Académique Documents
Professionnel Documents
Culture Documents
Uncertainty Management
Agenda
PERT Analysis
Monte-Carlo Analysis
A Roadmap
Uncertainty
Risks
Ambiguity
Central Assumption
Task Durations follow a Beta Distribution (right skewed)
Mean
(Average Time)
Mode
Minimum Time
(Optimistic Estimate) (Most likely time)
Maximum Time
(Pessimistic Estimate)
6
Central Assumption
Task Durations follow a Beta Distribution (centered)
Mean
(Average Time)
Minimum Time
(Optimistic Estimate)
Maximum Time
(Pessimistic Estimate)
7
Central Assumption
Task Durations follow a Beta Distribution (left skewed)
Mean
(Average Time)
Minimum Time
(Optimistic Estimate)
Mode
(Most likely time)
Maximum Time
(Pessimistic Estimate)
8
PERT
Program Evaluation and Review Technique
(PERT)
Developed by U.S. Navy in 1950s
Navy Special Projects Office
Polaris submarine weapon system
. With units of time as a common denominator, PERT quantifies knowledge
about the uncertainties involved in developmental programs requiring effort at
the edge of, or beyond, current knowledge of the subject effort for which little
or no previous experience exists. .
Willard Fazar (Head, Program Evaluation Branch, Special Projects Office, U. S. Navy),
The American Statistician, April 1959.
PERT Analysis
Step 1: Instead of asking for task durations, ask for
optimistic, pessimistic and most likely
Task
A
Optimisti
Most Likely
c
1
2
Pessimistic
3
11
13
15
10
12
10
15
23
14
11
16
21
35
10
14
10
PERT Analysis
Step 2: Calculate the mean and variance of each
task duration using the following formulas:
Pessimistic Optimistic
2
36
11
PERT Analysis
Step 2: Calculate the mean task duration, and the variance
of each task duration
Task
A
B
C
D
E
F
G
H
I
J
K
L
M
N
Optimisti
Mea
Most Likely Pessimistic
c
n
1
2
3
2
7
8
9
8
11
13
15
13
8
10
12
10
10
15
23
15.5
4
7
14
7.67
2
4
11
4.83
3
4
7
4.33
16
21
35
22.5
4
5
6
5
2
2
2
2
2
2
2
2
10.3
8
10
14
3
1
1
1
1
Variance
.11
.11
.44
.44
4.69
2.78
2.25
.44
10.03
.11
0
0
1
0
12
PERT Analysis
Step 3: Schedule the Project Using Mean Durations
2
Task A
2 days
0 slack
Earliest
Start
Earliest
Finish
ES
EF
LS
LF
10
10
23
23
33
Task B
8 days
0 slack
Task C
13 days
0 slack
Task D
10 days
0 slack
10
10
23
17.5
17.5 25.2
25.2 30
Task E
15.5
days
3 slack
5 20.5
Task F
7.67
days
3 slack
20.5 28.2
Task G
4.83
days
3 slack
28.2 33
23
33
Task J
5 days
42.5
slack
44.5 49.5
37.3 59.8
33 37.3
Task H
4.33
days
0 slack
33 37.3
Task I
22.5
days
0 slack
37.3 59.8
37.3 39.3
Task K
2 days
20.5
slack
57.8 59.8
Task L
2 days
45.8
slack
55.8 57.8
59.8 60.8
Task N
1 day
0 slack
7
17.3
Task M
42.5
days
44 slack
49.5 59.8
59.8 60.8
13
PERT Analysis
Step 4: Calculate Mean & Standard Deviation of
Critical Path
0
10
10
23
23
33
Task A
2 days
0 slack
Task B
8 days
0 slack
Task C
13 days
0 slack
Task D
10 days
0 slack
10
23
10
23
33
33 37.3
37.3 59.8
Task H
4.33
days
0 slack
33 37.3
Task I
22.5
days
0 slack
37.3 59.8
59.8 60.8
Task N
1 day
0 slack
59.8 60.8
Mean
()
13
10
4.33
22.5
60.8
Varianc
e
(2)
.11
.11
.44
.44
.44
10.03
11.6
3.4
Standar
d
Deviati
on
()
14
PERT Analysis
Step 5: Plot cumulative normal distribution
1
0.9
0.8
0.7
0.6
0.5
0.4
Cumulative Probability
0.3
0.2
0.1
0
Project Duration
15
PERT Analysis
Hint: To plot the cumulative normal distribution in
Excel, create (and then plot) the following Table:
X-Axis
Y-Axis
50
=NORM.DIST(50,,,1)
51
=NORM.DIST(51,,,1)
70
=NORM.DIST(70,,,1)
16
PERT Analysis
Step 5: Why Normal Distribution?
Central Limit Theorem: The sum of several
random variables with any (though identical)
distribution is approximately normally
distributed
17
PERT Analysis
Step 5: Plot cumulative normal distribution
1
0.9
0.8
There is a 90%
chance of the
project finishing in
less than 65 days.
0.7
0.6
0.5
0.4
Cumulative Probability
0.3
0.2
0.1
0
Project Duration
18
PERT Analysis
Step 5: Plot cumulative normal distribution
1
0.9
0.8
0.7
0.6
There is a 70%
chance of the
project finishing in
less than 63 days.
0.5
0.4
Cumulative Probability
0.3
In Excel, use
=NORMINV(0.70,60.8,3.4)
0.2
0.1
0
Project Duration
19
PERT Analysis
Step 6: Understand Uncertainty
Total Time until
Project is Due
Likelihood of
Finishing the
Project on Time
61
50%
62
60%
63
70%
64
80%
65
90%
69
99%
20
PERT Analysis
Step 7: Planning Project Buffers
Likelihood of
Finishing the
Project on Time
Project Buffer
(Days)
50%
61
60%
62
70%
63
80%
64
90%
65
99%
69
8
21
PERT - Criticisms
PERT assumes that the critical path remains
constant but uncertainty in task durations implies
that critical paths may change
PERT assumes that task durations are
independent of each other but since the same
people were involved in creating different estimates,
and since the same resources may be involved in
different tasks, this assumption may not hold
22
23
0.15
0.1
0.05
0
-0.05
Outcome
24
D2
Score
D1
D2
Score
D1
D2
Score
D1
D2
Score
D1
D2
Score
D1
D2
Score
25
0.15
0.10
0.05
0.00
Outcome
26
28
29
30
Monte Carlo
Simulation
Mean
60.8
61.3
Standard
Deviation
3.4
3.8
31
PERT Likelihood
of Finishing on
Time
Monte Carlo
Likelihood of
Finishing on Time
61
50%
50%
62
60%
59%
63
70%
68%
64
80%
75%
65
90%
82%
69
99%
97%
PERT underestimates
the project duration,
since the method ignores
that critical paths can
change!
32
Project B
Task
Task
Task
Task
Task
Task
Task
Task
Task
Task
Task
Task
Task
Task
33
Likelihood of
Being Critical
A
B
C
D
E
100%
80%
80%
80%
20%
F
G
H
20%
20%
100%
I
J
K
L
M
N
100%
0%
0%
0%
0%
100%
34
Contribution
A
B
C
D
E
1.1%
0.5%
2.9%
3.3%
2.1%
F
G
H
0.8%
0.5%
3.7%
I
J
K
L
M
N
84.9%
0%
0%
0%
0%
0%
35
Task Correlations
High, positive correlation
No correlation
36
1 = 10, 1 = 1
1 2 10 10 20
12 22 2 1.41
0.45
0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
0.45
0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
2 = 10, 2 = 1
= 20, = 1.41
37
1 = 10, 1 = 1
1 2 10 10 20
2 12 22 2 12
2 12 1
1 2 1
0.45
0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
= 0.5
0.45
0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
2 = 10, 2 = 1
= 20, = 1.73
3 1.73
38
Monte Carlo
Likelihood
Monte Carlo
Likelihood with
Correlations
61
50%
50%
52%
62
60%
59%
58%
63
70%
68%
64%
64
80%
75%
69%
65
90%
82%
74%
69
99%
97%
88%
Ignoring
correlations
between task
durations also
underestimates the
project duration
41
Key Take-Aways
Uncertainty
Simple variations in key project parameters can be
easily incorporated into planning
PERT Analysis
Simple and reasonably effective