linear springs with 9 N/m stiffness, 0.8 mm wire thickness, 14.5
mm outer diameter and 300 mm long (model scale) were used to represent the
horizontal spread mooring system as shown in Figure 3.8. Load cells were connected
between the model and the spring for measuring the mooring line tension at the fairlead.
It should be noted that the restraining system was pre-tensioned and clamped in a way
to ensure that no slacking of the wire occurred during the tests.
Figure 3.8: Mooring Line Arrangement for Model Tests
72
3.3.4 Laboratory Tests
3.3.4.1 General
The initial lab tests were conducted to calibrate all the waves used and to calibrate the
FPSO model to achieve sufficient draft and mass distribution. Then, after arranging the
experimental setup, free decay and static offset tests were conducted prior to the
seakeeping tests. Utmost care was taken during each phase to ensure the accuracy of
measurement and minimization of errors.
3.3.4.2 Wave calibration
Before starting the model tests, all the waves which are intended to be used in the tests
were calibrated at the model position in the absence of the model. The instantaneous
wave elevation was measured using the twin wire wave probes and each time the water
depth was changed, the wave probes were calibrated and the required water depth was
set with free surface set to zero position. During wave calibration, five wave probes
were mounted at least 2 m (model scale) apart, with one wave probe being at the centre
of the tank. Figure 3.9 shows the wave probe arrangement for wave calibration and
Figure 3.10 shows the wave calibration setup in the absence of FPSO model.
To enable the generation of wave conditions, it is necessary to know the

dimensionless Paddle Transfer Function (PTF) which relates the desired wave height
up on the model and the associated paddle movement. This relationship is dependent
on both water depth and frequency [124]. Hence it is mandatory to calibrate waves
with same wave height and period if they are used at different water depths. For regular
waves, the theoretical wave height and wave period is matched with the measured wave
height and wave period from the wave probe in the place of model, by adjusting the
gain factor in the HR wave maker software.
73
Figure 3.9: Wave Probe Arrangement for Wave Calibration
Figure 3.10: Wave probes in tank during wave calibration
74
For long-crested random waves, JONSWAP spectrum is used to represent the sea
state. Spectra is generated from the digital time history signal of the instantaneous wave
elevation output from the wave probe placed at the model position. Then keeping the
gain factor constant, PTF is adjusted until the energy density of measured and targeted
spectrums match [125]. Once the required PTF is found, it is saved for later use. This
will assure repeatability of the wave spectrum from one run to the next.
A thorough calibration of the current generation is also performed before the model
was placed in the basin. The current speed was measured with current meter covering
the model neighborhood. Several measurements were made over this grid to ensure
that the current speed is reasonably simulated and the current is reasonably steady and
uniform. In this study, the maximum current velocity used is 1 m/s for the water depth
of 1 m.
3.3.4.3 Calibration of model
Initially, the measurement of the model is taken to ensure that the model is constructed
with the specified dimensions. Then, the air weight of the model is taken in the absence
of any extra loads. The air weight of present FPSO hull model with cover is 17.1 kg.
Then the model was placed in the wave basin to check the draft in the absence of any
external loads. Once the initial draft was measured, additional weights are equally
distributed inside the hull model. The preferred ballasting technique is sand bags of
known weights. Small sand bags each weighing 1 kg was used to ballast the model to
get 50 % loading condition with a draft of 0.063015 m (model scale) and hull model
weight 30.9 kg (model scale).
The longitudinal centre of gravity of vessel can be found by using the 3-point mass
system [126]. The weight of a vessel is distributed along its length, acting downwards
over the entire structure. However, we consider all the weight to be acting vertically
75
downwards through one point which is the centre of gravity (CG). The vessel is placed
on two known weights 𝑃1 and 𝑃2 as shown in Figure 3.14, mostly 𝑃1 is measured by
keeping that end on weighing machine. Moving the vessel back and forth we find the
point where the vessel is balanced; this point is the centre of gravity. If the vessel is
perfectly even through its length, the centre of gravity will be exactly in the middle, if
it is not even, CG will be in such a position that the weight on one side will balance the
other. Once the distances 𝑙1 and 𝑙2 as shown in Figure 3.11 is measured, then
longitudinal CG at a distance 𝑋𝐶𝐺 from the aft end can be found by computing moment
about CG as following.
𝑃1
𝑋𝐶𝐺 = 𝑃 (𝐿𝑂𝐴 − 𝑙2 − 𝑙1 ) + 𝑙2 (3.1)
1 +𝑃2
Figure 3.11: Three Point Mass System
To measure the vertical CG of the model, the model was hanged at a universal
joint that is free to swing to perform an inclination test. By lifting the bow of the model,
the lifting load (𝐹𝑙𝑖𝑓𝑡 ) and the inclination angle (𝜃𝑖𝑛𝑐𝑙𝑖𝑛𝑎𝑡𝑖𝑜𝑛 ) were recorded along with
the distance from lifting point to rotational point (𝑑1 ). Then distance from CG to
76
rotational point, 𝑑𝑐𝑔 is obtained by substituting the data recorded to the following
formula,
𝐹 ×𝑑1
𝑙𝑖𝑓𝑡
𝑑𝑐𝑔 = 𝑊𝑒𝑖𝑔ℎ𝑡⁡𝑜𝑓⁡𝑚𝑜𝑑𝑒𝑙×sin (3.2)
𝜃 𝑖𝑛𝑐𝑙𝑖𝑛𝑎𝑡𝑖𝑜𝑛
3.3.4.4 Experimental setup
In the present study, the FPSO model with the horizontal mooring system was kept at
the centre of the wave tank. Four wave probes were used to measure the instantaneous
water surface elevation, each kept at the four sides of FPSO without obstructing the
view of the motion capture system. The maximum current which can be generated for
1m water depth was measured using Vectrino velocimeter. Five trackables were kept
on top of the FPSO to measure the displacement of the FPSO by reflecting the invisible
infrared light emitted by the Qualysis Oqus cameras. The suitably ballasted FPSO
model with 50% DWT loading condition was held on position using the horizontal
mooring system which consists of four soft springs of stiffness 9N/m connected with a
cable is used to hold the FPSO on position. Load cell was connected between the FPSO
and mooring line to measure the tension in the mooring line. In-place calibration of the
load cells over the expected measurement range was performed. A pretension of 2.804
N (model scale) was given on each mooring line while the FPSO was clamped to the
centre of the tank, to make sure that the FPSO is floating with its equilibrium position
as the centre of the tank, and to minimize the error in measuring the displacement of
FPSO. The spread moored FPSO was thus oriented along the centre line of the wave
tank with its bow facing the wave maker to simulate the head sea condition. The layout
of the experimental setup is as shown in Figure 3.12 and Figure 3.13 shows the FPSO
model in the wave tank.
77
Figure 3.12: Experimental Setup
Figure 3.13: FPSO Model in Wave Tank
78
3.3.4.5 Static offset test
The static offset test was conducted to obtain the mooring system stiffness. The
mooring line tensions were measured using the attached load cell-data logger system.
The mooring line tensions were recorded for every 2 cm (model scale) incremental
displacement of the model. Stiffness was obtained from the slope of the restoring force
and displacement plot.
3.3.4.6 Free decay test
The system natural period and damping ratio was found by conducting a free decay test.
An initial displacement was given to the restrained model in the desired DOF and
released to move freely. The free decay time series were recorded using the Qualysis
motion tracking system and the natural period in each DOF was obtained from the
respective time series plot.
3.3.4.7 Seakeeping tests
To investigate the dynamic motion responses of the model in the seakeeping condition,
regular as well as random waves, both long-crested and short-crested were generated.
The 6 DOF motion responses were captured using the Qualysis motion capture system.
To obtain the 6 DOF response amplitude operators for the model, the wave elevations
were generated and measured prior to the installation of the models as shown in section
3.3.4.2 by the wave probe placed at the same location where the models are now
installed. Long-crested regular and random waves were recorded for the duration of 3
minutes and 6 minutes respectively. Short-crested waves were generated and recorded
for 3 minutes [19]. JONSWAP spectrum with peak enhancement factor, 3.3 was used
to represent the random sea state.
79
Seakeeping tests were done for two purposes. One, for validating the numerical
model and the other, is to conduct parametric studies. Validation of the software
simulation model was done using the long crested white noise random wave generated,
as the water surface time series signal is infinitely unique, the statistical properties of
the waves are regarded as more similar to those found in nature. This should mean the
behaviour of the model under test should resemble the full-scale system more
accurately, especially in the extremes [127], [128]. Then the frequency dependent
motion RAOs for first order systems (linear) are obtained as [65]
𝑅𝑆 (𝜔)
𝑅𝐴𝑂⁡(𝜔) = ⁡ √ 𝑆(𝜔) (3.3)
The effect of water depth was studied under the action of regular waves and long-
crested waves were used to study the effect of mooring line azimuth angles on FPSO
motions. Also, a comparison study was done to identify the effect of short-crested
waves with directional spreading 𝑐𝑜𝑠 2 𝜃 in different mooring line azimuth angles.
The effect of water depth was studied by conducting sea keeping tests at 0.62 m,
0.70 m, 0.75 m, 0.85 m and 1 m. Table 3.3 shows the regular waves used for this study
and their calibrated values at each water depths. The wave module defined the regular
wave as the sine function. The motion RAOs when subjected to regular waves are
obtained as [65]
𝑅𝑒𝑠𝑝𝑜𝑛𝑠𝑒⁡(𝑡)
𝑅𝐴𝑂 = (3.4)
𝜂(𝑡)
The effect of mooring line azimuth angle was studied by varying mooring line
azimuth angle from 15o to 55o at the maximum water depth possible in the wave tank,
i.e. 1m. Both long-crested random waves and short-crested random waves were used
in the study. Table 3.4 shows the long-crested random waves used and Table 3.5 shows
the short-crested random waves used for this study.
80
Table 3.3: Regular Wave Series
Wave Targeted Measured
Wave Wave
Frequency Wave Height Wave Height
Series Period (s)
(Hz) (m) (m)
For Water Depth 1 m
Wave 1 1.25 0.8 0.04 0.0398
Wave 2 1.1 0.91 0.04 0.0399
Wave 3 0.9 1.11 0.04 0.0396
Wave 4 0.8 1.25 0.04 0.0392
Wave 5 0.6 1.66 0.04 0.0409
Wave Wave
series Period (s)
(Hz) (m) (m)
For Water Depth 0.85 m
Wave 6 1.25 0.8 0.04 0.0405
Wave 7 1.1 0.91 0.04 0.0401
Wave 8 0.9 1.11 0.04 0.0395
Wave 9 0.8 1.25 0.04 0.0386
Wave 10 0.6 1.66 0.04 0.0388
Wave Wave
series Period (s)
(Hz) (m) (m)
Wave 11 1.25 0.8 0.04 0.0381
Wave 12 1.1 0.91 0.04 0.0379
Wave 13 0.9 1.11 0.04 0.0398
Wave 14 0.8 1.25 0.04 0.0385
Wave 15 0.6 1.66 0.04 0.0381
Wave Wave
series Period (s)
(Hz) (m) (m)
Wave 16 1.25 0.8 0.04 0.0401
Wave 17 1.1 0.91 0.04 0.0377
Wave 18 0.9 1.11 0.04 0.0398
Wave 19 0.8 1.25 0.04 0.0409
Wave 20 0.6 1.66 0.04 0.0393
81
Wave Wave
series Period (s)
(Hz) (m) (m)
Wave 21 1.25 0.8 0.04 0.0402
Wave 22 1.1 0.91 0.04 0.0385
Wave 23 0.9 1.11 0.04 0.0427
Wave 24 0.8 1.25 0.04 0.0410
Wave 25 0.6 1.66 0.04 0.0383
Table 3.4: Long-crested Random Wave Series

Wave Wave Targeted Measured
series Period (s) Significant Significant
Wave Height Wave Height
(m) (m)
LCR1 0.7 0.05 0.0471
LCR2 0.8 0.05 0.0483
LCR3 0.9 0.05 0.0488
LCR4 1 0.05 0.0472
LCR5 1.2 0.05 0.0498
LCR6 1.5 0.05 0.0523
LCR7 1.7 0.05 0.0487
LCR8 2 0.05 0.0495
LCR9 2.3 0.05 0.0482
LCR10 2.5 0.05 0.0490
82
Table 3.5: Short-crested Random Wave Series
Wave Wave Targeted Measured
series Period (s) Significant Significant
Wave Height Wave Height
(m) (m)
SCR1 0.7 0.05 0.0528
SCR2 0.8 0.05 0.0480
SCR3 0.9 0.05 0.0478
SCR4 1 0.05 0.0522
SCR5 1.2 0.05 0.0512
SCR6 1.5 0.05 0.0526
SCR7 1.7 0.05 0.0481
SCR8 2 0.05 0.0479
SCR9 2.3 0.05 0.0525
SCR10 2.5 0.05 0.0491
3.4 Dynamic Analysis of FPSO
The numerical investigation of FPSO motion responses were performed using SESAM
suit of programs. Initially, the ship lines were generated using Rhinoceros 5 3D
software and then imported to SESAM Genie V5.3-10 for further modifications and
finite element mesh generation. The finite element mesh from SESAM Genie V5.3-10
(Tn.FEM) is given as input to the SESAM HydroD V4.5-08. RAOs were obtained by
performing a hydrodynamic analysis in SESAM HydroD V4.5-08. The RAOs
generated are stored in the Hydrodynamic results interface file (G1. SI) which is then
used for the time domain analysis in SESAM DeepC V5.0-06 along with the mesh
generated in Genie. The fully coupled dynamic analysis program SESAM Deep C V5.
0-06 gives the time series plot for 6 DOF FPSO motions. Figure 3.14 shows the
communication between the programs.
83
Figure 3.14: SESAM Communication
SESAM simulations were performed to conduct parametric studies on calibrated

and validated Berantai FPSO model (Spread Moored) and OTRC FPSO model (Turret
Moored). Using the verified modelling and simulation procedure, motion responses of
other FPSOs in Malaysia and Australia were also investigated. Validation curves and
parametric study results are given in Chapter 4. The modelling and analysis procedures
and the assumptions and theories used in the program are detailed in the following
sections.
3.4.1 FPSO Modelling
The vessel hull was lofted in Rhinoceros 5 3D with the corresponding hull dimensions
and imported to SESAM Genie V5.3-10 were the FPSO hull was modified and prepared
for further use in analysis. Some portion of the hull form was generated using the
guiding tool followed by a plate skinning operation. Genie allows creating a concept
model, from which the final finite element model will be created with refined meshing.
The concept model of Berantai FPSO developed in Genie V5.3-10 is shown in Figure
84
3.15. It was made sure that the mass distribution, C G, radius of gyration and draft of
the numerical model is same as the model used for physical model testing.
Figure 3.15: FPSO Concept Model
From the concept model, a panel model was generated with only one half of the
FPSO, located in the positive global coordinates. The model was initially combined to
be a single panel and then were divided at draft, fore and aft lines. Then the panels were
then divided at equal intervals 5 m in the three co-ordinates. This ensures high quality
panel model as per [129] . The wet surfaces were assigned and a load case was assigned
were the hydro pressure was acting throughout the wet surface of the hull pointing
towards the front side of the hull plates. The super element number was assigned as 2
and a finite element mesh was generated and exported. The panel model assigned with
the wet surface is shown in Figure 3.16.
In HydroD, to connect the mooring elements, a 2D Morison element should be

present in the structure. So, a Morison model was created by providing beams for the
connection of mooring lines from the FPSO to the seabed. A mesh activity was defined
with super element number 1 and the Morison model was exported. The structural
model with super element number 3 was developed by generating the symmetrical side
of hull and scaling the mass density to match the required mass distribution.
85
Figure 3.16: Panel Model Assigned with Wet Surface
Support conditions were provided so that the FPSO will act as a rigid body. The
plates were then divided at the maximum draft, aft and fore to create a balanced mesh.
The Morison and structural model were connected in a super element hierarchy and the
finite element mesh was generated for the FPSO structural model to be used for further
analysis in Hydro D V4.5-08. Figure 3.17 shows the finite element mesh generated for
Berantai hull.
Figure 3.17: Finite Element Mesh Generated
86
3.4.2 Frequency Domain Analysis
The hydrodynamic analysis was performed using SESAM HydroD V4.5-08 with
wadam wizard V8.2-02 in frequency domain to obtain the motion response amplitude
operators.
3.4.2.1 Assumptions and theories
The flow is assumed to be ideal and the free surface condition is linearised for the first
order potential theory. Only first order wave forces are considered in this study. The
global coordinate system is right handed with the origin in the still water level. The Z-
axis is normal to the still water level and the positive Z-axis is pointing upwards.
A combination of panel and Morison model (composite model) is used since

potential theory and Morison’s equation are applied to different parts of the hydro
model. The panel model is used to calculate the hydrodynamic loads and responses
from potential theory. A Morison model is used for calculation of hydrodynamic loads
from Morison’s equation. Mooring element is used to include external restoring forces
from weightless mooring lines with linear stiffness characteristics. Since Morison
element will contribute hydrostatic and Froude-Krylov force, as it has a certain volume,
very small diameter is used for the mooring lines to avoid this.
The radiation and diffraction velocity potentials on the wet part of the body surface
are determined from the solution of an integral equation obtained by using Green’s
theorem with the free surface source potentials as the Green’s functions. The source
strengths are evaluated based on the source distribution method using the same source
potentials. The integral equation is discretized into a set of algebraic equations by
approximating the body surface with several plane quadrilateral panels. The source
strengths are assumed to be constant over each panel. Two, one or no planes of
symmetry of the body geometry may be present. The solution of the algebraic equation
87
system provides the strength of the sources on the panels. The equation system, which
is complex and indefinite is then solved by an iterative method [70].
3.4.2.2 Analysis Procedure
The uncoupled hydrodynamic analysis was performed using HydroD V4.5-08 in

wadam wizard V8.2-02. Finite element models generated in SESAM Genie V5.3-10 is
used as input to HydroD V4.5-08. In the wadam wizard settings, composite model is
chosen to represent the structural model which is composed of both panel model and
Morison model. The wave directions taken for computing the response was given with
wave period ranging from 5 s to 25 s. Mooring lines were given by linear springs with
matching stiffness and pretension as the mooring system of physical model for
validation purpose. The wadam wizard generates the other half of the panel model to
form the FPSO hull and strip model is defined to be used in iterative roll damping. The
frequency domain analysis results were presented using Postresp V6.3-01. The Figure.
3.18 shows one of the spread moored FPSO model in HydroD V4.5-08 interface.
Figure 3.18: Spread Moored Model in HydroD V4.5-08
88
3.4.3 Time Domain Analysis
Fully coupled time domain analysis was performed using SESAM DeepC V5.0-06 in
time domain to obtain the time series of motion response in 6 DOF. DeepC is a package
of software programs, consisting of also the MARINTEK’s program RIFLEX and
SIMO [9].
3.4.3.1 Assumptions and theories
The mooring lines are discretised in to several beam elements in the finite element
modelling. The FPSO vessel is considered as rigid body and treated as a nodal element.
Linear wave potential theory is used all throughout in the present study. Modified
Morison equation incorporating relative velocity term is used to find the wave load on
mooring lines while diffraction theory is used to calculate the wave load on FPSO hull
[9]. The wind velocity is simulated in the time domain by use of a state space model
using NPD spectrum. The wind is directed only in the main direction and no transverse
gust is allowed while this assumption is used. The current is described by a profile with
specified direction and speed at different elevations. Linear interpolation is used to
explicitly define the current profile. The current is taken to be constant from the lowest
level specified to the bottom [130].
3.4.3.2 Analysis Procedure
The finite element mesh is imported to SESAM DeepC V5.0-06 along with the
hydrodynamic results interface file from HydroD V4.5-08. The interaction between the
wave and FPSO are described by a set of frequency dependent coefficients for inertia,
damping and excitation forces. These coefficients are obtained from the
diffraction/radiation analysis program Wadam in HydroD V4.5-08 which is converted
to a retardation function, and the frequency dependent force is included as a convolution
89
integral, introducing a memory effect in the time domain analysis. To convert from the
frequency to the time domain, the Kramers - Krönig relations are used (convolution
integrals).
Both spread mooring and turret mooring FPSO were used for the study. The
mooring lines were modelled by inputting material, sectional and structural properties.
Both free decay and sea keeping tests were performed using SESAM DeepC V5.0-06.
Free decay test was conducted to verify if the numerical model and the physical
model has the same mass distribution and hydrodynamic performance. This also helps
to check if the mooring system is reasonable. The simulation model was calibrated to
achieve the natural periods and damping ratios as later shown in the validation.
Currently SIMO assumes that the buoyancy of a vessel equals the vessel mass. This is
usually not correct because typically, lines also pull the vessel down (in addition to the
gravity force on the vessel). To correct this erroneous buoyancy force, a force should
be applied on the vessel centre of buoyancy. The force magnitude should equal the
difference between the vessel buoyancy and mass, and its direction should be the global
z-axis [93]. For translational degrees of freedom, a single force was applied in the
particular direction, whereas for rotational degrees of freedom, a force pair was given.
The best way to initiate a free decay analysis in DeepC, is to apply a horizontal force
to give the wanted offset. The force is applied for 20 s – 30 s, and then released. The
time and magnitude of the force will depend on the wanted offset and the stiffness of
the system. The offset will be nearly linearly dependent of time and magnitude; but,
from the trial and errors performed to achieve the required offset, the hydrodynamic
performance of the system remains the same and the natural period and damping ratios
are the same for different specified force and time set.
Once the model was calibrated, the same was used for conducting several
parametric studies and the motion responses were generated as time series. Figure 3.19
shows the FPSO model in SESAM DeepC interface.
90
Figure 3.19: FPSO Model in SESAM DeepC Interface
3.5 Operability Analysis
Operability analysis is conducted to identify downtime due to green water on FPSO

deck. The study uses motion responses calculated using dynamic analysis of FPSOs
and use that as an input to identify green water on FPSO. If the FPSO is prone to green
water occurrence, this will result in a downtime cost to be used in the LCCA. The below
sections detail the analysis technique and the chosen FPSOs for operability analysis and
site specific metocean conditions used.
3.5.1 Green water on FPSO and Downtime cost
Operability analysis based on green water phenomenon is conducted by calculating

freeboard exceedance of FPSO by conducting a linear diffraction analysis using
SESAM HydroD V4.5-08 by subjecting the FPSOs to site specific annual wave scatter
data. Results are also found by subjecting the FPSO to extreme metocean conditions
with 100-year return periods. Linear diffraction analysis is found to give good estimate
91
of probability of green water on FPSO [97]. The prerequisites for this analysis are the
FEM model of the vessel and site specific metocean data. The modelling and meshing
of the FPSO is as mentioned in section 3.4.1 using SESAM Genie V5.3-10. The finite
element mesh of the vessel is then used in SESAM HydroD V4.5-08 to generate vessel
motion RAOs and wave elevation RAOs. The procedure for hydrodynamic analysis in
SESAM HydroD V4.5-08 is detailed in section 3.4.2. Vessel headings taken for the
analysis are 180°, 165° and 150° based on the design basis for wave load analysis [131].
In the crossing sea conditions, green water can come from side as wells as on bow. So,
wave elevation RAOs are calculated at nodes at 5m interval along the side of vessel
from fore to aft at draft as shown in Figure 3.20. Only one half of the FPSO is
considered as the vessel is symmetrical in transverse direction and analysis is done at
maximum operating draft for each FPSO. The rest of the calculations are done using
SESAM Postresp V6.3-01, which is the post processing tool for SESAM HydroD V4.5-
08. The procedure to obtain freeboard exceedance is as following:
Step 1: Using SESAM Genie V5.3-10
Finite element mesh is created in SESAM Genie V5.3-10 and co-ordinates of points at
which wave elevation RAO must be calculated is recorded.
Figure 3.20: Points along the vessel hull at which wave elevation RAOs are calculated
in SESAM HydroD V4.5-08
92
Step 2: Using SESAM HydroD V4.5-08
Hydrodynamic analysis is performed in SESAM HydroD V4.5-08 with offbody points

as the co-ordinates of the points where wave elevation RAO should be calculated. The
results from SESAM HydroD V4.5-08 is viewed in SESAM Postresp V6.3-01 where
all outputs are based on the global co-ordinates where z axis is at the MWL.
Step 3: Using SESAM Postresp V6.3-01
Specific points at which wave elevation RAOs are obtained is automatically created in
SESAM Postresp V6.3-01, since offbody points are given in hydrodynamic analysis in
SESAM HydroD V4.5-08 mentioned in Step 2. A motion response amplitude variable
is created in this location to calculate absolute vertical motion of vessel with combined
effect of heave, roll and pitch as per Eq. 3.5. Global RAOs are calculated at the CG of
FPSO and 𝑥 and 𝑦 are the distance from CG to the specific point in x and y direction
respectively.
𝐶𝑜𝑚𝑏𝑖𝑛𝑒𝑑⁡𝑣𝑒𝑟𝑡𝑖𝑐𝑎𝑙⁡𝑚𝑜𝑡𝑖𝑜𝑛⁡𝑅𝐴𝑂⁡𝑜𝑓⁡𝐹𝑃𝑆𝑂⁡𝑎𝑡⁡𝑎⁡𝑝𝑜𝑖𝑛𝑡⁡ = 𝐺𝑙𝑜𝑏𝑎𝑙⁡𝐻𝑒𝑎𝑣𝑒⁡𝑅𝐴𝑂 +
(𝑦 × 𝐺𝑙𝑜𝑏𝑎𝑙⁡𝑅𝑜𝑙𝑙⁡𝑅𝐴𝑂) − (𝑥⁡ × 𝐺𝑙𝑜𝑏𝑎𝑙⁡𝑃𝑖𝑡𝑐ℎ⁡𝑅𝐴𝑂) (3.5)
Then the relative motion RAO is computed by taking the difference in wave elevation
RAO and combined vertical motion RAO at each specific point. A relative motion
spectrum is then created based on sea state from wave scatter diagram using JONSWAP
spectrum as shown previously in chapter 2, section 2.2.5.1 with 𝛾 as 3.3 which
represents the most hostile weather condition [131] for the results to be on conservative
side. From the created relative motion spectrum, Rayleigh distribution as shown in Eq.
3.6 is used to obtain the annual probability maximum and most probable maximum
relative motion response of FPSO as it is tested and validated by experiments [97].
−𝐹𝐵2
𝑃𝐸 {𝑅𝑚 > 𝐹𝐵} = 𝑒𝑥𝑝 {2𝑚 } (3.6)
0𝑅
93
To understand short term green water effects, exceedance was computed for a
probability of 0.01 in the case of extreme wave, wind and current conditions with 10-
year return period and most probable maximum relative motion was found for annual
wave scatter data corresponding to a probability of 0.63 [131].
Figure 3.21: Representation of Relative Wave Motion
From the obtained maximum relative motion, freeboard of the vessel at maximum
operating draft was subtracted to calculate the freeboard exceedance as shown in Eq.
3.7 as represented in Figure 3.21.
𝐹𝑟𝑒𝑒𝑏𝑜𝑎𝑟𝑑⁡𝐸𝑥𝑐𝑒𝑒𝑑𝑎𝑛𝑐𝑒 = 𝑀𝑎𝑥𝑖𝑚𝑢𝑚⁡𝑅𝑒𝑙𝑎𝑡𝑖𝑣𝑒⁡𝑚𝑜𝑡𝑖𝑜𝑛 − 𝐹𝑟𝑒𝑒𝑏𝑜𝑎𝑟𝑑 (3.7)
If the relative motion of FPSO exceeds the freeboard by less than 3 m, then the
susceptibility to green water is low. If it is between 3 m – 6 m, then it is medium
susceptibility and if it exceeds 6m, FPSO is highly susceptible to green water.
Downtime will occur if the freeboard exceedance is higher than 3 m [88]. If the vessel
is prone to risk from green water occurrence, then in the long term, downtime (days)
can be calculated by multiplying the probability of occurrence of green water with
corresponding annual joint probability of that particular wave height and time period
responsible for green water to the number of days in a year. Downtime cost per year
can then be calculated by multiplying number of days of downtime with price of oil per
barrel per day in USD.
94
3.5.2 FPSOs for operability analysis and site specific metocean data
The FPSOs in Australia and Malaysia chosen for the operability analysis based on green
water and subsequent Life-cycle cost study are shown in Table 3.6, along with their
structural, mooring and hull details. To account for the annual life-cycle cost in LCCA,
annual downtime cost needs to be calculated, that is if any exists. Hence approximate
downtime cost is calculated by using FPSO models generated with same dimension as
mentioned in Table 3.6 in the absence of original ship lines. Getting original ship lines
from the operators are difficult as they are confidential. However, these models are
analyzed for site specific wave conditions where they are operating.
For Malaysian oil fields, Omar et al [132] gives the joint annual probability
distribution of 𝐻𝑠 and 𝑇𝑝 in parts per thousand for selected zones in Malaysia. The
chosen FPSOs in Malaysian seas for operability analysis and LCCA are operating in
these selected zones. Operation location for these FPSOs are identified in these zones
and are marked on Malaysian oil fields map obtained from [2] as shown in Figure 3.22.
Zones are marked using the same name as in Omar et al for ease of identification of
metocean data used. The annual wave scatter data in terms of joint annual probability
in percentage is given in Table 3.7 – 3.9 for Malaysian locations. Percentage probability
is calculated as no. of occurrence of each wave divided by 1000 and multiplied by 100
(For example, 78 waves in 1000 occurrence means 7.8% or in fraction 0.078).
Generally, Malaysian seas have low wave heights with peak period in the range of
6 s to 7 s [132], whereas Australian seas have higher peak periods [133]. Annual wave
scatter data for Australian locations, where chosen FPSOs are operating are obtained
from Metocean View [133] in terms of annual joint probability distribution of 𝐻𝑠 and
𝑇𝑝 in parts per hundred thousand and the locations are as marked in Figure 3.23 on map
obtained from [134] .
95
Table 3.6: FPSOs for operability analysis and LCCA Study [2]
LEASED CONVERTED
HULL HULL HULL WATER
FPSO/FSO * DRAFT FB (YES-Y (C) OR MOORING
STATUS LOCATION LENGTH WIDTH DEPTH DWT(MT) DEPTH
NAME (m) (m) OR NO- NEWLY TYPE
(m) (m) (m) (m)
N) BUILT (N)
PERISAI
O MALAYSIA 264 41 22 13 9 127540 60 Y C ET
KAMELIA
KIKEH O MALAYSIA 337 55 27 21 6 273000 1350 Y C ET
CENDOR 7.6
O MALAYSIA 245 41 21.6 14 100020 63 Y C SM
II(ONOZO)
BERANTAI O MALAYSIA 207 32 17 12.6 4.4 55337 55 Y C SM
NINGALOO 9
O AUSTRALIA 238 42 24 15 101832 350 Y C SM
VISION
BUNGA 7
O MALAYSIA 233 43 19 12 87768 60 Y C STP
KERTAS
GLAS DOWR O AUSTRALIA 242 42 21 15 6 105000 344 Y N IT
MODEC 7.14
O AUSTRALIA 258 46 24 16.86 149686 492 Y C IT
VENTURE II
STYBARROW 12.3
N AUSTRALIA 265 48 24 11.7 140000 825 Y N IT
VENTURE
PYRENEES 7.2
O AUSTRALIA 274 48 23 15.8 143690 200 Y C IT
VENTURE
OKHA O AUSTRALIA 274 48 23 16.89 6.11 158000 78 N C RTM
NGANHURRA O AUSTRALIA 260 46 26 14 12 150000 400 N N RTM
*
O-Operating, D- Decom, Not Operating- N
96
Figure 3.22: Operation location for selected Malaysian FPSOs for Operability analysis and LCCA
(source: Offshore Magazine, 2013 & Omar et al [132])
97
Figure 3.23: Operation location for selected Australian FPSOs for Operability analysis and LCCA
(source: GEOATLAS, 2014)
98
Table 3.7: Annual wave scatter table for Zone A to be used for operability analysis of
Perisai Kamelia FPSO
Hs(m)
>3.0
2.5-3.0
2.0-2.5 0.10 0.10
1.5-2.0 0.10 1.20 4.30 0.20
1.0-1.5 0.20 0.04 7.50 15.60 1.70 0.10 0.10 0.02
0.5-1.0 0.80 26.40 31.30 3.60 0.40 0.10
0-0.5 4.00 2.20 0.02
Tp(s) 0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 9-10 >10
Table 3.8: Annual wave scatter table for Zone D to be used for operability analysis of
Berantai, Bunga Kertas and Cendor II FPSO
Hs(m)
>3.0
2.5-3.0 0.03
2.0-2.5 1 2
1.5-2.0 1.4 12.8 0.6
1.0-1.5 0.1 4.9 12.9 3.4 0.03
0.5-1.0 0.9 23.7 27.6 4.3 0.3 0.1
0-0.5 0.1 3.2 0.6
Tp(s) 0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 9-10 >10
Table 3.9: Annual wave scatter table for Zone I to be used for operability analysis of
Kikeh FPSO
Hs(m)
>3.0
2.5-3.0
2.0-2.5 0.1 0.6
1.5-2.0 1.3 6.7 2.2 0.03
1.0-1.5 5.1 20.8 5 0.6 0.02
0.5-1.0 0.02 12.7 31.7 9.6 0.8
0-0.5 1.4 1.3
Tp(s) 0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 9-10 >10
99
Table 3.10: Annual wave scatter table at location 21°S 114°E to be used for operability analysis of Nganhurra, Pyrenees Venture,
Stybarrow Venture and Ningaloo Vision FPSO
Hs(m)
0-0.5
0.5-1 0.19 0.18 0.16 0.12
1-1.5 0.25 1.00 2.41 2.92 1.94 0.89 0.36 0.46 0.14
1.5-2 0.21 0.11 0.12 0.15 0.31 1.11 4.10 8.28 6.74 3.14 1.18 1.37 0.28 0.34 0.16
2-2.5 0.42 0.25 0.14 0.31 0.68 2.07 6.77 9.55 5.70 1.93 1.67 0.46 0.35 0.19
2.5-3 0.12 0.36 0.15 0.14 0.23 0.55 2.17 5.34 4.97 2.06 1.44 0.43 0.20 0.12
3-3.5 0.14 0.12 0.31 1.45 2.36 1.63 1.09 0.26
3.5-4 0.25 0.70 0.85 0.66 0.18
4-4.5 0.13 0.23 0.39 0.11
4.5-5 0.10
5-5.5
Tp(s) <5 5-6 6-7 7-8 8-9 9-10 10-11 11-12 12-13 13-14 14-15 15-16 16-17 17-18 18-19 19-20
Table 3.11: Annual wave scatter table at location 19°S 116°E to be used for operability analysis of Okha and Modec Venture II
Hs(m)
0-0.5
0.5-1 0.21 0.72 1.62 1.66 1.17 0.51 0.23 0.26 0.10
1-1.5 0.16 0.38 0.24 0.20 0.22 0.44 1.31 4.83 8.65 6.87 3.34 1.27 1.44 0.27 0.36 0.17
1.5-2 0.99 0.94 0.42 0.20 0.38 0.73 2.37 7.11 9.02 6.00 2.11 2.02 0.48 0.37 0.14
2-2.5 0.31 1.28 0.75 0.17 0.23 0.19 0.43 1.90 4.70 4.39 2.09 1.53 0.42 0.25 0.14
2.5-3 0.60 0.87 0.15 0.19 0.97 1.61 1.34 1.07 0.29
3-3.5 0.54 0.16 0.30 0.38 0.48 0.14
3.5-4 0.19 0.16
4-4.5
4.5-5
5-5.5
Tp(s) <5 5-6 6-7 7-8 8-9 9-10 10-11 11-12 12-13 13-14 14-15 15-16 16-17 17-18 18-19 19-20
100
Table 3.12: Annual wave scatter table at location 10.5°S 125°E to be used for operability analysis of Glas Dowr FPSO
Hs(m)
0-0.5 0.13 0.17 0.11 0.14 0.58 1.64 1.93 1.47 0.58 0.25 0.30
0.5-1 0.16 0.56 0.93 1.18 0.68 0.38 0.42 1.40 5.11 11.05 11.79 6.26 2.31 2.55 0.38 0.56 0.16
1-1.5 1.06 2.22 2.30 1.55 0.43 0.28 0.28 0.72 3.13 6.95 5.63 2.36 2.20 0.47 0.36 0.15
1.5-2 1.24 3.23 1.21 0.40 0.21 0.25 0.11 0.25 1.09 1.56 1.10 0.97 0.24 0.11
2-2.5 0.93 1.18 0.18 0.11 0.14 0.15 0.19
2.5-3 0.55
3-3.5
3.5-4
4-4.5
4.5-5
5-5.5
Tp(s) <4 4-5 5-6 6-7 7-8 8-9 9-10 10-11 11-12 12-13 13-14 14-15 15-16 16-17 17-18 18-19 19-20
Table 3.13: Representative Extreme cyclonic condition in NWA [61]

Wave Wind Current
velocity
Hs(m) Tp(s)
(m/s) velocity (m/s)
100-year RP 14 14.5 44 1.9
1-year RP 3.2 8.5 14 0.6
101
The annual wave scatter data obtained from Metocean View is given in Table 3.10 –
3.12 in terms of their joint probability distribution in percentage. Percentage probability
is calculated as no. of occurrence of each wave divided by 10000 and multiplied by 100
(For example, 78 waves in 100,000 occurrence means 0.078% or in fraction 0.00078).
Downtime due to green water events using these wave scatter tables have been
performed for Malaysia and Australia. However, these data are only for wind generated
seas and does not consider the cyclonic conditions which North Western Australia is
prone to as given in Table 3.13. Hence analysis has also been conducted on chosen
Australian FPSOs to identify freeboard exceedance when subjected to 100-year
extreme cyclonic conditions to be used by design engineers as per [131].
3.6 Life - Cycle Cost Analysis of FPSO
Life - cycle cost analysis of FPSO is conducted based on the following procedure. The
cost data for FPSOs in Australia and Malaysia have been collected to compare the life
cycle cost of FPSOs with different mooring configurations and hull conditions. The
section below details the LCCA procedure adopted to calculate the LCCA for FPSOs
as shown in Figure 3.24.
3.6.1 Identification of design alternatives to be compared
The two main design alternatives which are going to be compared are mooring system
(spread/ ET/ IT/ RTM/ STP) and hull condition (newly-built/converted). The life-cycle
phases of an FPSO can be discretised as shown in Section 2.2.8. Out of the life cycle
phases identified, this study considers the capital cost for the whole project (C1),
operation and maintenance (C3+C4), downtime cost due to production loss in the event
of green water on FPSO deck (C6) and revenue from oil production (B9). The same is
shown in Figure 3.24.
102
FPSO
Design Alternatives
 Mooring System
 Hull Condition
Life-cycle Phases and Cost Data

 Capital
 Operation & Maintenance
 Downtime Cost
 Revenue from Oil Production
Life-cycle Cost for project life (N years)

 Real costs
 Discounted Costs
Net Present Value

 Sensitivity Analysis
NPV > or close to zero NPV < or far from zero

Minimum life-cycle cost Maximum life-cycle cost
v
Best design Alternative Second best design alternative
based on cost based on cost
Figure 3.24: LCCA Procedure for FPSO
3.6.2 Establishment of basic assumptions and determination of exact LCC

procedure to be adopted.
Assumptions on life expectancy of FPSO (N), cash flow timings, resale/residual values,
inflation, discount rate, source and reliability of data and comprehensiveness of life
cycle costing is the next step of the procedure.
103
3.6.2.1 Life expectancy of FPSO
Newly-built FPSOs are generally designed for 20-30-year fatigue life. But the
converted ones are usually designed for a short period of 10 years; hence two life- cycle
periods are chosen for this study, 10 years and 25 years.
3.6.2.2 Cash flow timings
Capital cost is included in the initial (zeroth)year, while annual operation and
maintenance cost, annual revenue from oil production and annual downtime cost (that
is if any exists as per the motion response of FPSO under wind generated sea condition
which can result in green water as shown in section 3.5.1) is included from the 1st year
through-to Nth year.
3.6.2.3 Discount rate
Inflation rate for Australian economy in 2017 can be taken as 1.9% from the Reserve
Bank of Australia [135]. Treasury bond rate of return of 2.8% based on a reasonable
10-year yield from Investing [136] and an average equity return rate of 7.003% taken
from Market Risk Premia [137]. Then the discount rate for Australian economy is
2.96% as per the method specified by Royal Institution of Chartered Surveyors (RCIS)
as shown in Eq. 3.66. The discount rate has been decreasing since 2011, from 5.56% to
2.96% in 2017 [138]. Inflation rate for Malaysian economy in 2017 can be taken as
3.7% from the Bank Negara Malaysia [139]. Treasury bond rate of return of 4.1%
based on a reasonable 10-year yield from Investing [136] and an average equity return
rate of 7.001% taken from Investing [136]. Then the discount rate for Malaysian
economy is 1.87% as per the method specified by RCIS as shown in Eq. 3.66. A
Discount-rate is required to assist with an understanding of the time value of money
such that a dollar today is worth more than a dollar in the future. The empirical formula
104
gives the discount rate based on the relevant input data as shown in Eq. 3.8, Eq. 3.9 and
Eq. 3.10 shows how to calculate: no risk return and average risk premium discount rate
from inflation rate, treasury bond rate of return and average equity return.
𝐷𝑖𝑠𝑐𝑜𝑢𝑛𝑡⁡𝑟𝑎𝑡𝑒 = 𝑁𝑜⁡𝑟𝑖𝑠𝑘⁡𝑟𝑒𝑡𝑢𝑟𝑛 + 0.5 × ⁡𝐴𝑣𝑒𝑟𝑎𝑔𝑒⁡𝑟𝑖𝑠𝑘⁡𝑝𝑟𝑒𝑚𝑖𝑢𝑚⁡𝑑𝑖𝑠𝑐𝑜𝑢𝑛𝑡⁡𝑟𝑎𝑡𝑒 (3.8)
𝑁𝑜⁡𝑟𝑖𝑠𝑘⁡𝑟𝑒𝑡𝑢𝑟𝑛 = 𝑇𝑟𝑒𝑎𝑠𝑢𝑟𝑦⁡𝑏𝑜𝑛𝑑⁡𝑟𝑎𝑡𝑒⁡𝑜𝑓⁡𝑟𝑒𝑡𝑢𝑟𝑛 − ⁡𝐼𝑛𝑓𝑙𝑎𝑡𝑖𝑜𝑛⁡ (3.9)
𝐴𝑣𝑒𝑟𝑎𝑔𝑒⁡𝑟𝑖𝑠𝑘⁡𝑝𝑟𝑒𝑚𝑖𝑢𝑚⁡𝑑𝑖𝑠𝑐𝑜𝑢𝑛𝑡⁡𝑟𝑎𝑡𝑒 =⁡
𝐴𝑣𝑒𝑟𝑎𝑔𝑒⁡𝑒𝑞𝑢𝑖𝑡𝑦⁡𝑟𝑒𝑡𝑢𝑟𝑛 − 𝑇𝑟𝑒𝑎𝑠𝑢𝑟𝑦⁡𝑏𝑜𝑛𝑑 (3.10)
3.6.2.4 Source and reliability of data
Data collection is a key step in the phases of LCCA application and utmost care has
been taken in obtaining them. The data collection in this study is described in the next
section.
3.6.3 Data Collection
The data collection for LCCA was done as per the method specified in Life-cycle
costing standard DS/EN ISO 15663-1 [140]. A data collection sheet was prepared and
sent to the potential data givers who are industry practitioners and experienced in
different phases of FPSO project. Cost data elements were kept simple as the detailed
cost breakdown for an FPSO project is confidential and unavailable. The data givers
were asked to provide the total capital cost of FPSO, operation cost, maintenance cost
and lease rate for leased FPSOs. From the various FPSOs for which cost data were
obtained, only the FPSOs in Australia and Malaysia were chosen for the life-cycle cost
study. Cost data were obtained from practitioners in PETRONAS, Malaysia,
CHEVRON, Australia, WOOD MACKENZIE Asset reports, published articles and
news letters on FPSO and other industry websites [141]–[152]. The maximum oil
105
production data are obtained from Offshore Magazine [2] [153] while downtime cost is
calculated from the motion response generated through dynamic analysis.
3.6.4 Cost Data for FPSOs
The FPSOs under study are noted as alternatively: spread mooring, internal turret
mooring, external turret mooring, submerged turret mooring and finally riser turret
mooring. Most of the new FPSOs have riser turret mooring while the older systems are
either spread moored or internal or external turret moored.
The capital cost includes the total cost of development of the project, i.e. the cost
of mooring, risers, wells, subsea and floater, cost of hull construction and EPC
(Engineering, Procurement and Construction) related expenditure which includes the
equipment cost for topside. The daily bare boat charter (BBC) rate and daily operation
and maintenance data were converted to annual data by multiplying with 2200 working
hours per year as the average billable working hours per year is 2200 [154]. The capital
cost, operation cost, maintenance cost and lease rate of FPSO are obtained from [141]–
[152].
Downtime cost due to green water events can be calculated by knowing FPSO
motion response responsible for the event by conducting a dynamic motion response
analysis as shown in section 3.4.2 and 3.5.1. Downtime cost per year can then be
calculated by multiplying number of days of downtime with price of oil per barrel per
day in USD and oil produced in barrels per day. It is assumed here that the FPSO
produces oil to its maximum capacity and then the revenue generated per year from oil
production can be obtained by multiplying maximum oil production capacity in barrel
of oil per day with 2200 working hours per year and cost of oil per barrel in USD. The
maximum oil production capacity of FPSOs can be obtained from Offshore Magazine
[2][153].
106
3.6.5 Life-cycle Cost Calculation for FPSO
The detailed cost data as per Eq. 2.18 has been adapted such that the life cycle cost for
FPSO is calculated based on capital cost (O1-outgoing 1), Operation and Maintenance
cost (O2- outgoing 2) over the life cycle period, Lease rate for FPSO (if the hull is
leased) (O3- outgoing 3) over the life cycle period, downtime cost calculated for green
water events (O4- outgoing 4) over the life cycle period and income from producing oil
(I1- Incoming 1) over the life cycle period. O2 over the life cycle period is calculated
as the sum of operation and maintenance cost from 1st year to Nth year, O3 is calculated
as the sum of lease rate of FPSO from 1st year to Nth year and O4 is calculated as the
sum of downtime cost of FPSO from 1st year to Nth year. O4 can be zero if there is no
downtime. I1 is calculated as the sum of income generated from oil production from 1st
year to Nth year which is taken as negative while calculating 𝐿𝐶𝐶𝐹𝑃𝑆𝑂 . All outgoing
cash flow as positive value and incoming cash flow as negative value. Therefore, life
cycle cost of FPSO is given by,
𝐿𝐶𝐶𝐹𝑃𝑆𝑂 = 𝑂1 +𝑂2 + 𝑂3 + 𝑂4 − 𝐼1 (3.11)
3.6.6 NPV Calculation for FPSO
Once capital costs, annual operation and maintenance cost, downtime cost and revenue
generated from oil production are calculated, real costs incurred per year (Rn) is
calculated, mainly consisting of annual operation and maintenance cost and downtime
cost as negative value and income from oil as positive value from 1st year to Nth year;
in the zero th year incorporating capital costs as negative value. Then discount factor as
shown in Eq. 3.12 is multiplied with every years’ real costs. If the discount factor for
year n is calculated, then
1
𝐷𝑖𝑠𝑐𝑜𝑢𝑛𝑡⁡𝐹𝑎𝑐𝑡𝑜𝑟𝑛 = ⁡ (1+𝑑𝑖𝑠𝑐𝑜𝑢𝑛𝑡⁡𝑟𝑎𝑡𝑒)𝑛 (3.12)
107
Hence for the zero th year, the discount factor will be 1 and gradually reduces as the life
of the FPSO expires. This takes into account the time value of the money. Then the
discounted present value (PVn) of the FPSO for each year is calculated as the product
of real costs incurred in that year (Rn) and Discount Factor for that year (Discount
Factorn) as shown in Eq. 3.13.
𝑃𝑉𝑛 = 𝑅𝑛 × 𝐷𝑖𝑠𝑐𝑜𝑢𝑛𝑡⁡𝐹𝑎𝑐𝑡𝑜𝑟𝑛 (3.13)
Finally, NPV is calculated as the sum of discounted present values from year zero to
year N as shown in Eq. 3.14.
𝑁𝑃𝑉 = ∑𝑁
𝑛=0 𝑃𝑉𝑛 (3.14)
A detailed spreadsheet was developed to compute the life-cycle costs and NPV for
various FPSOs.
3.6.7 Risk/Uncertainty Assessment
Sensitivity analysis has been carried out as per Whyte’s [103] method to calculate NPV
of best and worst systems by using discount rates higher and lower than the actual one;
in other words, to incorporate risks, a sensitivity analysis has been performed by
varying the discount rate from 2% to 10% and the procedure described in the above
section is applied to calculate NPV of the design alternatives. This study is highly
relevant as the market condition has seen some of its worst years recently. The usual
discount rate used for evaluating infrastructures were 7% in Australia previously, but
that’s when the treasury bond rate of return was 6.8%. But now the treasury bond return
is only 2.8% [155] [136]. Hence a sensitivity analysis by taking discount rate from 2%
to 10% would cover the worst and best market conditions while reflecting the effect of
the same on NPVs of FPSOs.
108
3.6.8 Selection of Design Alternative
The design with the NPV close to zero and minimum life-cycle cost is identified as the
cost effective FPSO alternative.
3.7 Chapter Summary
The numerical and physical model testing procedures adopted to evaluate the dynamic
responses of FPSO are explained along with the model details. The modelling
procedure, frequency domain simulations and time domain simulations carried out are
discussed in detail using the theories and assumptions used during the analysis.
Operability analysis to determine downtime cost due to green water events are then
discussed and finally, the life-cycle costing procedure adopted for FPSO is detailed
along with the chosen FPSOs in Malaysia and Australia for study. The results and
discussions of this study are detailed in the next chapter.
109
CHAPTER 4
RESULTS AND DISCUSSION
In this chapter, the motion responses of FPSO obtained using model testing and
software simulation methods are comprehensively presented in the prototype scale. At
first, the model testing results are discussed, then, the uncoupled and coupled analysis
performed using SESAM HydroD V4.5-08 and SESAM DeepC V5.0-06 is compared
with the model testing results conducted in UTP Offshore laboratory and published
results of experiments conducted at OTRC laboratory, Texas A&M. The calibrated
simulation model was used to perform parametric studies for metocean data, water
depth, hull parameters and mooring system details and the results are presented here.
Then, the motion RAOs, relative motions and subsequent downtime cost generated for
the FPSOs operating in Australia and Malaysia are presented. Also, the life cycle costs
and NPVs of the FPSOs in Australian and Malaysian waters with different mooring
systems and hull conditions are discussed along with the cost proportions due to each
life cycle phase. Finally, a discussion has been made of the factors affecting the
selection of an economic and efficient FPSO with good motion performance increasing
the operational time.
4.2 Experimental Results
The results of static offset test and free decay test is given in the below sections.
4.2.1 Static Offset Test
From the static offset test, the system spring constant was found to be 168 kN/m. The
force- excursion relationship of the system is shown in Figure 4.1. The linear regression
of the collected data is plotted along with the test results.
Figure 4.1: Force- excursion relationship of FPSO with horizontal mooring system
(test result in prototype Scale)
4.2.2 Free Decay Test
The natural period of the system was found by conducting free decay test. The results
shown in Table 4.1 are calculated by taking the average of time periods for the first few
sinusoidal excitations of the free decay series. Savitzky-Golay filtering method was
used in MATLAB to eliminate noise from the time series. The free decay graphs for
the horizontal motions are given in Figure 4.2 ~ Figure 4.7. Coupling effect from
horizontal mooring line restoring, damping and inertia forces [48][156] are observed
during the free decay tests. Especially in heave, it was observed that the heave motion
amplified after 111 s and then again decayed.
111
Table 4.1: Measured Natural Periods for FPSO [56]
Motion Natural Typical
Period (s) natural
Period (s)
Surge 103.5 >100
Sway 152 >100
Yaw 74.5 >100
Heave 9.25 5 - 12
Roll 25 5 - 30
Pitch 8.7 5 - 12
Figure 4.2: Free decay time series – Surge (test result in prototype scale)
Figure 4.3: Free decay time series – Sway (test result in prototype scale)
112
Figure 4.4: Free decay time series – Yaw (test result in prototype scale)
Figure 4.5: Free decay time series – Heave (test result in prototype scale)
Figure 4.6: Free decay time series – Roll (test result in prototype scale)
113
Figure 4.7: Free decay time series – Pitch (test result in prototype scale)
4.3 Validation of Uncoupled Frequency Domain Analysis
The frequency domain simulation performed using SESAM HydroD V4.5-08 for 50%
loading condition of the Berantai FPSO at a water depth of 100 m was compared with
the results from model tests performed at the UTP Offshore Laboratory. The vessel
details are given in Table 3.2. The vessel was subjected to head sea (wave direction –
180°) and crossing sea (wave direction – 150°) condition. The vessel was anchored to
position using four soft mooring lines which has limited influence on the vessel motion.
The properties of the spring are same as the one used for model testing. The vessel
responses are compared with the physical model testing results for the long crested
white noise wave condition with significant wave height 5 m and peak period 15 s.
White noise wave generated are more analogous to the real ocean surface and does not
have the constraint of periodicity [127], [128]. In reality, ocean waves are never regular
and each wave will be unique. This method of creating a water surface elevation time
history is not constrained by the requirement that the wave components are harmonics
of the primary wave. The motion RAOs are compared in Figure 4.8 ~ Figure 4.13. The
surge, heave and pitch motions of the vessel are studied since they are the prominent
degrees of freedom in head sea condition and the sway, roll and yaw motions are
investigated in crossing sea condition.
114
Figure 4.8: Surge RAO (Comparison of HydroD and experiment results)
Figure 4.8 illustrates the surge response of Berantai FPSO subjected to long crested
waves by experimental study and SESAM HydroD V4.5-08. From the comparison,
fairly good agreement could be observed with RMSD 0.16 with the experimental and
software simulation results. The comparison of heave responses is show in Figure 4.9
with RMSD 0.14 with the experimental and software simulation results. The pitch
motion RAO shown in Figure 4.10 has an RMSD value of 0.16 with the experimental
and software simulation results.
Figure 4.9: Heave RAO (Comparison of HydroD and experiment results)
115
Figure 4.10: Pitch RAO (Comparison of HydroD and experiment results)
Figure 4.11: Sway RAO (Comparison of HydroD and experiment results)
Figure 4.12: Roll RAO (Comparison of HydroD and experiment results)
116
Figure 4.13: Yaw RAO (Comparison of HydroD and experiment results)
It can be seen in Figure 4.12 that the Roll RAO is not reducing at large wave periods
(25 s). This is due to the resonance created when the vessel natural period equals the
wave period. The natural period of vessel in roll is also 25 s as seen in Table 4.1. Also
the roll damping is low as the FPSO model used for wave tank experiments do not have
bilge keel, which is usually present in real scenario [97]. Also, the damping produced
from horizontal mooring is also less when compared to catenary mooring [84].
The sway and roll RAOs shown in Figure 4.11 and Figure 4.12 has RMSD values
0.01 and 0.016 respectively with the experimental and software simulation results. The
Yaw motion RAO has RMSD value of 0.11 with the experimental and software
simulation results. The yaw motion RAO given in Figure 4.13 for software simulation
result shows a peak around 20 s. This is because the yaw and roll motions are
undermined in the experimental investigation. This is due to the use of horizontal
spread mooring system preventing large rotation. A similar study by Xie et al using
horizontal spread moored vessel [157] also reports the same for yaw and roll rotations
obtained using experimental study.
In the comparison, it is noticeable that the numerical results agreed fairly with the
experimental results. RMSD values found for the results comparing the model test and
commercial software program results are very low with a maximum of only 0.16 for
surge. It was seen that the commercial software results are smaller compared to the
experimental results. This is due to the assumptions used in the software simulation
117
that it does not consider the higher order forces that cause smaller responses occurring
especially at higher wave periods (low frequency region) [71]. Also, model testing
results are observed to be higher than software simulation in the lower wave periods
(high frequency region). It is due to the small responses of vessel because of the high
frequency excitation of the mooring lines [71], which again is not considered in the
simulation following the assumption that linearised wave is used.
4.4 Validation of Coupled Time Domain Analysis
The coupled time domain simulation performed using SESAM DeepC V5.0-06 and the
procedures adopted were verified by doing a validation by comparing the published
results from model tests performed at the OTRC wave basin and numerical simulations
by Kim et al [42].
4.4.1 FPSO and Metocean Conditions
The coupled analysis program is validated by using the published results by Kim et al
[42]. The prototype of turret moored FPSO was moored at 1828.8 m water depth using
12 chain-polyester-chain taut lines. The 12 mooring lines were divided into four groups
normal to each other and each group consists of lines 5° apart. Four equivalent mooring
lines were used in model testing with each mooring line having the combined effect of
three mooring lines in prototype. Since the length of the mooring lines were not able
to model as per the scale used, a truncated mooring system was used with equivalent
static surge stiffness of the prototype mooring design in the model testing. The main
dimensions of the FPSO are given in Table 4.2. The turret of the FPSO was located at
38.73 m (12.5 % LBP) behind the forward perpendicular of the FPSO. The mooring
line pretension is 1424 kN and the length of the mooring line is 2652 m. The other
mooring details are given in Table 4.3 & Table 4.4.
118
Table 4.2: Main Particulars of OTRC FPSO [42]
Description Unit Quantity

LBP m 310
Beam m 47.17
Depth m 28.04
Draft (80%) m 15.121
Displacement
MT 186051
(80%)
kxx at CG m 14.036
kyy at CG m 77.47
kzz at CG m 79.3
Table 4.3: Mooring Line Particulars for OTRC FPSO [42]

Segment 1: Chain
Length at anchor point m 121.9
Diameter cm 9.52
Dry Weight N/m 1856
Weight in Water N/m 1615
Stiffness AE kN 820900
Mean Breaking Load kN 7553
Segment 2: Polyester Rope
Length m 2438
Diameter cm 16
Dry Weight N/m 168.7
Weight in Water N/m 44.1
Segment 3: Chain
Length at anchor point m 91.4
Diameter cm 9.53
Dry Weight N/m 1856
Weight in Water N/m 1615
119
Table 4.4: Hydrodynamic coefficients of the chain and polyester rope [42]
Hydrodynamic Coefficients Chain Polyester Rope
Normal Drag 2.45 1.2
Normal Added Inertia 2 1.15
The experiments and simulations were conducted for the 100-year hurricane
condition in the Gulf of Mexico. The wave condition was given by JONSWAP
spectrum with a peak enhancement factor of 2.5 and to generate wind loading, NPD
spectrum is used. The metocean conditions are given in Table 4.5. The wind spectrum
generated is compared with the published spectrum and given in Figure 4.14. The
JONSWAP spectrum is given in Figure 4.15.
Table 4.5: Metocean Data used by Kim et al [42]
Wave
Significant Wave Height m 12.19
Peak Period s 14
Direction deg 180
Wind
Velocity at 10m above
m/s 41.12
MWL
Direction deg 150
Current
Profile
At free surface 0m m/s 0.9144
At 60.96m m/s 0.9144
At 91.44m m/s 0.0914
On the sea bottom m/s 0.0914
Direction deg 210
120
Figure 4.14: Wind Spectrum Comparison
Figure 4.15: Wave Spectrum Comparison
4.4.2 Natural Periods & Damping Ratios
As explained in section 3.5.3, free decay analysis was performed using SESAM DeepC
V5.0-06 to verify if the simulation model and the prototype has the same mass
distribution and hydrodynamic performance. A specified force of 50000 kN was given
in the respective directions for 20 s where the free decaying must be studied. The
simulation model was calibrated to achieve the natural periods and damping ratios as
given in Table 4.6. The free decay graphs are as shown in Figure 4.16 ~ 4.19
respectively. The comparison for natural periods and damping ratios shows that the
121
hydrodynamic behaviour of the simulated FPSO is matching well with the Kim et al
[42] model. The results obtained from SESAM DeepC V5.0-06 varies less than 5 %
from the experimental results and WINPOST simulations conducted by Kim et al [42].
The first peaks of the free decay curve were used in calculating natural period and
damping ratios.
Table 4.6: Comparison of Natural Periods & Damping Ratios
OTRC Experiment WINPOST (4

DOF SESAM mooring lines)
DeepC V5.0-06 (Published Results, (Published Results,
Kim et al [42]) Kim et al [42])
Natural Damping Natural Damping Natural Damping

Period Ratio Period Ratio Period Ratio
(s) (%) (s) (%) (s) (%)
Surge 204.2 2.1 206.8 3 204.7 4.4
Heave 11.2 6.6 10.7 6.7 10.8 11.8
Roll 12.4 2.3 12.7 3.4 12.7 0.7
Pitch 10.6 10.45 10.5 8 10.8 10.5
Figure 4.16: Free Decay Time Series – Surge (SESAM DeepC)
122
Figure 4.17: Free Decay Time Series – Heave (SESAM DeepC)
Figure 4.18: Free Decay Time Series – Roll (SESAM DeepC)
Figure 4.19: Free Decay Time Series – Pitch (SESAM DeepC)
123
4.4.3 Response Spectra
The motion spectrums generated using SESAM DeepC V5.0-06 after time domain
analysis were compared with motion spectrums developed by Kim et al [42] from the
model testing studies and the WINPOST- FPSO simulations in time domain. The
horizontal motions of the FPSO, surge, sway and yaw are dominant in the low
frequency whereas the vertical motions of the FPSO, heave, roll and pitch are
pronounced in wave frequency ranges. The spectral density curves for the six degrees
of freedom response of the OTRC FPSO is given in Figure 4.20 ~ Figure 4.25.
Figure 4.20 illustrates the surge response of OTRC FPSO subjected to long crested
waves by experimental and numerical study conducted by Kim et al [42] and SESAM
DeepC V5.0-06 in the present study.
Figure 4.20: Surge Motion Spectrum Comparison (SESAM DeepC and Kim et al
results [42])
124
Figure 4.21: Sway Motion Spectrum Comparison (SESAM DeepC and Kim et al
results [42])
Figure 4.22: Heave Motion Spectrum Comparison (SESAM DeepC and Kim et al
results [42])
125
Figure 4.23: Roll Motion Spectrum Comparison (SESAM DeepC and Kim et al
results [42])
Figure 4.24: Pitch Motion Spectrum Comparison (SESAM DeepC and Kim et al
results [42])
126
Figure 4.25: Yaw Motion Spectrum Comparison (SESAM DeepC and Kim et al
results [42])
The comparison of sway responses is shown in Figure 4.21 with average difference in
peak values of 25%. The heave motion RAO shown in Figure 4.22 shows an average
difference of 7% with the OTRC 1 experimental and software simulation results. Pitch
and yaw motion spectrum computed using software simulation shown in Figure 4.24
and Figure 4.25 respectively varies by an average of less than 25% with the published
experimental results.
Except roll, other degrees of freedom response are matching well with the published
experimental results by Kim et al [42]; though roll motions match well with the
WINPOST simulations by Kim et al [42]. The differences between the experimental
and numerical results can be attributed to the uncertainty related to wind and current
generation or the mismatch in numerical and physical model of the FPSO mooring
system, as truncated mooring system is used in the model testing at OTRC wave basin.
The difference in the roll motion can be corrected by giving additional roll damping
during simulation [42]. Despite the differences, the results look reasonable. There by
the calibrated model and the coupled analysis procedure can be repetitively used for
further investigative study of different parameters.
127
4.5 Parametric Study
Influence of metocean data, hull details and mooring line parameters are studied using
model testing, uncoupled frequency domain analysis in SESAM HydroD V4.5-08,
where mooring lines are not significant to determine RAO and coupled time domain
analysis in SESAM DeepC V5.0-06, where mooring line coupling effects are
predominant in deep water conditions. The following sections details the findings from
the extensive parametric study conducted on FPSOs and the configurations used.
4.5.1 Influence of Metocean Data on FPSO Motions
In this section, the influence of wave height and presence of current and wind on FPSO
motions are studied using the calibrated OTRC FPSO model and Berantai FPSO model.
These two FPSOs were given both turret moored and spread moored configuration to
identify the influence of mooring system in this scenario. Since the influence of
mooring lines are significant in this case, fully coupled time domain analysis in SESAM
DeepC V5.0-06 has been performed to study the same. The OTRC FPSO with
dimensions given earlier in Table 4.2 were moored using the mooring line details given
in Table 4.3 and Table 4.4 at water depth of 1828.8 m. Fully loaded Berantai FPSO
with DWT 57999 MT and draft 12.603 m, having mooring line details as shown in
Table 4.7 at water depth 1350 m is also used for the study. Mooring line details for
Berantai FPSO were chosen such that the natural periods of the FPSO are within the
typical range after conducting a free decay analysis in SESAM DeepC V5.0-06.
To study the first harmonic motion response of turret moored and spread moored
FPSO configurations in crossing sea condition under the influence of wind,
unidirectional random waves and current, wave height was varied from 4 m to 8 m,
while peak period ranged from 5 s to 25 s. Wave height was not increased beyond 8 m
as wave breaks at height more than 8 m for low wave periods [65] [158]. The current
velocity was 4.38 m/s acting at 210°, wind velocity was 36.91 m/s acting at 225° and
the wave was directed at 225°.
128
Table 4.7: Mooring Line Details for Berantai FPSO Model
Segment 1: Chain
Length at anchor point m 120
Diameter cm 23.5
Segment 2: Polyester Rope
Length m 2220
Diameter cm 9.2
Segment 3: Chain
Length at anchor point m 80
Diameter cm 23.5
In total, the significant motion responses are studied under 8 cases. They are:
 Case1: Motion response of turret moored OTRC FPSO subjected to wind,

wave and current as shown in Figure 4.26.
 Case2: Motion response of turret moored OTRC FPSO subjected to only

wave as shown in Figure 4.27.
 Case3: Motion response of spread moored OTRC FPSO subjected to wind,

 Case4: Motion response of spread moored OTRC FPSO subjected to only

wave as shown in Figure 4.29.
 Case5: Motion response of turret moored Berantai FPSO subjected to wind,

 Case 6: Motion response of turret moored Berantai FPSO subjected to only

wave.
129
 Case 7: Motion response of spread moored Berantai FPSO subjected to
wind, wave and current as shown in Figure 4.31.
 Case 8: Spread moored Berantai FPSO subjected to only wave as shown in

Figure 4.32.
Figure 4.26: Motion response of turret moored OTRC FPSO subjected to wind, wave
and current (SESAM DeepC results)
Figure 4.27: Motion response of turret moored OTRC FPSO subjected to only wave
(SESAM DeepC results)
130
Figure 4.28: Motion response of spread moored OTRC FPSO subjected to wind, wave
and current (SESAM DeepC results)
Figure 4.29: Motion response of spread moored OTRC FPSO subjected to only wave
131
Figure 4.30: Motion response of turret moored Berantai FPSO subjected to wind,
wave and current (SESAM DeepC results)
Figure 4.31: Motion response of spread moored Berantai FPSO subjected to wind,
wave and current (SESAM DeepC results)
132
Figure 4.32: Spread moored Berantai FPSO subjected to only wave (SESAM DeepC
results)
For cases 1 and 5, in the presence of wind, wave and current for turret moored
FPSOs, it is seen from Figure 4.26 and Figure 4.30 that, as the wave height increases,
the FPSO motions in the horizontal plane decrease while those in the vertical plane
increase. For per meter increase in wave height, surge motion decreases by 0.05 m to
2.81 m for turret moored OTRC FPSO and 0.01 m to 1 m for turret moored Berantai
FPSO. Sway motion decreases by 0.09 m to 3.6 m and yaw motion decreases by 0.02
rad to 0.9 rad for turret moored OTRC FPSO, per 1 m increase in wave height. Sway
motion decreases by 0.01 m to 0.8 m and yaw motion decreases by 0.01 rad to 0.05 rad
for turret moored Berantai FPSO, per 1 m increase in wave height. The reason is that,
the hydrodynamic damping in the wave are amplified due to the presence of wind and
current [62], [159], [84]. Also, larger wave heights cause a higher velocity of the floater
which in turn results in increased hydrodynamic damping [62], [160], [161]. This limits
the FPSO motions to small amplitudes in turret moored FPSOs at higher wave heights.
The fluctuations seen for the surge, sway and yaw motions of the turret moored
OTRC FPSO subjected to wind, wave and current for wave period 10 s occur when the
wave length is odd or even multiple of FPSO length. This results in the fluctuation of
the FPSO motions. In the presence of wave only, it can be seen from Figure 4.27 that,
133
for the wave period 25 s, all the FPSO motions other than heave, show unexpected
behaviour as functions of significant wave height and wave period. It is worthwhile to
note that Tp = 25 s or more is an exceptionally rare case. At the wave period of 25 s,
the wave length is relatively high when compared to the FPSO length and, since a turret
moored configuration is used, the FPSO being smaller compared to the wave, offers
nominal resistance to the first order wave excitations and, the motions fluctuate rapidly
in all the degrees of freedom [162]. This can also be attributed to the mooring line
stiffness. When compared to the Berantai FPSO motions, the OTRC FPSO motion
amplitudes in surge, sway and yaw are relatively higher. The lesser the mooring
stiffness, the higher will be the motion amplitudes. This lesser mooring stiffness along
with the higher wave period cause the unexpected motion behaviour in case 2.
However, in cases 3 and 7, where a spread moored configuration is used, it can be

seen in Figure 4.28 and Figure 4.31 that, above certain wave period, the FPSO motions
in horizontal plane increase even in the presence of wind and current. For surge, sway
and yaw, the FPSO motions increase for wave periods above 12 s for OTRC FPSO and
above 8 s for Berantai FPSO. When the wave period increases, the wave length
increases. If the wind velocity is not high, due to the large wave area, the effect of wind
on wave damping is very small. Also, since the FPSO is restrained in all directions, it
is more sensitive to first order wave excitations [57]. Hence, due to the drop in damping
and increased sensitivity to wave motion, the amplitudes of motion in a horizontal plane
increase for a spread moored configuration. For spread moored OTRC FPSO, the
surge, sway and yaw motions increases to a maximum of 1 m, 0.6 m, 0.02 rad
respectively per 1 m increase in wave height. For spread moored Berantai FPSO, the
surge, sway and yaw motions increases to a maximum of 0.7 m, 0.4 m, 0.01 rad
respectively per 1 m increase in wave height. The fluctuations in motion at certain
wave periods occur when the wave length becomes an odd or even multiple of FPSO
length. When wave length is half of FPSO length, cancellation of wave occurs, and
hence the FPSO motions decrease. It can be seen from Figure 4.26 and Figure 4.28 for
OTRC FPSO and Figure 4.30 and Figure 4.31 for Berantai FPSO that the sway and yaw
motions decrease significantly when spread moored configuration is used.
134
The surge and sway offsets are significantly reduced in the absence of wind and
current for spread moored OTRC FPSO. The same is applicable to Berantai FPSO in
the absence of wind and current. Also, when comparing the mooring system used, the
maximum significant surge response is 32.5 m in the case of turret moored FPSO as
shown in Figure 4.26 while the maximum significant surge response for spread moored
FPSO is 26.2 m as shown in Figure 4.28 for wave frequency ranges, 5 s - 25 s. This is
almost a 20% decrease when a spread mooring system is used. Similarly, for Berantai
FPSO, the surge response is higher when turret mooring is used as shown in Figure 4.30
and Figure 4.31.
In turret moored FPSOs, in the presence of wind, wave and current as seen in
Figure 4.26 and Figure 4.30, the horizontal offsets are maximum for high frequency
wave (at low wave periods) even though the FPSO motion decreases with increase in
wave heights due to the drift force from the influence of wind and current. The mooring
system allows the FPSO to weathervane when turret moored and hence resist the
combined effects of wind, wave and current for reducing motions. This is the main
reason why turret moored FPSOs are chosen over spread moored FPSOs for adverse
climates. The horizontal plane motions of turret moored FPSO are relatively higher
when compared to that of spread moored FPSO in the presence of wind, wave and
current due to the drift force. But the interesting fact is that, the horizontal FPSO
motions for turret moored FPSO decreases as wave height increases and becomes
comparable to those of spread moored FPSO; however, if we use a spread moored
configuration in adverse climates, the motions escalate resulting in mooring line
damage. This very nature of turret moored FPSOs makes them the preferred choice for
locations with hostile weather.
Irrespective of the mooring configuration, heave motion increases from 0.02 m to

1 m, roll motion increases by 0.01 rad to 0.08 rad and pitch motion increases by 0.001
rad to 0.01 rad, per 1 m increase in wave height in wave period ranges 5 s – 25 s. Heave,
Roll and pitch motions remain relatively similar for both mooring configurations
increases maximum by 1 m, 0.08 rad and 0.01 rad respectively per meter increase in
wave height, whether current and/or wind are present or not. It is also observed that
the rate of increase in heave motion is directly proportional to wave period ranging 5 s
135
– 25 s. Higher the wave period, higher will be the heave motion in the wave frequency
ranges.
4.5.2 Influence of Water Depth and Wave Periods on FPSO Motions
Influence of water depth on FPSO motions were investigated using model tests
conducted as described in section 3.3.4.7. The horizontally moored Berantai FPSO in
head sea was subjected to calibrated regular waves shown in Table 3.3. Tests were
conducted at 62 m, 70 m, 75 m, 85 m and 100 m (prototype scale). Surge, heave and
pitch motions were studied as they are the predominant motions for a spread moored
FPSO in head sea. The variations in surge, heave and pitch RAO are plotted for
different water depths and shown in Figure 4.33.
The plot shows that the mean values of surge and heave RAOs increase as the water
depth increases. At a wave period of 9.09 s, the surge RAO is minimum and then it is
increasing until a wave period of 12.5 s and then declining at the same wave period a
bit, again to increase until the wave period of 16 s. This pattern of deviation in surge
RAO is seen as the same for all the water depths. The heave RAO shows a general
trend in increase of its value from the wave period of 8 s and a slight decline in the
value for the RAO at the wave period of 11.11 s. The general trend in the variation of
heave RAO for all the wave periods for different water depths is the same. The pattern
of change in pitch RAO for different wave periods in the range 9.09 s to 12.5 s is seen
to be same as the surge RAO. The pitch RAO is minimum at the wave period of 9.09
s and then increasing until 12.5 s and then takes a sudden dip. The variation in pitch
RAO for different wave periods is also same for different water depths.
Most of the wave momentum is directed in the horizontal direction, which results
in an increase of FPSO motion in the horizontal direction as the wave period (or wave
length) increases. As the wave length gets very large when compared to the tanker
length, the tanker in effect becomes a particle floating on the surface of the wave and
as such offers no resistance resulting in large amplitude of motion [162].
136
Figure 4.33: Variation of Surge, Heave and Pitch Motion with Water Depth
(Experimental results in prototype scale)
137
The decrease in response during the wave period of 8 s to 9 s is because, at 8 s to 9
s, the ship length is approximately twice the wave length and the effect of waves on
FPSO will be reduced as the resulting wave will be very small due to cancellation. This
is mainly because at low wave periods, shorter waves are present which gets cancelled
due to large vessel. As the wave period increases, waves get longer and will have
significant effects on vessel motion and contribute to amplifying the vessel motion.
If the length of a ship is half the waves generated, the resulting wave will be very
small due to cancellation, and if the length is the same as the wavelength, the wave will
be large due to enhancement. At 8.5 s, the wave length is half of the ship length which
results in decrease in wave momentum and hence results in decrease in response.
The study covers water depths of domestic oil zones in Malaysia ( 62 m – Erb West,
70 m – PMO, 75 m – Baram Delta) [163] and in Australia (80 m- Montara) [2]. The
study also helps in understanding variation in heave motion in shallow waters which
can contribute to green water as seen later in section 4.7.1.
4.5.3 Influence of Mooring Line Azimuth Angle on FPSO Motions
Influence of mooring line azimuth angle on FPSO motions were investigated using
model tests conducted as described in section 3.3.4.7. The horizontally moored
Berantai FPSO in head sea was subjected to calibrated long crested and short crested
random waves shown in Table 3.4 and Table 3.5 respectively at 1 m water depth (model
scale). Surge, heave and pitch motions were studied as they are the predominant
motions for a spread moored FPSO in head sea. The variations in surge, heave and
pitch motions are plotted for different mooring line azimuth angles under the action of
long crested random waves as shown in Figure 4.34.
Figure 4.34 illustrates the variation of FPSO motion with mooring line azimuth
angle when subjected to long crested random waves. The surge motion was found to
be declining when the mooring line azimuth angle was increased to 30º from 15º and
then gradually increasing when the mooring line azimuth angle was increased again to
138
45º and 55º. The surge motion declines by 8%-25% when the mooring line azimuth
angle is changed to 30º from 15º as seen in Figure 4.34.
Angle under Long Crested Waves (Experimental results in prototype scale)
139
The surge motion increases by 5%-50% when the mooring line azimuth angle is
changed to 45º from 15º; increasing as wave period decreases. In the case of heave
motion, the motion is at its least when 15º mooring line azimuth angle is used, while it
is at its maximum when the mooring line azimuth angle is kept as 30º. Then again, the
heave motion reduces when the mooring line angle is increased from 30º to 45º and
slightly increases when the angle is increased to 55º. The pitch motion is the least
sensitive to the mooring line azimuth angle variation. Out of the three responses, surge
motion is highly sensitive towards mooring line azimuth angle and found to be
minimum at a mooring line azimuth angle of 30º.
Figure 4.35 illustrates the variation of FPSO motion with mooring line azimuth
angle when subjected to short crested random waves. At 8 s, the surge motion declines
by 22% when the mooring line azimuth angle was increased to 30º from 15º. For other
wave periods, the surge motion is minimum with negligible variation at mooring line
azimuth angles 15º and 30º. The surge motion increases by 1%-20% when the mooring
line azimuth angle is changed to 45º from 15º; increasing as wave period decreases.
Like the action of long crested waves, heave motion increases when the mooring line
azimuth angle is varied from 15º to 30º and then decreases when the mooring line
azimuth angle is 45º.
Hence it is visible that at high wave periods (low frequency waves), the surge and
heave motions are minimum at mooring line azimuth angle 15º and 45º. Pitch motions
are least sensitive to mooring line azimuth angle variation. Motions are always low
when the mooring line azimuth angle is 15º and to prevent high amplitude heave
motion, it is best to avoid mooring line azimuth angle 30º for FPSOs with similar
configuration.
The difference in amplitude of FPSO motion when subjected to short crested waves
and long crested waves are clearly visible from Figure 4.34 and Figure 4.35. The
motion amplitude in surge declines by 6% to 34% when short crested waves are used.
Except from wave periods 15 s – 20 s, heave and pitch motion increases in the presence
of short crested waves. Directional spreading sometimes increases the motions, loads
and accelerations, which might be due to the sensitivity of the motions to wave heading
angle [164].
140
Angle under Short Crested Waves (Experimental results in prototype scale)
141
4.5.4 Influence of Mooring Line Length on FPSO Motions
The mooring line length parametric study has been conducted using fully coupled
analysis in SESAM DeepC V5.0-06. The spread moored coupled Berantai Model with
mooring line details shown earlier in Table 4.7 at fully loaded DWT with draft 12.6 m
at a water depth of 1350 m was exposed to waves with significant wave height 6.3 m
and peak period 16 s acting at 225º, current with velocity 3.66 m/s acting at 210º and
wind of velocity 30.3 m/s acting at 225º. The mooring line length was varied from
2200 m to 2700 m. The six FPSO motions were studied and plotted against the ratio of
ship length to mooring line length. The ratio of ship length (SL) to mooring line length
(ML) varies from 0.074 to 0.086. If the mooring line length is below 2200 m, the
analysis fails as the mooring line breaks as it becomes too taut and when the mooring
line length was increased above 2700 m, the mooring line slackness increases and
doesn’t offer much resistance.
All the six FPSO motions were found to be increasing as the mooring length was
increased from 2200 m to 2700 m or as the ratio of ship length to mooring line length
decreases. The surge motion declines by 14% to 60% as the mooring line length was
reduced to 2200 m from 2700 m as seen in Figure 4.36. Also, the rate of reduction in
surge motion with mooring line length reduction is found to be inversely proportional
to the wave period.
The sway motion declines by 10% to 40%, the roll motion declines by 50% to 70%
and the yaw motion declines up to 75% when the mooring line length was reduced to
2200 m from 2700 m.
The pitch motion is least sensitive to the mooring line length variation with the
difference in FPSO motion falling under 6.4% when the mooring length was reduced
to 2200 m and the second least varying FPSO motion is the heave degree of freedom,
falling under 18% difference in FPSO motion as seen in Figure 4.36 ~ 4.37. Hence it
is best to keep the mooring line length minimum to minimise the FPSO motions with
appropriate pretension for similarly configured FPSOs.
142
Figure 4.36: Variation of FPSO Translational Motions with Mooring Line Length
143
Figure 4.37: Variation of FPSO Rotational Motions with Mooring Line Length
144
4.5.5 Influence of Spread Mooring Fairlead Location on Hull on FPSO Motions
The location of spread mooring fairleads on hull was changed to study the variation in
six degrees of freedom FPSO motions using SESAM DeepC V5.0-06. The metocean
data and mooring line parameters are the same as used in the mooring line length
parametric study in section 4.5.4 for water depth of 1350 m. The location of spread
mooring fairleads is varied from 12% to 26% of LOA of FPSO from aft and fore.
Surge motions are the minimum when the spread mooring fairleads are kept at 21%
of LOA from aft and fore and increases highly when the mooring fairleads are kept at
25% of LOA from aft and fore, i.e. when the mooring lines are connected near to the
mid ship as seen in Figure 4.38. Per percent increase in spread mooring position as
percentage of LOA from aft and fore causes a maximum of 0.16 m increase in surge
and sway motion.
Roll motion increases maximum by 1.1º and yaw motion by 1º per percent increase
in spread mooring position as percentage of LOA from aft and fore as seen in Figure
4.39.
The sway, roll and yaw motions are minimum when the fairleads are at 12% of
LOA from aft and fore and varies to a maximum increase of 20%, 28% and 25%
respectively when the fairleads are at 21% of LOA from aft and fore.
Heave motion remains nearly the same when the mooring fairleads are varied from
12% to 21% of LOA from aft and fore and increases only maximum 0.01 m per percent
increase in spread mooring position as percentage of LOA from aft and fore. Pitch
motion is the least sensitive towards spread mooring fairlead location. But under the
action of wave with peak period 6 s, the pitch motion of FPSO seems to suddenly
increase when the spread mooring fairleads are at 18% of LOA from aft and fore. This
might be because of the unexpected motion from the high frequency excitation of
mooring lines. Hence the location of mooring lines needs to be between 12% and 21%
of LOA from aft and fore to minimise the FPSO motions for similar configured FPSOs
in the wave frequency ranges.
145
Figure 4.38: Variation in Translational FPSO Motions with Location of Spread
Mooring Fairleads on Hull (SESAM DeepC results)
146
Figure 4.39: Variation in Rotational FPSO Motions with Location of Spread Mooring
Fairleads on Hull (SESAM DeepC results)
147
4.5.6 Influence of Hull Length to Beam Ratio on FPSO Motions
Based on the existing FPSOs in Malaysia and Australia, Length to Beam ratio of hull
ranges from 5.4 to 6.5. Uncoupled analysis using SESAM HydroD V4.5-08 has been
performed to study the influence of hull length to beam ratio in this range on 6 DOF
FPSO motions and the motion responses are shown in Figure 4.40 and Figure 4.41. The
modelling of vessels with length and beam shown in Table 4.8 has been done using
procedure mentioned in section 3.4.1. The hull depth 24 m and loading condition DWT
137852 MT were kept as same for all the FPSO models at water depth 1350 m with
heading 45º subjected to unidirectional random waves with significant wave height 6
m in the wave frequency ranges.
Table 4.8: Vessel dimensions for hull length to beam ratio parametric study
FPSO Hull Length to
Beam
Model Length Beam ratio
(m)
Name (m) (L/B)
FPSO1 233 43 5.419
FPSO2 243 44 5.523
FPSO3 253 45 5.622
FPSO4 263 46 5.717
FPSO5 283 49 5.776
FPSO6 303 51 5.941
FPSO7 313 52 6.019
FPSO8 323 53 6.094
Figure 4.40 illustrates the variation in surge, sway and heave motion when the
length to beam ratio of the hull is varied. Surge motion is found to be increasing with
the increase in length to beam ratio up to 2 m when length to beam ratio equals unity
and the rate of increase in surge motion is found to be proportional to the wave period
with rate of surge motion varying with the increase in length to beam ratio from 0.3%
to 32 % when the peak period is varied from 7 s to 20 s. At all the wave periods, surge
motion is found to be minimum when the length to beam ratio is 5.4 and the variation
in motion is negligible until the peak period is varied up to 12 s. Sway motion is found
to be declining with the increase in length to beam ratio when the peak period was
varied from 5 s to 12 s and then increasing proportional to the wave period up to 0.8 m
when length to beam ratio equals unity with rate of sway motion varying with the
148
increase in length to beam ratio from 6% to 15 % when the peak period is varied from
14 s to 20 s. Heave motion is declining with the increase in length to beam ratio,
reducing from 1% to 80% as the length to beam ratio is varied from 5.419 to 6.094.
Figure 4.41 shows the variation of roll, pitch and yaw motions with the length to
beam ratio of vessel hull. Roll motion increases with increase in length to beam ratio
up to wave period of 10 s and then declines with the increase in length to beam ratio up
to 60% for peak period of 20 s proportional to the wave period. Pitch motion is found
to be declining as the length to beam ratio is increased. Pitch motion is highly reduced
up to 161% as the length to beam ration is varied from 5.419 to 6.094. Yaw motion is
declining with increase in length to beam ratio up to 121 % for wave periods 5 s to 15
s. After 15 s, the yaw motion increases with increase in length to beam ratio for wave
periods up to 20 s a maximum of 8%.
It can be concluded that, for wave heights of 5 m and peak period ranging from 5 s
to 12 s, the surge, sway, heave, pitch and yaw motions are minimum for length to beam
ratio of 6.094 and for peak period ranging from 14 s to 20 s, the heave, roll and pitch
motions are minimum again at a length to beam ratio of 6.094. The horizontal motions,
surge, sway and yaw are found to be increasing with length to beam ratio after 12 s and
this is probably because mooring lines were not given to find the vessel RAOs and
could be reduced by giving proper mooring which can be verified by doing a coupled
analysis as a future scope of work.
For Malaysian seas, the most probable peak period is usually in the range of 6 s to
7 s [132]. It can be seen from Figure 4.40 ~Figure 4.41, for FPSOs having beam to
length ratio in the same range as the Malaysian and Australian FPSOs, the heave, roll
and pitch motion remains almost the same for peak period less than 7 s. This means that
for the Malaysian FPSOs, the combined vertical motion will be less for the Malaysian
metocean conditions and thus risk from green water is less when compared to
Australian FPSOs where the peak period of wave is higher as seen later in section 4.7.1.
As in higher wave periods, these FPSOs have higher motion in heave, roll and pitch as
seen in Figure 4.40 ~Figure 4.41. The Australian sea is prone to extreme cyclones with
longer wave periods [4].
149
Figure 4.40: Variation in Translational FPSO Motions with Length to Beam Ratio of
Hull (SESAM HydroD results)
150
Figure 4.41: Variation in Rotational FPSO Motions with Length to Beam Ratio of
Hull (SESAM HydroD results)
151
4.5.7 Influence of Hull Loading Condition on FPSO Motions
Influence of hull loading condition on FPSO motion is studied using the models
developed with dimension as shown earlier in Table 4.8 using SESAM HydroD V4.5-
08. The fully loaded FPSOs are having DWT 137852 MT. Parametric study was
conducted for 100%, 75%, 50% and 25% hull loading condition with respective drafts.
FPSO models with hull depth 24 m at water depth 1350 m with heading 45º were
subjected to unidirectional random waves with significant wave height 6.3 m and peak
period 16 s. The variation of FPSO motions with loading condition is shown in Figure
4.42 and Figure 4.43.
Surge motion declines up to 20%, sway motion declines up to 33%, heave motion
declines up to 11% and pitch motion declines up to 15% at full loading condition when
compared to 25% loading condition. Roll rotation increases up to 51% and yaw rotation
increases up to 38% at full loading condition when compared to 25% loading condition.
Surge, sway and heave motion declines by maximum 0.0074 m, 0.015 m and 0.007
m respectively per percent increase in loading condition. While pitch rotation declines
by maximum 0.005º per percent increase in loading condition. Roll motion increases
by maximum 0.02º and yaw motion increases by maximum 0.006º per percent increase
in loading condition.
Surge, sway, heave and pitch motion decreases as vessel loading increases and roll
and yaw motion increases with vessel loading for the given metocean data and vessel
dimensions (covering vessel dimensions operating in Australia and Malaysia). Roll and
yaw rotation are found to be increasing with loading condition could be reduced by
giving proper mooring and roll damping which can be verified by doing a coupled
analysis as a future scope of work.
152
Figure 4.42: Variation in Translational FPSO Motions with Hull Loading Condition
(SESAM HydroD results)
153
Figure 4.43: Variation in Rotational FPSO Motions with Hull Loading Condition
(SESAM HydroD results)
154
4.6 Motion Response of FPSOs with Similar Dimensions as Operating FPSOs in
Malaysia and Australia
The vessel RAOs of FPSOs with dimensions similar as the operating FPSOs in
Malaysia (shown in Table 4.9) and Australia (shown in Table 4.11) in 2016 are
computed using SESAM HydroD V4.5-08. For the respective FPSO dimensions, the
operating water depth, loading condition and operating draft is used for its operating
locations and the sea state was represented using JONSWAP spectrum. The extreme
metocean condition, 100 year wind, wave and current at the deepest oil field in
Malaysia, i.e. Kikeh oil field, has been used for the analysis of Malaysian FPSOs and
for Australian FPSOs, the extreme metocean conditions, 100 year cyclonic condition
of North West Australia has been used [61], so that the results are conservative to be
used in association with deep water analysis. The metocean data used for the analysis
of Malaysian FPSOs and Australian FPSOs at collinear condition, 315º are tabulated as
shown in Table 4.10.
Table 4.9: Operating FPSOs in Malaysia 2016 [2]
Dimension Hull Water

FPSO LOA B Operating DWT
similar as Depth Depth
Name (m) (m) Draft (m) (MT)
FPSO (m) (m)
FPSO A Berantai 207 32 17 12.6 55337 55

Bunga
FPSO B 233 43 19 12 87768 60
Kertas
FPSO C Cendor II 245 41 21.6 14 100020 63
FPSO D Kikeh 337 55 27 21 273000 1350

Perisai
FPSO E 264 41 22 13 127540 60
Kamelia
Table 4.10: Extreme Metocean Conditions used for the Analysis
Metocean Details Malaysia Australia[61]

Significant Wave Height (m) 6.3 14 m
Peak Period (s) 16 14.5
JONSWAP, γ 1.5 [68] 1.6
wind velocity (m/s) 15.58 44
Current Velocity (m/s) 3.66 1.9
155
Table 4.11: Operating FPSOs in Australia 2016 [2]
FPSO Name Dimension Hull Water
LOA B Operating
similar as Depth DWT(MT) Depth
(m) (m) Draft (m)
FPSO (m) (m)
FPSO F Armada
241 42 23 14.09 102123 135
Claire
FPSO G
Glas Dowr 242 42 21 15 105000 344
FPSO H Maersk
Nguima - 261 58 31 23 308490 340
Yin
FPSO I Modec
258 46 24 16.86 149686 492
Venture II
FPSO J Montara
274 43 24 16.7 146251 80
Venture
FPSO K
Nganhurra 260 46 26 14 150000 400
FPSO L Ningaloo
238 42 24 15 101832 350
Vision
FPSO M Northern
273 50 28 19 177529 380
Endeavour
FPSO N
Okha 274 48 23 16.89 158000 78
FPSO O Pyrenees
274 48 23 15.8 143690 200
Venture
The RAOs generated for the FPSOs with similar dimensions as operating FPSOs in
Malaysia and Australia are given in Figure 4.44 and Figure 4.45 respectively. It can be
observed from Figure 4.44 that the FPSO D with dimensions similar as Kikeh FPSO
has minimum motion response when compared to other FPSOs, having same
dimensions as Malaysian FPSOs for the Kikeh metocean condition, which shows the
suitability in choosing the particular hull dimensions for the location. Kikeh FPSO is a
converted FPSO and its motion performance is suitable for the Kikeh metocean
conditions. Specific roll damping was not provided in the analysis and it shows, except
FPSO C with dimensions of Cendor II, all other FPSOs in Malaysia are having similar
roll performance. The maximum surge and sway RAO falls below 1.1 m/m for the
Malaysian FPSOs. Maximum heave, roll, pitch and yaw are 0.9 m/m, 4.791 deg/m, 0.9
deg/m and 0.5 deg/m respectively.
156
Figure 4.44: RAOs of Operating Malaysian FPSOs (SESAM HydroD results)
157
Figure 4.45: RAOs of Operating Australian FPSOs (SESAM HydroD results)
158
Similarly, for FPSOs with same dimensions as Australian FPSOs shown in Figure
4.45, the maximum surge and sway RAO falls below 0.95 m/m. Maximum heave, roll,
pitch and yaw are 0.93 m/m, 4.5 deg/m, 0.8 deg/m and 0.4 deg/m respectively. The
RAOs generated for wave frequency ranges can be used for initial design of FPSOs in
these regions and in choosing roll damping methods and mooring system.
4.7 Green Water on FPSOs
Freeboard exceedance for chosen FPSOs as given in Table 3.6 to be used in LCCA are
calculated using the method detailed in section 3.5.1 to identify green water events, that
is if any exists for specific metocean conditions and the freeboard exceedance
calculated are approximate since ship lines were drawn using the corresponding FPSO
dimensions and scaled to proper DWT as original ship lines are unavailable for the
analysis and confidential data for the FPSO operators. However, the results indicate
whether the FPSO dimensions will result in low, medium or high susceptibility to green
water on FPSO deck. Also, no specific roll damping was given so that the result will be
on conservative side and the design engineers can choose appropriate roll damping
measures based on the results generated as in section 4.6.
4.7.1 Relative motion and freeboard exceedance for Malaysian and Australian
FPSOs
Approximate most probable maximum relative motion has been calculated for FPSOs
by using vessel models with dimensions similar to Malaysian FPSOs by subjecting
them to site specific metocean conditions obtained from Omar et al [132] for locations
shown in Figure 3.22. The maximum annual MPM relative motion for each FPSO
considering all the motions at specific points is shown in table format corresponding to
each joint annual probability of 𝐻𝑠 and 𝑇𝑝 . 𝐻𝑠 and 𝑇𝑝 are taken to be the threshold value
of each range during operability analysis. Table 4.12 shows the approximate annual
MPM relative motion for Perisai Kamelia FPSO in m corresponding to the wave scatter
159
Table 3.7. Table 4.13 ~Table 4.15 show the approximate annual MPM relative motion
for Berantai, Bunga Kertas and Cendor II FPSO respectively in m corresponding to the
wave scatter Table 3.8. Table 4.16 shows the approximate annual MPM relative motion
for Kikeh FPSO in m corresponding to the wave scatter Table 3.9.
Table 4.12: Annual MPM relative motion for Perisai Kamelia FPSO in m
Hs(m)
>3.0
3.0
2.5 0.79 0.86
2.0 0.31 0.53 0.64 0.69
1.5 0.07 0.23 0.33 0.40 0.48 0.52 0.56 0.60
1.0 0.05 0.16 0.22 0.27 0.32 0.34
0.5 0.02 0.08 0.11
Tp(s) 1 2 3 4 5 6 7 8 9 10 >10
Table 4.13: Annual MPM relative motion for Berantai FPSO in m

Hs(m)
>3.0
3.0 1.14
2.5 0.87 0.95
2.0 0.60 0.69 0.76
1.5 0.08 0.37 0.45 0.52 0.57
1.0 0.05 0.17 0.25 0.30 0.35 0.38
0.5 0.00 0.03 0.08
Tp(s) 1 2 3 4 5 6 7 8 9 10 >10
Table 4.14: Annual MPM relative motion for Bunga Kertas FPSO in m
Hs(m)
>3.0
3.0 1.10
2.5 0.83 0.92
2.0 0.56 0.66 0.74
1.5 0.08 0.36 0.42 0.50 0.55
1.0 0.06 0.18 0.24 0.28 0.33 0.37
0.5 0.00 0.03 0.09
Tp(s) 1 2 3 4 5 6 7 8 9 10 >10
160
Table 4.15: Annual MPM relative motion for Cendor II FPSO in m
Hs(m)
>3.0
3.0 1.07
2.5 0.81 0.89
2.0 0.54 0.65 0.71
1.5 0.07 0.34 0.40 0.49 0.53
1.0 0.05 0.16 0.22 0.27 0.32 0.36
0.5 0.00 0.02 0.08
Tp(s) 1 2 3 4 5 6 7 8 9 10 >10
Table 4.16: Annual MPM relative motion for Kikeh FPSO in m

Hs(m)
>3.0
3.0
2.5 0.71 0.81
2.0 0.49 0.56 0.64 0.69
1.5 0.32 0.37 0.42 0.48 0.51
1.0 0.05 0.15 0.21 0.24 0.28
0.5 0.02 0.08
Tp(s) 1 2 3 4 5 6 7 8 9 10 >10
Approximate most probable maximum relative motion has been calculated for
FPSOs by using vessel models with dimensions similar to Australian FPSOs by
subjecting them to site specific metocean conditions obtained from Metocean View
[133] for locations shown in Figure 3.23. The maximum annual MPM relative motion
for each FPSO considering all the motions at specific points is shown in table format
corresponding to each joint annual probability of 𝐻𝑠 and 𝑇𝑝 . Table 4.17 ~ Table 4.19
shows the approximate annual MPM relative motion for Nganhurra, Pyrenees Venture
and Ningaloo Vision FPSO respectively in m corresponding to the wave scatter Table
3.10. Table 4.20 ~Table 4.21 show the approximate annual MPM relative motion for
Okha and Modec Venture FPSO respectively in m corresponding to the wave scatter
Table 3.11. Table 4.22 shows the approximate annual MPM relative motion for Glas
Dowr FPSO in m corresponding to the wave scatter Table 3.12.
It can be seen from Table 4.12~ 4.22 that the maximum annual MPM relative
motion for these FPSOs are far below their freeboard given in Table 3.6 and will not
161
result in green water incidents when subjected to wind generated sea state given in
Table 3.7 ~Table 3.12.
Table 4.17: Annual MPM relative motion for Nganhurra FPSO in m

Hs(m)
1 0.46 0.47 0.46 0.44
1.5 0.62 0.69 0.71 0.68 0.65 0.61 0.56 0.52 0.44
2 0.44 0.51 0.61 0.71 0.83 0.92 0.94 0.91 0.87 0.82 0.75 0.69 0.64 0.59 0.55
2.5 0.64 0.76 0.89 1.04 1.15 1.18 1.14 1.09 1.02 0.94 0.87 0.80 0.74 0.68
3 0.77 0.91 1.07 1.25 1.38 1.42 1.37 1.31 1.22 1.13 1.04 0.96 0.88 0.82
3.5 1.06 1.25 1.60 1.53 1.43 1.32 1.21 1.12
4 1.74 1.63 1.51 1.39 1.28
4.5 1.83 1.56 1.44 1.35
5 1.73
Tp(s) 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Table 4.18: Annual MPM relative motion for Pyrenees venture FPSO in m
Hs(m)
1 0.28 0.30 0.31 0.31
1.5 0.40 0.42 0.45 0.46 0.45 0.44 0.42 0.40 0.36
2 0.48 0.51 0.52 0.52 0.53 0.57 0.60 0.61 0.60 0.58 0.56 0.53 0.50 0.47 0.45
2.5 0.64 0.65 0.65 0.66 0.71 0.75 0.77 0.76 0.73 0.70 0.66 0.63 0.59 0.56
3 0.77 0.79 0.78 0.79 0.85 0.90 0.92 0.91 0.88 0.84 0.61 0.75 0.71 0.67
3.5 0.86 0.86 1.03 1.01 0.97 0.93 0.88 0.83
4 1.16 1.11 1.06 1.00 0.95
4.5 1.25 1.19 1.13 1.07
5 1.25
Tp(s) 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Table 4.19: Annual MPM relative motion for Ningaloo Vision FPSO in m
Hs(m)
1 0.41 0.40 0.39 0.36
1.5 0.59 0.61 0.60 0.58 0.55 0.50 0.46 0.43 0.36
2 0.45 0.54 0.68 0.72 0.79 0.82 0.80 0.77 0.73 0.67 0.62 0.57 0.52 0.48 0.44
2.5 0.68 0.82 0.90 0.98 1.02 1.00 0.97 0.91 0.84 0.77 0.71 0.65 0.60 0.55
3 0.81 0.99 1.08 1.18 1.22 1.20 1.16 1.09 1.01 0.93 0.85 0.78 0.72 0.66
3.5 1.16 1.26 1.35 1.27 1.18 1.08 0.99 0.91
4 1.45 1.34 1.23 1.13 0.95
4.5 1.51 1.38 1.27 1.17
5 1.42
Tp(s) 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
162
Table 4.20: Annual MPM relative motion for Okha FPSO in m
Hs(m)
0.5
1 0.36 0.39 0.4 0.39 0.39 0.38 0.36 0.34 0.31
1.5 0.24 0.34 0.4 0.46 0.51 0.55 0.58 0.59 0.59 0.56 0.56 0.54 0.51 0.49 0.46 0.43
2 0.45 0.53 0.62 0.69 0.73 0.77 0.79 0.78 0.77 0.75 0.72 0.68 0.65 0.61 0.58
2.5 0.56 0.6 0.77 0.86 0.91 0.96 0.99 0.98 0.97 0.94 0.9 0.85 0.81 0.76 0.72
3 0.8 0.92 1.03 1.18 1.13 1.13 1.08 1.03 0.97
3.5 1.08 1.2 1.31 1.26 1.2 1.13
4 1.23 1.37
Tp(s) 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Table 4.21: Annual MPM relative motion for Modec Venture II FPSO in m
Hs(m)
0.5
1 0.39 0.41 0.41 0.4 0.38 0.36 0.33 0.3 0.26
1.5 0.23 0.33 0.4 0.47 0.53 0.58 0.62 0.62 0.6 0.57 0.53 0.49 0.46 0.42 0.39 0.36
2 0.44 0.53 0.63 0.71 0.78 0.82 0.82 0.79 0.76 0.71 0.66 0.61 0.56 0.52 0.48
2.5 0.55 0.66 0.79 0.89 0.97 1.03 1.03 0.99 0.95 0.89 0.82 0.76 0.7 0.65 0.6
3 0.79 0.95 1.06 1.19 1.14 1.07 0.99 0.91 0.84
3.5 1.04 1.24 1.25 1.15 1.06 0.98
4 1.26 1.42
Tp(s) 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Table 4.22: Annual MPM relative motion for Glas Dowr FPSO in m
Hs(m)
0.5 0.08 0.11 0.14 0.19 0.2 0.19 0.19 0.18 0.16 0.15 0.14
1 0.11 0.16 0.23 0.27 0.33 0.35 0.38 0.39 0.38 0.37 0.35 0.33 0.3 0.28 0.25 0.23 0.23
1.5 0.24 0.34 0.41 0.49 0.53 0.57 0.59 0.58 0.56 0.53 0.49 0.45 0.41 0.38 0.35 0.33
2 0.45 0.55 0.66 0.7 0.76 0.78 0.77 0.75 0.71 0.65 0.6 0.55 0.51 0.47
2.5 0.68 0.82 0.88 0.94 0.82 0.75 0.69
3 0.99
Tp(s) 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Also, it is seen from Table 4.12 ~ Table 4.22 relative motion of FPSO holds a linear
relation with wave height. This is because heave, roll and pitch motion increases with
wave height irrespective of the mooring used as seen in Figure 4.26 ~Figure 4.32. This
is also the reason why linear analysis produce good results and why only vessel model
can be used to simulate green water on FPSOs without mooring lines.
163
For FPSOs to be used in shallow waters, care should be taken while designing them,
as in shallow waters up to 100 m, heave and pitch motion increases as seen in Figure
4.33 which can make the FPSO susceptible to green water. For Malaysian seas, the
most probable peak period is usually in the range of 6 s to 7 s. It can be seen from Figure
4.40 ~Figure 4.41, for FPSOs having beam to length ratio in the same range as the
Malaysian and Australian FPSOs, the heave, roll and pitch motion remains almost the
same for peak period less than 7 s. This means that for the Malaysian FPSOs, the risk
from green water is less when compared to Australian FPSOs where the peak period of
wave is higher. As in higher wave periods, these FPSOs have higher motion in heave,
roll and pitch as seen in Figure 4.40 ~Figure 4.41. Hence utmost care should be taken
while designing FPSOs to be installed in Australian seas as 100-year extreme cyclones
shown in Table 3.13 can make them highly susceptible to green water as shown in Table
4.23 while extreme metocean conditions for the deepest oil field, that is Kikeh field as
shown in Table 4.10 will only leave the FPSO with a low susceptibility to green water
as seen in Table 4.24. The 100-year extreme cyclonic condition given in Table 3.13 is
representative of all of North Western Australia and not specific to the operating
location of chosen FPSOs for operability analysis.
Table 4.23: Annual maximum freeboard exceedance of FPSOs for extreme cyclonic
condition in NWA
FPSO Max
Dimension
Name Freeboard
similar as
Exceedance Susceptibility
FPSO
(m) to green water
FPSO F Armada High
10.07
Claire Susceptibility
FPSO G High
Glas Dowr 13.54
Susceptibility
FPSO H Maersk
High
Nguima - 8.80
Susceptibility
Yin
FPSO I Modec High
11.49
Venture II Susceptibility
FPSO J Montara High
11.42
Venture Susceptibility
FPSO K High
Nganhurra 9.20
Susceptibility
FPSO L Ningaloo High
10.46
Vision Susceptibility
164
FPSO Max
Dimension
Name Freeboard
similar as
Exceedance Susceptibility
FPSO
(m) to green water
FPSO M Northern High
9.36
Endeavour Susceptibility
FPSO N High
Okha 12.98
Susceptibility
FPSO O Pyrenees High
6.46
Venture Susceptibility
Table 4.24: Annual maximum freeboard exceedance of FPSOs for extreme metocean
condition in Kikeh field
Dimension Max Freeboard
FPSO
similar as Exceedance Susceptibility
Name
FPSO (m) to green water
LOW
FPSO A Berantai 2.53
Susceptibility
LOW
FPSO B Bunga Kertas 0.22
Susceptibility
LOW
FPSO C Cendor II 1.24
Susceptibility
LOW
FPSO D Kikeh 2.87
Susceptibility
LOW
FPSO E Perisai Kamelia -1.49
Susceptibility
4.7.2 Downtime cost due to green water
As seen from Table 4.12- 4.22 and discussed in section 4.7.1, there is no freeboard
exceedance for chosen FPSOs for LCCA study in Table 3.6 as freeboard for these
FPSOs are higher than the annual MPM relative motion. Therefore, there is no
downtime cost due to green water when annual wind generated wave scatter data is
considered. Even though the original ship lines were not used, from Table 4.12 ~ Figure
4.22 it can be seen that the MPM relative motion is not varying much for the different
FPSO dimensions and is expected to be in the same range even when original ship lines
are used to create vessel model. Also, for Malaysian FPSOs, it seen from Table 4.24
that even in extreme metocean conditions, FPSO is only susceptible to low level of
green water incidents. When considering Australian FPSOs, extreme cyclonic
165
conditions with 1-year return period as shown in Table 3.13 will not cause green water
incidents as the relative motion will be in the range as shown in Table 4.17 ~Figure
4.22, but, 100-year return period is causing high level of green water incidents. Also,
the 100-year extreme cyclonic conditions are representative for the whole of NWA and
not specific to the operating location for these FPSOs. However, this event occurring
in a 10-year life-cycle of FPSO (life-cycle period used in LCCA) is very low and the
downtime cost due to such event will be negligible as discussed later in this thesis
regarding the main cost driving components for FPSO life-cycle cost as capital,
operation and maintenance cost in section 4.9. Hence the downtime cost is
approximated to be zero due to green water incidents under wind generated sea
conditions.
4.8 Life-Cycle Cost of FPSOs in Malaysia and Australia
Table 4.25 gives the cost data of selected FPSOs given earlier in Table 3.6. The FPSOs
under study involve: spread mooring, internal turret mooring, external turret mooring,
submerged turret mooring and finally riser turret mooring. The cost data given in Table.
4.25 is obtained as per the procedure mentioned in section 3.6.4. As given in section
4.7.2, downtime cost due to green water incidents, when FPSOs are subjected to wind
generated sea is zero. Oil price for calculating revenue from oil production is taken as
71.5 USD per barrel as annual average of past 5 years [165]. The capital cost for the
chosen FPSOs are given in Figure 4.46. In these FPSOs, Kikeh FPSO with external
turret mooring has the maximum capital cost of 7195 million USD. Also, FPSOs with
riser turret mooring have comparatively high capital costs when compared to other
moored FPSOs with Nganhurra FPSO costing 5234 million USD. Out of the spread
moored FPSOs under study, Ningaloo Vision has the highest capital cost of around
1391 million USD, while Cendor II and Berantai have capital costs under 800 million
USD. Out of the internal turret moored FPSOs, Pyrenees venture has the highest capital
cost of 3359 million USD; the others having capital cost under 1125 million USD.
Okha and Nganhurra, two of the riser turret moored FPSOs have capital cost 4214
million USD and 5234 million USD respectively.
166
Table 4.25: Cost Data of selected FPSOs [2] [141]–[153]
FPSO/ FSO CAPITAL ANNUAL ANNUAL MAXIMUM ANNUAL
NAME COST OPERATION & BARE OIL REVENUE
(US MAINTENANCE BOAT PRODUCTIO FROM
DOLLARS) COST CHARTER N CAPACITY MAXIMUM OIL
(US DOLLARS) RATE OF FPSO PRODUCTION
(US (MBOPD) (US DOLLARS)
DOLLARS)
PERISAI KAMELIA 272100000 6750000 45000000 100 600233333
KIKEH 7195000000 7837500 59418700 120 720280000
CENDOR II 660000000 14726250 98175000 35 210081667
BERANTAI 800000000 7700000 49500000 30 180070000
NINGALOO VISION 1391000000 63454545 423030300 63 378147000
BUNGA KERTAS 150000000 3507595 7737400 30 180070000
GLAS DOWR 175100000 3000000 20000000 60 360140000

MODEC VENTURE 100 600233333
624000000 52700000 351333333
II
STYBARROW 1125000000 61071429 407142857 80 480186667
PYRENEES 96 576224000
3359000000 63333333 422222222
VENTURE
OKHA 4214000000 55600000 NA 30 180070000
NGANHURRA 5234000000 2806000000 NA 100 600233333
167
Figure 4.46: Capital Cost of FPSOs in Malaysia and Australia under study
The total life-cycle cost of FPSOs accounting only for capital, operation and
maintenance which are the expenses, are given in Figure 4.47 and the total life cycle
cost is maximum for Nganhurra FPSO with 75348 million USD for 25-year life-cycle
period, noted as a riser turret moored. Even though the capital cost was higher for
Kikeh FPSO, the life-cycle cost of Ningaloo Vision FPSO (SM), Modec venture II (IT),
Stybarrow FPSO (IT), Pyrenees venture (IT) and Nganhurra FPSO (RTM) is higher
when compared to the life-cycle costs of Kikeh FPSO (ET) with an average difference
in life-cycle costs of 16721 million USD for 25-year life-cycle period. Notably
Stybarrow and Nganhurra FPSO have newly built hulls. The riser turret moored FPSOs
have an average life-cycle cost of 40494 million USD for 25-year life-cycle period and
19032 million USD for the 10-year life-cycle period. Compared to the average life-
cycle cost of riser turret moored FPSOs, the average life-cycle cost of external turret
moored FPSOs are 13%, spread moored FPSOs are 16% and internal turret moored
FPSOs are 25% for 25 years life-cycle period. Whereas for 10-year life-cycle period,
the average life-cycle cost of spread moored FPSOs are 16%, external turret moored
FPSOs are 22% and internal turret moored FPSOs are 25% of that of riser turret moored
FPSOs. The rise in average life-cycle cost of external turret moored FPSOs in 25 years
168
are contributed by the initial high capital cost of Kikeh FPSO. Hence ranking the
costliest FPSO and associated mooring system in terms of its life-cycle cost: the riser
turret moored FPSOs are the costliest, followed by internal turret moored FPSOs for
both 10-year and 25-year life-cycle periods, based on the available FPSO cost data from
different reliable sources. The least expensive option for 10-year life-cycle period is
spread moored FPSO, while for 25-years, it is external turret moored FPSO.
Figure 4.47: Total Life-Cycle cost of FPSOs in Malaysia and Australia for 10-year
and 25-year life-cycle period
169
Figure 4.48: Cumulative Real Cost of FPSOs in Malaysia and Australia for 10-year Life-cycle Period
170
Figure 4.49: Cumulative Real Cost of FPSOs in Malaysia and Australia for 25-year Life-cycle Period
171
The cumulative real cost of FPSOs including capital, operation and maintenance
cost for 10-year life-cycle period is given in Figure 4.48 and for 25-year life-cycle
period is given in Figure 4.49. The rate of increase of cumulative costs per annum
ranges from 60 million USD to 500 million USD for spread moored FPSOs. The
average increase in cumulative costs per year for external turret moored FPSOs range
from 50 million USD to 70 million USD, whereas for internal turret moored FPSOs,
the annual rise in cumulative costs are from 20 million USD to 500 million USD.
The highest annual cost increase is for the riser turret moored FPSOs, ranging from
60 million USD to 3000 million USD. Market fluctuation in cost is not considered and
is assumed to be constant while calculating life-cycle cost. Hence to include market
fluctuations in the cost components a sensitivity analysis has been carried out in section
4.11.
4.9 Net Present Value of FPSOs in Malaysia and Australia
Net present value for FPSOs are calculated as per section 3.6.6 for different mooring
configurations at a discount rate of 2% for the capital cost and NPV comparison and
are shown in Figure 4.50. NPV calculated with only expenses, capital, operation and
maintenance are given in Figure 4.50. NPV calculated with expenses and revenue from
oil are shown in Figure 4.51.
The NPVs of FPSOs calculated with only expenses are all negative as the main
source of income from oil production is not included as in Figure 4.51. It can be seen
from the scatter diagram shown in Figure 4.50 that, maximum number of FPSO
configurations are having capital costs less than 2000 million USD and NPV greater
than -20,000 million USD. For these FPSOs, the capital costs are minimum and NPVs
are closer to zero. The other FPSOs which do not fall in this area are Pyrenees venture
with IT, Okha and Nganhurra with RTM and Kikeh with ET. Even though the capital
costs of these FPSOs are higher, the NPV of all the FPSOs except Nganhurra, is greater
than -10,000 million USD. Bunga Kertas, Glas Dowr and Perisai Kamelia FPSOs have
the highest NPV with values greater than -1000 million USD and capital costs less than
172
300 million USD. Also, it is to be noted that, these are turret moored FPSOs. For
spread moored FPSOs, the NPV falls between -1674 million USD to -5760 million
USD with capital cost less than 1391 million USD. The profitability of these FPSOs
can be accurately presented where the revenue obtained by the oil production is known,
and is shown in Figure 4.51.
only expenses)
Again, when the revenue from oil is included, the highest NPV is for Perisai
Kamelia, Glas Dowr, Modec Venture II and Bunga Kertas, starting from the FPSO with
highest NPV onwards and all of them being turret moored FPSOs. Hence it cannot be
generalised that all the turret moored FPSOs are costlier than the spread moored
versions when their life-cycle worth is considered. Also, it is seen from Figure 4.50
and Figure 4.51 that as the life-cycle period increases, the NPV decreases.
FPSOs with negative NPV even after the inclusion of income from oil production
are Okha and Nganhurra, which are riser turret moored FPSOs with lowest NPV from
the chosen FPSOs for study, followed by Pyrenees Venture, Ningaloo Vision, Kikeh
173
and Stybarrow FPSOs. This is mainly due to their high capital cost as seen in Figure
4.46. Downtime cost are negligible for all the FPSOs chosen for the study as seen in
section 4.7.2. Also, the trend in NPV variation over the years seems to be similar with
and without income from oil prices from Figure 4.50 ~Figure 4.51. Hence the main cost
driving factor for NPV is capital, operation and maintenance cost.
expenses and revenue from oil production)
4.10 Present Value of FPSOs in Malaysia and Australia
The present value of Australian FPSOs for a (calculated built-up) discount rate of
2.96% and Malaysian FPSOs for a (calculated built-up) discount rate of 1.87% for 10-
year life-cycle period and 25-year life-cycle period is given in Figure 4.52 ~Figure 4.55.
Present value of FPSOs are calculated for all expenses and income. The variation in the
present worth of these FPSOs for the future cash flows are depicted over the number of
years of respective life-cycles. The present worth increases as the life decreases except
174
for Nganhurra and Ningaloo Vision, as in the long term, the higher capital investment
is retrieved through income from oil. Whereas for other FPSOs, the operation and
maintenance cost is proving to be high in the long term, resulting in lower present
values.
cycle period
cycle period
175
It can be seen from Figure 4.52 ~ Figure 4.55 that the present value takes a sudden
rise in the year 1 and then slightly decreases towards the end of life of the asset from
first year of operations except for Nganhurra and Ningaloo Vision for the reasons
mentioned above. After the initial year, the present value decreases from 10 million to
1 million USD per year for the various FPSOs over the life period.
cycle period
cycle period
176
The present value graph can be used to identify the sum of the future cash inflows
discounted for inflation and interest for a period of 10 years and 25 years to represent
the value of this money in present day dollars. As per the study, for Australian FPSOs,
except Nganhurra and Ningaloo Vision, the present value of all other FPSOs are greater
than 41 million USD for a period of 10 years and 27 million USD for a period of 25
years. Similarly, for Malaysian FPSOs, the minimum present value is 96 million USD
for 10 years and 73 million USD for a life period of 25 years.
4.11 Sensitivity Analysis
Sensitivity analysis is performed to identify which FPSO configuration should be

adopted for any given interest rate without additional calculations as per the procedure
mentioned in section 3.6.7. The calculated discount rate for the present Australian and
Malaysian economy is 2.96% and 1.87% respectively. For sensitivity analysis
(including risk factor in analysis), the discount rate is varied from 2% to 10% which
presents the best and worst scenarios in the market rates. The NPV of FPSOs in
Malaysia and Australia for 10-year life-cycle period is plotted along with the respective
discount rates as shown in Figure 4.56 ~ Figure 4.59. Also, NPV of FPSOs for discount
rate of 2% is plotted against their capital cost to show the relation between capital cost
and their respective NPVs. The expenses and revenue from oil is included in the life-
cycle cost and NPV calculation.
Figure 4.56 shows the NPV profile for spread moored FPSOs. As the discount rate
is increased from 2% to 10%, the net present value decreases up to a rate of 50 million
USD per percent discount rate. For external turret moored FPSOs, the NPV decreases
up to 30 million USD per percent discount rate as seen in Figure 4.57. Figure 4.58
shows the decrease of NPV as 50 million USD to 100 million USD per percent increase
in discount rate for internal turret moored FPSOs. Nganhurra, which is a riser turret
moored FPSO shows an increase in NPV, as in the long term, the higher capital
investment is retrieved through income from oil as seen in Figure 4.59.
177
Figure 4.56: NPV Profile for Spread Moored FPSOs
Figure 4.57: NPV Profile for External Turret Moored FPSOs
178
Figure 4.58: NPV Profile for Internal Turret Moored FPSOs
Figure 4.59: NPV Profile for Riser Turret Moored FPSOs
179
In general, as the discount rate increases, the net present value of the asset decreases
as the risk associated with the investment increases. Also, it can be seen from Figure
4.56 ~ Figure 4.59 that, NPV decreases as capital cost increases. The NPV profile
reflects the net present worth of these FPSOs, at a varying market situation and can be
used as a reference in the initial estimate of similar configured FPSOs.
4.12 Cost Proportions
FPSOs with similar dimensions and dead weight tonnages are compared to find the
difference in cost proportion for FPSOs with different mooring configurations and hull
conditions. A discount rate of 2% is used for the comparison purpose and NPV is
calculated for all expenses (Capital, operation and maintenance) and income from oil
production.
4.12.1 Cendor II and Glas Dowr FPSO
Figure 4.60 compares Cendor II FPSO with spread mooring and converted hull to Glas
Dowr FPSO with internal turret mooring and newly built hull for 10-year analysis
period. Table 4.26 gives the cost and dimensions of these FPSOs. Even though
converted hull is used for Cendor II, the initial cost of Cendor II FPSO is higher than
Glas Dowr FPSO. Again, the operation and maintenance cost of FPSO with a converted
hull and spread mooring, i.e. Cendor II FPSO is higher than Glas Dowr FPSO.
Figure 4.60: Cost Proportions of Cendor II and Glas Dowr
180
It is to be noted that, the NPV of Glas Dowr is more with value -3151 million USD for
10-year life period. NPV of Cendor II is 386 million USD. So, even though Glas Dowr
FPSO has internal turret and newly built hull, over the whole life period, it proves to be
the better option when compared to the spread moored Cendor II FPSO.
Table 4.26: Cost, Dimensions and DWT of Cendor II and Glas Dowr
FPSO Cendor II Glas Dowr
Cost
Capital Cost (Million USD) -660 -175
Operation and Maintenance Cost (Million USD) -1129 -230
Revenue from Oil Production (Million USD) 2294 3933
NPV (Million USD) 386 3151
Dimensions
DWT 100020 105000
Hull Length (m) 245 242
Hull Width (m) 41 42
Hull depth (m) 21.6 21
4.12.2 Nganhurra and Stybarrow FPSO
Figure 4.61 compares Nganhurra with riser turret mooring to Stybarrow FPSO with
internal turret with cost and dimensions as shown in Table 4.27. Both are having newly
built hulls, while Stybarrow FPSO is leased, Nganhurra FPSO is owned. Nganhurra
having riser turret mooring, costs more than Stybarrow with internal turret.
Figure 4.61: Cost Proportions of Nganhurra and Stybarrow Venture
181
The operation and maintenance cost of Nganhurra FPSO is also higher when
compared to Stybarrow Venture. The NPV of Stybarrow FPSO is -621 million USD
which is higher than Nganhurra FPSO with NPV -24552 million USD for 10-year life
period.
Table 4.27: Cost, Dimensions and DWT of Nganhurra and Stybarrow Venture
FPSO Nganhurra Stybarrow
Cost
NPV (Million USD) -24552 -621
Dimensions
DWT 150000 140000
Hull depth (m) 26 24
4.12.3 Okha and Pyrenees Venture FPSO
Figure 4.62 compares Okha FPSO with riser turret mooring to Pyrenees Venture with
internal turret with cost and dimensions as shown in Table 4.28. Both are having
converted hulls but Pyrenees Venture is on lease while Okha is owned. Even though
both these FPSOs are having same dimensions and nearly same DWT, the initial cost
varies by 25%, Okha costing more than Pyrenees venture as Okha is owned and having
riser turret mooring.
Figure 4.62: Cost Proportions of Okha and Pyrenees Venture
182
The maintenance cost is higher for Pyrenees venture with internal turret mooring
when compared to Okha. NPV of Okha FPSO is -2947 million USD, while that of
Pyrenees Venture is -2069 million USD. The NPV of Pyrenees Venture is more than
that of Okha even though it is leased, having internal turret mooring and higher
maintenance and operation cost.
Table 4.28: Cost, Dimensions and DWT of Okha and Pyrenees Venture
Pyrenees
FPSO Okha Venture
Cost
NPV (Million USD) -2947 -2069
Dimensions
DWT 158000 143690
4.12.4 Perisai Kamelia and Glas Dowr FPSO
Figure 4.63 compares Perisai Kamelia with external turret mooring and converted hull
to Glas Dowr FPSO with internal turret mooring and newly built hull with details shown
in Table 4.29. The initial cost and operation and maintenance cost of Perisai Kamelia
FPSO is higher when compared to Glas Dowr FPSO. Even after having newly built
hull, Glas Dowr FPSO have NPV more than that of Perisai Kamelia proving to be the
better option.
Hence among the FPSOs compared, internal turret moored FPSOs are found to be
having higher NPV. Internal turret moored FPSOs are found to be profitable in the long
run even after having newly built hull in some cases and in some cases, being leased.
183
Figure 4.63: Cost Proportions of Perisai Kamelia and Glas Dowr
Table 4.29: Cost, Dimensions and DWT of Perisai Kamelia and Glas Dowr
Perisai
FPSO Kamelia Glas Dowr
Cost
NPV (Million USD) 5150 3151
Dimensions
DWT 127540 105000
4.13 Cost and Motion Performance of FPSOs
Excessive motion of FPSOs can lead to production down time and thus result in loss of
profit from the project. Hence motion and cost of the FPSOs are interlinked to such an
extent that, a basic dynamic response study is required while choosing the cost effective
FPSO options suitable to the oil field. Most of the cost related decisions are taken
without considering the performance of the FPSO. As seen from section 4.12, even
though the initial investments are higher for FPSOs with newly built hull and turret
mooring system, their net present values are higher than their counter parts with
converted hull and other type of turret/spread moorings.
184
Turret moored FPSOs are preferred mostly in environment with extreme weather
conditions due to their weathervaning capabilities and spread moored FPSOs are
preferred in calm weather condition due to their comparatively lower CAPEX. The
main reason behind the same is established through the present study. It was shown in
section 4.5.1 that the horizontal motions are decreasing for higher wave heights in the
presence of wind and current when turret moored FPSOs are used and the same is
increasing when spread moored FPSOs are used. Also, from the life-cycle cost
analysis, it was seen that the average life-cycle cost is minimum for spread moored
FPSOs. Whereas, the NPV of these spread moored FPSOs are lower than some of the
external turret moored and internal turret moored FPSOs as seen in Figure 4.50 and
Figure 4.51. Even though, the capital cost is minimum, when comparing FPSOs with
similar dimensions and DWT, spread moored FPSOs with converted hull are shown to
have higher OPEX as seen in Figure 4.50 and Figure 4.51 resulting in much lower NPV
than the turret moored FPSO with newly built hull. This may be due to the use of
converted hulls which are not specifically designed for the metocean conditions. This
emphasizes the need for a site specific dynamic motion response study of the converted
hull to be used to minimise the future operational down time and cost. As mentioned
before, most of the oil companies use converted tankers for small projects, even then,
the use of appropriate tanker can lead to huge profit in terms of its life-cycle cash flows.
Also, during the motion response study, 100- year return period should be used
because, when the structure is not designed for 100-year condition, they may induce
much greater vessel motion than the 1 year return extreme storm condition, causing
operation shut downs and structural integrity problems. The RAOs generated and
shown in Figure 4.44 and Figure 4.45 can be used as reference while choosing similar
configured FPSOs for Malaysian and Australian oil fields as they are generated by
subjecting FPSOs to extreme weather condition for 100 years. For FPSOs with same
dimension as operating FPSOs in Malaysia and Australia, downtime cost due to green
water events in the presence of wind generated sea is approximately zero. Since NWA
is prone to extreme cyclonic conditions, a future study can include site specific swells
as well.
185
From LCCA, the riser turret moored FPSOs are the costliest, followed by internal
turret moored FPSOs for both 10-year and 25-year life-cycle periods, based on the
available FPSO cost data from different reliable sources. The least expensive option for
10-year life-cycle period is spread moored FPSO, while for 25-years, it is external turret
moored FPSO. Among the FPSOs compared in section 4.12, internal turret moored
FPSOs are found to be having higher NPV. Internal turret moored FPSOs are found to
be profitable in the long run even after having newly built hull in some cases and in
some cases being leased.
The range of other significant parameters to be adopted to minimise the vessel

motions are mentioned from section 4.5.2 ~ section 4.5.7. The parametric charts
developed for motion response variation of FPSOs and NPV profiles for FPSOs can be
used as benchmarks for future FPSO projects, with similar configuration and
dimension, in changing market condition and oil price.
4.14 Chapter Summary
In this chapter, the motion RAOs in 6 DOF obtained using software simulation based
on 3D potential theory were compared with experimental results to verify the modelling
and analysis procedure adopted and to calibrate the simulation model for parametric
study. Verification was done for both coupled and uncoupled analysis procedures. The
parametric study results were presented identifying the range of mooring line length,
mooring line azimuth angle, spread mooring fairlead location, hull length to beam ratio
and hull loading condition at which the FPSO motion is minimum. Influence of
metocean conditions and water depth on FPSO motions were also presented. Motion
RAOs for operating FPSOs in Malaysia and Australia subjected to 100-year extreme
weather conditions were also presented. Downtime due to green water and subsequent
downtime cost is evaluated. Subsequently, the life cycle costs and NPVs of the FPSOs
in Australian and Malaysian waters with different mooring systems and hull conditions
were discussed along with the cost proportions due to each life cycle phase. Finally, a
discussion has been made on the factors affecting the selection of FPSO in terms of its
cost and motion performance.
186
CHAPTER 5
CONCLUSIONS AND RECOMMENDATIONS
5.1 Conclusions
The simulation procedure adopted is capable to predict the wave frequency motion
responses with reasonable accuracy as shown in section 5.1.1 and the relative motions
found by the dynamic analysis of FPSOs leads to the calculation of subsequent
downtime cost due to green water events in the presence of wind generated sea as given
in section 5.1.4. These downtime cost are used as an input to the life-cycle cost analysis
of FPSOs along with other cost data, capital cost, operation and maintenance and
income generated from oil production collected from industry personnel, asset reports,
offshore magazine and industry news. Section 4.13 discusses the cost and motion
results obtained through this research. This is a unique attempt to bring the cost and
motion response of FPSO under one umbrella enabling the design engineers to make
logical decisions supported by research data while choosing initial configuration of
FPSO for a particular oil field under specific metocean conditions.
This is enabled by conducting extensive parametric studies covering metocean

conditions, hull parameters and mooring line parameters for spread moored and turret
moored FPSOs. The parametric charts help in understanding the trend and variation in
motion amplitude in 6 DOF with the chosen parameters. Parametric study was
conducted using experimental test, uncoupled and coupled dynamic analysis using
SESAM suit of programs. The gaps in previous FPSO parametric studies are identified
in Table 2.2, critical literature review and the present study fills the gap in FPSO motion
parametric studies as shown below in section 5.1.2 and 5.1.3.
The life-cycle cost and NPV of FPSOs for 10-year and 25-year life period are
calculated and the cost comparison for FPSOs with similar capacities fills the gap in
187
knowledge by revealing the best economical FPSO option and associated mooring
system for the chosen life cycle periods in section 5.1.5. Comparison was made between
FPSOs with newly built/converted hull and spread/ET/IT/RTM mooring system.
Thus section 5.1.1 – 5.1.5 concludes the research in alignment with the five research
objectives given earlier in section 1.7. The problems discussed earlier in section 1.6
pointed towards reducing the 6 FPSO motions and life-cycle costs to achieve increased
operational time and maximum productivity. In the iterative FPSO design procedure,
the initial step is the sizing of an FPSO vessel to define a hull geometry and mooring
system and the initial configuration is chosen such that the motion criteria are under
limit when it comes to vessel displacement and air gap to avoid green water
phenomenon. The research contributes heavily to the selection of FPSO configuration
with the motion performance studied for similar configured FPSOs operating in
Malaysia and Australia, parametric charts developed and life-cycle cost calculated. The
results of the research could help in the selection of appropriate initial hull geometry
and mooring system so that the burden of iteration and modelling procedure could be
reduced. The life-cycle cost aspects of FPSO will help take logical cost decisions in the
early phase of FPSO project, bringing managerial aspect to the engineering work and
reduce monetary wastage later in the life-cycle of FPSO ensuring maximum
productivity and performance.
5.1.1 Adequacy of simulation model and simulation procedure adopted to predict

wave frequency motion responses
Software simulation model is developed for uncoupled analysis and from the
comparisons discussed in Chapter 4, it can be concluded that the software simulation
procedure for the uncoupled frequency domain analysis has produced response results
agreeing closely with the trend of experimental results. Also, the magnitudes of the
results have been found to be having very low RMSD value, maximum being 0.16 in
188
surge compared with experimental results. Hence, this simulation procedure can be
very well adopted for the parametric study where mooring line details are not key.
Also, software simulation model is developed for coupled analysis, when mooring
lines are significant to study the variation in motion response of FPSO. Coupled
software simulation procedure has been validated against the experimental tests
conducted by Kim et al [42] at the OTRC wave basin. The developed simulation model
could predict the natural frequencies of the FPSO with less than 5% error. The motion
responses were found to have acceptable accuracy with less than 25% difference with
the experimental results produced by Kim et al [42].
5.1.2 Suitability of spread moored and turret moored FPSOs in extreme weather
conditions
To assess the suitability of turret moored and spread moored FPSO in extreme weather,
wave height parametric study was conducted with and without the presence of wind
and current. Heave, Roll and pitch motions remain relatively similar for both spread
mooring and turret mooring configurations with or without wind and current. It was
also observed that the rate of increase in heave motion is directly proportional to wave
period. The higher the wave period, the higher will be the heave motion. This can
contribute to higher vertical combined motion and subsequent relative motion as seen
in section 4.7.1, Table 4.12 ~Figure 4.22. So, at higher wave heights, FPSOs are prone
to risk from green water and thus downtime.
In the absence of wind and current, all the six motions for turret moored and spread
moored FPSOs increase. In reality, wind and current exists and the amplitude of
horizontal plane motions of turret moored FPSO are relatively higher when compared
to that of spread moored FPSO in the presence of wind, wave and current due to the
drifting force. Sway and yaw motions are significantly reduced when spread moored
189
system is used. But, the horizontal FPSO motions for turret moored FPSO decreases
as wave height increases and becomes comparable to those of spread moored FPSO;
however, if we use a spread moored configuration in adverse climates, the motions
escalate resulting in mooring line damage. This is the main reason for preferring turret
moored FPSO over spread moored FPSO in adverse climates and this very behaviour
shows the weathervaning nature of the turret moored FPSOs in extreme weather
conditions. The metocean parametric charts developed could be used to obtain motion
response amplitudes of similar configured FPSOs in varying wave heights and make
decision while choosing mooring system based on metocean conditions.
5.1.3 Effect of water depth, mooring line and hull parameters on FPSO motions
The water depth parametric study covers water depths of domestic oil zones in Malaysia
(62 m – Erb West, 70 m – PMO, 75 m – Baram Delta) [163] and in Australia (80 m-
Montara) [2]. In surge, heave and pitch the mean RAO increases as the water depth
increases from 62 m to 100m (as per model tests). For FPSOs to be used in shallow
waters, care should be taken while designing them, as in shallow waters up to 100 m,
heave and pitch motion increases resulting in high combined vertical motion of FPSO
which can make the FPSO susceptible to green water and subsequent downtime.
At high wave periods (low frequency waves), the surge and heave motions are
minimum at mooring line azimuth angle 15º and 45º. Motions are always low when
the mooring line azimuth angle is 15º. To prevent high amplitude heave motion, it is
best to avoid mooring line azimuth angle 30º for FPSOs with similar configuration. The
six degrees of freedom FPSO motions decline with an increase in the ratio of hull length
to mooring line length. It is best to keep the mooring line length minimum to minimise
the FPSO motions with appropriate pretension for similarly configured FPSOs. The
FPSO motions are minimum when the mooring line fairleads are located at 12% to 21%
of LOA from aft and fore. The FPSO motions increase heavily when the spread mooring
190
fairleads are kept closer to mid ship. Pitch motions are least sensitive to mooring line
parameters.
For Malaysian seas, the most probable peak period is usually in the range of 6 s to
7 s [132]. It can be seen from Figure 4.40 ~Figure 4.41, for FPSOs having beam to
length ratio in the same range as the Malaysian and Australian FPSOs, surge, sway and
yaw motion remains almost the same for peak period less than 7 s and heave and pitch
reduces as hull length to beam ratio increases. This means that for the Malaysian
FPSOs, for hull length to beam ratio of 6.094, the combined vertical motion will be less
for the Malaysian metocean conditions and thus risk from green water is less. The peak
period of wave is higher in Australian seas and the Australian sea is prone to extreme
cyclones with longer wave periods [4] as seen in section 4.7.1. Again, at higher wave
periods, the heave, pitch and roll motion decreases with increase in hull length to beam
ratio. Thus, risk from green water will be less when hull length to beam ratio of 6.094
is used. However, surge and sway motion increases in higher wave periods with
increase in hull length to beam ratio, which again could be handled using weathervaning
FPSOs. Surge, sway, heave and pitch motion decreases as vessel loading increases and
roll and yaw motion increases with vessel loading for the given metocean data and
vessel dimensions. Increase in roll could be controlled by providing additional roll
damping. Even though, the vertical motions will be minimum at full loaded condition,
care should be taken while designing to avoid green water effects and operational
downtime at higher wave heights as the FPSO will have the least freeboard and
maximum draft in fully loaded condition.
5.1.4 Downtime cost due to green water effects for site specific conditions of
Malaysia and Australia
Downtime cost due to green water effects under location specific wind generated sea
condition is evaluated to be zero for FPSOs with dimensions shown in Table 3.6 using
191
location specific annual wave scatter diagram approach. Relative motions are found for
FPSOs with similar dimensions as operating FPSOs in Malaysia and Australia as given
in Table 3.6 and it was found that freeboard for those FPSOs in maximum draft is higher
than the relative motions in section 4.7. However, for 100-year extreme operating
conditions, FPSOs are highly susceptible to green water in Australian sea, while FPSOs
in Malaysian seas have only low susceptibility. The downtime cost is calculated to be
included for the 10-year and 25-year life-cycle cost study of FPSOs in Table 3.6 and
the 100-year extreme conditions might not occur during those periods or probability of
those events are very low. Hence a future work is recommended to study downtime due
to green water under location specific swells due to cyclones for annual conditions in
Australian seas.
5.1.5 Cost effective FPSO configurations for 10-year and 25-year use in Malaysia
and Australia
Comparing the total life-cycle cost, riser turret moored FPSOs are the costliest,
followed by internal turret moored FPSOs for both 10-year and 25-year life-cycle
periods, based on the available FPSO cost data from different reliable sources. The least
expensive option for 10-year life-cycle period is spread moored FPSO, while for 25-
years, it is external turret moored FPSO. Turret moored FPSOs are shown to have
higher NPV including and excluding revenue from oil price among the FPSOs used for
LCCA and the main cost driving factor for NPV is identified to be capital, operation
and maintenance cost.
Among the FPSOs compared in section 4.12, internal turret moored FPSOs are
found to be having higher NPV. Internal turret moored FPSOs are found to be profitable
in the long run even after having newly built hull in some cases and in some cases,
being leased.
192
Also, as the discount rate increases, the net present value of the asset decreases as
the risk associated with the investment increases. And, it can be seen from Figure 4.56
~ Figure 4.59 that, NPV decreases as capital cost increases. The NPV profile reflects
the net present worth of these FPSOs, at a varying market situation and can be used as
a reference in the initial estimate of similar configured FPSOs.
Hence considering both motion performance and cost of FPSO configurations,

spread moored FPSOs are preferred option for short term use in benign ocean
environment due to less life cycle cost as seen in section 5.1.5 and reduced motion
amplitude as mentioned in section 5.1.2, whereas turret moored options are effective in
the long run in extreme metocean conditions for higher NPV, less total life-cycle cost
for 25 years as mentioned in section 5.1.5 and weathervaning nature as seen in section
5.1.2.
5.2 Recommendation for Future Work
This research was aimed to study the motion responses and cost of FPSO to enable
better selection of FPSO configuration to have increased productivity and better
performance. The results of this study are region specific to Malaysia and Australian
sea for chosen FPSO dimensions. The following studies should help in the ultimate
endeavor for a better understanding of this topic:
 Sea keeping performance for FPSOs under multi directional waves and higher
order waves could be studied varying the structural and metocean parameters
including both mooring and risers in the analysis.
 Downtime cost for FPSOs in Australia could be calculated due to green water
incidents when FPSO is subjected to site specific swells due to extreme cyclonic
conditions.
193
 Obtaining original ship lines from FPSO operators are difficult due to company
policies. However, the simulation procedure adopted could be used to conduct
dynamic analysis, downtime calculation and LCCA of any FPSOs and original
ship lines can be used for the same upon availability in future.
 The need for LCCA while choosing mooring system is recommended to be

included in the standards for design of floating platforms while choosing hull
and mooring system.
194
REFERENCES
[1] J. K. Paik and K. T Anil., Ship Shaped offshore Installations Design , Building and
Operation, Cambridge University Press, 2007.
[2] C. Barton, H. Hambling, E. K. Albaugh, B. Mahlstedt and D. Davis. (2017, August).

2017 Worldwide Survey of Floating Production, Storage and Offloading (FPSO) Units.
Offshore Magazine [Online]. Available: www.offshore-mag.com/maps-posters.html
[3] T. Moan, J. Amdahl, X. Wang and J. Spencer, “Risk Assessment of FPSOs, with
Emphasis on Collision, ” ABS Tech. Paper, pp. 200 – 229, 2002.
[4] J. Xia (2013, July). Proper FPSO hull and mooring designs can improve production
uptime. Offshore Magazine [Online]. Available: https://www.offshore-
mag.com/articles/print/volume-73/issue-6/engineering-construction-
installation/proper-fpso-hull-and-mooring-designs.
[5] L. T. England, A. S. Duggal and L. A. Queen, “A Comparison Between Turret and

Spread Moored FPSO for Deep Water Field Developments,” Deep Offshore Technology
, pp. 1-23, 2001.
[6] D. M. Mattos and C. F. G. Mastrangelo , “PORTRAIT OF FPSO USE OFFSHORE

BRAZIL, ” FPSO Workshop Proc., Houston, TX, pp. 54-63, 2000.
[7] “Rationalisation of FPSO design issues, ” Noble Denton Europe Ltd., London, Offshore
Technology Report 2000/097, 2001.
[8] B. Buchner, “Green Water From the Side of an Weathervaning FPSO,” Proc. of
OMAE99, Newfoundland, Canada pp. 1–11, 1999.
[9] SESAM USER MANUAL, DeepC Theory, Marintek and DNV, 2005.
[10] J. A. Pinkster and G. F. M. Remery, "The Role of Model Tests in the Design of Single
Point Mooring Terminals ," Offshore Technology Conference, Texas, OTC 2212, pp
679 - 703, 1975.
[11] J. E. W. Wichers, “A Simulation Model for a Single Point Moored Tanker,” Ph.D.
dissertation, Tech. Univ. Delft, 1988.
195
[12] D. L. Garrett, “Coupled analysis of floating production systems,” Ocean Eng., vol. 32,
no. 7, pp. 802–816, 2005.
[13] G. A. Gratsos, H. N. Psaraftis, and P. Zachariadis, “Life cycle cost of maintaining the
effectiveness of a ship’s structure and environmental impact of ship design parameters:
An update,” Int. Conf. Des. Oper. Bulk Carriers, pp. 169–182, 2009.
[14] E. P. D. Mansard and B. D. Pratte, "Moored Ship Response in Irregular Waves," Coastal
engineering, pp. 2621–2640, 1982.
[15] S. K. Chakrabarti, “Physical model testing of floating offshore structures,” Proc. MTS
Dyn. Position. Conf., 1998.
[16] C. Fournier and D. Anglin, “The Use of Moored Ship Response Modelling Coupled
With Pianc Vessel Motion Criteria for the Estimate of Port Downtime the Case of
Complejo Portuario De Mejillones , Chile,” Canadian Coastal Conf., pp. 1–14, 2003.
[17] B. V. Korvin-Kroukovsky, "Investigation of ship motions in regular waves,"

Transactions of the society of Naval Architects and Marine Engineers, vol. 63, pp. 386-
435, 1955.
[18] N. Salvesen, E. Tuck, and O. Faltinsen, “Ship motions and sea loads,” Trans. SNAME,
vol. 78, pp. 250–287, 1970.
[19] J. N. Newman, "The theory of ship motions," Advances in Applied Mechanics, vol. 18,
pp. 221- 283, p. 1978.
[20] P. D. Sclavounos, "The diffraction of free-surface waves by a slender ship," Journal of

Ship Research, vol. 1, no. 28, pp. 29-47, 1984.
[21] Chang, M.S, "Computations of three-dimensional ship-motions with forward speed",

Proc. 2nd Int. Conf. on Numerical Ship Hydrodynamics, Berkeley, USA, pp. 124-135,
1977.
[22] R. B. Inglis and W. G. Price, "A three dimensional ship motion theory- comparison
between theoretical predictions and experiment data of the hydrodynamic coefficients
with forward speed," Transactions of the Royal Institution of Naval Architects, vol. 124,
1982.
196
[23] R. B. Inglis and W. G. Price, "A three dimensional ship motion theory: Calculation of
wave loading and responses with forward speed," Transactions of the Royal Institution
of Naval Architects, vol. 124, pp. 183-192, 1982.
[24] G. X. Wu and R. E. Taylor, R.E, "The numerical solution of the motions of a ship
advancing in waves," Proc. 5th Int. Conf. on Numerical Ship Hydrodynamics,
Hiroshima, Japan, pp. 529-538, 1989.
[25] X. B. Chen, L. Diebold and Y. Doutreleau, "New green-function method to predict

wave-induced ship motions and loads," Proc. 23rd Symp. on Naval Hydrodynamcis, Val
de Reuil, France, pp. 66-81, 2000.
[26] C. W. Dawson, "A practical computer method for solving ship-wave problems," Proc.
2nd Int. Conf. on Numerical Ship Hydrodynamics, Berkeley, USA, pp. 124-135, 1977.
[27] Y. T. Fan and P. A. Wilson, "Time-domain non-linear strip theory for ship motions,"
Int. Journal of Maritime Eng., pp. 33-47, 2004.
[28] J. H. Vugts, “The hydrodynamic coefficients for swaying, heaving and rolling cylinders
in a free surface,” Int. Shipbuild. Prog., vol. 15, pp. 251–276, 1968.
[29] J. N. Newman, “The Exciting Forces on Fixed Bodies in Waves,” J. Sh. Res. Soc. Nav.
Archit. Mar. Eng., vol. 6, pp. 1–10, 1962.
[30] J. Journée, “Quick strip theory calculations in ship design,” PRADS’92 Conf. Pract.
Des. Ships, vol. I, pp. 5–12, 1992.
[31] S. N. Das and S. K. Das, “Determination of coupled sway, roll, and yaw motions of a
floating body in regular waves,” Int. J. Math. Math. Sci., no. 41, pp. 2181–2197, 2004.
[32] K. P. W. Hem Lata and Thiagarajan, “Comparison of added mass coefficients for a
floating tanker evaluated by conformal mapping and boundary element methods,”
Mech. Eng., vol. 1, pp. 1388–1391, 2007.
[33] T. Momoki, T. Kneko and T. Fukasawa, "On the Calculation Method of Surface
Pressure Distribution of a Container Ship in Wave," Proc. Twenty-second ISOPE
Conference, Greece, pp 961 - 967, 2012.
[34] T. Jiang and G. Lloyd, “Motion Prediction of a Single- Point Moored Tanker Subjected
197
to Current, Wind and Waves,” Journal of Offshore Mechanics and Arctic Engineering,
vol. 112, pp. 83-90, 1990.
[35] G. Feng, X. Jiang, H. Ren, and Q. Zhang, “Motion Response Analysis of Multi-point
Moored FPSO,” Proc. 25th ISOPE, Hawaii, USA, pp. 1666–1674, 2015.
[36] J. M. Heurtier, P. Buhan, E. Fontane, C. Cunff, F. Biolley and C. Berhault, "Coupled

Dynamic Response of Moored FPSO with Risers," Proc. of the Eleventh ISOPE
Conference, Norway, Vol I, pp. 319 – 326, 2001.
[37] J. E. W. Wichers, H. J. Voogt, H. W. Roelofs and P. C. M Driessen,. DeepStar-CTR

4401-Benchmark Model Test, Technical Report No. 16417-1-OB, MARIN,
Netherlands, 2001.
[38] J. E. W. Wichers and P. Devlin, “Effect of Coupling of Mooring Lines and Risers on
the Design Values for a Turret Moored FPSO in Deep Water of the Gulf of Mexico,”
Proc. 11th Int. Offshore Polar Eng. Conf., Stavanger, Norway, pp. 480–487, 2001.
[39] Y. Lou and S. Baudic, "Predicting FPSO responses using model tests and numerical
analysis," Proc. of the Thirteenth ISOPE, USA, pp. 167-174, 2003.
[40] Y. M. Low and R. S. Langley, “Time and frequency domain coupled analysis of
deepwater floating production systems,” Appl. Ocean Res., vol. 28, no. 6, pp. 371–385,
2006.
[41] A. Tahar and M. H. Kim, “Hull/mooring/riser coupled dynamic analysis and sensitivity
study of a tanker-based FPSO,” Appl. Ocean Res., vol. 25, no. 6, pp. 367–382, 2003.
[42] M. H. Kim, B. J. Koo, R. M. Mercier and E. G. Ward, “Vessel/mooring/riser coupled

dynamic analysis of a turret-moored FPSO compared with OTRC experiment,” Ocean
Eng., vol. 32, no. 14–15, pp. 1780–1802, 2005.
[43] A. Priyanto, Erwandi and Samudro, “Surge Forces of a FPSO at Lee Side of Submerged
Plate in Regular Waves,” 2nd Reg. Conf. on Vehicle Eng. and Tech., Kuala Lumpur, pp.
1 - 6, 2008.
[44] C. L. Siow, J. Koto, H. Yasukawa, A. Matsuda, D. Terada, C. G. Soares and M. Z. M

Samad, “Experiment Study on Hydrodynamics Characteristic of Rounded- Shape
FPSO,” The 1st Conf.on Ocean, Mechanical and Aerospace -Science and Eng., pp. 1–
198
6, 2014.
[45] C. L. Siow, J. Koto, H. Yasukawa, A. Matsudad, D. Teradad, C. G.Soarese, A. Incecikf,

M. A. G. Pauzi, “Full paper Mooring Effect on Wave Frequency Response of Round
Shape FPSO,” Jurnal Teknologi, vol. 5, no. 74, pp. 59–68, 2015.
[46] A. O. Vázquez-hernández, G. B. Ellwanger, and L. V. S. Sagrilo, “Long-term response

analysis of FPSO mooring systems,” Appl. Ocean Res., vol. 33, pp. 375–383, 2011.
[47] Y. Rho, K. Kim, C. Jo, and D. Kim, “Static and dynamic mooring analysis - Stability of
floating production storage and offloading ( FPSO ) risers for extreme environmental
conditions,” Inter J Nav Arch. Oc Engng, vol. 5, pp. 179–187, 2013.
[48] E. Fontaine, P. Orsero, A. Ledoux, R. Nerzic, M. Prevosto, and V. Quiniou, “Reliability

analysis and Response Based Design of a moored FPSO in West Africa,” Struct. Saf.,
vol. 41, pp. 82–96, 2013.
[49] Q. W. Ma, S. Yan, D. Greaves, T. Mai, and A. Raby, “Numerical and experimental
studies of Interaction between FPSO and focusing waves,” Proc. of the 25th ISOPE,
Hawaii, USA, 2015.
[50] J. Chen, Y. Sun, and P. Zhang, “Dynamic Response Analysis of FPSO Based on
SESAM,” Adv. Mater. Res., vol. 694–697, pp. 267–270, 2013.
[51] C. Ji, Y. Cheng, Q. Yan, and G. Wang, “Fully coupled dynamic analysis of a FPSO and
its MWA system with mooring lines and risers,” Appl. Ocean Res., vol. 58, pp. 71–82,
2016.
[52] C. Kang, C. Lee, S. Jun, and Y. Oh, “Fatigue Analysis of Spread Mooring Line,”
International Journal of Geological and Environmental Eng., vol. 10, no. 5, pp. 504–
510, 2016.
[53] J. T. Lopez, L. Tao, L. Xiao, and Z. Hu, “Experimental study on the hydrodynamic
behaviour of an FPSO in a deepwater region of the Gulf of Mexico,” Ocean Eng., vol.
129, pp. 549–566, 2017.
[54] S. Hong, J. Lew, D. Jung, H. Kim, D. Lee, and J. Seo, “A study on the impact load
acting on an FPSO bow by steep waves,” Int. J. Nav. Archit. Ocean Eng., vol. 9, pp. 1–
10, 2017.
199
[55] DNV, “DeepC Coupled Analysis Tool. A white Paper. Rev 3,” 2004.
[56] Global Performance Analysis of Deepwater Floating Structures, Recommended

Practice DNV-RP-F205, October, 2010.
[57] M. Caire, C. E. S. de Souza, and J. P. R. Cortina, “The influence of wind-wave energy

spreading on the riser system response of a spread-moored FPSO,” Mar. Syst. Ocean
Technol., no.10, pp. 26-37, 2015.
[58] Y. M. Low and R. S. Langley, “A hybrid time/frequency domain approach for efficient
coupled analysis of vessel/mooring/riser dynamics,” Ocean Eng., vol. 35, no. 5–6, pp.
433–446, 2008.
[59] Dynamic Loading Approach for Floating Production , Storage and Offloading (FPSO)
Installations, ABS, May 2010.
[60] O. M. Faltinsen, Sea loads on ships and offshore structures, Cambridge University
press, vol. 1. pp. 223–228, 1990.
[61] J. Xia, “FPSO Design to Minimise Operational Downtime due to Adverse Metocean
Conditions off North West Australia,” Deep Offshore Technol., pp. 1–15, 2012.
[62] S. K. Chakrabarti , Offshore Structure Modeling, Advanced Series on Ocean

Engineering Vol.9, World Scientific Publishing Co. Ptc. Ltd, Singapore, 1994.
[63] R. Mercier, “FPSO Responses in Gulf of Mexico Environments,” OTRC Report, 2007.
[64] A. S. Duggal, C. N. Heyl, and P. F. Poranski, “Terra Nova FPSO: Integration of model
tests and global analysis,” JMR03, pp. 1-6, 2000.
[65] S. K. Chakrabarti, Hydrodynamics of Offshore Structures , Computational Mechanics

Publications, Heidelberg, 1987.
[66] R. L. Wiegel and J. W. Johnson, “Elements of Wave Theory,” Coast. Eng., no. 1893,
pp. 5–21, 1950.
[67] Y.M. E. Abbas, “Studies on the non-linear interactions associated with moored semi-
submersible offshore platforms,” Ph.D Thesis, Universiti Teknologi PETRONAS, 2011.
[68] R. S. Sharizal Amurol, Kevin Ewans, “Measured Wave Spectra Offshore Sabah &
200
Sarawak, Malaysia,” in Offshore Technology Conference Asia held in Kuala Lumpur,
Malaysia, 2014, no. March, pp. 1–13.
[69] J. Morison, "The force exerted by surface waves on piles," University of California,
Dept. of Engineering, Fluid Mechanics Laboratory, 1949.
[70] SESAM USER MANUAL, Wadam, DNV, 2010.
[71] “Review of model testing requirements for FPSO’s, ” BMT Fluid Mechanics Ltd.,
United Kingdom, Offshore Technology Report 2000/123, 2001.
[72] R.V. Whitman, “Evaluating Risk in Geotechnical Engineering”, ASCE Convention and
Exposition, 1981
[73] T. Cepowski, “The modeling of seakeeping qualities of Floating Production , Storage

and Offloading ( FPSO ) sea-going ships in preliminary design stage,” Polish Maritime
Research, vol. 17, no. 4, pp. 3–12, 2010.
[74] X. Li, J. Yang, and L. Xiao, “Motion Analysis on a Large FPSO in Shallow Water,”
Proc. of the 13th ISOPE, Hawaii, USA, pp. 235–239, 2003.
[75] S.Q. Wang, S. Y. Li and X. H. Chen, “Dynamical Analysis of a Soft Yoke Moored
FPSO in Shallow Waters,” Proc. of the sixth Int. conf. on Asian and Pacific Coasts,
Hong Kong, China. pp. 956 - 963, 2011.
[76] C. G. Soares, N. Fonseca and R. Pascoal, “Experimental and Numerical Study of the
Motions of a Turret Moored FPSO in Waves, Journal of Offshore Mechanics and Arctic
Engineering, Vol. 127,pp. 197 - 204, 2005.
[77] J. Munipalli and K. Thiagarajan, “Effect of Wave Steepness on Yaw Motions of a

Weathervaning Floating Platform,” 16th Australasian Fluid Mechanics Conference,
Gold Coast, Australia, pp. 1007–1011, 2007.
[78] T. R. Kannah and R. Natarajan, "Experimental Study on the Hydrodynamics of a

Floating, Production, Storage, and Offloading System," Journal of waterway, port,
coastal and ocean engineering, no.132, pp. 66-70, 2006.
[79] T. R. Kannah and R. Natarajan, “Effect of Turret Location on the Dynamic Behaviour
of an Internal Turret Moored FPSO System,” Journal of Naval Architecture and Marine
201
Engineering, pp. 23 - 27, 2006.
[80] A. Yadav, S. Varghese and K. P. Thiagarajan, “Parametric Study of Yaw Instability of

a Weathervaning Platform,” 16th Australasian Fluid Mechanics Conference, Gold
Coast, Australia, pp. 1012 - 1015, 2007.
[81] N. Baghernezhad, P. Edalat, and M. Etemaddar, “Hull Performance Assessment and

Comparison of Ship-Shaped and Cylindrical FPSOs With Regards To : Stability , Sea-
Keeping , Mooring and Riser Loads In Shallow Water,” Int. Journal of Maritime
Technology, Vol. 8, pp. 1 - 13, 2017.
[82] M. O. Ahmed and A. Musaad, “Effects of mooring lines and water depth parameters on
the dynamic motion of a turret moored FPSO, ” Engineering Challenges for Sustainable
Future, 2016.
[83] K. C. Ewans, L. Vanderschuren, P. S. Tromans, "FPSO Conference-Estimating Wind-

Sea and Swell for FPSO Operability," Journal of Offshore Mechanics and Arctic Eng.,
no. 128,pp. 314-321, 2006.
[84] A. Hassan, M. Downie, and A. Incecik, “Wave drift forces and responces in deep water
and extreme environmental conditions,” Proc. HYDRALAB III Jt. User Meet., pp. 3–6,
February 2010.
[85] C. T. Stansberg, E. M. Hermundstad, J. R. Hoff, R. Baarholm, "Wave Drift Forces in

Current," Proc. OMAE, France, 2013.
[86] M. J. Teles, M. Benoit, A. A. P. Silva, "Modelling Combined Wave-Current Flows

Using a RANS CFD Solver with Emphasis on the Effect of the Turbulent Closure
Model," Advances in Hydroinformatics , Springer Hydrogeology, Singapore, pp. 473-
485, 2014.
[87] “Analysis of green water susceptibility of FPSO/FSU’s on the UKCS, ” BOMEL Ltd.,
United Kingdom, Offshore Technology Report 2001/005, 2001.
[88] B. Buchner, “The Impact of Green Water on FPSO Design,” Offshore Technology
Conference, Texas, USA, pp. 45–57, 1995.
[89] B. Buchner, Green Water on Ship-type Offshore Structures, Grafisch Bedrijf Ponsen &
Looijen by, Wageningen, The Netherlands, 2002.
202
[90] K. B. Nielsen, “Numerical Prediction of Green Water Loads on Ships, ” Ph.D Thesis,
Technical University of Denmark , Lyngby, Denmark, 2003.
[91] E. Ballard, N. Barltrop, E. Falkenberg, S. Fyfe, C. G. Soares, B. Iwanowski, T.

Kleefsman, “ Summary Report on Desing Guidance and Assessment Methodologies for
Wave Slam and Green Water Impact,” Safe Flow, Rep. no. 15874-1-OE, February,
2004.
[92] K. M. T. Kleefsman, G. E. Loots, A. E. P. Veldman, B. Buchner, T. Bunnik and E.

Falkenberg, “ The Numerical Simulation of Green Water Loading Including Vessel
Motions and the Incoming Wave Field, ” Proc. of 24th Int.Conf. on Offshore Mechanics
and Arctic Engineering, Halkidiki, Greece, pp. 1 - 12, 2005.
[93] H. Lu, C. Yang, and L. Rainald, “Numerical Studies of Green Water Effect on a Moored
FPSO,” Proc. of the Eighteenth Int.Offshore and Polar Engineering Conf., Vancouver,
Canada, 2008.
[94] N. Tao, W. Yun, L. Bingnan, T. Shentao and Z. Yongkang, “State-of-the-art of green

water on FPSO in harsh environment,” Key Engineering Materials, 2011.
[95] E. Akandu, A. Incecik, and N. Barltrop, “The Susceptibility of FPSO Vessel to Green
Water in Extreme Wave Environment,” The 1st Conf. on Ocean, Mechanical and
Aerospace, pp. 82–87, 2014.
[96] R. V. Veer and A. Boorsma, “Towards an improved understanding of Green water

exceedance at the Bow of an FPSO, ” 35th Int. Conf. on Ocean, Offshore and Arctic
Engineering, Busan, South Korea, Vol 1, 2016.
[97] R. Werter, “Green water along the side of an FPSO,” MSc Thesis, Delft University of
Technology, 2016.
[98] X. Zhang, H. Wolgamot, S. Draper, W.Zhao and L. Cheng, “The role of overtopping
duration in greenwater loading Xiantao,” 33rd IWWWFB workshop, Guidel-Plages,
France, pp. 1–4, 2018.
[99] B. Buchner and J. L. Garcia, “Design aspects of Green Water Loading on FPSOs ,” The
22nd Int. Conf. on Offshore Mechanics & Arctic Engineering, Cancun, Mexico, pp. 1–
11, 2003.
203
[100] R. J. Van Der Wal and G. De Boer, “Downtime Analysis Techniques for Complex
Offshore and Dredging Operations,” Proc. 23rd Int. Conf. on Offshore Mechanics and
Arctic Engineering, Vancouver, Canada, pp. 1–9, 2004.
[101] E. B. Djatmiko, Murdjito and I. Prasetyawan, “Operability analysis of an FPSO on the

basis of its slamming and green water,” The 5th Biennale Conf. on Marine Technology,
Makassar, Indonesia, pp. 1 -16, 2006.
[102] D. C. Corea, A. C. de Oliveira, E. A. Tannuri and S. H. Sphaier, “Comprehensive

Downtime Analysis of DP- Assisted Offloading Operation of Spread Moored Platforms
in Brazilian Waters,” Proc. of the 32nd Int. Conf. on Ocean, Offshore and Arctic Eng.,
Nantes, France, pp. 1 -8, 2013.
[103] A. Whyte, Life –cycle cost analysis for built assets: LCCA framework, VDM Verlag
Dr. Muller, Germany, 2011.
[104] K. V. Oeveran K. V. and M. Wilks, "Life cycle costing qualification document," Rev 3
, pp 1 – 7, 2009.
[105] I. Thalji, J. P. Liyanage and M. Hjollo, "Scalable and Customer – Oriented Life Cycle
Costing Model: A Case Study of an Innovative Vertical Axis Wind Turbine Concept
(Case – VAWT)," Proc. Twenty – second ISOPE Conference, Rhodes, Greece, pp 423-
425, 2012.
[106] J. Emblemsvag, Life-cycle costing, John Wiley a sons, USA, ISBN 978-0-4713-5885-
5, 2003.
[107] J. Roubal, ‘Product life cycle cost management--necessary tool for industrial
companies," The Free Library 01, January 2009.
[108] K. Nam, D. Chang, K. Chang, T. Rhee and I. B. Lee, "Methodology of Lifecycle Cost
with Risk Expenditure for Offshore Process at Conceptual Design Stage," Energy, No.
36, pp 1554 – 1563, 2011.
[109] L. C. Santos, G. P. Gareia and V. D. Casas, “Methodology to study the life cycle cost
of floating offshore wind farms,” Energy Procedia, pp 1- 8, 2013.
[110] G. B. Howell, A. S. Duggal, C. Heyl, and O. Ihonde, “Spread Moored or Turret Moored
FPSO’s for Deepwater Field Developments,” Procedings Offshore West Africa, pp. 1–
204
21, 2006.
[111] D. Kayrbekova, “Activity-Based Life-Cycle Cost Analysis, ” PhD Thesis, University

of Stavanger, Norway, 2011.
[112] H. A. Kurniawati, W. D. Aryawan, A, “Long-term fso/fpso charter rate estimation 1),

KAPAL, Vol. 13, No.1, pp. 7–12, 2016.
[113] J. M. Cabrera-miranda, P. M. Sakugawa, R. Corona-Tapia, J. K. Paik, “On design

criteria for a disconnectable FPSO mooring system associated with expected life- cycle
cost,” Ships Offshore Struct., vol. 13, no.4, pp. 432 - 442, 2018.
[114] L. Greene and B. Shaw, “The steps for successful life cycle cost analysis," Proc. IEEE,
Ch 2881, pp.1209-1216, 1990.
[115] D.J.O. Ferry and R. Flanagan (1991). Life cycle costing – a radical approach, CIRIA
report no.122, London, UK.
[116] S. Kirk and A. Dell’Isola, Life cycle costing for design professionals, 2nd ed., McGraw-
Hill, 1995.
[117] A. Al Hajj, "Simple cost-significant models for total life-cycle costing in buildings,"
Ph.D. dissertation, University of Dundee, UK, 1991.
[118] A. Ashworth, "Life Cycle Costing:predicting the unknown," Building Engineer, vol. 71,
no. 3, pp. 18-20, 1996.
[119] H. Barringer and D. Weber, "Life Cycle Cost Tutorial," Fifth Int. Conf.on Process Plant
Reliability, Houston, TX, no.3, 1996.
[120] J.K. Hwang, M. I. Roh, and K. Y. Lee, “Detailed design and construction of the hull of
a floating, production, storage and off-loading (FPSO) unit,” Ships Offshore Struct., vol.
5, no. 2, pp. 93–104, 2010.
[121] HR Multi-element wave generation system with AC derives and dynamic wave
absorption for Universiti Teknologi PETRONAS user manual , Wallingford - UK CQR
4187, 2008.
[122] QTM User Manual, 2011.
205
[123] M. O. A. Ali, “Numerical and Experimental Studies on the slow drift motions and the
mooring line responses of truss spar platforms, ” Ph.D Thesis, Universiti Teknologi
PETRONAS, 2012.
[124] Laboratory instrumentation and software Wave generation software manual, HR

Wallingford.
[125] H. R. Wallingford Ltd., “FPSO Response in Long and Short Crested Seas,” OTC Report,
no.18, 2002.
[126] A. Ruina and R. Pratap, “Center of mass and gravity,” Introd. to Statics Dyn., vol. 1, pp.
78–91, 2010.
[127] MARINET, Best Practice Manual for Wave Simulation, 2015.
[128] Instrumentation and software for physical models, HR Wallingford.
[129] DNV GL, “SE-23 FPSO Global Strength and Spectral Fatigue Analysis - Workshops,”
Kuala Lumpur, 2014.
[130] SESAM User Manual DeepC, Marintek and DNV, 2010.
[131] Structural Design of Offshore Ships, DNV-OS-C102, 2012.
[132] O. Yaakob, F. E. Hashim, K. M. Omar, A. H. M. Din, and K. K. Kong, “Satellite-based

wave data and wave energy resource assessment for South China Sea,” Renew. Energy,
vol. 88, pp. 359–371, 2016.
[133] Metocean Solutions. (2018). Hindcast. [Online]. Available:www.metoceanview.com.
[134] Offshore Magazine. (2013, 22 July). Offshore Map Atlas E-book. [Online].
Available:www.offshore-mag.com.
[135] Reserve Bank of Australia. (2018, 24 April). Measures of Consumer Price Inflation.
[Online]. Available: https://www.rba.gov.au/inflation/measures-cpi.html.
[136] Fusion Media Limited. (2018, May). 10 year Bond Yield. [Online]. Available:
https://au.investing.com/rates-bonds.
[137] Fenebris. (2018, March 31). Application of Valuation Parameters in Practice. [Online].
206
Available: http://www.market-risk-premia.com.
[138] State of Victoria, Australia. (2018, May 2). Wage Inflationa and Discount Rates.
[Online]. Available: https://www.dtf.vic.gov.au/financial-reporting-policy/wage-
inflation-and-discount-rates.” .
[139] Bank Negara Malaysia. (2018, May). Malaysia Inflation Rate. [Online]. Available:
http://www.bnm.gov.my.
[140] Petroleum and natural gas industries – Life cycle costing – Part 1: Methodology,
DS/EN ISO 15663-1, 2007.
[141] S. Thomas, private communication, March 2017.
[142] H. B. Taib, private communication, March 2017.
[143] Bluewater Holding B.V. (2014, june 30). Half-year report. [Online]. Available:
http://stamdata.no/documents/NO0010697485_IB_20140829.pdf.
[144] Rigzone. (2016, August 22). Malaysia’s Perisai Gets 6 Months Charter Extension for
FPSO Perisai Kamelia. [Online]. Available:
http://www.rigzone.com/news/oil_gas/a/146264/malaysias_perisai_gets_6_months_ch
arter_extension_for_fpso_perisai_kamelia.
[145] Wood Mackenzie, (2016, June). Enfield Area Key facts. [Online]. Available:
https://www.woodmac.com/store/
[146] Wood Mackenzie, (2015, July). Cossack-Wanaea-Lambert-Hermes Key facts. [Online].

Available: https://www.woodmac.com/store/
[147] Wood Mackenzie, (2017, February). Mutineer Exeter Area Key facts. [Online].
Available: https://www.woodmac.com/store/
[148] Aurelia Energy N. V. (2012). Annual report pursuant to Section 13 or 15(d) of the
securities Exchange Act of 1934. [Online]. Available:
https://www.sec.gov/Archives/edgar/data/1172374/000110465906021216/a06-
8054_120f.html.
[149] Petrofac. (2011, January 31). Petrofac to Lead Development of Berantai Field For
Petronas. [Online]. Available: https://www.petrofac.com/en-gb/media/news/petrofac-
207
to-lead-development-of-berantai-field-for-petronas/
[150] PETROLIAM NASIONAL BERHAD (PETRONAS). (2012). Reimagining Energy PE

2011 Annual Report. [Online]. Available: http://www.petronas.com.my/investor-
relations/Documents/annual-report/AnnualReport_FinancialStatement_PE2011.pdf.
[151] Wood Mackenzie, (2016, October). Kikeh Key facts. [Online]. Available:
https://www.woodmac.com/store/
[152] L. S. Wah and M. M. Chandrashegaran. (2012, June 25). FORGINGAHEAD Annual

Report 2012. Perisai Petroleum Teknologi Bhd. [Online]. Available:
http://ir.chartnexus.com/perisai/docs/AR/2012.pdf.
[153] C. Barton, H. Hambling, E. K. Albaugh, B. Mahlstedt and D. Davis. (2016, August).

2016 Worldwide Survey of Floating Production, Storage and Offloading (FPSO) Units.
Offshore Magazine. [Online]. Available: www.offshore-mag.com/maps-posters.html.
[154] P. C. Tulsian, Cost Accounting, Tata Mc-Graw Hill Publishing company Ltd, New
Delhi, 2011.
[155] M. Terrill and H. Batrouney. (2018, February 27).Why the discount rate for major
projects needs to be lowered. [Online]. Available:
https://www.macrobusiness.com.au/2018/02/discount-rate-major-projects-needs-
lowered/
[156] H. Ormberg, H. Lie, and C. T. Stansberg, "Coupled analysis of offshore floating

systems," Numerical Models in Fluid-Structure Interaction, Chapter 10, vol. 18, WIT
Press, pp. 389-429, 2005.
[157] Z. T. Xie, J. M. Yang, Z. Q. Hu, W. H. Zhao and J. R. Zhao, The horizontal stability of
an FLNG with different turret locations, Int. Journal of Naval Architecture and Ocean
Eng., No. 7, pp 244--258, 2015.
[158] A. Muzathik, W. Wan Nik, M. Ibrahim, and K. Samo, “Wave Energy Potential of
Peninsular Malaysia,” J. Eng. Appl. Sci., vol. 5, no. 7, pp. 11–23, 2010.
[159] H. M. Morishita and J. R. Souza Junior, "Dynamic Behaviour of a DICAS FPSO and
Shuttle Vessel under the Action of Wind, Current and Wave," Proc. Of the Twelth
ISOPE, Kitakyushu, Japan, pp. 142-150, 2002.
208
[160] D. Faulkner, M. J. Cowling and A. Incecik, A, Integrity of Offshore Structures- 4,
Elsevier Appl. Science, Elsevier Science Publishers Ltd., England, 1990.
[161] T. Ishihara, P. V. Phuc, H. Sukegawa, K. Shimada and T. Ohyama, "A study on the
dynamic response of a semi submersible floating offshore wind turbine system part 1:
A water tank test," ICEW 12, Cairns, Queensland, Australia, pp. 2511-2518, 2012.
[162] B. J. Gernon, J. Y. K. Lou, "Dynamic Response of a Tanker Moored to an Articulated

Loading Platform," Ocean Engineering, vol.14, no.6, pp.489 – 512, 1987.
[163] Design of Fixed Offshore Structures, PTS 34.19.10.30, 2010.
[164] R.G. Standing, S.J. Rowe, and W.J. Brendling, "Jacket Transportation Analysis in
Multidirectional Waves," 18th Annual Offshore Technology Conf., Houston, Texas,
May, 1986.
[165] Statista. (2018). Average annual Brent crude oil price from 1976 to 2018. [Online].
Available: https://www.statista.com/statistics/262860/uk-brent-crude-oil-price-
changes-since-1976/
209
PUBLICATIONS LIST
Published Papers
1. Rini N., Kurian V.J. & Whyte A., (2018), ‘Floating Production Storage and Offloading
Systems cost and motion performance: a system thinking application’, Frontiers of
Engineering Management.
2. Rini N., Kurian V.J. & Whyte A., (2016), ‘Dynamic Behaviour of FPSOs in the Kikeh-
Field under Different Loading Conditions’, ARPN Journal of Engineering and Applied
Sciences, vol-11, no-4, pp.2302-2307, ISSN 1819-6608.
Conference Proceedings
1. Rini N, Whyte A, & Kurian J, (2015), ‘Floater Responses: Life-Cycle Considerations

Off West Australia’, Proc of the 1st International Conference on Decommissioning &
Offshore & Subsea Structures, UK, pp.96-101, ISBN: 978-0-9930121-1-2.
2. Kurian J, Rini N & Whyte A, (2015), ‘Mooring-Line Serviceability for Floating
Production Storage & Offloading (FPSO) Facilities’, Proc of the 1st Int. Conference
on Decommissioning & Offshore & Subsea Structures, UK, pp.34-39, ISBN: 978-0-
9930121-1-2.
3. Gadja P, Whyte A, Rini N & Kurian J, (2015), ‘Offshore Substructure Design-
Specification Options Comparison using Life-Cycle Cost Analysis’, pp.56-59, ISBN:
978-0-9930121-1-2.
4. N. Rini, V.J. Kurian, M.S. Liew & A. Whyte., (2016), 'Coupled Analysis for the Effect
of Wave Height on FPSO Motions', Engineering Challenges for Sustainable Future, pp
69-73, ISBN 978-1-138-02978-1.
5. V.J. Kurian, N. Rini, M.S. Liew & A. Whyte., (2016), 'FPSO response and water depth:
A study using model tests', Engineering Challenges for Sustainable Future, pp 51-54,
ISBN 978-1-138-02978-1.
6. Rini Nishanth, Velluruzhathil John Kurian & Andrew Whyte, ‘Coupled motion
responses of turret moored and spread moored FPSO under the Influence of wave, wind
and current,’ pp 217, Proceedings of the International Conference on Ships and
Offshore Structures (ICSOS), Hamburg, Germany, 2016.
210

Dynamic Response and Life-Cycle Analysis of Floating Production Storage and Offloading Systems

Uploaded by

Copyright:

Available Formats

Dynamic Response and Life-Cycle Analysis of Floating Production Storage and Offloading Systems

Uploaded by

Document Information

Original Title

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

Dynamic Response and Life-Cycle Analysis of Floating Production Storage and Offloading Systems

Uploaded by

Copyright:

Available Formats

School of Civil and Mechanical Engineering

Dynamic Response and Life-Cycle Analysis of Floating

This thesis is presented for the Degree of

To my husband and Son

To my mother and sister To my grand parents

To my family members To my friends

I am immensely grateful to Dr. Andrew Whyte, Associate Professor and Head of

I would like to express my deep and sincere gratitude to my research supervisors,

I am thankful to my previous supervisor Dr Kurian Velluruzhathil John for not only

I am grateful to Curtin University and Universiti Teknologi PETRONAS staff,

Institute of Technology PETRONAS Sdn Bhd. / Curtin University

© RINI NISHANTH, 2018

Institute of Technology PETRONAS Sdn Bhd

All rights reserved.

Figure 1.1: Worldwide Distribution of FPSO 2017 ....................................................... 2

1.1 Chapter Overview

1.2 Background of FPSO

1.2.1 History of FPSO

Figure 1.1: Worldwide Distribution of FPSO 2017

Figure 1.2: Various Parts of FPSO

1.2.3 Advantages of FPSOs

Advantages of FPSOs are

• Early Production and huge storage capacity.

• Easy to remove and reuse.

• Reduced upfront investment.

• Can be used in any water depth.

• Abandonment costs are less than for fixed platforms.

• Ample deck space reducing risk of oil spilling.

1.3 Relevance of Cost and Motion Analysis of FPSO

Extreme FPSO motion can lead to significant operational downtime, disrupting

1.4 Relevance of Dynamic Analysis of FPSO

Spread mooring system helps to maintain a fixed orientation of FPSO in global

1.6 Problem Statement

The number of floaters operating in Malaysian and West-Australian waters is mounting

Hull motions should be minimised to achieve increased operational time and

1.7 Objectives of the Study

1. Validation of software simulation model and simulation procedures by

1.8 Scope of the Study

The scope of the study is as follows:

Figure 1.3: Six Degrees of Freedom of FPSO

 FPSO is assumed to be producing oil to its maximum capacity and life-cycle

1.10 Overview of the Thesis

2.1 Chapter Overview

2.2 Reported Studies

2.2.1 Dynamic Analysis of SPM and Ships

2.2.1.1 Model testing on SPM and ships

2.2.1.2 Numerical investigation of SPM and ship motions

Vugts [28] conducted experimental and theoretical investigations to determine the

2.2.2 Dynamic Analysis of FPSO

Tahar et al [41] developed a hull/mooring/riser coupled analysis program for a

2.2.3 FPSO Motion Response Using Uncoupled and Coupled Analysis

In the traditional methods of vessel response calculation, an uncoupled approach to

• The damping from mooring lines on LF vessel motion must be simplified;

2.2.4 FPSO Motion Response using Model Testing

2.2.4.1 Modelling of FPSO

2.2.4.2 Modelling of mooring lines

2.2.5 Hydrodynamics of FPSO

2.2.5.1 Representation of wave

The JONSWAP spectrum, which was derived from fetch-limited measurements