NETWORK ANALYSIS

INTRODUCTION AND IDENTIFICATION OF CRITICAL PATH

Network analysis is the organized application of systematic reasoning for planning,
scheduling and controlling, practical situations where many separate tasks (which
collectively make up the whole project) can happen simultaneously or
consecutively such that it is difficult to establish the relationships between the
separate jobs. The technique can be applied to any purposeful chain of events
involving the use of time, labour and physical resources although it is usually
related to large industrial or commercial projects of a complex nature where the
scale of operation gives rise to correspondingly greater financial and administrative
problems.
Network analysis involves the following three basic steps:
(i)

It defines the job to be done.

(ii)

It integrates the various activities in a logical time sequence.

(iii)

It controls the progress of project plan.

Main techniques of network analysis are as under:

(a) Critical Path Analysis (CPA) aka Critical Path Method (CPM)
(b) Project Evaluation and Review, Technique (PERT.)
Both the techniques were developed in USA during the late 1950s.PERT was
developed by US Navy Engineers to plan and control the huge Polaris Submarine
Programme. CPM was developed by du Pont and Remington Rand Companies to
help the process of scheduling maintenance of chemical plants. Both the
techniques have been applied successfully to improve efficiency of execution of
large projects within pre-determined time and cost limits.
Page 1 of 28

Some of the more common applications of network analysis occur in projects

scheduling for:
(a) The construction of buildings, bridges, factories, etc.
(b) Budgeting and auditing procedures;
(c) Missile development;
(d) The installation of a computer system;
(e) Launching a new project;
(f) Strategic and tactical military planning
BASIC TERMS USED IN NETWORK ANALYSIS
Project
A project is a combination of interrelated activities that must be executed in a
certain order before the entire task (work) is completed.
Activity
Any individual operation that consumes time, money or other resources and has an
end and beginning is called an activity. An activity is represented in a network by
an arrow, the tail of which represents the beginning while the head represents the
completion of the activity. It is shown as under:-

Event
An event in a network diagram is a time-oriented reference point that signifies the
end of one activity and the beginning of another. An event that immediately
Page 2 of 28

follows another event is called a successor event while the one that comes
immediately before another event is called a predecessor event. An event is
represented in a network by a circle or node. It is shown as under:

Arrow Diagram
An arrow diagram is a network in which the branches are constructed as if they
were arrows. The head of the arrow is used to indicate the direction of flow
through the network. When the arrows are joined head to tail they show the
required precedence relationships between the nodes or events of the network.
Network
A network is a graphic representation of a projects operation and is composed of
activities and events (nodes). In a network, events are identified by numbers i.e. 1,
2, 3, 4. while the letters A, B, C, D represent the activities.
A network may appear as under: -

Page 3 of 28

CONSTRAINTS IN A NETWORK PROJECT

All the projects consist of certain activities that can begin only after certain others
are finished. In fact, the entire project may be considered as a series of activities
which may begin only after another activity or activities are finished. In a network
schedule, these types of relationships are called constraints and are represented by
inequalities. For example, A<B will indicate that the activity A must be completed
before the start of the activity B.
Parallel Activities
In practice, there arise situations when an activity cannot be started unless another
activity is started and can start when the other activity has been partially completed
(not fully completed). As an example, one may consider activities like ... one
representing the construction of a wall and the other denoting its plastering.
Plastering can start when a portion of the wall is ready and then go on concurrently
with the construction of the wall and finish only after the construction of the wall
has finished. This has been shown in Fig. 2

Dummy Activities
There are situations when two or more activities can be performed simultaneously.
For example, in Fig. 3 (a), a segment of the network activities B and C have the
Page 4 of 28

same start and end events. In such cases it becomes necessary to define a dummy
activity so as to distinguish the activities having common starting and finishing
times for establishing a proper precedence relationship. Such an artificially created
dummy activity is shown by a dotted line in Fig. 3 (b).

A dummy activity is that activity which does not consume time or resources. A
dummy activity is represented on a network by a dotted arrow. It is shown as
under:
A dummy activity is created for the following reasons:
(i)

To make activities with common starting and finishing events

distinguishable,

(ii)

To identify and maintain the proper precedence relationship between

activities which are not connected by events.

It may be observed that the introduction of a dummy activity in Fig. 3 (b) does not
affect the basic message conveyed by Fig3 (a). It simply clarifies the precedence
relationship. There is no real activity associated with the dummy arrow. The
dummy activity simply notes that activity E cannot begin until C has been
completed whereas E can begin after B has been completed.

Page 5 of 28

Loops and Dangling

Whenever an activity is represented as going back in time, a closed loop would
occur. This is shown in Fig. 4, which is simply the structure similar to Fig. 3. (b)
with the activity C reversed in its direction.

Thus one property of a correctly constructed network diagram is that it is noncyclic.

Quite often, there arise situations when some activities, other than the initial
(starting) and final (ending), do not have any successor events. See for example,
Fig. 5 below:-

The activity C does not have any successor event. Such a situation is known as
Dangling in network analysis.

Page 6 of 28

To avoid such situations, it should be kept in mind that all events except the first
and last of the whole project must have at least one entering and one leaving
activity.
Rules For Constructing the Network
For the construction of a network, we make use of the following rules:
(a) Each activity is represented by one and only one arrow. This implies that no
single activity can be represented twice in the network.
(b) No two activities can be identified by the same end events. This implies that
there must be no loops in the network.
(c) Time follows from left to right. Arrows pointing in opposite directions must
be avoided.
(d) Arrows should be kept straight and not curved or bent.
(e) Avoid arrows which cross each other.
(f) Use dummies freely in rough draft but final network should not have any
redundant dummies.
(g) Every node must have at-least one activity preceding it and at least one
activity following it, except for the nodes at the very beginning and at the
very end of the network. The beginning node has no activities before it and
the ending node has no activities following it.
(h) Only one activity may connect any two nodes. This rule is necessary so that
an activity can be specified by giving the numbers of its beginning and
ending nodes.
After constructing the network, the next step is to assign a number to every event
and to place the same inside the node circle. The number sequence should be such
Page 7 of 28

so as to reflect the flow of the network. Various steps involved in numbering the
events may be summarized as follow:Step 1.

The starting event which has all outgoing arrows and no incoming
arrow is numbered 1 or 0.

Step 2.

If the initial event is numbered 0, then delete all the arrows coming
out from event O. This will convert some more events (at least

one)

into initial events. Number these events as 1, 2, 3

Step 3.

Remove all the arrows going out from these numbered events to
create more initial events. Assign the next numbers to these

events.
Step 4.

Continue the procedure of assigning the numbers, until the last event
has been assigned a number i.e. until the event which has all

arrows

coming in and no arrow going out has been numbered.

The above procedure of assigning the numbers to various events of a network is

known as Fulkerson Rule.
Example I
Draw a network for the simple project of erection of steel works for, a shed. The
various elements of project are as under: Activity

Description

Predecessor

code
A
B
C
D

Erect site workshop

Fence site
Bend reinforcement
Dig foundation

A
B

Page 8 of 28

E
F
G
H
I
J
K

Fabricate steel work

Install concrete plant
Place reinforcement
Concrete foundation
Paint steel work
Erect steel work
Give finishing touch

A
B
C, D
G, F
E
H, l
J

Solution
The resulting network is shown in the figure given below.

N.B THERE IS NOTHING WRONG WITH ACTIVITY E

Requirements for Network Analysis
If network methods are to be successfully applied, certain information
requirements must be met. These requirements are common to both CPM and
PERT, and we must, therefore, identify them before discussing either of these
techniques.
Arrow Diagram
Page 9 of 28

As we have already seen, network analysis begins with the construction of an

arrow diagram (or network). This device is an important part of the descriptive
phase of the job-project analysis because it provides management with a picture of
the activity-event sequence.
Time Calculations,
Since both CPM and PERT are time-oriented techniques, estimates of the time
required to perform specified activities are essential. Each activity that is defined in
the network must be given a time-to-completion value.

Critical Path
An unbroken chain of activity arrows in a network which connects the initial event
to some other event is called a path. The path consisting of the sequence of those
events and activities that require the maximum time in the completion of the
project is known as critical path of the project. The activities associated with the
critical path are called critical or bottleneck activities. A critical path shall be
denoted by heavy (double) line arrows in a network. A specimen of the critical path
is shown in Fig. 6

Page 10 of 28

One of the primary objectives of both CPM and PERT is to identify the critical
path of the network. As seen above, this path is the sequence of activities or events
along which there is no free time, i.e. there is no room for slippage in schedule
without delaying the entire project, it is readily identified because it is the sequence
that requires the greatest amount of time from start to finish.
Finding the Critical Path
(a) Enter the time taken for each activity on the network.
(b) Find the earliest time at which each event can take place. Work
systematically from the smallest event number to the largest in numerical
order calculating the earliest time which each event can take place. An event
is the end of one or more activities for each of the activities ending in that
event add together the earliest event time for the start of the activity and the
time taken for the activity. The earliest event time will be the largest of the
numbers found in this way for the particular event.
(c) Find the latest time at which each event can take place. To do this start with
the last event and work backwards from the highest to the lowest event
number. An event is the start of one or more activities. For each activity,
starting from a particular event, subtract the time taken for the activity from
the latest event time for the end of the activity. The latest event time will be
the smallest number found in this way for the particular event.
(d) The critical path is found by joining all events for which the earliest event
time (EET) equals the latest event time (LET).
Page 11 of 28

(e) All activities lying on the critical path are called critical activities because
any delay in their execution will lead to a delay in the completion of the
entire project.
FLOAT OR SLACK
Slack is with reference to an event and float is with respect to an activity. In other
words, is used with PERT and float with CPM, but they may be interchangeably
used in general practice. Float or slack means extra time over and above its
duration which a non-critical activity can consume without delaying the project.
Total Float
The total float for an activity is given by the total time which is available for
performance of the activity, minus the duration of the activity.
Free Float
This is that part of the total float which does not affect the subsequent activities. In
other words, free float is the time by which the completion of any activity can be
delayed beyond the earliest finish time without affecting any subsequent activity.
Independent Float
If all the preceding activities are completed at their latest, in some cases, no float
may be available for the subsequent activities which may, therefore, become
critical. On the other hand, in some cases, even after completion of the activities at
their latest, there may still be some float left. The part of the float, which remains
unaffected by utilization of float by the preceding activities and does not affect the
succeeding activities is called independent float.
Page 12 of 28

The knowledge of floats helps the management in determining the flexibility of the
schedule and the extent to which the resources will be utilized on different
activities.
Float time can be calculated as under:Total float = LFT - EST - D
Free float =EFT-EST-D
Independent float = EFT - LST -D
Where:
EST = Earliest starting time
LFT = Latest finishing time
LST Latest starting time
EFT = Earliest finishing time
D = Activity duration.
Distinction with CPM and PERT
The main points of distinction between CPM and PERT are as below:i.

CPM is activity oriented, i.e. CPM network is built on the basis of activities.
Also results of various calculations are considered in terms of activities of
the project. On the other hand, PERT is event oriented.

ii.

CPM is a deterministic model, i.e., it does not take into account the
uncertainties involved in the estimations of time for execution of a job or an
activity. It completely ignores the probabilistic element of the problem.
PERT, hover, is a probabilistic model. It uses three estimates of the activity
time: optimistic, pessimistic and most likely with a view to take into account
time uncertainty. Thus the expected duration of each activity is probabilistic
Page 13 of 28

and expected duration indicates that there is fifty present probability of

getting the job done within that time.
iii.

CPM places dual emphasis on time and cost and evaluates the trade-off
between project cost and project time. By deploying additional resources, it
allows the project manager to manipulate project duration within certain
limits so that project duration can be shortened at an optimal cost. On the
other hand, PERT is primarily concerned with time. It helps the manager to
schedule and coordinate various activities so that the project can be
completed on scheduled time.

It is difficult to say with certainty which approach is commonly used. The nature
of the project generally dictates the use of an approach. CPM is commonly
used for those projects with are repetitive in nature, (e.g. construction projects)
and

where one has prior experience of handling similar projects. PERT is

generally used for those projects where time required to complete various activities
is not known apriori. Thus, for planning and scheduling research and development
projects PERT is widely used. For such projects it is very necessary that the
project is completed in time, PERT can determine the probabilities of completing
various stages of such projects. It can identity the activities

which have high

potential of causing delays in completing the project on scheduled date so that the
manager can take preventive measures to reduce possible delays to maintain the
project schedules.
Advantages of Network Analysis
(a) Network analysis helps to determine the objective of the project.

Page 14 of 28

(b) The method enforces planning because data from many sources must be
collected and collated before being logically put together to give the
network:
(c) Areas of responsibility are specifically defined, the relationship between
activities is clearly shown, and the network reveals the interactions of all
participants:
(d) The technique provides for simple communication, because the network
diagrams and charts are easily understood by non-specialists:
(e) Control is simplified, because network analysis permits the use of
management by exception, whereby the management need act only when the
situation is out of control:
(f) The technique is equally applicable to large and small scale operations:
(g) The technique is simple and easy to apply, en by people with a
nonmathematical background.
(h) The system lends itself easily to computers, and many computer
manufacturers provide standard packages of network analysis routines their
equipment.
Limitations of C.P.M
Following are three major limitations of CPM:
i)

CPM is based on the assumption of known time for each of the activity in
the project which may not be true in real life.

ii)

For determining the time estimates, CPM does not incorporate statistical
analysis.

iii)

When certain changes are introduced in the network, the entire evaluation
of the project is to be repeated and a new critical path is to be found out.
Page 15 of 28

Limitations of PERT
The important limitations of PERT are as follow:
i)

PERT is based on the time estimates rather than known time for each
activity.

ii)

It emphasizes only on time and not n costs,

iii)

It is not practicable for routine planning of recurring activities.

iv)

In PERT, the calculation of probabilities is carried out on the assumption

of independent activities and as such the distribution of total time to be
normal. But in real life this may not be true.

v)

For active control of a project, it requires frequent updating and revising

of the PERT calculations.

Page 16 of 28

NET WORK ANALYSIS

CRASHING

OF PROJECTS

COST ANALYSIS

AND

RESOURCE

SCHEDULING
Introduction
In order to include the cost aspects in project scheduling we have to define the
cost-duration relationships for the various activities in the project. Generally, the
cost of a project includes direct costs and indirect costs. The direct costs are
associated with the individual activities and the indirect costs are associated with /
the overhead costs such as administration or supervision costs.
Crashing of protects means to shorten the project completion by spending extra
resources. In real life, it is always possible to employ more resources. Cost rations
are important since activities will only be speeded up at a cost. Analysis of the
direct costs would normally be appropriate in determining whether the additional
cost is worthwhile in terms of a shorter completion time.
Time-cost Optimization Procedure
The process of shortening a project is called crashing and is usually achieved by
adding extra resources to an activity Project crashing involves following steps:
Step 1:- Critical path.
Find the normal critical path and identify the critical activities.
Step 2:-Cost slope.
Calculate the cast slope for the different activities by using the formula

Cost slope =

Crash cost Normal Cost

Page 17 of 28

Normal Time Crash Time

Step 3:-Ranking.
Rank the activities in the ascending order of cost slope.
Step 4: - Crashing.
Crash he activities in the critical path as per the ranking, i.e., activity having lower
cost-slope would be crashed fist to the maximum extent possible. Calculate the
new direct cost by cumulative adding the cost of crashing to the normal cost.
Step 5:- Parallel crashing.
As the critical path duration is reduced by the crashing in step 3, other paths also
become critical, i.e., we get parallel critical paths. This means that project duration
can be reduced duly by simultaneous crashing of activities in the parallel critical
paths.
Step 6:- Total cost.
Crashing as per step 3 and step 4, one reaches a point when further

crashing

is

either not possible or does not result in the reduction of crashing of project
duration (This can happen when activities which can be crashed lie in the noncritical path). For the different project durations total cost is found out. Total cost is
got by adding corresponding fixed cost to the direct cost, and the direct cost is got
by adding the expediting/crashing cost cumulatively to the normal cost.
Example 1
For the network given below find the optimum cost schedule for the completion of
the project:

Page 18 of 28

Solution
Making use of normal time, the resulting network is shown below in the figure
given

From the above figure, we note that the critical path comprises the activities. (2,3),
(3,4), (4,5) and (5,6) with expected project length as 38 days. Associated with 38
days, the minimum direct project cost is Sh. 60 + 75 + 90 + 100 + 50+ 40+ 50 + 70
= Sh 535.
For the optimum cost schedule, first compute the cost slope (incremental cost
crashing) as follows:
Page 19 of 28

Activity

Cost

slope

1-2
2-3
2-4
3-4
3-5
3-6
4-5
5-6

cost)
(120-60)/10-8)
(150-75)/(9-6)
(150-90)/(7-4)
(140-100)/6-5)
(80-50)/9-7)
(70-40)/(10-8)
(70-50)/(6-4)
(110-70)/(7-5)

(Incremental
30
25
20
40
15
15
10
20

Crashing of Jobs
In order to reduce the project duration, crash at least one of the jobs in the critical
path. This is being done because crashing of the job not on the job not on the
critical path does not reduce the project length.
Step 1. On critical path as the minimum cost slope is for the job (4 - 5), it is
cheapest to crash (shorten) this job at an extra cast of Sh.10 per day. Since the time
avai1able for crashing this job is two days, the revised network is as followed:

Page 20 of 28

Duration of the project now is 36 days and total cost = Sh. 535 + Shs. 10 x 2 = Shs.
555.
Step 2:- Having crashed job (4 - 5) to the maximum, the next step would be to
crash job (5 - 6) for two days at an extra cost of sh. 20 per day Thus, we have:

Duration of the project is now 34 days and total cost = Sh555 + Sh.20x2 = Sh595

Step 3. Now crashing the job 2- 3 by 3 days

Page 21 of 28

Durationoftheprojectnowis3l days and total cost = Shs 595 + Sh. 25 x 3= Shs 670.
Step 4:- Further crashing the job (1 - 2) by 2 days, we get

Duration of the project now is 29 days and total cost Shs 670 + Shs 30 x 2 = Shs
730.
Step 5:- Finally crashing the job (3 - 4) by one day find that there are two critical
path ( 1- 2-3- 4- 5- 6 and 1- 2- 3- 5- 6. These are shown in the network given
below.
Page 22 of 28

Duration of project now becomes 28 days and total cost =Sh730 + 1 x sh40 =
sh770
As further crashing of remaining jobs does not reduce the project duration
optimum duration of project is 28 days having the total cost as Sh. 770.

With the help of the CPM technique, crash the project to its minimum length at the
lowest possible direct cost, assuming there is no indirect or opportunity cost.
Page 23 of 28

Solution
Using the normal time duration, the network diagram is given below.

The critical path of the project is B - D F and duration of the project is 20 weeks
with total cost at Sh 36,000. The cost slope of the activities for the above network
is computed as follows:
Activity
A
B
C
D
E
F

Cost slope
(14000 10000)/(6-6)
(8000 5000)/(4-3)
(5000-4000)/3-2)
(6000-1000)/8-3)
(13000-9000)/(14-6)
(800-7000)/8-4)

2000
3000
1000
1000
500
250

Crashing of Activities
Since the activities lying on the critical path control the project duration, crash the
activities lying on the critical path.

Page 24 of 28

Step 1. First we crash that activity of the critical path which involves minimum
cost slope. Since the activity F gives the minimum cost slope, compress its
duration from 8 to 6 weeks with an additional cost of Sh. 2 x 25 Sh. 500. The
revised network is:

Duration of project is now 18 weeks and total cost Sh. 36,000 + Sh. 2 x 250 = Sh.
36,500.
Step 2. Now there are two parallel critical paths. We choose the minimum c slope
which lies on any of the critical paths. So, we compress the duration activity F
from 6 to 4 weeks and the duration of activity E from 14 to 12 weeks with
additional cost of sh.2x50c) + Sh 2 x 250. The revised network is:

Page 25 of 28

Duration of the project is now 16 weeks and the total cost Sb. 36,500 + Sh.500 x 2
+ Sb. 250 x 2 = Sh. 38,000
Step 3. Crashing now the activity D from 8 to 5 weeks and activity E from 12 to 9
weeks with an additional cost of Sh. 3 x 1,000 + Sh 3 x 500, the revised network is:

Duration of the project now is 13 weeks and the total cost Sh. 38,000 + Sh. (1,000
+ 500) x 3 = Sh. 42,500.
Page 26 of 28

Step 4. Now, since all the activities are lying on the critical path activity A from 6
to 5, C from 3 to 2, D from 5 to 3 and E from 9 additional Cost:
Of Sh. 2,000+ Sh. 1,000 + sh. 1,000 x 2 + sh. 500 x 2
The revised network is

Duration of the project now is 11 weeks and total cost

= Sh. 42,500 + Sh. (2,000 + 1,000 + 2,000 + 1,000) Sh. 48,500.
Step 5. Finally crashing the activity A from 5 to 4 weeks and B from 4 weeks,
there is an additional cost of Sh. 1 x 2,000 + Sh. I x 3,000 the network is:

Page 27 of 28

Duration of the project now becomes 10 weeks and total cost = Sh. 48500 + Sh.
2000 + Sh. 3000 = Sh. 53500
Since no further crashing is possible, minimum length of the project is 10 weeks
and the lowest possible cost is Shs. 53500.

Page 28 of 28