Multi-Criteria
Decision Making
Submitted To: Dr. Rajesh Rajagopal
Submitted By:
Introduction to Decision Making
● Decision making is the process of making choices by defining a decision, collecting
facts, and evaluating alternative resolutions.
● Decision Making is preeminently a human function. We humans make several
decisions knowingly or unknowingly everyday.
● Decision Making as a structured tool is a complex process of:
1. Selecting criteria
2. Determining alternatives
3. Gathering, evaluating, and processing information.
4. Producing and evaluating partial or immediate results.
5. Reconsidering criteria, alternatives and information on the basis of achieved results.
6. Repeating the process until an actionable outcome, i.e. a decision has been
reached
An example of Decision Making
A school committee is tasked with allocating a fixed number of scholarships to
students based on their performances on the subjects they are being taught e.g.
mathematics, computer science, biology etc.
Decision problem is
● To rank all students from best to worst
● To select the top students as recipients of a scholarship
Types of Decision Problems
1. Choice problem.
2. Sorting problem.
3. Ranking problem.
4. Description problem
The decision making for such problems is difficult because these problems involve
multiple criteria that affect the decision making process.
Systematic approach to decision making involving several criteria: MCDM
MULTI CRITERIA DECISION MAKING
Multiple Criteria Decision Making (MCDM) is about methods for making decisions
where multiple criteria (or objectives) need to be considered together, in order to
rank or choose between the alternatives being evaluated.
The criteria can be both qualitative and quantitative in nature.
These methods have been applied in fields ranging from but not limited to health,
finance, environmental management, urban planning, robotics etc
MCDM serves as an interface between Decision Makers and Analysts, guiding them
to reach a decision when multiple and often conflicting criteria are involved
Key aspect of MCDM is that the final decision need not be the best possible one
but one that is acceptable by all stakeholders
STEPS IN A TYPICAL MCDM PROCESS
1. Identify the problem
2. Formulate the problem
3. Construct the evaluation
4. Reach a final recommendation
SUBSTEPS IN MCDM
FORMULATE THE PROBLEM
IDENTIFY THE PROBLEM
- Identify the decision alternatives and
- Identify the stakeholders
their criteria
- Identify the context of the problem
- Identify the type of decision
- Identify the objective of the problem
decision and its respective
- Managing Multiple DMs and their
properties
different perspectives
SUBSTEPS IN MCDM
CONSTRUCT THE EVALUATION MODEL REACH A FINAL RECOMMENDATION
Choice of Mathematical Model and its Recommendation is presented to the DM
Tuning
- Validates Recommendation
- To reflect the perspective of the DM A
- Revisits precious steps in order to
resolution method needs to be selected
refine the solution
- In order to provide a recommendation to - Asks for additional supporting
the decision problem analysis
Popularity of Different MCDM Methods
Analytical Hierarchy Process
(AHP)
1. AHP method is the most famous MCDM method
2. The AHP method looks at the problem in three parts:
a. Goal
b. Criteria
c. Alternatives
3. AHP algorithm is basically composed of two steps: a. Determining the
relative weights of the decision criteria. b. Determining the relative
rankings (priorities) of alternatives.
Elimination Et Choice Translating
Reality (ELECTRE)
1. Electre method uses Concordance and Discordance Indexes to perform a
pairwise comparison among alternatives. These pairwise comparisons gives the
leading alternative, when one alternative is compared with another alternative.
2. Usually the Electre Methods are used to discard some alternatives to the
problem, which are unacceptable. After that we can use another MCDM to
select the best one.
3. The Advantage of using the Electre Methods before is that we can apply
another MCDM with a restricted set of alternatives saving much time.
4. The ELECTRE method has several unique features generally not found in other
MCDM methods; these are the concepts of outranking and thresholds.
Analytic Network
Process (ANP)
1. ANP can model complex decision problems,
whereas AHP is not sufficient.
2. ANP allows for feedback connections and loops.
3. In ANP, criteria, sub-criteria, and alternatives
are treated equally as nodes in a network.
Technique for Order of Preference by
Similarity to Ideal Solution (TOPSIS)
1. It is a method of compensatory aggregation
that compares a set of alternatives by
identifying weights for each criterion,
normalising scores for each criterion and
calculating the geometric distance between
each alternative and the ideal alternative,
which is the best score in each criterion.
2. This technique is based on the comparison
between all the alternatives included in the
problem,Thus, if one alternative is removed or
added, depending on the situation, the whole
Other Methods:-
DEMATEL
● It is especially useful in analyzing the cause and effect relationships among
components of a system.
● The DEMATEL can confirm interdependence among factors and aid in the
development of a map to reflect relative relationships within them.
● IT can be used for investigating and solving complicated and intertwined
problems.
VIKOR
● The character of VIKOR method is that the VIKOR method can determine a
compromise solution, by using utility weight, the attitude of decision-makers
can be reflected by adjusting utility weight.
PROMETHEE
● PROMETHEE (Preference Ranking Organisational Method for Enrichment
Evaluation) is used in the study to select the best sustainable concept
Conclusion
MCDM: Multi-criteria decision making (MCDM) is an approach designed for the
evaluation of problems with a finite or an infinite number of choices.
AHP method is the most famous and widely used MCDM method.
