IE4248

Project Planning & Control

Introduction

Program Evaluation & Review

Technique
(PERT)
 Course Outline
Course Component Week 1 Week 2 Week 3 Week 4 Week 5 Week 6 Week 7
Wk Begin 27-Jan 03-Feb 10-Feb 17-Feb 24-Feb 03-Mar 10-Mar

Online
Feb Bank
Lecture 1 No Lecture 1 PM Roles & Assessment 1 Managing & No Lecture 1
Holiday PERT
Mon 17.00 - 18.00 Project Scope 25% Budgeting

Lecture 2 Course
Project Project No Lecture 2 Goldratt &
Tues 17.00-18.00 Overview & Planning Scheduling
Selection Crashing Levelling
Introduction

Network PERT/CPM
Video Tutorials Introduction Project Project
Diagrams & Activity
Released to MS Project Selection Crashing
CPM Sequencing
Class Update

This weeks Lab on PERT & Levelling is now uploaded

to Brightspace

This weeks Reading

Chapter 5 Project Management in Practice
Wk 5 Critical Path Analysis and PERT Charts

Online Assessment 2 24th March (25%)

Chp III & IV Project Management in Practice, Mantel, S., Meredith, J., Shafer,
S. & Sutton, M. (2008/11). John Wiley & Sons.
Labs & lecture notes Associated with the above chapters

Reading from Wk 3 and 4

 Project Evaluation &
Review Technique
(PERT)

Chapter 5
 Recap

Up to now you have learned about:

• Definition and constraints of a project
• Project Selection Techniques:
• PM Roles & Responsibilities
• Project Organisational Structure & Project Plan
• Developing a Work/Cost break down structure (WBS)
• Network Diagraming (AOA and AON Networks)
• Critical Path Analysis (CPM)
• Project Crashing
 Overview
Todays Lecture
– Program Evaluation and Review Technique (PERT)
– Stochastic Estimates of Project Duration
– Building in contingencies
Program Evaluation and Review Technique (PERT)

The Program (or Project) Evaluation and Review

Technique, commonly abbreviated PERT, is a
statistical tool, used in project management, that is
designed to analyze and represent the tasks involved in
completing a given project. PERT is commonly used in
conjunction with the critical path method (CPM).
When people refer to PERT they are frequently referring
to the technique of calculating probabilistic activity
times and using these to calculate the probability of
completing the project on time.
The Normal Distribution
A normal distribution is a very important statistical data
distribution pattern occurring in many natural phenomena,
such as height, blood pressure, lengths of objects produced by
machines, etc. Certain data, when graphed as a histogram (data
on the horizontal axis, amount of data on the vertical axis),
creates a bell-shaped curve known as a normal curve, or normal
distribution.

CLT implies that if we are able to make some reasonably

good estimates of how long activities take we can assume
the values fit in to a distribution that looks like this:
Normal Distribution
The normal distribution has the following important
characteristics: (1) the curve has a single peak; (2) it is
bell-shaped; (3) the mean (average) lies at the centre
of the distribution, and the distribution is symmetrical
around the mean; (4) the two tails of the distribution
extend indefinitely and never touch the horizontal
axis; (5) the shape of the distribution is determined by
its mean (µ) and standard deviation (σ).

As with any continuous probability function, the area

under the curve must equal 1, and the area between
two values of X (say, a and b) represents the
probability that X lies between a and b as illustrated
on Figure 1. Further, since the normal is a symmetric
distribution, it has the nice property that a known
percentage of all possible values of X lie within
± a certain number of standard deviations of the
mean, as illustrated by Figure 2. For example, 68.27%
of the values of any normally distributed variable lie
within the interval (µ - 1 σ, µ + 1 σ).
 Normal Distribution
The probability of the normal
distribution as given in previous
slide is difficult to work with in
determining areas under the
curve, and each set of X values
generates another curve as long
as the means and standard
deviations are translated to a new
axis, a Z- axis, with the
translation defined as:

The new variable Z is normally distributed

with a mean of zero and a standard
deviation of 1. Tables of areas under this
standard normal distribution have been The resulting values, called Z-
compiled and widely published so that areas values, are the values of a new
under any normal distribution can be found variable called the standard
by translating the X values to Z values and normal variate, Z. The translation
then using the tables for the standardized process is depicted in Figure 3.
normal.
 to  4tm  t p
Expected time, te: te 
6
2
t p  to  t p  to 
Standard Deviation:
SD  Variance:Var  
6  6 
Estimating Time

to = the most optimistic time to

complete an activity – assumes
nothing goes wrong.
tp = the pessimistic time to complete
an activity – assumes nothing goes
right.
tm = the most likely time (Normal) to
complete an activity.
te = most expected (Estimated) time
PERT Process
PERT Process
1. Calculate te for each activity.
2. Understand and calculate the Critical Path.
3. Use the sum of the variances of the activities
on the critical path to calculate the
probability of completing the project within a
specified time.
Example Pessimistic time
Normal time Estimated time
Optimistic time
Example

7
D

2 I
A C F

B E G J
1 4 5 8 9

6
AOA Diagram
 Critical Path: ACEFIJ

Example 8.2 5.3 13.5

7
D 28.9 20. 34.2
7
0.0 8.2 8.2 34.2 9.8 44.0
25.2 9.0 34.2
0.0 0.0 8.2 34.2 0.0 44.0
2 C 25.2 0.0 34.2
I
A 8.2 9.8 18.0
F 25.2 5.0 30.2
8.2 0.0 18.0
B J
E 39.0 13. 44.0
1 4 5 8 8 9
G
0.0 6.0 6.0 18.0 7.2 25.2 44.0 6.2 50.2

12.0 12. 18.0

0
18.0 0.0 25.2.
H 44.0 0.0 50.2

25.2 16.8 42.0

27.2 2.0 44.0

6
Example Pessimistic time
Normal time Estimated time Identify the Paths
Optimistic time
1.ADIJ
2.ACEFIJ (Critical
Path)
3.ACEGL
4.ACEHJ
5.BEFIJ
6.BEGJ
7.BEHJ

Expected
Critical Path
time = 50.2
days
Sum of Variance on Critical Path = 0.25 + 0.25 + 0.25 + 0.11 + 0.69 +
0.25 = 1.8
√1.8 = 1.34 - We use the SD to find the probability of completing
within a specified time
Problem
What is the probability of competing the project in 53 days?

Probability = 0.9817

There is a 98.17%
probability that
the project will
complete within
53 days.
 Note: Z Score sign is important
Problem
What is the probability of competing the project in 48 days?

The tables are telling us that there

is a 94.95% probability that we will
not complete the project in 48
days.

Alternatively

•Probability = 1- 0.9495 = 0.0505

There is a probability of 5.05% that

the project will complete within 48
What about the other paths?
In order to have a proper estimate of achieving the project in 53 days you must
also factor in the non critical paths.
This can be done by calculating the probability for each of the other paths
achieving 53 days and getting the product of the results

Therefore the Total probability of project being delivered in 53 days is

=1 x 0.982 x 1 x 0.996 x 1 x 1 x 1 = 0.978

There is a 97.8% probability that he project will complete within 53 days

Predicting Completion Date with confidence
Suppose you need to know the
completion date with a certain
confidence (say 95%)
By working backwards from the
tables the Z score for 95% can be
calculated
In our example Z score in the
tables for 95% confidence is
1.645.

In other words We can be 95%

confident that the project will
be complete in 52.4 days
Contingency
Using PERT there is only a 50%
probability of completing the project
within the expected time te
We need to build in a contingency.
 If we know that there is a 50% probability
that the project will be completed in 50.2
days by how much would we need to extend
the schedule to be 100% sure (or as close
as possible) that it will be completed.
Contingency

The highest value we can read from the tables is

4.00, or a probability of 99.997% (1-.00003) that
the project will be completed.
Using this value:

In other words we need a contingency of 5.36 days

to increase the chances of completing the project
from 50% to close to 100%.
Non Critical Path Activities
In the example we have used Fathom the Paths
there are 6 non critical paths:
1.ADIJ
ADIJ, ACEGJ, ACEHJ, BEFIJ, BEGJ, 2.ACEFIJ (Critical Path)
BEHJ 3.ACEGL
4.ACEHJ
To assess their impact on the 5.BEFIJ
6.BEGJ
critical path they must be 7.BEHJ
followed to the point where
they become critical:
To do this we must identify the
tasks that are common and the
tasks that are different from
the critical path (ACEFIJ)
Example Critical Path: ACEFIJ

7
D

2 I
A C F

B E G J
1 4 5 8 9

Fathom the Paths

H
1.ADIJ
Potential Critical Path: ACEHJ 2.ACEFIJ (Critical Path)
3.ACEGL
4.ACEHJ
6 5.BEFIJ
6.BEGJ
AOA Diagram 7.BEHJ
Non Critical Path Activities
For example ACEHJ:
 A,C,E,J are common; For this Fathom the
path to become critical H must Paths
take longer than F+I
 H duration= 16.8; H variance = 1.ADIJ
2.25; 2.ACEFIJ
 F+I duration = 18.8 (Critical Path)
Probability that H is greater 3.ACEGL
than 18.8? 4.ACEHJ
5.BEFIJ
 (Z =18.8-16.8/√2.25) 6.BEGJ
Z = 1.33 -> 0.9082 7.BEHJ
This indicates that h will be equal to or less than 18.8
days 90.82 percent of the time, which means that it will
be greater than FI only about 9.18 percent of the time.

9.18% chance of ACEHJ

becoming the critical path
Summary of PERT

Using PERT puts the focus of attention on

time alone to the detriment of project
activities.
If PERT is conducted by a planning
department there may be disagreement
about what is actually achievable.
PERT has to be completed after resource
levelling and taking other constraints into
account – it is based on all activities
starting as soon as possible.
PERT assumes independent task durations
Excel is often used to simulate PERT data.
Other Methods

Graphical Evaluation and Review

Technique (GERT)
 combines flowgraphs, probabilistic
networks, and decision trees
 allows loops back to earlier events
and probabilistic branching