INTEGRATED SCHEDULING SYSTEM - AN APPROACH D-5

By

Joseph R. Lorenzo, Manager Dikshit Sharad.

Planning, Scheduling & and
Senior Consultant
Coordinating Information & ManagementSciences
Member, PACE
Ebasco Services Incorporated
Two Rector Street
NewYork, NewYork 10006

1. Introduction
CPM-basedscheduling methods are no novelty No one would possibly bother to becomeinvolv-
these days. Most of us have had someexposure to ed in sophisticated managementsystems if the
CPMor PERT, and possess varying degrees of ex- following needs were not experienced:
perience from understanding and delight to dismay,
discomfort and dispair. There are several systems . A logical plan for the work,
in use today -- someare offered by the computer
manufacturers, someare modified versions of manu- . An assurance of feasible schedule,
facturers' programs, and someare specially tailor-
madefor specific purposes. . A realistic estimate of resources,
A scheduling system serves its purpose very . A practical allocation of resources,
well when it integrates the elements of time and
resources. It is an excellent tool for inter- . A dependable cash flows.
discipline communications, for sharing and assign-
ing responsibility. As an integral part of corpor- The computer becomesquite indispensable:
ate information systems, it is an effective control
apparatus. . To generate and contain massive data,
A thoughtfully conceived work plan proves to . To process, manipulate and sort data,
be dependable after its built-in logic is tested
against available resources for feasible schedul- . To provide managementreports,
ing. This paper discusses resource planning and
demonstrates an approach to arrive at resource- . To maintain timeliness and accuracy.
constrained schedules.
For convenience of illustrating a typical
scheduling application, we shall consider the
power plant construction projects for which the
2. System Criteria system would provide three principal categories
of information:
Planning and scheduling, as a control tool,
has becomean accepted discipline or function, to . Time and resource data for Project Management
fill the needs of project managementwhich revolve to make strategy decisions and to satisfy
around the control of time, resources, costs. The alternate goals, such as, client requests,
modern computer-oriented techniques to meet such recovery from delays, and regulatory impacts.
needs have reached a degree'of sophistication
which is difficult to ignore. They confer two . Short-term, priority schedule guidance for
precious attributes on the managementprocess -- participating disciplines to properly utilize
speed and accuracy. their manpower.

CPM-basedtime control methods simply have . Key measurementdata so that the work progress
not proved to be consonant with long-established can be correctly evaluated.
cost recovery procedures used by constructors in
general. However, in large industrial projects We shall not enter into the detailed mechanics
such as power plants it is universally acknowledged of the computerized scheduling system, however, a
now that computerized CPMscheduling is economic- brief mention of the basics of CPMmight be rele-
ally feasible and practically inevitable. cost vant. A CPMnetwork, or a logic diagram, is a
control methods have two primary objectives: a) graphical representation of project activities plan-
to obtain production rates for use in estimating ned in a logical sequence. The critical path, or
new projects and in forecasting remaining work on the longest activity sequence, naturally requires
on-going projects, and b) to check the actual pro- maximumattention in terms of applying resources.
duction costs of current projects against the esti- The charts showing Typical Componentsof a Logic
mated costs. The real trick for total control Diagram and Illustrative Project Network may help
then lies in the prudent use of computer for inte- in understanding the concept. Only in case of
grating schedules with costs. small simplified logic diagrams we can compute
 OVERVIEWOF INTEGRATED
SCHEDULINGSYSTEM

ACTlON PLAN Y*rn LiErnCD EARLY& LATE

DATE BOUNDARIES

RESWRCE
CONSTRAINED.
SCMEDVLE

early/late dates for starting and completing various the work plan reasonably reflects the real
activities, and corresponding slacks. However, for world. The action plan, in the form of logic
a construction network, complexities arise out of diagrams, is fed into the CPM machinery.
external restraints such as design drawings, equip- Based upon the confines of project scope and
ment deliveries and work authorization by regula- interdependencies of project activities and
tory bodies, and out of internal tie-ins, such as upon the known key milestone dates, we get
among concrete slabs and equipment installation. the early and late date boundaries within
When total project activities requiring monitoring which the activities must start and finish.
and control get into 18,000 to 20,000 range, the The CPM outputs are standard for network
use of computer is inevitable. methods, and are basic to the determination
of schedule flexibility and trends.

3. System Overview b. Resource Planning (Resource Planning

Method - RPM)
The Overview of Integrated Scheduling System
diagram is put together so as to create an impres- Now the additional information on esti-
sion of an assembly-line production process, where mated resource requirements to carry out the
the final product is a resource-constrained sche- activities, in terms of quantities of mate-
dule! The raw materials required at various stages rial,manpower and costs can be cranked into
of the process are indicated. Also shown is the the process. The resource planning module
recycling from the Resource Planning & Scheduling examines each category of resources required
Method (RPSM) shop through to the Resource Planning over the entire project period to fulfill the
Method (RPM) shop involving the resource availabil- early date-based work schedule as well as to
ity planning done iteratively for optimization, or meet the late date-based work schedule. Re-
refining; of the product. source envelopes, or pairs of S-shaped curves,
are developed to review cumulative early start
There are, indeed, three levels of (data) pro- and late finish usage of resources. As can
cessing which various project input elements are be seen, only the upper and lower limits are
subjected to before the output is produced for established so far.
analysis, interpretation and use.
C. Action Planning (Resource Planning &
Scheduling Method - RPSM)
a. Time Planning (Critical Path Method-CPM)
To arrive at a practical action schedule,
The first step is to determine if all we have to relate the required resources to
the planning criteria are valid, if the net- what are actually available now and as best
work model represents the project scope and can be projected for later periods. This in-
165
 INITIAL STAGE

INTEGRATED SCHEDULING SYSTEM RESOURCES

AVAILABILITY INPUT

LOGIC REVIEW
COMMENTS
PREPARATION OF &MANPOWER INITIAL
LOGIC D’GMS. ESTIMATES
- - -

-” -7
Y
I

CONSTRUCTION LOGIC

UPDATE STAGE

INTEGRATED SCHEDULING SYSTEM

DATA COLLECTION
ON HOME OFFICE UPDATE ANALYSIS & ISSUE OF

OVERALL
SCHEDULE-RELATED\ MANAGEMENT ACTION
AS NECESSARY
QH:ii+,
PROJECT
7

\
\
\
\
\
COST/SCHEDULE 1
DATA COLLECTION DATA PROCESSING
FROM FIELD
OPERATIONS

formation is entered into the resource plan- Planning & Scheduling (P&S) has the prime res-
ning and scheduling program to test the poss- ponsibility to initiate and upkeep the scheduling
ible allocation against each activity. The operations on a project. Networks for engineering/
feasibility of staying within the early-late procurement work are prepared on the basis of past
band is examined by going back through the experience, specific project scope, d.rawings/speci-
RPM program as often as necessary for a fications lists, engineering studies, licensing
satisfactory optimum. At this point, the requirements, and so on. Lead discipline engineers,
resource-constrained schedule is ready. design supervisors, coordinators, purchasing staff
and similar members of the project team are con-
sulted for supplementary information.
4. Initial Scheduling Phase
Construction personnel, with the same set of
Basically, in any scheduling operation, there data, prepares worksheets/sketches to reflect the
are two phases of work: Initial and Update. Since strategies thought out in the field and makes cer-
the procedure to carry out these phases involves a tain that the logic structure complies with the
cross-discipline work effort, a look at the Action cost/production control system run on jobsite com-
Responsibility Chart will facilitate an unaerstand- puter. P&S engineer, on receiving these sketches,
ing of it. composes CPM networks in standard format, and.
166
incorporates them in the master scheduling system
by connecting engineering/design/procurement
c> Should the start of structural steel
erection depend on the completion of all
activities with pertinent construction activities. underground piping work?
The initial networks are reviewed by engineering
and construction personnel. P&S incorporates all d) Must all electrical openings in the floor
comments into the networks, prepares input sheets, slab bespecified before starting the
gets activity cards keypunched, obtains initial detailed reinforcement design?
trial runs, eliminates mechanical errors and
verifies the output against approved networks.
Having thus converted the networks into a prelimi- The project team studies the critique, and
nary error-free schedule, a schedule analyst from problem reconciliations are made in project meet-
P&S prepares a critique to be analyzed by cogni- ings. Following final agreements on disputable
zant team members. The critique covers the follow- logic, firm activity durations and estimated re-
ing: sources are applied to the networks. In case of
engineering/procurement activities, only manpower
. Each path having slack equal to or less than resources, as man-days for up to five skill cate-
the difference between the calculated project gories employed for each activity, are applied.
completion date and the desired one is sepa- For construction, up to ten resources as applicable
rated out. for quantities, craft skills and dollars are per-
mitted. The input data are developed by Estimat-
. Activities forming such a path are listed ing/Cost Engineering in conformity with standard
chronologically with their early finish dates. code numbering system for appropriate detail or
sumunaxy level of the work breakdown structure.
. Questionable or problematic aspects in the
logic of each path are pointed out. For After all iterations are completed through
example: RPM and RPSM, with the added information on re-
source availability, the schedule is finalized
a) Is it necessary to complete electrical and distributed. A critique is again prepared by
design before beginning miscellaneous the CPM analyst, in which current and potential
steel design? problem areas are highlighted. It is made a part
of project management reports, which also include
b) Must the pumps be purchased before, simplified overview networks, physical accomplish-
after or together with motors? ment graphs and similar information.

INTEGRATED SCHEDULING SYSTEM

ACTION RESPONSIBILITY CHART

RESPONSIBLE FUNCTIONAL UNIT PRIMARY ACTIVITY

1. PLANNING & SCHEDULING a. Preparation of CPM networks

b. Computerization - input sheets. debugging printxwts
c. Analvw of computer o”tp”t at all stages
d. Distribution of reports

2. COORDINATING a. Liaison among project team members, meetings, timely informatlon flow
b. Assistance to Proiect Manager as required

3. PROJECT TEAM a. Discipline Engineers provide, review & validate work plan reflected in the networks; estimate & assign manpower
to actlwties
b. Purchasing provides input on lead tfmes for material &equipment procurement
c. Estimating works with work charge & accounts codes, furnlshes information on quantities &resources requirements
d. Proiect Engineer monwrs mtegrity of engmeering & design related data and major project acwities

4. COST ENGINEERING a. Home offlce englneer attends to engineering & related services and manpower distribution to pertfnent activities
b. Field engmeer provideslmon~tors estimates. forecasts, cash flow requirements for construction planning. scheduling,
producnon control as related to networks.

5. CONSTRUCTION a. Field office works out detalled plan. sends logic sketches to home office, provades all necessary input data;
constders alternatwe strategnes as requwed & updates all data regularly pertainmg to resources allocation, etc.
b. Home office reviews logic, ensures proper engineerlnglprocurement restraints, supports all computerization & related
efforts, maintams correlation of scheduling system wth constructnon management system

6. PROJECT MANAGER a. Takes overall responsibility of project planning, scheduling. cost controlling
b. Directs all key decislonmaking procesres for the proiect management

NOTE.
There are other functions. as well as activities involved in the total project effort, but only those associated with the
schedulmg system are highlighted here.

167
5. Update Cycle The flow of information that takes place
during initial and update phases among all people
Periodic review of the progress status, re- involved in the scheduling process is identified
assessment of operating plan logic and resource in the diagram titled Data Flow, Integrated Sche-
requirements become a part of the update cycle. duling System.
A monthly update of status and a quarterly revi-
sion of logic as required are considered adequate.
6. System Outputs
Each discipline lead engineer/design super-
visor marks up the network sheet relating to his Outputs are available from the system on
work, or the print-out sort, with progress. The optional basis, as various sorts by dates, by
coordinator gathers information on activities resources, by responsibility codes, etc. These
which include purchasing, expediting, client's codes have a S-digit alphanumeric field, which
reviews and authorizations and. passes on to P&S. can be used in multiple ways to identify the
Networks are modified to reflect work completion, party responsible to carry out each activity, to
additions/deletions, and logic changes. When signify the nature of work (design drawings, engi-
construction starts, field input is received and neering specifications, concrete construction), to
incorporated. Latest remaining activity durations, denote the area/location/system and so on. Activ-
quantities, manpower forecasts and feasible man- ities are grouped together for every j-digit code
ning/loading levels are collected and used in applied to the networks. Additional fields of up
RPM-RPSMprograms. to lo-14 digits are provided for using further
sortable coding of job number, work charge/cost
Preliminary update runs are analyzed and dis- accounting numbers, specification/inquiry/purchase
cussed in project meetings for data correction and order/drawing numbers, etc., as applicable.
schedule acceptance.
The most typically used outputs are sampled
All initial and update phase activities are here for an illustrative purpose:
carried out under the general direction of the
Project Manager, or Assistant Project Manager a) I-J List, is a standard activity listing by
designated for the purpose. The maintenance of the predecessor (I) events, arranged in a
networks, resolution of problems in running the numerical order. A J-I list of activities
logic through the system, issuance of analyses and arranged in a numerical order of successor
schedule reports are in the hands of a Lead Plan- (J) events is also available as a common
ning Engineer. He gets the help and support of option.
the Project Coordinator for timely collection of
data, and for all follow-up information on criti- b) Critical Activity List, is an essential re-
cal items. port for critical analysis of activities

DATA FLOW
INTEGRATEDSCHEDULING SYSTEM
/ \
PROJ. ENGINEER

- LEAD DISCIPL. ENGRS.

4
CPM LOGIC
FOR REVIEW. DESIGN SUPERVISORS
SCHEDULE
CRITIOUES
ACCOUNT CODES.
WORKBREAKDOWN ESTIMATES/F0 RECASTS.
&PLAN INFO. RESOURCES SI I ACCOUNT/ COMMENTS,

C”Nbl”“Lll”N
RECONCILIATION
CONSOLIDATED ON PROBLEMS
CONSTR. LOGIC,
RESOURCES ASSTT. PROJ. MGR.
HOME OFFICE
_^.. ^v....^-.^.. PLNN’G & SCHED. ENGR.
ENGINEER a INPUT DATA SHEETS,
FINALIZED I COORDINATOR I ACTIVITY CARDS, ETC.
NETWORKS WITH
TIE-INS
CONSTR. LOGIC GUIOANCE & INFO. ON
SKETCHES, CONST. CONTROL
UPDATE lNF0 SYSTEM, REVIEWS a-
2: 1,
map / \
Y,o z
a?% COMPUTER PRINTOUTS COMPUTER
( FIELD PLANNER , , y;z _ STAFF
%ZZ

Y L /
\ A
n, IT.nF.Cl-nCF
--, -. “--. -
t MANAGEMENT KEY DECISIONS
& OTHER DIRECTIVES I & AUTHORIZATION
t
PROJECT. CORPORATE, 7 - --
I CLIENT I

168
 I.1 LIST
(EXPANDED FOR RESOURCES)

PREO WCC RES ACTIYITY OURA- EARLY CRITICAL TOTAL EARLY CRITICAL SCWCMP OATE
WENT EVENT (0) * . . . * . * . . . . OESCR,fs,,ON . * . . . . * . . . . . f TION START STAN1 SLACK FINISH FINISH DATE TYPE

2011 2012 MP CHECK FHB BASE SLAB MM EL 28.5 316106 2.1 2, OCT 74 0, MAR 76 71.0 11 NO” 74 22 MAR 76
6407 lOE406 4

ZOll 2029 64P INT COMM FHB BASE SLAB MAS EL 28.5 316105 2.1 27 OCT 74 07 MAR 75 71.0 11 NOV 74 22 MAR 76

2015

2016
2016

201s
“EP

44P
CLT

OC4
RVW

RVN
FHB

FHB
BASE

BASE
SLAB

SLAB
M EL 28.5

FIN&SECT EL28.S
315105

316105
3.0

2.1
07

07
MAR

MAR
4435
76
76 06

12 APR
19
APR 76

75
56.6

57.4
28

22
MAR

MAR
76

76
,
2,

2,
APR

Ay+’
75
, :.J/
2017 2010 64s REV OSN FHB BASE SLAB EL 28.5 316105

548 INC COM+.SO FHB BASE SLABMASEL 28.5 316105

2022 MP F’REL FHB BASE SLAB PLN&SECT EL 28.5 316105

2029 2015 640 INC COM FHB BASE SLAB MAS EL 28.5

A
2113 54R INC COMtSO FHB
_..---/-‘- / / ~MD.r~OAOTYORS.r,

RESOURCE NAME

WEEKS

16143 16144 84K “NO DWGS NUCLEAR PPG PENETRATlOFJ 3225 6.6 29 NO” 76 02 OEC 76 0.5 14 JAN 77 17 JAN 77

16144 15146 53K R”W VDR DWG NUCLEAR PIPE PENETR 3225 4.3 14 JAN 77 l,.lAN,, 06 13 FE6 77 16 FE6 77

16146 15069 AGS SPEC NUCLEAR PIPE PEN AGS INTEF CHECK 3226 3.1 13 FE6 77 16 FE6 7, 0.5 07 M’lkR 77 10 MAR 7,

,614, 15174 8ml FABRICATE NUCLEAR PPG PENETRATION 3225 62.0 07 MAR 7, 10 MAR 7, 0.6 06 MAR 78 ON MAR 78

15069 15147 0 0.0 0, MAR 77 ,OMAR,, 06 07 MAR 77 10 MAR 77

16174 16176 83N DELIVER NUCLEAR PPG PENETRATION 3225 21 06 MAR 78 09 MAR 78 0.6 21 MAR p. 24 MAR 78

15176 93626 D 0.0

93525 93526 LE.40 TIME 4.3 I::

93526 93627 CRM INSTALL MECHANICAL PENETR RB “NIT 1 12 43, 21.7

93558 93669 CRM INSTALL RB REClRCULATlON PPG UN 1 9.1 52.0

93527 93568 D

93559 >

141%
L SORT KEY
,415,

14614

21163

falling behind schedule. Gn the first out- slack. If the difference between calculated
put sheet, the difference between the cal- and desired end dates, for instance, is 4.3
culated or natural completion date of the weeks (i.e., one month), it meansthat the
project and the desired deadline for the pro- logic requires the total project duration to
ject is pointed out. This system does not extend 4.3 weeks beyond the prescribed aead-
allow what is commonlyknown as the negative line. According to general CPMpractice, this
slack, since it is not logically possible to will ordinarily be the maximumnegative slack,
apply resources to activities which are lag- equal to (-4.3) weeks. Consequently, the most
ging behind. (In such a case, the late finish critical path, indicated by (0.0) slack is to
date is earlier than the early finish date!) be considered as having the highest negative
The critical activity list contains groups of slack (-4.3). The subsequently listed paths
activities in a numerical slack sequence, would read 0.1, 0.2, etc., which would really
starting with the most trim path, or zero imply (-4.2), (-4.1) and decreasing so on.
Thus all paths having slacks that read up to lative quantities required to perform all
4.3 weeks are critical. activities that consumeman-days for the
particular skill, by their early start dates,
The particular activity that may have contri- and also by the critical or late finish dates.
buted to the delay on each of the critical It also calculates the average figures and
paths can be traced. Raving identified the tabulates them. The results are computer-
existence, origin and extent of such a path, drawn to give the S-curve plot, for each re-
the corrective action is decided upon. The source/skill category, of cumulative man-days
schedule analyst points out the need for such versus months over the complete project
corrective action in his critiques, which may period. The data furnished by this plot form
involve changing of crew size, adding an the early start-late finish envelope. It is
extra shift, devising new logic for the work, analyzed in the context of managementobject-
or shifting of priorities. The critical ives and basic resource limitations to deter-
activity list is printed out in order of slack mine the month-by-month manning or resource
value as a primary sort, and is subsorted application to be ultimately loaded on each
secondarily in a chronological order. activity.

Typical S-Curves, have a unique significance The S-curves are finalized through the RPSM
as they have to be used even during the data program by additional inputs relating to each
processing stages of RPMand RPSM. At the resource category and identifying the avail-
end. of CPMruns, the timing framework of the ability for different lengths of time within
project is clearly established and given out the total project duration. For instance, we
in a tabulated form. The RPMprogram is have to specify that from the initial time tl
meant to set up resource limits imposed by up to a time t2, "X" number of mechanical
the manning requirements of the network. It designers, or carpenters, will be available;
runs through the CPMfile, sumsup the re- or that "XV cubic yards of concrete will be
sources of each category, e.g., man-days for placed. And, for period t2-t , and period
a certain skill, for every month remaining on t3 to the end of the project 2he numbersmay
the project. The system works out the cumu- be "Y" and "Z" respectively. A coding scheme

TYPICAL S-CURVES
(FOR EACH SKILL/RESOURCE CATEGORY)

EARLY START

TIME BOUNDARIES

TOTAL MANPOWER
,-REQUIRED

RESOURCE ENVELOPE

PROJECT DURATION (YEARS)

170
to account for all such variables is devised d) DO List, available in a tabulated form or in
and used as applicable. For simplicity, we a bar chart format, is really the immediately
may restrict ourselves to the manning problem. useful operating tool for the supervisors.
The pattern of manpower allocation is then DO lists are prepared for any span of time
clearly visible on the S-curve. Initially, requested (60 days, 100 days, 365 days) or,
it gives the band of feasible distribution if not specified, for the entire project dura-
over time on early and late basis. Any tion. A go-day or quarter-year period is
planned manning curve lying outside of this recommended as being most practical. DO lists
so-called resource envelope presents non- provide easy guidance to the job engineer,
feasibility. the field supervisor, the project manager or
the client, in the form of a list of all items
expected of him and his work force in the next
When the planned manning lies above the Early time interval. These items can be further
Start distribution curve,.too many nonproduct- grouped, as desired, chronologically, by work
ive idle men will be engaged on the project category, such as: drawings to be signed out,
notenough work being Possible according to the specifications to be finalized, and vendor
network-logic. SimXrly, if the plan-curve drawings to be reviewed; or concrete/form-
lies below the Late Finish curve, the manpower work/rebar work by pour. They also show the
loading is lower than can be permitted for schedule for each task to meet management
timely completion of the project. The pro- objectives within the resources allocated to
ject, being undermanned, will have a delayed him. As this schedule is based on the logic
completion. The slope of the S-curve at any governed by all prerequisites as well as on
point indicates how many men are required at the estimated/forecasted/plannedresources,
that time. The flat slope near the project the supervisor can carry on his work with a
beginning indicates a slow buildup in work desirable degree of confidence. Since the DO
force. The labor applied afterwards increases lists reflect the effect of all preceding
rapidly to a peak, stays at that level and at activities on the current schedule, the
the end phases out quickly. If the buildup supervisor can evaluate their impact for
is delayed too long, the work force has to changes, if any, to be made to his immediate
rise more abruptly as seen on the Late Finish work plan, on a continuing basis.
curve than what the Early Start would allow
for staying on the project schedule. The The supervisor (design or construction) may
Early-Late curves, being drawn for each skill, further detail his work plan to meet the daily
indicate how critical it is in a time-and- requirements as outlined in the DO list. Such
availability sense, rather than in a work- detailed requirements would cover determina-
content sense. tion of crew sizes, their work locations,
needs for tools and equipment, and the like.
The horizontal distance between the two curves Such details could not and should not be in-
at a given manpower loading level indicates cluded in the master schedule. The extent of
the Time Slack for the skill during the period details built into the networks should depend
spanned. A relatively low demand is suggested on the practicality of monitoring actual per-
if this distance is large. If the distance, formance on them. Resides, a certain amount
however, is small, then the demand is high, of freedom and flexibility must be available
the skill is critical, probably demanding a to the supervisor to best let him employ his
high degree of efficiency and coordination to personal know-how, style and originality in
utilize it effectively. The vertical dis- maneuvering the ways and means of getting
tance, in the same way, between the curves things done. The master project schedule
signifies the Manning-Slack, or the possible should be used only to channelize rather than
variation in the loading level at that partic- regiment the abilities of professionals.
ular time of evaluation. The skill category
is critical for small manning slack as the Among other outputs that merit some mention
availability is close to demand. But if the are the tabulations referring to Skill-Usage (or
manning slack is large, shortages can probably resources actually applied) versus Availability
be absorbed. S-curves are easily established (or resources planned for use). For each skill,
for other than human resources, so as to re- or resource, a month-by-month statement for the
present linear feet of pipe to be installed, remainder of the project is available to also com-
tons of structural steel to be erected, or pare the usage requirement against the real pro-
dollars to be expended. jected application of resources. For months where
the usage exceeds availability, the usage entry is
The planned availability may be influenced by flagged. Preceded symbols (like * an asterisk, or
physical or policy limitations. Requisite + a plus sign) are employed as indicators to denote
skilled manpower may just not be available, various situations, for example, when a particular
or it might turn out to be that the capacity skill/resource applied to a particular activity in
of the concrete batch plant is limited. Per- a DO list is of secondary and not primary nature.
haps the resources could be provided, but While attempting to schedule an activity, if a
have to be spread over several projects. nonfeasible resource situation is encountered,
Management may not want to increase the avail- the system assigns the early date to it, flags it,
ability by authorizing scheduled overtime be- and continues to schedule the next activities.
cause of premium pay involved, or perhaps the Whethersuch local conditions have a global impli-
client may prefer a lower cash flow provision cation or not is determined while analyzing the
at the time. sorts and preparing critiques for review.
71
 00
(PRINTED FORMATI

44P ac-4 REW-KEY ONE LINE WlRlNGDlAG 5001w

446 cc4 RVW CDNTMT “Es%. LOCKS+ HATCH 5501

44P oc4 RVW SBLC SYS FD 216150

44P oc4 APP MAIN STEAlVl FL DIAG

44P OC4 APP EXTR STEAM FL DlAG

,567, 15673 448 GC4 RVW REACTOR FEED PUMP 3.0 30 APR 74 30 APR 74

15931 15949 448 GC4 RVW M.S. REL VAS. 3.0

35264 36266 f-l-

/
8615 8616 _ --A-
21x84 2,305 j ,---
.,‘I’
21104 *1106 _A’
I’
,-’
_-’ I--- CPM GENERATEO
/-

L --RESD”RCE CONSTRAINED
FlNlSH DATE
I- CRlTlCAL
MlN”SSCHEDULED
FINISH ,IN WEEKS1

-RESOURCE CONSTRAINED
START DATE

-CPM GENERATED

TrPlCAL
DOLIST
PAGE 1
ENASCO SERVICES 64700 PERT RUN THURSDA” 8/15/74. TIME IS 12 00 A.M
LISTING OF AC*l”lTlES BY SKILL

DUMMY ACTlVlTlES ARE OMITTED FROM LISTING.

JIJL AUG SEP OCT NOV OEC JAN

PRED WCC RES ACTIVITY
EVENT EVENT (0).. . . . . . . . . . OESCR,PT,ON
. . . . . *. . . . . . S30122011200123012201120012
17418418529630630741851952 SLK
SCH/CMP
OATE
PR,M,
SEC

.?600226003 448 GC4 RVW STAND.BY DG 6061 XXXXXX 62

21005 21006 44P Qc-4 REVW.KEY ONE LINEWIRINGDIAG 500100 sxxxx 10

13034 13950 44P REV RADWASTE BLOG GA’S sxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx 101 29 JAN 75 * *
L
20575 20576 44P OC4 RVW RPCS SYS ID 216130 .XXXX .......................................... 63

21104 21105 44P OC-4 REVW-CONCEPT.PRTECT.RELAY DIAG 500110 xxxx ............ 6

26719 26707 448 OC4 RVW STUCTURAL STL 5521 xxxxxx 73

.................................

21224 21225 44P NC-4 REVW-STDBY DIESEL&EN LOGIC DIAG 501120 xxxx ........... ..: ............ 57

10223 10269 448 (l.c.4 RVW RADW-HORI & VERT CENT PMP 2041 xxxxxx .......................... 4,

13431 13432 44P GC4 APP DOMEST WIR SYS FL DIAG 214155 .... xxxx 73
.......................

13306 13307 44P GC4 APP AIR EVACUATlON FL DIAG 214122 XXXX .................... 72

29303 26307 448 GC4 RVW RB ELEVATOR 5504 XXXXXX ............. 36

21024 21025 44P OC 4 REVW-460V. MCC ONE-LINE DIAGRAM 500150 xxxx .......... 6

15901 15903 446 OC4 APRVL ISOLATION VLVS 3162 xxxx .... 43
...................
I
21074 21075 44P GC 4 REVW-INST. VITAL A/C l.LINE DlA 501100 ‘r ..-- ~ xxxx 0

21124 21125 44P GC 4 REVW-AUX. D/C ONE LINE WIRE DIAG 60,105 xxxx 0

31205 31240 44P OC-4 APPRV SCHEM&LOGIC DIAGTURBGEN 134110 xxxx 46 23 JAN 75

26239 29243 MB GC4 RVW POOL LINERS 5505 06 FEB 75

35306 36309 44P GC 4 APPR HVAC CONTROL DIAG. SH 1 616150

I j/I /
I
I--

SCHEDULE SLACK CRITICAL FIN,SH

“S” PREVIOUSLY SCHEDULED MlNUS “X-5 = SCHEDULE CRITICAL MlNUS -tl DATE FOR BARS
STARTED EARLY START (l/2 WEEKS) PERIOD (l/2 WEEKS, SCHEDULE FINISH EXTENDING BEYOND
(DOTS = l/2 WEEKS) TABLE BOUNDARY

172
 TYPICAL RBSOURCE ENVRLOPS AND AVERAGE MAN LOADIK:

LRTE FINISH
ERRLY START
RVG HEN X LOO

RES -- 1-11135 RESOURCE RNQLYSIS

EFF 05/31/74

TYPICAL COMPONENTS OF A LOGIC DIAGRAM

SUCCEEDING EVENT]

PREL. DESIGN DETAIL DESIGN TO CONSTRUCTION

---+--,
13.0
VENDOR INFORMATION 1
------------J
1 ACTIVITY
/
DURATION IN WEEKS \
1 MONTH = 4.3 WEEKS 1

13 004
SHEET NUMBER2
DEFINITIONS L SECIIJENTIAL NUMBER

ACTIVITY - A TIME CONSUMING ELEMENT IN PROGRAM AND REPRESENTED ON NETWORK BY ARROW. TIME FLOWS
FROM TAIL T O HEAD: NOT SCALED. EVENT SUCCEEDING ONE OR MORE ACTIVITIES NOTCONSIDERED
COMPLETE UNTIL ALL PROCEEDING ACTIVITIES ON WHICH I T IS DEPENDENT ARE COMPLETED,
EVENT - MEANINGFUL, SPECIFIED ACCOMPLISHMENT IN THE PROGRAM. RECOGNIZABLE AS A PARTICULAR
HAPPENING, A POINT IN TIME, REPRESENTED BY CIRCLE. CONSUMES NO RESOURCES, NO TIME.
PRECEDING EVENT - AN EVENTTHATMUST BE COMPLETED BEFORE THE FOLLOWING ACTIVITY CAN BE STARTED, ASTARTING
EVENT FOR ONE OR MORE ACTIVITIES.
SUCCEEDING EVENT - AN EVENT THAT CAN NOT BE ACCOMPLISHED UNTIL THE PRECEDING ACTIVITY IS COMPLETED, AN
ENDING EVENT FOR ONE OR MORE ACTIVITIES.
RESTRAINT - A PREREDUISITE ACTIVITY WHICH HAS T O BE COMPLETED BEFORE A GIVEN ACTIVITY CAN BE STARTED.
INTERFACE - DEPENDENCY OR CONNECTING ACTIVITY BETWEEN TWO SETS OF ACTIVITIES, OFTEN A “DUMMY”
ACTIVITY CONSUMING NO TIME, NO RESOURCES.
CRITICAL PATH - THAT STRING OF INTERDEPENDENT ACTIVITIES, WHICH TAKES THE MOST TIME FROM STARTOF THE
FIRST THROUGH COMPLETION OF THE LAST.

173
START ILLUSTRATIVE PROJECT NETWORK
l-1-75

\
\ END
10-l-76

KEY
A THRU J AREACTIVITIES”
01 THRU 13 ARE “EVENTS”
I
(
‘,i’ -,;~ElyF:“l”“
\
---IS A “CRITICAL PATH”

POSSIBLE PATHS
A-B.C = 17 MONTHS
A-F-C-H = 21 MONTHS
0.E-F-G.H = 20 MONTHS
D-E-I-J = 19 MONTHS

J9I.L
THIS CRITICAL PATH HAS NO
“SLACK”. DELAYS UP TO 4,1,2
MONTHS ON OTHER PATHS SUCH
AS A-B-C, D-E-F-G-H 8 D-E-I-J
012345678 9 10 11 12 13 14 15 16 17 18 19 20 21 WILL NOT AFFECT THE END
DATE.
I I I I I I I I I I I I I I I I I I I I , I
MONTHS

NETWORK SCHEDULING
AND
ALLOCATION/LEVELLING

o I I .
..:A,, - ’ ’ ‘O ”

B. RESOURCE LEVELLING
7. Discussion Probably the most significant trend has been
toward expanded flexibility in scheduling compu-
The integrated scheduling system we have re- tations and in report formats. The new Burroughs
viewed so far is essentially no different from PROMIS program and the IBM PMS/360 appear to con-
over a hundred others as far as the basics are tain high degrees of flexibility, including report
concerned, Most systems, developed by researchers generators that permit the user to design the con-
in the field of deterministic (as against prob- tent and layout of his own reports. Such programs
abilistic) models for network scheduling and re- are expensive to develop, and they indicate a
source allocation, have faced shortcomings in growing demand for tailor-made or customized net-
attacking the time/cost tradeoff problem, such as: work processing in a wide variety of applications.
optimality is not guaranteed, or restrictive The integrated scheduling system, as an approach
assumptions are made about activity time-cost re- that is, can thus result from an expanded or
lationship, or resource requirements and availabil- modified version of an existing standard scheduler
ity are not taken into account. Consequently, they Program, suitably enhanced to fulfill specific
do not provide a satisfactory basis for project needs of the nature of business, and the demands
management. It is in this respect that the system for decision tools to serve the top management.
meets the objective, namely, to serve as an effi- Such enhancement projects, depending upon the
cient management tool for project control. urgency for developing the sophisticated tools,
and upon the talent and resources available within
the user's organization, can be undertaken by the
"Many models consider only one type of re- systems development group with the help of in-house
source (usually manpower, or money). Some work management consulting group, if any. Otherwise,
with stated resource constraints, some with un- outside consultants with related expertise could
limited resources. A few models allow variable be retained for help. The concepts outlined here
resource availability, while alnost all of them lend themselves readily to multiple projects con-
assume that activity resource requirements are trol. It must be emphasized, however, that the
constant over time. Few models permit interrup- resource allocation feature, for successful imple-
tion of activities or the possibility of perform- mentation, necessitates wholehearted cooperation
ing activities at several resource levels or with from every participating function of the organiza-
different methods. Finally, most of the research tion. Nothing serves its cause better than an all-
deals with single project situations. In spite out support from top management.
of this diversity, which merely reflects the com-
plexity of reality, the models can be classified In the conclusion it may be stated that inte-
in a convenient way according to the criterion grated or not, a control system, no matter how
used for schedule evaluation, or objective func- sophisticated, is no substitute for professional
tion. 'Minimum duration models', in simple or ingenuity, managerial ability and personal intuition
generalized form, attempt to obtain a schedule
for which the shortest feasible project duration
is achieved. 'Resource levelling models' aim at
minimizing variations in the level of resources
used. Finally, 'time/cost tradeoff models' have
the minimization of total project cost as their
goal."

The quotation is from an e:ctensive study made

in November 1973 by Amar Alain I:haker on optimality
in project networks. Among his concluding remarks,
he points out that as a rule tb? models he examined
fell far short of reality and tnat even if simplify-
ing assumptions are made, the problems of resource Bibliography
allocation are extremely difficult to solve. The
scope of this paper does not allow us to go into 1. Alpert, R. and Orkand, D. S., A_Time-Resource
mathematical algorithms of the network problem Trade-Off Model for Aiding Management Deci-
solutions. A brief bibliography is appended at sions, Technical Paper No. 12, Operations Re-
the end for those interested in further academic search, Inc., Silver Spring, Maryland, 1962.
pursuit in this area. For our purpose here, how-
ever, four sets of diagrams are shown to exemplify 2. Bigelow, C. G., Bibliography on Project Plan-
the resource allocation, resource levelling, opti- ning and Control by Network Analysis: 1959-
mization of manpower loading and smoothing out of 1961, Operations Research, 10, 5 (September-
resource requirements by interactive computer- Eber), 1962, pp. 729-732.
aided process.
3. Chaker, A. A., Optimal Scheduling and Resource
Moder and Phillips, in their 1970 textbook on Allocation in Project Networks, Department of
project management with CPM and PERT, have also Civil Engineering, University of Illinois at
noted that certain advanced features in CPM techno- Urbana-Champaign,Urbana, Illinois, 1973.
logy have increased in popularity compared to pro-
grams with PERT probability computations; and that
resource allocation procedures, although still pre- 4. Davis, E. W. and Heidorn, G. E., An Algorithm
senting extremely difficult prcblems, are continu- for Optimal Project Scheduling Under Multiple
ing to be developed and made available on a pro- Resource Constraints, ManagementScience, Vol.
prietary basis. 17, No. 12, August 1971, pp. B%03-B816.
175
5. Goldberg, 8. R., An Algorithm for the Sequen- a. Martino, R. L., Resource Allocation and Sche-
tial Solution of Schedule Networks, Operations dulin@;, Project Management and Control Series,
Research, l-2, 3 (May-June), 1964, pp. 499-502. Vol. III, American Management Association,
New York, 1965.
6. Herroelen, W. S., Resource-Constrained Pro-
ject Scheduling - the State of the Art, 9. Moder, J. J. and. Phillips, C. R., Project
Operational Research Quarterly, Vol. 23, Management with CPM and PERT, Second Edition,
~0. 3, September 1972. Van Nostrand Reinhold Company, New York, 1970.

7. Lerda-Olberg, S. Bibliography on Network-Based 10. Weist, J. D. and Levy, F. K., A Management

Project Planning and Control Techniques: 1962- Guide to PERT/CPM, Prentice-Hall, Inc.,
l&$ Operations Research, 14,5 (September- Englewood Cliffs, New Jersey, 1969.
October), 1966, pp. 925-931.

176