Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 22

TABLE OF CONTENTS

1 DECLARATION

2 ACKNOWLEGEMENT

3 ABBREVIATIONS

4 ABSTRACT

5 PROBLEM STATEMENT/PURPOSE
6 INTRODUCTION
7 Examples of MCDM problems in real life
8 SAW METHOD
9 Technique for Order Preference by Similarity
to Ideal Solution (TOPOSIS METHOD)
10 Analytic Hierarchy Process (AHP)
11 How to Use the Analytical Hierarchy Process
(AHP)
12 REFERENCES
ABBREVIATIONS

1 MCDM - Multi-Criteria Decision-Making

2 MADM- Multiple Attribute Decision Making

3 MODM- Multiple Objective Decision Making

4 SAW - Simple additive weighting

5 AHP - Analytic Hierarchy Process

6 TOPOSIS - Technique for Order of Preference by Similarity


to Ideal Solution

7 V+ - Ideal best

8 V- - Ideal worst

9 S+, S- - Euclidian distance from ideal best to ideal worst

10 PS - Performance score
ABSTRACT

Multi-criteria decision making (MCDM) is a branch of operations research


(OR). The algorithms of the popular MCDM processes (AHP, SAW and
TOPSIS) are explained and their applications in selection of best car,
selection of best mobiles and selection of best land. Multi-Criteria Decision-
making (MCDM) has the potential for improving all areas of decision-
making in engineering, from design to manufacture, but is especially
beneficial for applications in high technology market sectors, where product
differentiation and competitive advantage are often achieved by just very
small gains in material performance. It is also best for middle class families
to choose the best alternatives from various number of alternatives.

Multi-criteria decision-making in general follows six steps including,


(1) problem formulation
(2) identify the requirements
(3) set goals
(4) identify various alternatives
(5) develop criteria
(6) identify and apply decision-making technique
INTRODUCTION

Decision-making on day-to-day basis is a common human practice that


essentially requires one to choose a best alternative among many. Modern-
day decision-making has evolved over the years with early developments
dealing with multi-objective optimization approach in the field of
operations research.

Multi-criteria decision making (MCDM) also referred to as multiple


criteria decision analysis (MCDA), is a research area that involves the
analysis of various available choices in a situation or research area which
spans daily life, social sciences, engineering, medicine, and many other
areas. MCDM is one of the most popular decision-making tools utilized in
various fields.

Since 1950s, multi-criteria decision-making (MCDM) has been practiced


actively by theoretical and applied scientists to test the potential capability
of mathematical modeling of decision-making problem.

Application of multi-criteria decision-making (MCDM) theory is the use of


computational methods that incorporate several criteria and order of
preference in evaluating and selecting the best option among many
alternatives based on the desired outcome. It is applied to different fields to
obtain an optimum solution to a problem where there are many parameters
to consider that cannot be decided by the users’ experiences. The
application gives a ranking result based on the selected criteria, their
corresponding values, and assigned weights.

In real-life problems, there are many critical parameters (criteria) that can
directly or indirectly affect the consequences of different decisions. Stakes
are always high whenever human life is in danger, so it is always important
to make the right decisions. When deciding whether to use a particular
medication, treatment, or medical equipment, not only are the problems with
multiple criteria very complex, but multiple parties are also deeply affected
by the effect.
PROBLEM STATEMENT

MCDM in real life problem solving is important as in many situations of


real-life problems, you would have to evaluate many criteria for making
a decision.
Very often we need to choose something we desire from a few promising
choices. While renewing your wardrobe, you need to
 Decide first which type of new dresses you would require and then,
 Actually, buy the clothing items from the market.

Overall objective of renewing your wardrobe was not very well defined.
You only felt that some of your clothing items are old and not anymore
you would like to use them as also you would like to get hold of a few
items that are in fashion and missing in your wardrobe.
At the first step you had inspected your available dresses and decided on
the types and numbers of new dresses to buy by using your judgment and
conditions of the items available with you. This is an important process
of defining your target objective more specifically, and is a necessary
step for any kind of detailed buying activity.
Thus, we follow the steps in buying or choice cases:
 Analyse and define target objective in as much detail as possible.
For daily life buying or choosing cases, this may be a short step,
though for organizational choice activity, this may be a long-drawn-
out detailed step. Generally speaking, this step of deciding what to
buy is the first sub-step under the category of Problem definition.
The better you define your problem before going into solving
process, more are the chances of your success.
 Knowing the type of items you would buy, now you think a bit and
decide when and from where you would buy the items. This is
the initial information analysis step to further focus on the actual
buying process. In daily life buying activities this won’t take a long
time, But in organizational procuring and acquiring cases, a detailed
analysis and decision making usually will be required. Generally
prospective buying sources will be more than one. Though we are
specifying a separate step for clarity, this in fact is a second sub-
step in the category of Problem definition.
 Choosing or selecting: This is the main thing you have to do now.
You already know what to buy and from where to buy. You arrive
at a particular shop and start selecting (to keep things simple we
assume your target shop to be only one).

In an MCDM problem, you have thus three major items to consider,


 Goal or objective: you have to analyse and decide on your goal or
objective as precisely as possible. There are real life situations
where objectives may be more than one. Such problems are
classified as Multi-objective Multi-criteria Decision-Making
problems. This first step is important because if you are not too sure
of your objective to start with, your chances of reaching the desired
solution will be bleak.
 Set of criteria or preferences: based on which you would select
from the set of choices to fulfil your objective. You need to form
this set of preferences as clearly as possible. Additionally, the set
of preferences should be comprehensive, should not be too large
in number and most importantly, should not depend on each
other. In other words, the set of evaluation criteria should be
exhaustive and independent of each other as also should not be too
many to make the evaluation process confusing. In real life, a
golden rule is to select the few most important criteria – not too
many. Already the problem is subjective and complex. With too
many criteria, the choice evaluation may go out of hand altogether.
 Set of Choices or possibilities: from which you would choose: in
the same way that you identify the most suitable set of evaluation
criteria, you would choose the promising set of choices
exhaustively but not too large in number. The final evaluation of
each choice against the set of evaluation criteria should be simple
and clear enough to understand the process of selection as also the
suitability of the final selection. To reduce the number of choices,
in many cases an intermediate step or technique of short-listing is
applied where the number of choices is large.
Examples of MCDM problems in real life
 MCDM for Wardrobe renewal: We have already touched the
problem of renewing your wardrobe. As a first activity you need to
decide what and how many to buy. This may not be the final decision
but it is necessary to devote some time on examination of your
existing wardrobe and analysing the new requirements to form an
idea of your requirements before you go to the shop to buy. When in
a shop, as you have already identified the choice types as also your
preferences on choosing, you would go through a process of one or
two stages of short-listing before choosing the items fulfilling your
desire. While going through the selection process, you may further
refine your preferences and even modify your objective to some
extent.
 MCDM for Selecting employees: in any organization this is the job
of HR department. Depending on the job specification, suitable
candidates are selected out of many aspirants. If applicants are large
in number, selection is carried out through a number of stages using
elimination and short-listing techniques. The final stage is always the
most thorough in identifying evaluation criteria and the actual
evaluation process. Where creativity is the crucial desired ability,
advanced organizations may ignore academic performance and
directly judge the creativity of the candidates. Taking up a routine
path, another organization with same requirement of creativity may
eliminate candidates not having a minimum academic performance
thus missing potential candidates with great creativity. This is
interdependence of the real objective and identifying the criteria of
choice. Without a clear understanding of this interdependence, a sub-
par selection is probable.
SIMPLE ADDITIVE METHOD (SAW)
In decision theory, the weighted sum model (WSM), also called weighted
linear combination (WLC) or simple additive weighting (SAW), is the best
known and simplest multi-criteria decision analysis (MCDA) / multi-criteria
decision-making method for evaluating a number of alternatives in terms of
a number of decision criteria.
Simple Additive Weighting (SAW) which is also known as weighted linear
combination or scoring methods is a simple and most often used multi
attribute decision technique. The method is based on the weighted average.
An evaluation score is calculated for each alternative by multiplying the
scaled value given to the alternative of that attribute with the weights of
relative importance directly assigned by decision maker followed by
summing of the products for all criteria. The advantage of this method is
that it is a proportional linear transformation of the raw data which means
that the relative order of magnitude of the standardized scores remains
equal.
STEPS INVOLVE IN SAW METHOD TO SOLVE MCDM
PROBLEM: -
a. Determine the criteria that will be used as a reference in
making decisions, namely Ci.
b. Determine the suitability rating of each alternative for each alternative.
c. Making a decision matrix based on the criteria (Ci), then normalizing
the matrix based on the equation adjusted for the type of attribute (profit
attribute or cost attribute) in order to obtain a normalized matrix R.
d. The final result is obtained from the ranking process, namely the addition
and multiplication of the normalized matrix R with the weight vector so
that the largest value is chosen as the best alternative (Ai) as a solution.
The formula for carrying out the normalization is as follows:-
𝑋𝑖𝑗 /max 𝑋𝑖j if j is the benefit attribute
Rij= {
min 𝑋𝑖𝑗/ 𝑋𝑖𝑗 if j is the cost / non beneficial attribute

Where Rij is a normalized performance rating; Xij is the attribute value of


each criterion; Max Xij is the greatest value of each criterion; Min Xij is the
smallest value of each criterion; Benefit is the greatest value is the best; Cost
is the smallest value is the best. Rij is the normalized performance rating of
the alternatives Ai on attribute Cj; i = 1,2,…, m and j = 1,2,…, n. The
preference value for each alternative (Vi) is given as:
n
𝑉𝑖 = ∑ 𝑤𝑗 𝑅ij
J=1

Where Vi is the ranking for each alternative, Wj is the weighted value of


each criterion; Rij is the normalized performance rating value. A larger Vi
value indicates that the alternative Ai is preferred.
Technique for Order Preference by Similarity to Ideal Solution
(TOPOSIS METHOD)

Technique for Order Preference by Similarity to Ideal Solution


(TOPSIS) came in the 1980s as a multi-criteria-based decision-making
method. TOPSIS chooses the alternative of shortest the Euclidean distance
from the ideal solution and greatest distance from the negative ideal
solution.
To make this definition easier, let’s suppose you want to buy a mobile phone,
you go to a shop and analyse 5 mobile phones on basis of RAM, memory,
display size, battery, and price. At last, you’re confused after seeing so many
factors and don’t know how to decide which mobile phone you should
purchase. TOPSIS is a way to allocate the ranks on basis of the weights and
impact of the given factors.

 Weights mean how much a given factor should be taken into


consideration (default weight = 1 for all factors). like you want RAM
to have weighed more than other factors, so the weight of RAM can be
2, while others can have 1.
 Impact means that a given factor has a positive or negative impact.
Like you want Battery to be large as possible but the price of the
mobile to be less as possible, so you’ll assign ‘+’ weight to the battery
and ‘-‘ weight to the price.
This method can be applied in ranking machine learning models on basis of
various factors like correlation, R^2, accuracy, Root mean square error, etc.
The main advantage of the method is that it limits the subjectivity introduced
by policy makers bringing it down mainly to the stage of determining the
weights of the criteria. Other benefits of the method include
– a simple calculation procedure that is easy to program.
– the results obtained for all the solutions can be visualized with the use of
a polyhedron, at least for any two dimensions.
STEPS TO IMPLEMENT THE TOPOSIS METHOD :-

STEP 1:-Calculating Normalized Matrix and weighted Normalize matrix.


We normalize each value by making it: where m is the number of rows in the
dataset and n is the number of columns. I vary along rows and j varies along
the column.

STEP 2 Calculating Ideal Best and Ideal worst and Euclidean distance for
each row from ideal worst and ideal best value. First, we will find out the
ideal best and ideal worst value: Now here we need to see the impact, i.e. is it
‘+’ or ‘-‘ impact. If ‘+’ impact Ideal best for a column is the maximum value
in that column and the ideal worst is the minimum value in that column, and
vice versa for the ‘-‘ impact.
Now we need to calculate Euclidean distance for elements in all rows from
the ideal best and ideal worst, Here diw is the worst distance calculated of
an ith row, where ti,j is element value and tw,j is the ideal worst for that
column. similarly, we can find dib, i.e. best distance calculated on an ith row.

STEP 3:- Calculating Topsis Score and Ranking. Now we have


Distance positive and distance negative with us, let’s calculate the Topsis
score for each row on basis of them.
TOPSIS Score = diw / (dib + diw) for each row
Now rank according to the TOPSIS score, i.e. higher the score, better the
rank.
Analytic Hierarchy Process (AHP) METHOD

The Analytic Hierarchy Process (AHP) is a method for organizing and


analyzing complex decisions, using math and psychology. It was developed
by Thomas L. Saaty in the 1970s and has been refined since then.

It contains three parts: the ultimate goal or problem you're trying to solve, all
of the possible solutions, called alternatives, and the criteria you will judge
the alternatives on. AHP provides a rational framework for a needed decision
by quantifying its criteria and alternative options, and for relating those
elements to the overall goal.

Drawbacks of AHP
The AHP method has its own issues. The method involves higher level
mathematics. It is based on the concept of eigen vectors. It is for this reason
that performing the calculations pertaining to AHP on an Excel sheet are an
ordeal. However, of late software tools have been developed that can perform
the calculations. The managers therefore just need to be aware of the AHP
process, the calculations are automated.
How to Use the Analytical Hierarchy Process (AHP)

Step 1: Define Alternatives


The AHP process begins by defining the alternatives that need to be
evaluated. These alternatives could be the different criteria that solutions
must be evaluated against. They could also be the different features of a
product that need to be weighted to better understand the customers
perception. At the end of step 1, a comprehensive list of all the available
alternatives must be ready.

Step 2: Define the Problem and Criteria

The next step is to model the problem. According to AHP methodology, a


problem is a related set of sub problems. The AHP method therefore relies on
breaking the problem into a hierarchy of smaller problems. In the process of
breaking down the sub-problem, criteria to evaluate the solutions emerge.
However, like root cause analysis, a person can go on and on to deeper levels
within the problem. When to stop breaking the problem into smaller sub
problems is a subjective judgement.

Step 3: Establish Priority amongst Criteria Using Pairwise Comparison


The Fundamental Scale [Saaty, 1980]

Intensity
of
Importance
Definition Explanation
on an
absolute
scale

Two activities contribute equally to the


1 Equal Importance
objective

Moderate importance of Experience and judgment


3
one over another strongly favour one activity over
another
Essential or strong Experience and judgment
5
importance strongly favour one activity over
another
An activity is strongly favoured and its
7 Very strong importance
dominance demonstrated in practice

The evidence favouring one activity


9 Extreme importance over another is of the highest possible
order of affirmation

Intermediate values
2,4,6,8 between the two adjacent When compromise is needed
judgements
The AHP method uses pairwise comparison to create a matrix using saaty fundamental
scale.

Step 4: Derive the properties of weight criteria

Now need to verify this pair-wise matrix is consistency or not, for this the
following procedure need to be adapted.
The values obtained from the cost, safety, comfort and mileage are need to be
estimated and their sum also calculated as shown above.
Step 3: Consistency Check

The consistency ratio should be less than 0.1, then only the pair wise
comparison satisfies. Then the weights will be takes as the values from the
A2 matrix.
Step 4: Derive overall priorities and Final decision
Weight (Cost) = 0.059246; Weight (Safety) = 0.4828; Weight (Comfort) =
0.31385;
Weight (Mileage) = 0.1441, these are the finalized weights for the problem
and the weights overall sum should be equal to one.

Now, for the car selection based on the criteria a decision matrix need to be
framed, as shown below.

Generally, the Cost of car is a non-beneficial attribute (lower the better) the
remaining Safety, comfort and mileage are beneficial attributes (higher the
better). Need to perform the calculations as shown here.
The final synthesis of model is shown here, the weights and the normalized
values later need to estimate the final value for ranking.

CAR1 =(0.75*0.059246)+(1*0.4828)+(0.93*0.31385)+(0.81*0.1441)= 0.93


68
CAR2=(1.0*0.059246)+(0.75*0.4828)+(1.00*0.31385)+(0.93*0.1441)=0.86
CAR3=(0.86*0.059246)+(0.63*0.4828)+(0.87*0.31385)+(1.0*0.1441)=0.76
86
Finally, CAR 1 is selected as the best alternative among four.
REFERENCES

[1] Ali Jahan, ... Marjane Bahamians, in Multi-criteria Decision Analysis


for Supporting the Selection of Engineering Materials in Product Design
(Second Edition), 2016
[2] Are MCDM methods useful? A critical review of Analytic
Hierarchy Process (AHP) and Analytic Network Process (ANP
(Reviewing editor)
Article: 1623153 | Received 30 Dec 2019, Accepted author version posted
online: 26 May 2019, Published online:31 May 2019
[3] https://www.passagetechnology.com/what-is-the-analytic-hierarchy-
process.
[4] Multicriteria decision making (MCDM): A framework for research and
applications
www.researchgate.net/publication/224570929_Multicriteria_decision_m
aking_MCDM_A_framework_for_research_and_application

[ 5] The Journal of Multi-Criteria Decision Analysis (JMCDA) was


launched in 1992, with an explanatory byline ‘Optimization, Learning and
Decision Support’ added with a restructuring of the editorial board in 2009
https://onlinelibrary.wiley.com/journal/10991360

You might also like