Professional Proofreading Assignment

This document is a professional proofreading assignment that discusses the process of multi-criteria decision making using AHP (Analytic Hierarchy Process) and Data Envelopment Analysis (DEA). It provides definitions, advantages, and calculations for both methods. The document also highlights the advantages and disadvantages of AHP and DEA models.

1
Professional proofreading assignment
Student’s Name
Course
Institution’s Name
Date

2
Professional proofreading assignment
A. Multi-Criteria Decision Making – AHP
The Multi-Criteria Decision Making (MCDM) is a "method of optimizing complex decisions
with multiple criteria or objectives" and can be divided into two types (Zimmermann, 1991):
Types Description
Multiple Objective Decision
Making
- The search for alternatives that best satisfy the given
purpose in an infinite set of alternatives
- Weighting method, ε-constraint method, multipurpose
linear programming method, etc.
Multiple Attribute Decision
Making
- Determine the order of preference for a finite set of
alternatives
- Target Achievement Evaluation Method, Multifactor
Utility Function Method, Rating Model, Outranking
Method, AHP, etc.
The MODM technique assumes that the alternatives are infinite and finds an option that minimises the
deviation from the set of goals in this infinite set of other options. On the other hand, decision-making
provides a number of alternatives that reflect the most preferred choices among these limited
alternatives. Therefore, this study discusses in depth the process of solving problems concerning the
multi-criteria decisions using AHP (Analytic Hierarchy Process).
1. Definition and Advantage of AHP and Researches of its Application
It is a method to divide the whole process of decision making into several stages and then to
analyse and interpret it systematically to help rational decision-making. The areas that are being utilised
are the government's policy decisions, the establishment of various election strategies, conflict
management and dissolution, preliminary feasibility study and feasibility assessment, budget allocation,
R & D project selection, new product development, Performance evaluation, economic analysis, and so
on (T. Saaty (1977, 1980, 1988, 1995). The advantages of AHP are simple to express decision-making
processes as well as efficient in terms of time and cost which can improve the quality of decision-

3
making. Besides, decision-makers can form a mutual consensus on various interests (Ramanathan,
2001).
2. Pre-Acquisition of AHP
AHP is based on some basic axioms (Vargas, 1990). These axioms are also crucial in the
process of applying real AHP.
Axiom Name of Process Description
1 Reciprocal
Comparison
A mutual comparison of two decision-makers must be
possible and should be able to indicate the degree of
importance. The degree of this importance must establish the
inverse condition. That implies that B would be 1/x times as
important as A given that A is x times important to B.
2 Homogeneity The degree of importance must be expressed through a set of
scales within a limited range. That is, there must be a
comparable range of criteria between the comparison objects.
3 Independence The same level of factors that assess relative importance
should not be related to each other in terms of characteristics
or content.
4 Expectation It is assumed that the hierarchical structure has a complete
structure that meets the rational expectations of decision
makers. That is, the hierarchy must contain all the
considerations that are considered in the decision. On the
other hand, since the number of levels is large and the
hierarchical structure becomes deeper, calculation complexity
is caused.
3. Process of AHP
AHP analysis is performed through the following procedure (Saaty, 1990).
Stage Description
1 Brainstorming Define the problem exactly and clarify the requirements of the problem.
2 Structuring It looks at all the factors related to the problem and constructs a
hierarchy that covers the highest level of the problem from the goal to
the mid-level evaluation item selection and placement to the lowest level
of comparison of the alternatives.

4
3 Weighting In order to determine the degree of importance of the subordinate items
at the lower level based on one item in the middle level, a pairwise
comparison between the items is conducted for all of the subordinate
items, The relative importance of the subordinate evaluation items is
created by the comparison matrix.
4 From the comparison matrix obtained in Step 3, we obtain the relative
estimated weights among the evaluation items and then examine the
consistency of the answers. If there is no consistency, review the results
of the twin comparisons and adjust them to be consistent. Inconsistency
ratios are used as a measure of the consistency of responses. In general,
if the inconsistency ratio is less than 10%, there is no problem in the
consistency of judgment, and if it is more than 20%, the consistency
problem is reviewed.
5 Repeat steps 3 to 4 for the evaluation items at all levels in the hierarchy
set in step 2.
6 Measurement The process of multiplying the relative weights of the evaluation criteria
at a certain level with the relative weights of the dependent criteria in the
lower level. The measurement is taken from the highest level to the
lowest level in sequence and then adding the relative weights of the
alternatives obtained by the evaluation criteria to each alternative And
the relative weight between the alternatives, taking the evaluation criteria
together.
7 Comparing the evaluation scores of the alternatives obtained in Step 6,
the alternative that has the highest score is selected.
8 Feedback Review the overall consistency of the assessment results to determine if
there is a lack of consistency in comparative judgments or if there is a
mistake in the hierarchical structure of the problem from the outset.
B. Data Envelopment Analysis
1. Definition of Data Envelopment Analysis (DEA) and its Advantages
The most significant features of the data envelopment analysis are the decision-making units
(DMU) Unit) in terms of performance after it was first proposed by (Charnes, Cooper, and Rhodhes,
1978). Research using DEA has been conducted in the service industry, public institutions, financial
institutions, shipping and logistics, and manufacturing.
Name of Model Description

5
CCR The proposed CCR model by Charnes et al. (1978) can measure
the DMUs relative efficiency by taking into account a significant
amount of inputs and outputs. This is because some factors such as
constant return to scale (CRS) This is because the efficiency
analysis is based on the linear analysis plan model of the method
of converting the output into a single scale. Therefore, the
efficiency score is only expressed as technical efficiency of the
combination of the scale effect and the technical performance. The
CCR model seeks to maximise the ratio of DMUs under simple
constraints that the weighted sum of outputs for the sum of
weighted inputs of DMUs should not exceed 1 and that the
weights of each input and output element are more significant than
zero Linear planning model. The scale unprofitable is the fact that
not all companies are operating at the optimum range due to the
appropriate model when all companies operate at the optimal
scale, or incomplete competition and financial constraints in
reality.
BCC Banker et al. (1984) presented the BCC model by extending the
CCR model by assuming the VRS (variable Returns to Scale). The
BCC model additionally includes a linear scale with no sign
constraint to allow full-scale variability. In other words, the value
obtained through the BCC model shows only pure technology
efficiency by separating the scale index from the DMU efficiency.
The combination of scale efficiency and pure technical efficiency
is because of the CCR model derived efficiency whereas that
derived from the BCC model is purely technical efficiency. Using
the efficiencies obtained from the CCR and the BCC models,
technology efficiency can be divided into Scale efficiency of pure
technology.
The advantage is that DEA can handle multiple inputs and multiple outputs, which is a nonparametric
method that does not require assumptions concerning the technique of the production function that can
be directly compared between DMUs, and the units of measurement of inputs, and outputs can be
different.

6
2. Calculation of Efficiency
DEA is a 'multi-factor productivity analysis model' that overcomes these problems, which are deployed
in determining the relative DMUs efficiency of similar character. At this time, the efficiency score
between various input and output factors is calculated by the following equation.
Efficiency= weighted average ∑ of outputs
weighted average ∑ of inputs
Farrell (1957) defined efficiency as the relation between the outputs to the input used by the production
organisation. The author used the method to determine the efficiency of the production concept as a
measure of the effectiveness of the production organisation, respectively. The ability of an enterprise or
a public service provider to produce the maximum output at a given input is referred to as technical
efficiency, and the ability to determine the optimal input combination in terms of production factor
prices is called allocative efficiency.
The concept of efficiency in the input space represents the quantity of two production factors
x1 and x2 for producing a single unit of the y outputs, is as follows. An upper right part of the curve SS
'is the production possibility set whose level of output is fixed to 1 unit, and the curve SS' is an efficient
subset of the producible set, frontier), and straight line AA 'is the isochronous line reflecting the price of
the production factor. Since the production Q produces a quantity of output y equal to P while using
only the positive OQ / OP level that is used by the production organisation P, two production factors, x1
and x2, are defined as technical efficiency of P, Is between 0 and 1. Also, since Q 'produces the same
amount as Q, it has the same technical efficiency, and at the same time, Q' can produce two equal
amounts at the cost of OR / OQ lower than Q, so this ratio is called allocation efficiency.

7
AHP and DEA are models developed to yield a final alternative when there are multiple
alternatives. DEA measures relative efficiency rather than absolute efficiency. It is a nonparametric
method, which makes it difficult to perform statistical tests. As DMU increases, the calculation becomes
complicated, and sensitivity to variable selection is pointed out. On the other hand, AHP has the
disadvantage that it is difficult to objectively evaluate each alternative according to the importance of
selection criteria, and bureaucratic characteristics are not reflected in a few opinions. In this study, the
DEA model overcomes the disadvantage that many DMUs can have the highest efficiency by applying
the AHP model to differentiate rankings. The disadvantage of the AHP analysis that the subject of the
evaluator is involved is that the objectivity of the DEA model is utilised respectively.