Project Risk

Management
Dr W A Tipu
Analytical Hierarchy
Process
Multi-criteria
Decision Making
Analytic Hierarchy
Process

• The Analytic Hierarchy Process (AHP), is a procedure

designed to quantify managerial judgments of the
relative importance of each of several conflicting
criteria used in the decision-making process.
Case Study
While you are working as an active team member in PMO of a construction company. You help senior management
deciding new business

Owing to rise in awareness among masses of projects and people’s demand of higher life standards has brought an
opportunity of more projects.

Your organization has sighted 3 projects:

• Project A is road project
• Project B is high-rise building
• Project C is mass transit bus service lane in urban city

You must choose only one project based on three parameters:

• Profitability
• Technical knowledge available with your company
• Client’s reputation.
Case Study

Project C has 100 million profit

Project A has 80 million profit

Project B has 40 Million profit

Case Study

Your company has:

• Full experience in road sector
• Few projects in buildings
• No project in mass transit bus lane though you have hired a new project manager who has vast experience in that

Client for:
• Project B is very reliable
• Project C is very ambitious and may have some loss of support during the course of work
• Project A is challenging.

Guidelines from top management are that you must go for maximum profitability and then the client’s attitude
should be considered whereas technical knowhow can be acquired from market;

Find out which project is acceptable and less risky?

Step1: List
the Overall
Goal,
Criteria, and
Decision
Alternatives.
Step2: Develop • Rate the relative importance between each pair of decision alternatives.
• The matrix lists the alternatives horizontally and vertically and has the
a Pairwise numerical ratings comparing the horizontal (first) alternative with the
vertical (second )alternative.
Comparison • Ratings are given as follows:
Matrix
 Intermediate numeric ratings of 8, 6, 4, 2
can be assigned

Step2: For example, a numeric rating of 8 will

imply extremely preferred to very strongly
Develop a preferred
Pairwise A reciprocal rating (i.e., 1/9, 1/8, etc.) is
Comparison assigned when the second alternative is
preferred to the first
Matrix
The value of 1 is always assigned when
comparing an alternative with itself.
 In terms of profit:

• Project C has 100 million profit

Pairwise • Project A has 80 million profit

• Project B has 40 Million profit.

Comparison Team has decided that in terms of profitability:

Matrix: • Project C

Profitability • Is moderately preferred to Project A

• Is very strongly preferred to Project B.
• Project A
• Is strongly to very strongly preferred to Project B.
Pairwise
Compariso
n Matrix- Priority Value Assigned
Profitabilit
Same 1
y
Weakly Preferred 2
Moderately Preferred 3
Strongly Preferred 5
Very Strongly Preferred 7
Extremely Preferred 9
Pairwise Comparison Matrix-
Profitability

Project A Project B Project C

strongly to
Project A Same Not Moderately
Very Strongly
Not strongly to
Project B Same Not Very Strongly
Very Strongly
Project C Moderately Very Strongly Same
Pairwise Comparison Matrix-
Profitability

Project A Project B Project C

Project A 1 6 1/3

Project B 1/6 1 1/7

Project C 3 7 1
Pairwise Comparison Matrix-
Profitability

Project A Project B Project C

Project A 1.00 6.00 0.33

Project B 0.17 1.00 0.14
Project C 3.00 7.00 1.00
Step3: Develop a Normalized Matrix

• Divide each number in a Pairwise Comparison Matrix-Profitability

Project A Project B Project C

column of the pair wise
Project A 1.00 6.00 0.33
comparison matrix by its
Project B 0.17 1.00 0.14

column sum. Project C 3.00 7.00 1.00

Column Sum 4.17 14.00 1.48

Step3: Develop a Normalized
Matrix
Normalized Pairwise Matrix-Profitability

Project A Project B Project C

Project A 0.24 0.43 0.23

Project B 0.04 0.07 0.10

Project C 0.72 0.50 0.68

 Average each row of the normalized

Step4: matrix

Develop These row averages form the

the priority vector of alternative

preferences with respect to the
criterion
Priority
Vector The values in this vector sum to 1.
Step4: Develop the Priority Vector

Normalized Pairwise Matrix-Profitability

Project A Project B Project C
Project A 0.24 0.43 0.23
Project B 0.04 0.07 0.10
Project C 0.72 0.50 0.68

Sum 1.00 1.00 1.00

 Priority Vector-
Profitability
Step4:
Develop
the Project A 0.30
Priority Project B 0.07
Vector
Project C 0.63
 The consistency of the subjective input in
the pairwise comparison matrix can be
measured by calculating a consistency
ratio.

Step5:
Calculate the A consistency ratio of less than 0.1 is
Consistency good.

Ratio
For ratios which are greater than 0.1, the
subjective input should be re-evaluated.
Step 5: Calculate the Consistency
Ratio

Step1 Step2: Step3:

For each row of the pairwise For each row, divide its weighted Determine the average, λ max, of
comparison matrix, determine a sum by the priority of its the results of step 2.
weighted sum by summing the corresponding (row) alternative.
multiples of the entries by the
priority of its corresponding
(column) alternative.
Step 5: Calculate the Consistency Ratio

Compute the consistency index, CI, of the n

Step 4: alternatives by:
• CI = (λmax - n)/(n - 1)
Step 5: Calculate the Consistency Ratio

• Step 5: Number of Alternatives (n) Random Index (RI)

• Determine the random
3 0.58
index, RI, as follows:
4 0.9

5 1.12

6 1.32

7 1.24

8 1.41
Step 5:
Calculate • Step 6
the • Compute the consistency ratio: CR =
Consistency CI/RI.
Ratio
Computing the consistency Ratio
• Multiply the pairwise comparison matrix and priority vector

Pairwise Matrix Priority

Vector
1.000 6.000 0.333 X 0.298
0.167 1.000 0.143 0.069
3.000 7.000 1.000 0.632
Computing the
consistency Ratio Pairwise Matrix
1.000 X 0.298 + 0.298 + 0.416 + 0.925
6.000 X 0.069 + 0.211 =
0.333 X 0.632 =

• Multiply the pairwise comparison

0.167 X 0.298 + 0.050 + 0.069 + 0.209
1.000 X 0.069 + 0.090 =
matrix and priority vector 0.143 X 0.632 =
3.000 X 0.298 + 0.894 + 0.486 +
7.000 X 0.069 + 0.632 = 2.013
1.000 X 0.632 =
Computing
the
consistency
Ratio 0.925/0.298 = 3.10
0.209/0.069 = 3.02
• Divide these number by
their priorities to get: 2.013/0.632 = 3.18
 Average the above results to get λmax = 3.10

Computing Compute the consistence index, CI, for two terms

CI = (λmax - n)/(n - 1) = (3.10 - 3)/2 = 0.05
the
consistency
Ratio Compute the consistency ratio, CR, by CI/RI, where RI = .58 for 3 factors:
CR= CI/RI = 0.05 /.58 = 0.086

Since the consistency ratio, CR, is less than .10, this is well within the
acceptable range for consistency.
 Experience of your company
• Full experience in road sector
• Few projects in buildings
• No project in mass transit bus lane though you
have hired a new project manager who has vast
Pairwise experience in that.
ComparisonMatrix:
Technical Team has decided that in terms of
Knowledge Technical Knowledge of Project A:
• Strongly preferred to Project B
• Extremely preferred to Project C.

In turn Project B is very strongly

preferred to Project C.
Pairwise ComparisonMatrix: Technical
Knowledge

Pairwise Comparison Matrix-Technical

Knowledge
Project A Project B Project C
Project A 1.00 5.00 9.00
Project B 0.20 1.00 7.00
Project C 0.11 0.14 1.00
Normalized Matrix: Technical
Knowledge
Pairwise Comparison Matrix-Technical Knowledge
Project A Project B Project C
Project A 1.00 5.00 9.00
Project B 0.20 1.00 7.00
Project C 0.11 0.14 1.00

Column Sum 1.31 6.14 17.00

Normalized Matrix: Technical
Knowledge
Normalized Pairwise Matrix-Profitability
Project A Project B Project C
Project A 0.76 0.81 0.53
Project B 0.15 0.16 0.41
Project C 0.08 0.02 0.06

Sum 1.00 1.00 1.00

 Priority Vector-
Profitability

Project A 0.702
Priority
Vector Project B 0.242

Project C 0.056
Pairwise Comparison Matrix: Client
Project B is very reliable
Project C is very ambitious and may have some loss of
Client for: support during the course of work
Project A is challenging.

Team has decided that in terms of Strongly preferred to Project C

Client’s Reputation Project B is: Very strongly preferred to Project A.

In turn Project C is moderately preferred to Project A.

 Pairwise Comparison Matrix-Technical
Knowledge

Project A Project B Project C

Pairwise Project A 1.00 0.14 0.33
ComparisonM
atrix: Client Project B 7.00 1.00 5.00
Project C 3.00 0.20 1.00

Column
11.00 1.34 6.33
Sum
Normalized Matrix: Client

Normalized Pairwise Matrix-Profitability

Project A Project B Project C

Project A 0.09 0.11 0.05

Project B 0.64 0.74 0.79

Project C 0.27 0.15 0.16

Sum 1.00 1.00 1.00

 Priority Vector-
Profitability

Priority Project A 0.083

Vector:
Client Project B 0.724

Project C 0.193
Pairwise Comparison Matrix: Criteria

Go for maximum profitability

Guidelines from top management Second Priority is the client’s attitude
are that: Technical knowhow can be acquired from market.

Team has decided that in terms of Strongly preferred to Client Attitude

Criteria that Profitability is: Extremely preferred to Technical Knowledge

Client’s Attitude is moderately preferred to Technical Knowledge.

Pairwise Comparison Matrix: Criteria
Pairwise Comparison Matrix-Technical Knowledge
Profitability Client Technical
Knowledge
Profitability 1.00 5.00 9.00
Client 0.20 1.00 3.00
Technical
Knowledge 0.11 0.33 1.00

Column Sum 1.31 6.33 13.00

Normalized Matrix: Criteria
Normalized Pairwise Matrix-Profitability
Technical
Profitability Client
Knowledge
Profitability 0.76 0.79 0.69
Client 0.15 0.16 0.23
Technical
0.08 0.05 0.08
Knowledge

Sum 1.00 1.00 1.00

 Priority Vector-
Profitability

Profitability 0.748
Priority
Vector
Client 0.180

Technical
Knowledge 0.071
 Step6: Develop a
Priority Matrix

After steps 2 through 5 has

been performed for all criteria,
the results of step 4 are The column entries are the
summarized in a priority matrix priority vectors for each
by listing the decision criterion.
alternatives horizontally and
the criteria vertically.
Overall Priority Vector

• The overall priorities are determined by multiplying the

priority vector of the criteria by the priorities for each
decision alternative for each objective.
Overall Priority Vector

Pairwise Comparison Matrix-Overall

Profitability Client Technical Criteria
Knowledge
X
Project A 0.298 0.08 0.70 0.748
Project B 0.069 0.72 0.24 0.180
Project C 0.632 0.19 0.06 0.071
Overall Priority
Vector

• As we can see from the Table

that the highest score (0.512)
Project A 0.288
is of "Project C“ i-e Mass
Transit Bus Service Lane Project B 0.200
• So, this project appears to be
the overall recommendation. Project C 0.512
Thank you