Subject:Correction Request PERT Distribution Question

Respected Sir/Madam,

In the question about the distribution used in PERT analysis, the answer key
marks Normal distribution as correct. However, standard textbooks, including:

Project Planning and Control with PERT and CPM by Dr. B.C. Punmia & K.K.
Khandelwal, and

Introduction to Operations Research by Hillier & Lieberman

clearly explain that Beta distribution is used to model activity durations in

PERT using three estimates (optimistic, most likely, pessimistic). Normal
distribution is only used later to approximate the project duration, not individual
tasks.

Since the question doesn't specify "project duration," the technically accurate
answer should be Beta distribution.

A few supporting textbook pages are attached for reference.

Thank you!
PROJECT PLANNING AND CONTROL

WITH

AND
PERT
CPM

Dr. [Link]
[Link]
 PROJECT PLANNING AND cONTROL
WITH

PERT
AND

CPM

By
Dr. B.C. PUNMIA
B.E. (Hons.), M.E. (Hons.), Ph.D.
Formerly,
DEAN FACULTY OF ENGINEERING
M.B.M. ENGINEERING COLLEGE,
JODHPUR
AND
K.K. KHANDELWAL
B.E. (Hons.), I.A.S.

LAXMI PUBLICATIONS (P) LTD

BANGALORE " CHENNAI COCHIN " GUWAHATI
HYDERABAD JALANDHAR KOLKATA LUCKNOw
MUMBAI RANCHI
NEW DELHI
 Pages
4.4. Steps in development of network 71
4.5. Work breakdown structure 79
4.6. Hierarchies 82
4.7. Illustrative examples 85
Problems 90

CHAPTER 5
PERT: TIME ESTIMATES
5.1. Introduction 91
5.2. Uncertainties : Use of PERT 92
5.3. Time estimates 93
5.4. Frequency distribution 95
5.5. Mean, variance and standard deviation 98
5.6. Probability distribution 101
5.7. Beta distribution 104
5.8. Expected time 106
Problems 112

CHAPTER 6
PERT:TIME COMPUTATIONS
6.1. Introduction 115
6.2. Eariiest expected time 115
6.3. Formulation for T 118
6.4. Latest allowable occurrence time 122
6.5. Formulation for TL 125
6.6. Combined tabular computations for Tg and TL 129
Problems 132

CHAPT'ER 7
PERT:NETWORK ANALYSIS
7.1. Slack 134
7.2. Critical path 137
7.3. Illustrative examples 138
7.4. Probability of heeting scheduled date 144
Problemns 155

CHAPTER 8
CPM: NETWORK ANALYSIS
8.1. Introduction 158
8.2. CPM: process 159
8.3. CPM: networks 161
8.4. Activity time estimate 163
8.5. Earliest event time 164
8.6. Latest allowable occurrence time 169
104 PERT : TIME ESTIMATES

18 -6
6
= 2 days.
Hence we conclude, in general, that standard deviation is
given by
tp - to
...(5.6)
6

or variance tp -to .[5.6 (a)

It is seen that the standard deviation is affected by the
relative distance from the most optimistic estimate to the most
pessimistic estimate. It is not influenced by the most likely
estimate (t).
The above method of calculating standard deviation is
approximate. A more exact method is by frequency distribution,
explained in example 5.1. However, in PERT problems, the
emphasis is one-time, non-repetitive projects for which there is no
history of the activity. Hence we must base computations for o on
the given time estimates of the estimator. If the estimator feels them
his range of to and tp includes about all the possible values under
the curve, then the standard deviation can be computed from
Eq. 5.6 with reasonable accuracy.
5.7. THE BETA DISTRIBUTION
The beta distribution is a typical type of probability distribu
tion, which fits well for PERT analysis. A beta distribution is the
one which is not symmetrical about its apex. Fig. 5.6 shows two beta
distributions, one having skew to the left (beta distribution for
optimistic estimator) and the other having skew to the right (beta
distribution for the pessimistic estimator).
The originators of PERT were interested in finding that type
of probability distribution which satisfies the following conditions :
[Link] distribution should have a small probability of reach
ing the most optimistic time (shortest time).
2. The distribution should have a small probability of reach
ing the most pessimistic time (longest time).
3. The distribution should have one and only one most likely
time (i.e. unimodal) which would be free to move between the two
extremes mentioned in 1 and 2 above.
 PERT : TIME ESTIMATES
106

5.8. EXPECTED TIME

The three time estimates t,, (optimistic time), tp (pessimistic
time) and t, (most likely time) are identified on the Beta-distribu
tion, The variance and standard deviation can be computed using to
and tp. However, one must combine the three time estimates into
one single time-the average time taken for the completion of the
activity or job. This average time or single workable time is com
monly called the expected time and is denoted by tg. Ifthe exact shape
of the probability distribution curve is known, the average time or
expected time could be accurately calculated. However, since the
precise curves are never available (specially for non-repetitive jobs)
we must use approximation. This is done algebraically, using a
weighted average derived by statisticians. In computing the ex
pected time, a weightage of 1 is given to the optimistic time to,
weightage of 4 to the most likely time (t) and weightage of 1 to the
most pessimistic time (tp).
to + 4t, t tp ...(5.7)
Thus, t = 6
The above expression for t, based on weighted average
method, is reasonable since the chance of completion of the job in to
or tp is much less than the most likely time (t).
Let us take examples of estimated times of completion of two
jobs A and B, as under.
to tt tp (days)
Job A 4 6 11
Job B 5 10 12
The expected time for these jobs are
to +4t +tp 4 +(4 x6) +11 = 6.5 days
6 6
to + 4t+tp 5+ (4 x 10) + 12
6
= 9.5 days
6
Thus, for job A, the expected time falls to the right of the most
likely time, though the curve has skew to the left [Fig. 5.7 (a)]. For
job B, the expected time tg falls to the left of the most likely time,
though the curve has skew to the right.
One important point should be noted about the expected time
tg. The expected time t represents the average value while the most
 The McGraw HillCompanles
detailsailsWitsh CD-RO
SPECIAL
INDIAN
00 ProtM. Fo
EDITION e inside.

Frederick S. Hillier
Gerald J. Lieberman

IMTRODUCTION TO EIGHTH EDITION

UL 180+

OPERATIONS
RESEARCH
CONCEPTS AND CASES

For sale in India, Pakistan, Nepal, Bangladesh, Sri Lanka &Bhutan only.
hil61217_ch22.qxd 4/29/04 05:58 PM Page 22-1

CH A PTE R

Project Management
with PERT/CPM

ne of the most challenging jobs that any manager can take on is the management of
a large-scale project that requires coordinating numerous activities throughout the
organization. A myriad of details must be considered in planning how to coordinate all
these activities, in developing a realistic schedule, and then in monitoring the progress of
the project.
Fortunately, two closely related operations research techniques, PERT (program eval
uation and review technique) and CPM (critical path method), are available to assist the
project manager in carrying out these responsibilities. These techniques make heavy use
of networks (as introduced in the preceding chapter) to help plan and display the coordi
nation of all the activities. They also normally use a software package to deal with all the
data needed to develop schedule infornmation and then to monitor the progress of the pro
ject. Project management software, such as MS Project in your OR Courseware, now is
widely available for these purposes.
PERT and CPM have been used for a variety of projects, including the following types.
1, Construction of a new plant
2. Research and development of a new product
3. NASA space exploration projects
4. Movie productions
5. Building a ship
6. Government-sponsored projects for developing a new weapons system
7. Relocation of a major facility
8. Maintenance of a nuclear reactor
9. Installation of a management information system
10. Conducting an advertising campaign
PERT and CPM were independently developed in the late 195Os. Ever since, they
have been among the most widely used OR techniques.
The original versions of PERT and CPM had some important differences, as we will
point out later in the chapter. However, they also had a great deal in common, and the two
techniques have gradually merged further over the years. In fact, today's software pack
ages often include all the important options from both original versions.
22-1
hil61217_ch22.qxd 4/29/04 05:58 PM Page 22-15

22.4 DEALING WITH UNCERTAIN ACTIVITY DURATIONS 22-15

Question 6: For the other activities, how much delay can be tolerated without delaying
project completion? These tolerable delays are the positive slacks given in
the middle column of Table 22.3.

22.4 DEALING WITH UNCERTAIN ACTIVITY DURATIONS

Now we come to the next of Mr. Perty's questions posed at the end of Sec. 22.1.
Question 7: Given the uncertainties in accurately estimating activity durations, what is
the probability of completing the project by the deadline (47 weeks)?
Recall that Reliable willincur a large penalty ($300,000) if this deadline is missed. There
fore, Mr. Perty needs to know the probability of meeting the deadline. If this probability
is not very high, he will need to consider taking costly measures (using overtime, etc.) to
shorten the duration of some of the activities.
It is somewhat reassuring that the PERTICPM scheduling procedure in the preceding
section obtained an estimate of 44 weeks for the project duration. However, Mr. Perty un
derstands very well that this estimate is based on the assumption that the actual duration
of each activity will turn out to be the same as its estimated duration for at least the ac
tivities on the critical path. Since the company does not have much prior experience with
this kind of project, there is considerable uncertainty about how much time actually will
be needed for each activity. In reality, the duration of each activity is a random variable
having some probability distribution.
The original version of PERT took this uncertainty into account by using three dif
ferent types of estimates of the duration of an activity to obtain basic information about
its probability distribution, as described below.

The PERT Three-Estimate Approach

The three estimates to be obtained for each activity are
Most likely estimate (m) estimate of the most likely value of the duration,
Optimistic estimate (o) = estimate of the duration under the most favorable conditions,
Pessimistic estimate (p) = estimate of the duration under the most unfavorable
conditions.
The intended location of these three estimates with respect to the probability distribution
is shown in Fig. 22.6.
Thus, the optimistic and pessimistic estimates are meant to lie at the extremes of what
is possible,whereas the most likely estimate provides the highest point of the probability

IFIGURE 22.6
Model of the probability
distribution of the duration
of an activity for the PERT Beta distribution
three-estimate approach:
m = most likely estimate,
o= optimistic estimate, and
p= pessimistic estimate.
0

Elasped time
hil61217_ch22.gxd 4/29/04 05:58 PM Page 22-19

22.4 DEALING WITH UNCERTAIN ACTIVITY DURATIONS 22-19

I FIGURE 22.7
o=9
The three simplifying
approximations lead to the 47 - 44
probability distribution of the =

3 = 1
duration of Reliable's project
being approximated by the
normal distribution shown
here. The shaded area is the Project duration
portion of the distribution 44 47
(in weeks)
that meets the deadline of (Mean) (Deadline)
47 weeks.

TABLE 22.6 Calculation of , and oz for Reliable's project

Activities on Mean Critical Path Mean Variance

1
A 2

B 4 1
C 10 4
4
4

F 1
8 1
5 1
4
6

Project duration up 44 =9

Now we are ready to determine (approximately) the probability of completing Reli

able's project within 47 weeks.

Approximating the Probability of Meeting the Deadline

Let

T= project duration (in weeks), which has (approximately) a normal distribution

with mean u, = 44 and variance o=9,
d= deadline for the project =47 weeks.
Since the standard deviation of Tis o, =3, the number of standard deviations by which
d exceeds u, is

47 - 44
K, = = 1,

Therefore, using Table A5.1 in Appendix 5 for a standard normal distribution (a normal
distribution with mean 0 and variance 1), the probability of meeting the deadline (given
the three simplifying approximations) is
P(T<d) =P(standard normal s K,)
= | - P(standard normal > K)= 1 -0.1587 0.84.