SSG 311:
Mathematical
Modelling for
Artificial Intelligence
Systems
MODULE 2
Analytic Hierarchy Process
Intended Learning Outcomes
At the end of this course, you should be able to do the
following:
1. Know what a decision system is
2. Know what AHP is
3. Understand why AHP is important in their daily activity
4. Know the various steps involve in an AHP system
5. Implement AHP in manufacturing system
Table of Content
What is AHP
General Idea
Example:
Ranking Scale for Criteria and Alternatives
Pros and Cons of AHP
3
Analytic Hierarchy Process(AHP)
The Analytic Hierarchy Process (AHP) is a structured
technique for organizing and analyzing complex decisions.
It is a practical approach to solving complex decision problems
involving the comparisons of attributes or alternatives.
It was developed by Thomas L. Saaty in the 1970s.
Application in group decision making.
4
Analytic Hierarchy Process (Cont.)
AHP states that if you have a decision problem there is one
thing that is obvious (i.e. you have choices or alternatives),
hence you need to know the contributory factors, you need to
arrange your problems in hierarchical form, compare your
factor two at a time, compare the attributes and the alternatives
with respect to attributes two at a time, at the end you get the
global view.
5
Application of AHP
Wide range of applications exists:
◦ Selecting a car for purchasing
◦ Deciding upon a place to visit for vacation
◦ Deciding upon an MBA program after graduation.
◦ Deciding upon the best overall decision.
6
General Idea
AHP algorithm is basically composed of two steps:
1. Determine the relative weights of the decision criteria
2. Determine the relative rankings (priorities) of alternatives
Both qualitative and quantitative information can be compared
by using informed judgments to derive weights and priorities.
7
Steps for AHP
1. Develop the hierarchical structure for the decision problem.
2. Determine the relative weights of each alternative with
respect to the characteristics and sub characteristics in the
hierarchy.
3. Determine the overall priority score of each alternative.
4. Determine the indicators of consistency in making pair wise
comparisons of the characteristics and alternative.
5. Make a final decision based on the result.
8
Hierarchical Steps for decision and
alternatives
Objective
Several
Attribute A Attribute B Attribute C layers of
attributes
One layer of
Alternative 1 Alternative 2 alternative
Compare the attributes two at a time with respect to the objective.
Compare the alternative with respect to to each9other.
Ranking Scale for Criteria and Alternatives
10
FLOW CHART OF AHP METHODOLOGY
Specify Objective
Identify Alternatives
Identify Relevant Factors
Develop or Revise matrix of
pair wise comparison of factors
Compute Normalised Relative
weights of Factors
Compute consistency Ratio of
pair wise comparison of factors
NO
Is 0<Ratio<0.01?
A
A
Develop or Revise matrix of pair wise
comparison of Alternatives with respect to
factors
Compute consistency Ratio of pair wise
comparison Alternative Select Next Factor
NO
Is 0<Ratio<0.01?
YES
Have the Alternatives been compared with NO
respect to all factors?
YES
Compare overall Desirability weight of
each Alternative
12
Select Best Alternative
13
14
15
Knowledge and historical data is required.
Attributes
A B C
Attributes A 1 XAB XAC
B XBA=1/XAB 1 XBC
C 1/XAC 1 /XBC 1
16
If CR < 0.10, the Rankings are consistent.
If CR >= 0.10 , the comparisons should be recalculated.
Random Consistency Index
(RI)
25
Which design?
Material Cost (C ) (comparison matrix)
Plate Welds Plate Rivets Casting
Plate Welds 1 1 0.33
Plate Rivets 1 1 0.33
Casting 3 3 1
Sum 5 5 1.667
Normalized C
Plate Welds Plate Rivets Casting P
Plate Welds 0.2 0.2 0.2 0.2
Plate Rivets 0.2 0.2 0.2 0.2
Casting 0.6 0.6 0.6
26 0.6
where P is Design Alternative Priorities
Weighted Sum (W or P*C)
1 1 0.33 0.2 0.6
1 1 0.33 0.2 0.6
3 3 1 0.6 1.8
Ws * 1/W
0.6 5 3
0.6 5 3
1.8 1.67 3
CI= (3-3) /2=0
27
CR= (0/0.58) = 0
Final Rating Matrix
Transpose the final rating matrix
29
This can be represented graphically
0.6
0.5
0.4
0.3
Series1
0.2
0.1
0
Welds Rivets Casting 30
More about AHP: Pros and Cons
• It allows multi criteria decision making.
• It is applicable when it is difficult to formulate criteria
evaluations, i.e., it allows qualitative evaluation as well as
Pros quantitative evaluation.
• It is applicable for group decision making environments
•There are hidden assumptions like consistency.
Repeating evaluations is cumbersome.
• Difficult to use when the number of criteria or
alternatives is high, i.e., more than 7.
Cons
•Difficult to add a new criterion or alternative
• Difficult to take out an existing criterion or alternative,
since the best alternative might differ if the worst one is
excluded. 31
GROUP PROJECT
As the presidential adviser, “Using AHP, advice the federal
government on how to go about managing the present
economic situation such that the masses will not be affected
negatively (for instance the removal of subsidy, the increase in
prices of commodities and inflation)
THIS PROJECT WILL BE PRESENTED ON THE 4TH OF DECEMBER
2023.
32
ASSIGNMENT - ALL
USING AHP METHODOLOGY, DESIGN HOW YOU MAKE YOUR CHOICE IN THE FOLLOWING
PROBLEMS:
1. WETHER TO TAKE UP YOUR SIWES INTERN WITH A SOFTWARE COMPANY
2. WETHER TO TAKE UP TEACHING APPOINTMENT WITH FEDERAL GOVERNMENT AS
A CAREER
33
REFERENCE
Expert Systems Applications in Engineering and Manufacturing by Adedeji
B. Badiru, 1992. Prentice Hall, pp. 36 – 55
Analytic Hierarchy Process for Project selection by Indian Institute of
Technology Kanpur. Prof. Raghunanda Sengupta
34
