NETWORK TECHNIQUES FOR PROJECT

MANAGEMENT

STRUCTURE
8.0 Objective
8.1 Introduction
8.2 PERT/CPM : Background and Development
8.3 Development of Project Network
8.4 Time Analyasis
8.4.1 Time Estimation
8.4.2 Determination of Critical Path
8.4.2.1 Calculate the Earliest Occurrence Time (EOT) for each Event
8.4.2.2 Calculate the Latest Occurrence Time (LOT) for each Event
8.4.2.3 Calculate the Slack for each Event
8.4.2.4 Obtain the Critical and Slack Paths
8.4.2.5 Calculate the Activity Floats
8.0 OBJECTIVE
After reading this lesson, you should be able to
a) Explain the role of network techniques in project management.
b) Make use of the network techniques for planning scheduling and controlling
the different activities of the project.

1
8.1 INTRODUCTION
Projects are successful if they are completed on time, within budget, and to performance
requirements. Management of any project involves planning, coordination and control
of a number of interrelated activities with limited resources, namely men, machines,
money and time. Furthermore, it becomes necessary to incorporate any change from
the initial plan as they occur, and immediately know the effects of the change. Therefore
the managers are compelled to look for and depend on a dynamic planning and schedule
system which will not only produce the best possible initial plan and schedule, but will
also sufficiently dynamic to react instantaneously to changed in the original plan and
schedule. The question of such a dynamic system/ technique led to the development of
network analysis. It provides a framework which :
➢ defines the job to be done,
➢ integrates them in a logical time sequence and finally,
➢ affords a system of dynamic control over the progress of the plan.
Network analysis is a generic name for a number of associated project planning and
control procedures that are all based on the concept of network. PERT, an acronym for
Program Evaluation and Review Technique and CPM, an acronym for Critical Path
Method are the two widely used techniques of project management that were developed,
independently and simultaneously, during the 1950s. The network analysis underlying
PERT and CPM helps to support the three phases of effective project management.
Planning
➢ identify the distinct activities,
➢ determine their durations and interdependencies,
➢ construct a network diagram,
➢ determine minimum overall project duration (using the network diagram),
and
➢ identify the tasks critical (i.e. essential) to this minimum duration.

2
Scheduling
➢ construct schedule (‘time chart’),
➢ schedule contains start and finish times for each activity, and
➢ evaluate cost-time trade-offs (evaluate effects of putting extra money, people
or machines in a particular task in order to shorten project duration).
Controlling
➢ monitor/control project by use of network diagram,
➢ follow progress of the various activities ; and
➢ make adjustment where appropriate.
8.2 PERT/CPM : BACKGROUND and DEVELOPMENT
PERT and CPM- both techniques use similar network models and methods are have the
same general purpose. They were developed during the late 1950s. PERT was originally
developed by the U S Navy’s Special Product Office in cooperation with the consulting
firm of Booz, Allen and Hamilton. It was developed as a network flow chart to facilitate
the planning and scheduling of the Polaris Fleet Ballistic Missile Project, a massive
project with about 250 contractors and about 9000 sub contractors and its application
is credited with saving two years from the original of five years required to complete
the project. Designed to handle risk and uncertainty, PERT is eminently suitable for
research and development and programmes, aerospace projects, and other projects
involving new technology. In such projects the time required for completing various
jobs or activities can be highly variable. Hence the orientation of PERT is ‘probabilistic’.
CPM, is akin to PERT. It was developed (Independently) in 1956-57 by the Du Pont
Company in the US to solve scheduling problems in industrial settings. CPM is
primarily concerned with the trade-off between cost and time. It has been applied mostly
to projects that employ fairly stable technology and are relatively risk free. Hence its
orientation is ‘deterministic’.
As both PERT and CPM approaches to Project Management use similar network models
and methods, the term PERT and CPM are sometimes used interchangeably or
collectively as PERT-CPM methods. The differences between those tools come from

3
how they treat the activity time. PERT treats activity time as a random variable whereas
CPM requires a single deterministic time value for each activity. Another difference is
that PERT focuses exclusively on the time variable whereas CPM includes the analysis
of the time/Cost trade-off.
The PERT/CPM is capable of giving answers to the following questions to the project
manager :
➢ when will the project be finished ?
➢ when is each individual part of the scheduled to start and finish ?
➢ of the numerous jobs in the project, which one must be timed to avoid being
late ?
➢ is it possible to shift resources to critical jobs of the project from other
non-critical jobs of the project without affecting the overall completion
time of the project ?
➢ among all the jobs in the project, where should management concentrate its
efforts at one time ?
Methodologically, PERT/CPM were developed from traditional GANTT Charts used
for scheduling and reviewing the progress of activities. Developed by Harry Gantt in
1916, these charts give a time line for each activity. They are used for planning,
scheduling and then recording progress against these schedules.
Basically there are two basic types of Gantt Charts : Load Charts and Project Planning
Charts.
Load Charts : This type of chart is useful for manufacturing projects during peak or
heavy load periods. The format of the Gantt Load Chart is very similar to the Gantt
Project Planning Chart, but, Load Chart, uses time as well as departments, machines or
employees that have been scheduled.
Project Planning Chart
It addresses the time of individual work elements giving a time line for each activity of
a project. This type of chart is the predecessor of the PERT. As it can be seen in the
Figure, it is really easy to understand the graph, but in developing it you need to take

4
into consideration certain precedence relationship between the different activities of
the project. On the chart, everyone is able to see when each activity start and finishes
but there is no possibility to determine when each activity may start or if we can start a
particular activity before finishing the immediate predecessor activity. Therefore, we
need somehow know the precedence relationship between activities. This is the main
reason for using the PERT/CPM tools instead of using exclusively Gantt Charts.
Widely diverse kind of projects can be analyzed by the techniques of PERT/CPM. In
fact they are suitable for any situation where :
(a) the project consists of well-defined collection of activities or tasks.
(b) the activities can be started and terminated independently of each other,
even if the resources employed on the various activities are not independent.
(c) the activities are ordered so that they can be performed in a technological
sequence. Thus precedence relationships exist which preclude the start of
certain activities until other are completed. For instance, road levelling
cannot start unless the roadbed is laid.
We now proceed to discuss the techniques to provide answers to the types of questions
stated earlier. The initial step in each of these is to portray the given project graphically
by means of network, which provide the basic tool for analysis.
8.3 DEVELOPMENT OF PROJECT NETWORK
Basic to network analysis is the networks diagram. Both the methods of PERT and
CPM graphic representation of a project that it is called “Project Network” or “Project
Diagram” or “CPM Diagram”, and it is used to portray graphically the interrelationships
of the elements of a project and to show the order in which the activities must be
performed. A simple network chart for a ‘Seminar Planning Project’ is shown in Figure
8.1 as an example.

5
 Figure 8.1 Project Network

In order to represent a project network, two basic elements are used :

A circle called “node”, represents an event. An event describes a checkpoint. It does

not symbolize the performance of work, bit it represents the point in time in which the
event is accomplished.

An arrow, called “arc”, represents an activity-a recognizable part of the project

involving mental or physical work and requiring time and resources for its completion.
The network will try to reflect all the relationships between the activities.
Since activities are the basic building blocks of a network diagram, it is necessary to
enumerate all the activities of the project. For this purpose, it is helpful to break the
project in several steps. The number of steps, of course, would depend on the magnitude
and complexity of the project. For industrial projects generally a two-step procedure
would suffice. In the first step, the major parts of the project are identified and in the
second step the activities of each major part are delineated. Activities should be so
defined that they are distinct, reasonably homogeneous tasks for which time and
resources requirement can be estimated.

6
Once the activities are enumerated it is necessary to define for each activity, the
activities, which precede it, the activities which follow it, and the activities which can
take place concurrently. Given this information, the network diagram, showing the
logical relationship between activities and events may be developed following either
the forward method or the backward method.
The forward method begins with the initial events, marking the beginning of the project,
and proceeds forward till the end event is reached. The backward method begins with
the end event and works backwards till the beginning event is reached.
Rules for constructing a project network :
Three simple rules govern the construction of a project network :
1) Each activity must be represented by only one directed arc or arrow.
2) No two activities can begin and end on the same two nodes circle. A situation
like the one shown in the following figure is not permissible.

3) There should be no loops in the network. A situation like the one shown in
the figure given below is not permissible.

Another element to represent a project network is a “dummy activity”.

Tasks that must be completed in sequence but that don’t require resources or completion
time are considered to have event dependency. These are represented by dotted lines
with arrows and are called dummy activities. To explain it, we will consider the following
example :

7
 ACTIVITY IMMEDIATE PREDECESSOR
A .........................
B .........................
C A, B
D B
The temptation is to represent these relationships as :

But then we have broken the second earlier mentioned. To show that activities A and B
precede C, whereas activity B precedes activity D, we use a dummy activity as shown in
the following figure.

A dummy activity may also be used to represent a constraint necessary to show the
proper relation ship between activities. As shown in the following figure, activities A
and B must be completed before activity C can be start, only activity B must be completed
before activity D can start.
To construct a project network, first of all, we need a list of activities, showing the
precedence relationships between the different activities involved is shown in Table
8.1 as an example.

8
 Table 8.1 Activities of the Project ‘Launching a New Product’
ACTIVITY NAME IMMEDIATE DURATION
PREDECESSOR (months)

A Market analysis ................... 1

B Product Design A 3

C Manufacturing study A 1

D Select best product design B, C 1

E Detailed marketing plans D 1

F Manufacturing process D 3

G Detailed project design D 3

H Test prototype G 1

I Finalize product design F, H 1,5

J Order components I 1

K Order production equipment I 3

L Install production equipment K 2

Figure 8.2 shows the network with the Earliest Start time, Earliest Finish time, Latest
Start time and Latest Finish time of the activities (these will be discussed later in
the lesson).
Because each activity must have a unique pair of starting and ending nodes, we must
use a dummy activity to draw the first four activities, as shown in the figure. Constructing
a project network is a trial-and-error process. It usually takes two or three attempts to
produce a neatly constructed network.

9
Figure 8.2 Network of the Project ‘Launching a New Product’

8.4 TIME ANALYSIS

Once the logic and details of the project network have been established, time estimates
must be assigned to each activity. With this representation we can determine the
minimum completion time for the project i.e. the critical path and the critical activities
and the slack or float of other activities, so that we can find the activity schedule i.e.
when each activity should start and when it may be completed. For discussing these
aspects of network analysis we will use the simple project shown in Figure 8.3.
8.4.1 Time Estimation
Assigning time to individual activities is essential in order to analyze a network.
Therefore an estimate must next be made how long each activity will take for its
completion. This is done by discussing with the people responsible for the completion
of the specific activities. In CPM analysis the activity time estimates are deterministic
i.e. time of various activities are known so we have only one time for each activity.
A distinguishing feature of PERT is its ability to deal with uncertainty in activity
completion times. For each activity, the model usually includes three times estimates:
Optimistic time (a) - generally the shortest time in which the activity can be completed
under ideal, favorable conditions. It is common practice to specify optimistic times to
be three standard deviations from the mean so that there is approximately a 1% chance
that the activity will be completed within the optimistic time.

10
Most likely time (m) - the completion time under the normal conditions, having the
highest probability. Note that this time is different from the expected time.
Figure 8.3 Network with Three Time Estimates (in weeks)

Pessimistic time (b) - the longest time under worst, externally unfavorable conditions,
which an activity might require. Three standard deviations from the mean is commonly
used for the pessimistic time.
PERT assumes a beta probability distribution for the time estimates. For a beta
distribution, the expected time for each activity can be approximated using the following
weighted average :
Expected time = (Optimistic + 4 x Most likely + Pessimistic) / 6
te=(a+4m+b)/6
This expected time might be displayed on the network diagram as shown in Figure
8.3
8.4.2 Determination of Critical Path
Once the network diagram with single time estimates has been developed, the following
computational procedure may be employed for determining the critical path/s, event
slacks, and activity floats.
8.4.2.1 Calculate the Earliest Occurrence Time (EOT) for each Event.
The EOT of an event refers to the time when the event can be completed at the earliest.
Looking at event we find that the since the paths leading to it, viz, (1-2-4) and (1-3-4)
take 15 weeks and 20 weeks, respectively, the EOT of event 4 is 20 weeks. In general

11
terms, the EOT of an event is the duration of the longest path (from the beginning event
whose EOT is set at 0) leading to that event. The EOTs of various events in our illustrative
project are shown in Figure 8.4. It may be noted that in Figure 8.4 a circle represents an
event. The upper half of the circle denotes the event number, the left quarter in the
lower half denotes the EOT, and the right quarter in the lower half denotes the Latest
Occurrence Time, (LOT) a term described the later.
The EOT of the end event obviously represents the minimum time required for
completing the project. To obtain the EOT of various events we start from the beginning
event and move forward towards the end event. This computational procedure is referred
to as the forward pass. In this computation we assume that each activity starts
immediately on the occurrence of the event preceding it. Hence the starting and finishing
time for various activities obtained from this computation are the Earliest Starting
Time (EST) and the Earliest Finishing Time (EFT).
The general formula for EOT is :
EOT (i) = Max [Eot(k) + d(k-i)]
where EOT (i) = earliest occurrence time of event i
EOT (k)=earliest occurrence time of event k (k precedes i and there may be
several k’s)
d (k-i) = duration of activity (k-i)
The maximisation shown is done considering all activities (k-i) leading to event node i
have been completed.
The formulae for EST and EFT are :
EST (i-j) = EOT (i)
EFT (i-j) = EST (i-j)+d(i-j)
where EST (i-j) = earliest starting time for activity (i-j)
EOT (i) = earliest occurrence time of event (i)
EFT (i-j) = earliest finishing time for activity (i-j)
d(i-j) = duration of activity (i-j)

12
8.4.2.2 Calculate the Latest Occurrence Time (LOT) for each Event.
The LOT for an event represents the latest allowable time by which the event can occur,
given the time that is allowed for the completion of the project (occurrence of end
event). Normally the time allowed for the completion of the project is set equal to the
EOT of the end event (In other words, the project is supposed to be completed at the
earliest possible time). This means that for the end event the LOT and EOT are set
equal. The LOT for various events is obtained by working backward for the end event.
This procedure is known as the backward pass. The LOT for event 4 in our illustrative
project, for example, is equal to the LOT for event 5, the end event, minus the duration
of the activity (4-5), which connects event 4 with 5. Since the LOT for event 5 is 28
weeks and duration of activity (4-5) is 2 week the LOT for event 4 is 26 weeks (28-2).
This represents the latest time by which event 4 should occur to enable the project to
be completed in 28 weeks. Likewise, the LOT for other events can be calculated by
moving backward. The LOT for various events is shown (in the right quarter of the
lower half of event nodes) in Figure 8.4
Figure 8.4 Network with EOT and LOT of Events

The general formula for LOT is :

LOT(i) = Min [LOT(i) - d(i-j)]
where LOT(i) = latest occurrence time of event i
LOT(i) = latest occurrence time of event j (ij follows i & there may be several
j’s)
d(i-j) = duration of activity (i-j).

13
The minimization shown here is done with respect to all activities (i-j) starting from i
Given the LOT for various events we can calculate the Latest Finishing Time (LFT) and
Latest Starting Time (LST) for various activities
The formulae for LFT and LST are :
LFT (i-j) = LOT (i)
LST (i-j) = LFT (i-j) -d (i-j)
where LFT (i-j) = latest finishing time for activity (i-j)
LOT (j) = latest occurrence time of event (j)
LST (i-j) = latest starting time for activity (i-j)
d (i-j) = duration of activity (i-j)
8.4.2.3 Calculate the Slack for each Event
The slack for an event is the difference between its LOT and EOT. The slacks for various
events of our illustrative project are shown in Table 8.2
Table 8.2 Event Slack
Event LOT EOT Slack = LOT - EOT
1 0 0 0
2 13 13 0
3 18 12 6
4 26 20 6
5 28 28 0
8.4.2.4 Obtain the Critical and Slack Paths
A path is a sequence of activities that leads from the starting node to the finishing
node. The critical path parts with the beginning event, terminates with the end event,
and is marked by events, which have a zero slack. This is obviously the path on which
here is no slack, no cushion. Other paths are slack paths with some cushion. The critical
path for our illustrative project is (1-2-5). Dark arrows in Figure 8.5 indicate it.

14
 Table 8.3 Critical and Slack Paths
Path Activities Duration Path Slack
1-2-4-5 1-2, 2-4, 4-5 17 28 - 17 = 11
1-3-4-5 1-3, 3-4, 4-5 22 28-22 = 6
1-2-5 1-2, 2-5 28 0

Figure 8.6 Critical Path in the Network

The critical path is the longest path from the beginning event to the end event. Since the
end can be reached, i.e., project completed, only when this longest path is traversed,
the minimum time required for completing the project is the duration on the critical
path. The duration on the critical path of our project is 28 weeks; this is the minimum
time required completing the project. (It is already indicated by the EOT of event 5,
the end event.)

15