Unit II

Project Evaluation & Review Technique

(Project Scheduling with uncertain
activity timings)
 PERT Uncertain activity timings
PERT is based on the assumption that an activitys duration follows a probability
distribution instead of being a single value.

Three time estimates are required to compute the parameters of an activitys

duration distribution:

optimistic time (a): the time the activity would take if things did go well

most likely time (m ): the consensus best estimate of the activitys duration

pessimistic time (b): the time the activity would take if things did not go well
 PERT Uncertain activity timings
The probabilistic data for project activities generally follow Beta distribution.
The formula for mean () and variance (2) of Beta distribution are given below.

Mean (expected time): = a+4m+b

6
2
Variance: 2= a-b
6

The range for the time estimates if from a to b. The most likely time will be
anywhere in between the range from a to b.

The expected project completion time = i, where i is the expected time of the
ith critical activity.

The variance of the project completion time = i 2, where i 2 is the variance of

the ith critical activity in the critical path.
 PERT Uncertain activity timings
As a part of statistical analysis, one may be interested in knowing the probability
of completing the project on or before a given due date or knowing the expected
project completion time if the probability of completing the project is given .

For the purpose of this analysis, the Beta distribution is approximated to standard
normal distribution whose statistic is given below.
z = (x )/,
where x actual project completion time
expected project completion time
standard deviation of the expected project completion time

Therefore, P(x C) represents the probability that the project will be completed
on or before the C time units.

This can be covered into the standard normal statistic z as follows:

P[(x )/] [(C )/] = P(z(C )/)
 PERT Analysis
Draw the network.

Analyze the paths through the network and find the critical path.

The length of the critical path is the mean of the project duration probability
distribution which is assumed to be normal.

The standard deviation of the project duration probability distribution is

computed by adding the variances of the critical activities (all of the activities that
make up the critical path) and taking the square root of the resulting sum.

Probability computations can now be made using the normal distribution table.
 PERT A case study for insight
1) Consider the following data of a project with uncertain activity timings.
Duration (weeks)
Activity Predecessor(s) a m b
A - 1 2 3
B - 2 2 8
C A 6 7 8
D B 1 2 3
E A 1 4 7
F C,D 1 5 9
G C,D,E 1 2 3
H F 1 2 9
(a) Construct the project network.
(b) Find the expected duration and variance of each activity.
(c) Find the critical path and the expected project completion time.
(d) What is the probability of completing the project on or before 20 weeks?
(e) If the probability of completing the project is 0.8, find the expected project
completion time.
 PERT A case study for insight
Solution:
(a) As a first step, by using the information given in Table, we interlink the various
activities and thus achieve the project network

E G

C
H

D F
B
 PERT A case study for insight
Solution:
(b) As a second step, by using the information given in Table, we find out the mean
time (average time) and variance for all the activities of the project.

The formula for the mean time and variance for the activities are as given below

Mean (expected time): = a+4m+b

6
2
Variance: 2= a-b
6
 PERT A case study for insight
(b) The expected duration and variance of each activity are as shown in the
following table.

Time Duration Mean

Activity a m b Duration Variance
A 1 2 3 2 0.11
B 2 2 8 3 1.00
C 6 7 8 7 0.11
D 1 2 3 2 0.11
E 1 4 7 4 1.00
F 1 5 9 5 1.78
G 1 2 3 2 0.11
H 1 2 9 3 1.78
 PERT A case study for insight
Solution:
(c)The calculation of critical path is to be done on the basis of mean time of the
activity, Early start & Early finish and Latest start & Latest finish for the project
network model.

E G

C
H

D F
B
 PERT A case study for insight

E 2 6 G 9 11

4 9 15 2 15 17

A 0 2 H 14 17

2 0 2 C 2 9 3 14 17

7 2 9

F 9 14
D 3 5
5 9 14
B 0 3 2 7 9

3 4 7
 PERT A case study for insight
Float for various activities of the present example:

Activity EF LF Float Critical

A 2 2 0 Y
B 3 7 4
C 9 9 0 Y
D 5 9 4
E 6 15 9
F 14 14 0 Y
G 11 17 6
Critical Path : A-C-F-H
H 17 17 0 Y
Mean Duration: 17 weeks
The activities for which float is zero are critical activities and their path gives the critical
path
 PERT A case study for insight
(d) The sum of the variances of all the activities on the critical path is 3.78 weeks (.11 +
.11+1.78 + 1.78)
Therefore = 3.78 = 1.945 weeks

P(x 20) = P[(x )/ (20 17)/1.945]

= P(z 1.542)
= 0.9382 (Value obtained from standard normal table).
Therefore, the probability of completing the project on or before 20 weeks is 0.9382 i.e.
93.82%

(e) P( x C) = 0.8
P [(x - )/ (C )/ ] = 0.8
P [z (C 17)/ 1.945] = 0.8
From the standard normal table, the value of z is 0.84 when the cumulative probability is
0.8. Therefore, (C 17)/1.945 = 0.84

Hence, C = 0.84 * 1.945 + 17 = 18.63 weeks = 19 weeks (Pessimistic).

The Project will be completed in 18.63 weeks (approximately 19 weeks) if the probability
of completing the project is 0.8.
 PERT Practice problem
P1) Consider the following data of a project with uncertain activity timings.

Duration (weeks)
Activity Predecessor(s) a m b
A - 4 5 12
B - 1 1.5 5
C A 2 3 4
D A 3 4 11
E A 2 3 4
F C 1.5 2 2.5
G D 1.5 3 4.5
H B,E 2.5 3.5 7.5
I H 1.5 2 2.5
J F,G,I 1 2 3
(a) Construct the project network.
(b) Find the expected duration and variance of each activity.
(c) Find the critical path and the expected project completion time.
(d) What is the probability of completing the project on or before 16 weeks?
(e) If the probability of completing the project is 0.9, find the expected project completion time.
 PERT Practice problem
P2) Consider the following data of a project with uncertain activity timings.

Duration (weeks)
Activity Predecessor(s) a m b
A - 4 4 10
B - 1 2 9
C - 2 5 14
D A 1 4 7
E A 1 2 3
F A 1 5 9
G B, C 1 2 9
H C 4 4 4
I D 2 2 8
J E, G 6 7 8
(a) Construct the project network.
(b) Find the expected duration and variance of each activity.
(c) Find the critical path and the expected project completion time.
(d) What is the probability of completing the project on or before 12 weeks?
(e) If the probability of completing the project is 0.85, find the expected project completion time.
 CPM & PERT Benefits
Benefits
Especially useful when scheduling and controlling large projects

Straightforward concept and not mathematically complex

Graphical networks help highlight relationships among project activities

Critical path and slack time analyses help pinpoint activities that need to be
closely watched

Project documentation and graphics point out who is responsible for various
activities

Applicable to a wide variety of projects

Useful in monitoring not only schedules but costs as well

 CPM & PERT Benefits
CPM/PERT can answer the following important questions:
When will the entire project be completed?

What are the critical activities or tasks in the project?

Which are the noncritical activities?

What is the probability the project will be completed by a specific date?

Is the project on schedule, behind schedule, or ahead of schedule?

Is the money spent equal to, less than, or greater than the budget?

Are there enough resources available to finish the project on time?

If the project must be finished in a shorter time, what is the way to accomplish
this at least cost?
 CPM & PERT Limitations
Limitations
Project activities have to be clearly defined, independent, and stable in their
relationships

Precedence relationships must be specified and networked together

Over emphasis on critical paths

Deterministic CPM model

Activity time estimates are subjective and depend on judgment

PERT assumes a beta distribution for these time estimates, but the actual
distribution may be different

PERT consistently underestimates the expected project completion time due

to alternate paths becoming critical