Topic 3

Probabilistic Activity Times

PERT

Lecture: PhD. Đinh Bá Hùng Anh

Tel: 01647.077.055/090.9192.766
Mail: [email protected]
Contents

Probabilistic Activity Times

Probability Analysis of a Project Network

Project Crashing

Example Problem

3-2
Probabilistic Activity Times

■ Activity time estimates usually cannot be made with certainty.

■ PERT used for probabilistic activity times.
■ In PERT, three time estimates are used: most likely time (m), the
optimistic time (a), and the pessimistic time (b).
■ These provide an estimate of the mean and variance of a beta
distribution: Three-Point Method
2
b - a
 
variance: v   
 6 
 

mean (expected time)

t  a  4m  b
6

a m b
3-3
Probabilistic Activity Times

Figure 3.1: Network for Installation Order Processing System 3-4

Probabilistic Activity Times

Table 3.1: Activity Time Estimates for Figure 3.1 3-5

Probabilistic Activity Times

Figure 3.2: Earliest and Latest Activity Times 3-6

Probabilistic Activity Times

Expected Project Time and Variance

■ Expected project time is the sum of the expected times of the
critical path activities.
■ Project variance is the sum of the critical path activities’ variances
■ The expected project time is assumed to be normally distributed
(based on central limit theorem).
■ In example, expected project time (tp) and variance (vp) interpreted
as the mean () and variance (2) of a normal distribution:

 = 25 weeks
2 = 62/9
= 6.9 (weeks)2
3-7
Probability Analysis of a Project Network

■ Using the normal distribution, probabilities are determined by

computing the number of standard deviations (Z) a value is from
the mean.
■ The Z value is used to find corresponding probability in Table
3.2.

3-8
Probability Analysis of a Project Network

P(Z < 0,12) = (0,12) = 0,5478

3-9
Probability Analysis of a Project Network
Table 3.2

3-10
Probability Analysis of a Project Network

Figure 3.3: Normal Distribution of Network Duration

3-11
Probability Analysis of a Project Network

What is the probability that the new order processing system will
be ready by 30 weeks?
µ = 25 weeks
2 = 6.9  = 2.63 weeks
Z = (x-)/ = (30 -25)/2.63 = 1.90
Z value of 1.90 corresponds to probability of 0.9713 in Table 3.2.
Probability of completing project in 30 weeks or less = 0.9713.

3-12
Probability Analysis of a Project Network

Figure 3.3: Probability the Network will be completed in 30 weeks or less

3-13
Probability Analysis of a Project Network

■ A customer will trade elsewhere if the new ordering system is

not working within 22 weeks. What is the probability that she
will be retained?
Z = (22 - 25)/2.63 = -1.14
■ Z value of 1.14 (ignore negative) corresponds to probability
of .8729 in Table 3.2.
■ Probability that customer will be retained is .1271

3-14
Probability Analysis of a Project Network

Figure 3.4: Probability the Network will be completed in 22 weeks

or less.
3-15
Project Crashing

■ Project duration can be reduced by assigning more resources

to project activities.
■ However, doing this increases project cost.
■ Decision is based on analysis of trade-off between time and
cost.
■ Project crashing is a method for shortening project duration
by reducing one or more critical activities to a time less than
normal activity time.

3-16
Project Crashing

Figure 3.5: The Project Network for Building a house

3-17
Project Crashing

Crash cost & crash time have a linear relationship:

Total Crash Cost $2000

Total Crash Time 5 weeks
 $400 / wk

Figure 3.6
3-18
Project Crashing

Table 3.3
3-19
 Project Crashing

Figure 3.7: Network with Normal Activity times and weekly crashing
costs 3-20
Project Crashing

As activities are crashed, the critical path may change and several
paths may become critical.

Figure 3.8: Revised

Network with activity 1
crashed
3-21
Project Crashing

■ Project crashing costs and indirect costs have an inverse

relationship.
■ Crashing costs are highest when the project is shortened.
■ Indirect costs increase as the project duration increases.
■ Optimal project time is at minimum point on the total cost
curve.

3-22
Project Crashing

Figure 3.9: The Time-Cost Trade-Off

3-23
Project Crashing

Example

j 1 2 3 4 5 6 7 8 9 10 11 12 13 14

8, 10, 12,
Prio. - 1 1 2 3 3 4 5,6 5,6 7
9 11
11
13

pj 5 6 9 12 7 12 10 6 10 9 7 8 7 5

cashj 3 5 7 9 5 9 8 3 7 5 4 5 5 2

cj 7 2 4 3 4 3 4 4 4 5 2 2 4 8

cjcash 20 25 20 15 30 40 35 25 30 20 25 35 20 10
3-24
Project Crashing

Maximum Processing Times

6 12 10
2 4 7 9
5 10
12 10 8 5
1
6 9 12 14
9 7 7
3 11 13
7 6

5 8

3-25
 Project Crashing

6 12 10 9
processing time
2 4 7 10 8
5
12 10 12 5
S 1
6 9 14 T
9 7 7
3 11 13
7 6

5 8

longest path  56  1  3  6  9  11  12  14
3-26
Project Crashing

Maximum Processing Times

Pmax = 56
6 12 10
2 4 7
9
5 10
12 10 8 5
1
6 9 12 14
9 7 7
3 11 13
7 6
5 8

Minimum cut sets: {1}, {3}, {6}, {9}, {11}, {12}, {14}
3-27
Project Crashing

Critical Path Subgraph (Gcp)

C1=7
C6=3 C9=4 C12=2 C14=8
1
6 9 12 14

3 11
C3=4 C11=2

Cut sets: {1},{3},{6},{9}, Minimum cut set with

{11},{12},{14}. lowest cost: {11} and {12}.

j 1 2 3 4 5 6 7 8 9 10 11 12 13 14
cj 7 2 4 3 4 3 4 4 4 5 2 2 4 8
3-28
Project Crashing

Activity N.time C.time N.Cost C.Cost Crash cost/pd Crash by Crashing cost
Project 56 37
1 5 3 7 20 6.5 2 13
2 6 5 2 25 23 1 23
3 9 7 4 20 8 2 16
4 12 9 3 15 4 3 12
5 7 5 4 30 13 0 0
6 12 9 3 40 12.33 3 37
7 10 8 4 35 15.5 2 31
8 6 3 4 25 7 0 0
9 10 7 4 30 8.67 3 26
10 9 5 5 20 3.75 4 15
11 7 4 2 25 7.67 3 23
12 8 5 2 35 11 3 33
13 7 5 4 20 8 2 16
14 5 2 8 10 .67 3 2
TOTALS 56 247

3-29
 Project Crashing

6 12 10 9 processing time
2 4 7 10 7
5
12 10 12 5
S 1
6 9 14 T
9 7 7
3 11 13
7 6
5 8

longest path  55 : 1  3  6  9  11  12,13  14

3-30
Project Crashing

Critical Path Subgraph (Gcp)

C1=7
C6=3 C9=4 C12=2 C14=8
1
6 9 12 14

3 11 13
C3=4 C11=2 C13=4

Minimum cut sets: {1},{3},{6}, Minimum cut set with

{9}, {11},{12,13},{14}. lowest cost {11}

C11 = 2 < C0 = 6  net saving of 4  P11 = 6.

j 1 2 3 4 5 6 7 8 9 10 11 12 13 14
cj 7 2 4 3 4 3 4 4 4 5 2 2 4 8
3-31
Project Crashing

Critical Path Subgraph (Gcp)

C1=7
C6=3 C9=4 C12=2 C14=8
1
6 9 12 14

3 11 13
C3=4 C11=2 C13=14

Reduce processing time Reduce processing

next on job 6 time

3-32
 Project Crashing
6 12 10 9 processing time
2 4 7 10 7
5
12 10 12 5
S 1
6 9 14 T
9 6 7
3 11 13
7 6

5 8

longest path  54 : 1  3  6  9  11  12,13  14

or : 1  2  4  7  10  12  14
minimum cut sets :
1, i, j: i  2,4,7,10,12  j  {3,6,9,11}, 12,13, 14
minimum cut set with lowest costs : 2,11
c 2  c11  4  c 0  6  net savings of 2  p2 : 5*, p11 : 5 3-33
Project Crashing
job 2 hits 5min 12 10 9 processing time
minimum 2 4 7 10 7
5
12 10 12 5
S 1
6 9 14 T
9 5 7
3 11 13
7 6

5 8

longest path  53 : 1  3  6  9  11  12,13  14

or : 1  2  4  7  10  12  14
minimum cut sets :
1, i, j: i  4,7,10,12  j  {3,6,9,11}, 12,13, 14
minimum cut set with lowest costs : 4,11
c 4  c11  5  c 0  6  net savings of 1  p 4 : 11, p11 : 4 * 3-34
 Project Crashing
5min 11 10 9 processing time
2 4 7 10 7
5
12 10 12 5
S 1
6 9 14 T
9 4min 7
3 11 13
7 6

5 8

longest path  52 : 1  3  6  9  11  12,13  14

or : 1  2  4  7  10  12  14
minimum cut sets :
1, i, j: i  4,7,10,12  j  {3,6,9}, 12,13, 14
minimum cut set with lowest costs : 4,6
c 4  c 6  6  c 0  6  no net savings  STOP 3-35
Example Problem

Problem Statement and Data

Given this network and the data on the following slide, determine the
expected project completion time and variance, and the probability
that the project will be completed in 28 days or less.

3-36
Example Problem

Problem Statement and Data

3-37
Example Problem

Solution
Step 1: Compute the expected activity times and variances.
2
b -
a
t  a  4m  b
 
v  
6 6 



3-38
 Example Problem

Step 2: Determine the earliest and latest activity times & slacks

3-39
Example Problem

Step 3: Identify the critical path and compute expected

completion time and variance.
 Critical path (activities with no slack): 1  3  5  7
 Expected project completion time: tp = 9+5+6+4 = 24 days
 Variance: vp = 4 + 4/9 + 4/9 + 1/9 = 5 (days)2

3-40
 Example Problem

Step 4: Determine the Probability That the Project Will be

Completed in 28 days or less (µ = 24,  = 5)
Z = (x - )/ = (28 -24)/5 = 1.79
Corresponding probability from Table 3.2, is .9633 and P(x  28)
= 0.9633.

3-41
Home work

Precedence Relationships Chart

Immediate
Activity Optimistic Most Likely Pessimistic
predecessor
A - 76 86 120
B A 12 15 18
C B 4 5 6
D G 15 18 33
E D 18 21 24
F A 16 26 30
G C, F 10 13 22
H D 24 18 32
I A 22 27 50
J D, I 38 43 60
3-42