Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
By
MASTER OF SCIENCE
IN
PRODUCTION ENGINEERING
By
Ayman Mostafa Mohamed Abd AL-khalek
B.Sc. Industrial Engineering
A Thesis Submitted to the Faculty of Engineering at Fayoum University in
Partial Fulfillment of the Requirements for the Degree of
MASTER OF SCIENCE
IN
PRODUCTION ENGINEERING
Professor of Mathematical,
Mathematical and Physics
Department, Faculty of
Engineering, Fayoum University
2014
By
Ayman Mostafa Mohamed Abd AL-khalek
B.S.c Industrial Engineering
A Thesis Submitted to the Faculty of Engineering at Fayoum University in
Partial Fulfillment of the Requirements for the Degree of
MASTER OF SCIENCE
IN
PRODUCTION ENGINEERING
Approved by the
Examining Committee :
Prof. Dr. Hazim Aly Attia, Chairman
ii
TABLE OF CONTENTS
LIST OF FIGURES.....
Vii
Xv
11
13
14
14
15
16
18
19
19
20
21
23
24
27
27
iii
28
28
30
32
33
33
34
34
35
36
38
40
40
41
42
44
46
49
49
50
54
54
57
59
59
69
72
73
73
73
78
78
79
80
80
81
82
83
85
86
86
86
88
88
88
90
91
93
94
97
97
102
104
109
109
110
vi
LIST OF FIGURES
Fig. 2.1: The BSC Framework....
23
27
38
43
Fig. 4.1: Schematic Diagram for The Fuzzy (GMM) -TOPSIS Integrated
Approach................
53
57
58
60
66
67
71
77
78
81
90
97
101
102
102
106
107
107
108
vii
108
LIST OF TABLES
Table 1.1: Summary of The Literature Survey.....
13
36
37
39
41
46
51
55
55
56
56
58
59
62
62
Table 4.10: The Weights Optained from Fuzzy (GMM) of Main Criteria.
62
Table 4.11: represent priority vectors for criteria, sub-criteria and alternatives.
63
Table 4.12: Overall Weight Vector for The Alternatives with Respect to The
Criteria......
63
64
64
64
65
65
65
66
66
67
68
viii
68
70
70
71
83
89
89
95
98
98
Table 6.6: The Weights Optained from Fuzzy (GMM) of Main Criteria
98
99
99
99
100
100
103
104
Table 6.14: The Performance of Both Faculties of Engineering (A) and The
Milestone (B)....
ix
104
ABBREVIATIONS
ABC
Activity-Based Costing
AHP
ANP
BCBC
Cost
CCMA
CEO
CI
Consistency Index
COM
Cost of Material
CMP
CR
Consistency Ratio
CT
Customer
DC
Design Concept
DEA
DEMATEL
Ergonomic
ED
Euclidean Distance
EFQM
ERP
ETD
Easy To Dismantle
ETR
Easy To Repair
ETS
Easy To Storage
ETT
Easy To Transfer
ETU
Easy To Use
EVM
FAHP
FL
Financial
FPM
FPPM
FUNARBE
GMM
GRA
HBR
IP
Internal Process
ISO/IEC
IT
Information Technology
KIC
Fracture Toughness
KPI
L&G
LLB
LW
Light Weight
Maintenance
MADM
MC
Management Commitment
MCDA
MCDM
Multi-Criteria Decision-Making
xi
MODM
NSE
No Sharp Edge
Performance
PPM
R&D
RI
ROA
Return On Assets
ROCE
ROE
Return On Equity
ROI
Return On Investment
SBU
Safety
SAW
SCM
SF
Strong Framework
SPP
ST
Stability
SWOT
TD
Total Deviation
TOPSIS
TQM
USA
UWSENE
xii
LIST OF SYMBOLS
A1
Alternative 1
A2
Alternative 2
aij
a11
a1n
an1
ann
c1
Criterion 1
c2
Criterion 2
C*
fj
Criterion function
fc
Compromise solution
f*
Closest solution
Lower limit
Medium limit
NV
Normalized Vector
rij
Si
S*
S-
Upper limit
Ui
W1
Weight of criterion 1
W2
Weight of criterion 2
xiii
ACKNOWLEDGEMENTS
Primarily, I dedicate this dissertation to ALLAH by whose grace all things are
possible. Without his constant presence, I would not have been able to complete
this thesis.
My parents have always supported my desire to seek further knowledge, and
gave me the means to do so. I would like to thank them for their love and support
over the years.
I am very lucky to have my thesis done under the supervision of Professor
Hazem Aly Attia for his guidance, patience, and continual support.
I would like to thank my supervisor, Dr. Mohammed Fahmy, for his commitment
to helping me, his leadership, and for being such an inspiration.
Special thanks to Prof. Dr. Attia Hussein Gomaa and Prof. Dr. Mostafa Abdeen
for there time and efforts on examining the present work.
To my good friend, Hagag Maher, thank you for being at my side throughout the
years of this work. We were a great team. I will not forget that.
Last but not the least; I would like to express my deep appreciation for those who
helped me throughout my way to finish my thesis. Without their great assistance,
encouragement and follow up, this work would not have been achieved.
xiv
ABSTRACT
cause-effect
relationships,
implementation,
time-dimension,
estimate BSC indicators weights. The new proposed model is modeled and
analyzed using MATLAB.
The new prioritization model solves the main problem of BSC technique with
high accuracy. Verification of the proposed Fuzzy (GMM) - TOPSIS model is
performed in several cases.
The second objective of this thesis adds new perspective to treat limitation in
balanced scorecard implementation. The concept of management commitment is
the degree of members' recognition of organizational goals and values that they
are willing to work extraordinarily hard to help the organization complete its
goals. The origin BSC, perspective "management commitment" and prioritization
technique Fuzzy (GMM) - TOPSIS are the new integrated BSC model.
The third objective of this thesis establishes the development of BSC within an
organization and applies practical research. The proposed BSC model is applied
in Faculty of engineering (A) and results of the performance are found out. The
results show that, the balanced scorecard is a successful and acceptable tool for
organizations performance measurement and excellent tool to translate
organizations strategies to measurable objectives. Results of proposed Fuzzy
(GMM) - TOPSIS prioritization model are compared with other reported
research works in the literature. Sensitivity analysis for the model provides the
smallest ED among other model which means more accurate results of priority.
The results from Fuzzy (GMM)-TOPSIS model found out that management
perspective has the first priority which means that it is the most important
component of the five balanced scorecard perspectives of the faculty
performance. The most significant advantage of the use of the balanced scorecard
is that it provides a wider development of metrics that are closely connected to
the strategic goals of organization (here faculty of engineering).
xvi
CHAPTER 1
INTRODUCTION AND PROPLEM STATEMENT
1.1 Introduction
Organizations have always found it difficult to balance pressing operational
concerns with long term strategic priorities. The tension is crucial: World-class
processes won't prompt to success without the right strategic direction, and the
best strategy in the world will get nowhere without strong operations to execute
it. Considering the importance of strategic planning in organizations and
producing the competitive advantage in them and actually, now a day the
organization is moving in a competitive, and complex environment and there is a
transaction among them. The senior managers and all those looking for
comprehensive picture of present situation of the company a clear understanding
of present situation of the company and a clear understanding of its future image
need some information more than just Standards in financial operation to assess
the strategic operation and long term view of the company and also to achieve
operational strategies. [1]
Miscellaneous types of tools are offered for this process, Balanced Scorecard is a
proper tool for evaluating and designing of operational strategies. This tool was
introduced by Kaplan and Norton in 1992, for the first time. BSC is a conceptual
frame work and its function is to translate strategic objectives of a company into
a set of operational attributes. These indices are usually selected from four
financial, customer, internal processes and
perspectives. Many attributes were used for the advancement of the company in
the direction of its perspective. Some other attributes are used for evaluation of
company development in accessing to long-term objectives. Furthermore, BSC
helps the managers to identify the lagging and leading attributes in their company
[2].
The traditional BSC contains four categories (i.e., financial, customer, internal
business processes, and organizational learning and growth activities), Kaplan
and Norton [3]. This study added a new perspective to treat limitation in
balanced
scorecard
implementation.
This
perspective
is
management
using
the
balanced
scorecard
could
effectively
enhance
accomplishment of strategic goals and performance; this was affirmed by DerJang & Hsu-Feng. Kaplan & Norton in their research explained the development
process of the BSC in order to ensure attainment of organization goal using
banks and insurance industries. The BSC is a tool that links strategies to
organization goals. According to Ali-Rahimi, balanced scorecard provides a
mechanism to align the activities and processes of different groups with long
term goals of the organization [8]. He combined the EFQM and BSC models to
improve the performance of the organization. Bernard in his studies found out
that only one of third of large companies uses the BSC but about one of fifth has
adopted the latest version of the BSC which makes use of Strategy Maps. To
attain this, strategic plans are made by the organization such that the roles of
employees are clearly stated in line with expected performance goals. Strategy
according to Horngreen is making a specification of how the organization can
Non-profit organizations usually take their mission based on reducing their costs,
improving quality and doing their works more efficiently, hence the greatest
difference between businesses and non-profit organizations is in their missions
Kettunen [21]. Therefore when using the BSC in the field of nonprofit
organizations and public sector, the financial perspective ought to have a inferior
role. Apart from the banking sector, many researchers were impressed by the
BSC. Greiling performed an explorative empirical study on a sample of 20 nonprofit organizations in the social services sector in Germany [22]. In a study
conducted by Kollberg and Elg, the authors attempted to identify the major
characteristics of the BSC application in health care organizations in Sweden
[23]. Sunisa Atiwithayaporn and Wanchai Rivepiboon developed a model
consisting of the well-known standard procedure; ISO/IEC12207, and the
Balanced Scorecard to fulfill missing dimensions for software projects [24].
the general body of BSC and deals with the all perspectives [48]. Hossein
Yavarian et. al. proposed an integrated approach for the balanced scorecard
(BSC) and Data envelopment analysis DEA [49]. Reza Abbasi et. al. suggests
that an interested balanced scorecard tool using the AHP -TOPSIS technique for
weighting and ranking the alternatives[50]. Saeid Saeida Ardekani et. al. studied
Performance evaluation of Ceramic and Tile Industry by using
Balanced
scorecard (BSC) interested with Fuzzy analytic hierarchy process (FAHP), Fuzzy
VIKOR [51]. Javad Dodangeh et. al. demonstrated a model for selection and
ranking of strategic plans in Balanced Scorecard using Topsis method and Goal
Programming model [1]. Javad Jassbi et. al. discussed A Fuzzy DEMATEL
framework for modeling cause and effect relationships of strategy map in
balanced scorecard [52]. Gwo-Donq Wu et. al adopted an integrated approach for
balanced scorecard and Fuzzy hierarchy, analytic network process (ANP) for
new product development projects selection[53]. Mehrzad Navabakhsh et. al.
used integrated model of balanced scorecard and analytic hierarchy process
AHP- SAW to evaluate appraise Iran- agency insurance companies [54].
As it was stated in the previous section, there are many MADM methods
available in the literature. Each method has its own characteristics. There are
many ways can classify MADM methods. One way is to classify them according
to the type of the data they use. That is, we have deterministic, stochastic, or
fuzzy MADM methods. However, there may be situations which involve
combinations of all the above (such as stochastic and fuzzy data) data types.
Another way of classifying MADM methods is according to the number of
decision makers involved in the decision process. Hence, we have single decision
maker MADM methods and group decision making MADM. In this study we
concentrate our attention on single decision maker deterministic MADM
methods. Finally, it should be stated here that there are many other alternative
ways for classifying MADM methods. However, the previous ones are the most
widely used approaches in the MADM literature. To achieve the best possible
result from a more effective performance, this study compared between them and
selected a suitable method to measure the weights.
10
hierarchy
for
non-profit
organizations.
Regarding
the
perspectives is problematic and the design techniques are poor in illustrating the
dynamics of a system (absence of feedback loops).The present work will take
this draw back into consideration as will be shown later.
The frame work of balanced scorecard is divided into four perspectives Finance,
Customer, Internal Processing, Learning and Growth. This study developed a
new
perspective
"management
commitment"
which
will
support
the
Profit
No.
Non-profit
Application
Automotive industry
Information Technology
SCM
R&D of Energy
Commercial Bank
Commercial Banks
Performance Evaluating of
Electronic dep.
Commercial Bank
social services
Health Care Services
ISO/IEC12207 software
projects
Developing a new
perspective for BSC
Tool
Educational
AHP-BSC
Fuzzy-BSC
DEA-BSC
AHP-TOPSIS-BSC
Fuzzy(AHP-VIKOR)-BSC)
Fuzzy-DEMATEL-BSC
Fuzzy-ANP-BSC
AHP-SAW-BSC
Add new perspective
13
References
11
12
13
14
17
18
19
20
22
23
24
25, 26, 27, 28, 29, 30,
31, 32, 33, 34, 35, 36,
37, 38, 39, 40, 41, 42,
43, 44, 45
47, 48, 49
companies,
based
on
their
experience
of
Balanced
Scorecard
underlying
factor
behind
these
failures
is
ineffective
analysis (DEA), AHP, ANP, Fuzzy theory, with BSC model to evaluating
projects performance as shown in previous section. Although a good amount of
research work to develop BSC model, and prioritizing the perspectives of BSC
by using different MCDM methods, there is still a need to employ a new simple
model and sensitive and systematic mathematical approach to guide the decision
maker in taking an appropriate decision.
The methodological approaches in the numerous case studies of BSC
implementation projects vary significantly in the sequence, content and number
of implementation steps and phases. These approaches are applicable to
particular companies and market segments rather than attempting to provide
generalized knowledge. For example Letza [63] uses a six step approach
methodology for implementing the BSC, Ahn [64] uses a six phases approach to
implement a BSC, and Brewer [65] proposes a four step value dynamics
framework for translating strategy into measures. Lohman et al. [66] Proposes a
nine step approach for BSC implementation and Alexandros et al. [67] uses a six
phases approach along to main axis to implement BSC.
Depending on these the previous studies, this work proposes ten steps integrated
model for evaluation of organizations performance. Integrated model uses BSC
as strategic management system to measure and analyze the organization
performance. BSC perspectives and its indicator weights estimated and
optimized using the proposed Fuzzy (GMM) - TOPSIS prioritization model.
15
Appendix B lists the questionnaires which used in the practical case study.
Appendix C the published papers of this thesis.
17
CHAPTER 2
PERFORMANCE MEASUREMENT USING BALANCED
SCORECARD
2.1
Introduction
2.2
BALANCED SCORECARD
employees,
owners and
stakeholders
[71]. If performance
measurement is executed properly, the correct data generated will inform the user
19
of the information what the status of the business is; how it is faring; and where it
is going. A performance-measurement system enables an enterprise to plan,
measure, and control its performance according to a pre-defined strategy. In
short, it enables a business to achieve the desired results and to create
shareholder value. An example of a multi-perspective, performance-management
tool used in organizations is the BSC that offers four perspectives on measure
performance.
22
24
purely logical
(e.g.
relationships developed
from
2.3
The BSC as per Kaplan and Norton [3] is made up of four perspectives; these
four perspectives are driven by the company vision and mission. This ensures
that the measures and goals in each perspective lead the company to its ultimate
vision. Figure 2.2 illustrates the relationship between the various components of
the BSC. These will be discussed in detail.
VISION
Mission
Customer
Financial
Internal process
stakeholders, and provide direction for the future state of the organization. The
organization should scan the future and define its role in it. Once the vision
becomes clear, the organization can work toward achieving it by developing the
strategic plan [76]. Kaplan & Norton [3] agree that a shared ultimate goal or
strategy that has gained consensus and translates the direction in which the
organization wishes to head, is the starting point whence a BSC can be created.
2.3.2
Every organization has a mission, a purpose, a reason for being. Often the
mission is the reason why the organization first came into being to meet a need
identified before. A solid mission statement should accurately state why an
organization exists, and what it hopes to achieve in the future. It articulates the
essential nature, values and actions taken in an organization.
A mission statement defines the reason for existence of an organization. It
embodies its philosophies, goals and ambitions. Any organization that attempts
to operate without a mission statement runs the risk of drifting without having the
ability to verify that it is on its intended course. A clear business mission contains
the purpose of the organization; its values, strategy, and scope; and its standards
and behaviors [76].
viewed as the focus of the BSC it is almost as though the other three perspectives
are in essence answerable to the financial perspective.
Customer Perspective
In today's competitive markets, the key emphasis for most executives will be on
the customer. According to Mackay, many organizations have taken up the
challenge of focusing on customer satisfaction, identifying customer needs and
re-engineering their business capabilities from the customer interface. The
customer perspective defines how the organization differentiates itself from
competitors to attract, retain, and deepen relationships with targeted customers.
The value of the customer perspective is crucial because it helps an organization
connect its internal processes to improved outcomes with its customers [7]. The
customer perspective is at the core of any business and is crucial to long term
improvement of company performance. The customer-based virtuous circle,
whereby investment in employee training leads to improved service quality;
which in turn results in higher customer satisfaction leading to increased
customer loyalty, which boosts revenues and margins.
The customer perspective typically includes several core or general measures
derived from the desired successful outcomes of a well-formulated and
implemented strategy. These core measures may include overall indicators such
as customer satisfaction, customer complaints, customers lost/won, sales from
new products, and on-time delivery [2]. Measures related to customers include
results from customer surveys, sales from repeat customers, and customer
profitability. The customer perspective is a core of any business strategy which
describes the unique mix of product, price, service, relationship, and image that a
company offers [7].
2.4
Performance Drivers
2.5
33
CHAPTER 3
BALANCED SCORECARD WEIGHTING TOOL
3.1 Introduction
Balanced Scorecard has measures performance of an organization to link
objectives by its strategic goals. The organizations in different carrier and
different stage of growth have own different important KPIs. So, the important of
KPIs will different from stage to stage and organization to another. Therefore,
estimating weights for balanced scorecard perspectives and measures
performance which reflect organization importance degree of the different
perspectives is required. The organization redistributes weights in BSC as well as
total performance. There are many different theories and methods can use to
estimate BSC weights. Each method has its own basic concept, aim, advantages
and disadvantages. Which one is chosen by management or decision makers for
assessing performance depends on the status and type of the organization. To
achieve the best possible result from a more effective performance, this study
compares between them and chooses a suitable method to measure the weights.
We limit our topic to the normative analysis and the prescriptive analysis, since
the descriptive analysis (or called behavior decision research) is especially
highlighted by the fields of psychology, marketing, and consumer research. On
the other hand, the normative analysis and the prescriptive analysis are the
concentration in the fields of decision science, economics, and operations
research (OR).
Decision making is extremely intuitive while considering the single criterion
problems, since we only need to choose the alternative with the highest
preference rating. However, when decision makers evaluate the alternatives with
the multiple criteria, many problems, such as weights of criteria, preference
dependence, and conflicts among criteria, seem to complicate the decision
problems and should be overcome by more sophisticate methods [78].
Multiple criteria decision making (MCDM) refers to making decisions in the
presence of multiple, usually conflicting criteria. There are two groups within the
MCDM, multi attribute decision making (MADM) which deals with selection
problems and, multi objective decision making (MODM) which deals with
design problems [79].
MODM studies decision problems in which the decision space is continuous. A
typical example is mathematical programming problems with multiple objective
functions. The first reference of this problem is, also known as the "vectormaximum" problem. On the other hand, MADM concentrates on problems with
discrete decision spaces. In these problems the set of decision alternatives has
been predetermined.
The main question is, therefore, how to choose the appropriate methodology in
decision making. Table 3.1 lists specific operational requirements for a MADM
technique [80].
Table 3.1: Prerequisites of MCDA Techniques [80]
Prerequisites of MCDA techniques
Weights elicitation
Justification
To provide preference information between
the evaluation criteria
Critical threshold values, veto
To operation the assimilative capacity of the
environmental, economic, resource, and social
base
Comparability
To perform an integrated comparison between
the different actions
Qualitative
and
quantitative To handle the mixed information
information
Rigidity
To give robust results
Group decision-making
To include a diverse audience of stakeholders
Graphical representation
To render the outcome understandable
Ease of use
To familiarize the Decision Makers (DMs)
with the decision-making process
Sensitivity analysis
To enhance the transparency of the procedure
Variety of alternatives
To incorporate all possible courses of action
Large number of evaluation criteria
To embrace all different aspects
Consensus seeking procedures
To reach up a global compromise
Incorporation of intangible aspects
To be capable of taking into account hidden
dimensions of the problem
Hierarchy of scale
To decrease the ambiguities and provide for
explicit consistency
Concrete meaning for parameters used
To improve the reliability of the process
Learning dimension
To acknowledge and accept new information
revealed during the evolution of the procedure
Temporal aspects
To consider the emergency of the situation
and clarify long- and short-term concerns
36
Definition
Similar importance (SI)
Moderate importance (MI)
2,4,6,8
Intermediate values
Explanation
Two requirements are of equal value
Experience slightly favors one requirement
over another
Experience strongly favors one requirement
over another.
A requirement is strongly favored and its
dominance is demonstrated in practice.
The evidence favoring one over another is
of the highest possible order of affirmation.
When compromise is needed
37
After these steps, each method has individual mathematical equation to compute
the weights as following:
(3.1)
To calculate the Eigen value ( max), multiply on the right matrix of judgments
by the priority vector, obtaining a new vector.
Secondly, Calculate the Consistency Index (CI).
The CI can be calculated using the eq.3. 3.
(3.3)
Selecting the appropriate value of random index (RI) table 3.3 is according to the
matrix size.
Table 3.3: Random Index of Analytic Hierarchy Process
:
Size of
matrix (n)
Random
index (RI)
10
11
12
0.58
0.9
1.12
1.24
1.32
1.41
1.45
1.49
1.51
1.59
39
Where the term [-1/ln(N)] provides the condition 0 < Ej < 1 to be fulfilled. The
weight of criteria Xj, Wj can be determined by eq. 3.7 as following:
(3.7)
~~~~
2, 4 ,6 ,8
Definition of linguistic
variables
Similar importance (SI)
Moderate importance (MI)
Intense importance (II)
Demonstrated importance (DI)
Extreme importance (EI)
Intermediate values
Fuzzy
number
(L,M,U)
(L,M,U)
(L,M,U)
(L,M,U)
(L,M,U)
(L,M,U)
User define
=
=
=
=
=
=
(
(
(
(
(
(
,1,
,3,
,5,
,7,
,9,
, ,
)
)
)
)
)
)
Where: L is the lower limit, M is the medium limit, U is the upper limit.
Then, the geometric mean method for finding the final fuzzy weights of each
criterion can be formulated by eq. 3.8, 3.9 as follows:
,
Wi = ri * (r1+r2++rn)-1
(3.8)
Where
ri = (ai1* ai2*.ain)1/n
(3.9)
After determining the weights for lower, median and upper comparison matrices,
the weighting average is computed to get the final weights.
problems and the best alternative can be derived by the following eq.3.10, 3.11:
(3.10)
Ui (x) =
(3.11)
and
Wjrij(x)
Where
Ui (x) : the utility of the ith alternative.
wj : the weights of the jth criterion.
and
rij(x) is the normalized preferred ratings of the ith alternative with respect to the
jth criterion.
In addition, the normalized preferred ratings rij(x) of the ith alternative with
respect to the jth criterion can be defined by:
For benefit criteria, rij(x) = xij/xj*, where xj* =max xij , 0 rij(x) 1
For cost criteria, rij(x) = 1/xij / 1/xj* = min xij / xij
where xij is the normalized preferred ratings of the ith alternative with respect to
the jth criterion.
42
(3. 12)
Within the VIKOR method L1,j (as Sj in Eq. 3.12.) and L,j (as Sj in Eq. 3.12.) are
used to formulate ranking measure. The solution obtained by minj Sj is a
maximum group utility (majority rule), and the solution obtained by minj Sj is a
minimum individual regret of the opponent.
The compromise solution Fc is a feasible solution that is the closest to the ideal
F*, and compromise means an agreement established by mutual concessions, as
is illustrated in Figure 3.2 by:
f1 = f1* - f1c and f2 = f2* - f2c.
(3.13)
(b) Compute the values Sj and Rj, j=1,2,., J by the following eq.3.14, 3.15.
Sj =
wi (fi* - fij)/(fi*-fj-)
Rj = max [
(3.14)
wi (fi* - fij)/(fi*-fj-)]
(3.15)
(c) where wi are the weights of criteria, expressing their relative importance.
(d) Computer the value QJ , j= 1,2,, J by the following eq.3.16.
Qj = v (Sj S*)/(S- S*) + (1 v)(Rj R*) /(R- R*)
(3.16)
Where
S * = min Sj , S- = max S j
R* = min Rj , R- = max Rj
44
Steps of TOPSIS
(3.17)
(3.18)
(3.19)
(3.20)
Step 4: Calculate the separation measures for each alternative. The separation
from the ideal alternative is:
S i* =
(vij vj*)2 ]
i = 1, , m
(3.21)
(vij vj' )2 ]
i = 1, , m
(3.22)
Ci* 1
45
(3.23)
Entropy
Fuzzy(GMM)
SAW
TOPSIS
VIKOR
Decision Making
Individual and
group
Group
Group
Group
Methodology
Creating
hierarchical
structure
And pair wise
comparison
matrices
Creating network
structure and
Pair wise comparison
matrices
Creating matrix
structure and calculating
distance to positive and
negative ideal point
Areas of Usage
To support decision
making for
complexity
To support decision
making for
complexity
To support decision
making for complexity
To support decision
making for
complexity
easy to adapt
easy to adapt
easy to adapt
Creating matrix
structure and
comparing pairs
of alternatives to
form an
outranking
relation
To support
decision
making for
complexity
not easy to adapt
need
No need
No need
No need
No need
No need
Weighting System
Pair Wise
comparison
Using (GMM)
Method.
No specific
Method.
No specific method.
Linear or
Vector normalization
No specific
method.
Criteria Evaluation
Tangible and
intangible
criteria
Tangible and
intangible
criteria
Tangible and
intangible
Criteria
High Sensitivity
analysis.
Flexibility of the
software package
Adaptability/Flexibility
Consistency
Measurement
Advantages
Easy to use;
scalable; hierarchy
structure can easily
adjust to fit
many sized
problems; not data
intensive.
Disadvantages
46
weights of criteria such as AHP and Fuzzy. Other methods used Linear or
Vector normalization methods to determine ranking of criteria such as
SAW, TOPSIS and VIKOR.
7. AHP method can't solve the problem of multi criteria decision making
revealed don't always reflect the real situation, result obtained may not be
logical.
10. TOPSIS method is more difficult than SAW but it more sensitive the
47
From this analysis Fuzzy (GMM) and TOPSIS methods have a lot of positive
points. Therefore, this research proposes the integration of the Fuzzy (GMM)
and TOPSIS methods in an attempt to obtain a mixture of strong characteristics.
48
CHAPTER 4
PROPOSED AND VERIFIED THE PRIORITIZATION
MODEL
4.1 Introduction
The third chapter studied the group of MADM methods that calculate weights
and ranking the alternatives in terms of advantages and disadvantages. This
chapter proposes an integrated decision-making approach based on fuzzy
linguistic variables and geometric mean method integrated with TOPSIS
(technique for order performance by similarity to ideal solution) framework. This
model will help manager to estimate the perspectives of balanced scorecard
weights and rank its indicators. Arranging the alternatives is depending on
detremine the weights so that, accuracy in calculating weights is recommended.
This chapter comparing the methods, in which weights are calculated, using
euclidean distance of each method separately. Thus, examining the methods,
which are ranking the alternatives, using Approximation method. This
Approximation method studies the sensitivity analysis of ranking methods
according to the changes occuring in the weights. The results of these
comparisons are verifying the proposal of the integration of the Fuzzy (GMM)
and TOPSIS methods.
This chapter provides four numerical examples to illustrate and verify the
potential applications of the proposed models. The illustration shows the
advantages of the model presented and verifies the effectiveness of the model
with previous work.
The objectives are usually conflicting and therefore, the solution is highly
dependent on the preferences of the decision-maker and must be a compromise.
Each method has strong and weakness characteristics, so future techniques
integrated methods to get better characteristics. Therefore, this research proposes
the integration of the Fuzzy (GMM) and TOPSIS methods in an attempt to obtain
a mixture of strong characteristics.
50
Table 4.1: The Pair-wise Comparison of Linguistic Variables Using Fuzzy Numbers
Intensity of fuzzy
scale
~
1
Definition of linguistic
variables
Similar importance (SI)
Fuzzy
number
(L,M,U)
= ( ,1,)
~
3
(L,M,U)
= ( ,3,)
~
5
~
7
~
9
(L,M,U)
= ( ,5,)
(L,M,U)
= ( ,7,)
(L,M,U)
= ( ,9,)
Intermediate values
(L,M,U)
= ( , , )
~~~~
2, 4 ,6 ,8
User define
Where: L is the lower limit, M is the medium limit, U is the upper limit.
Next, from the information of the pair wise comparison, we can form the fuzzy
positive reciprocal matrix as follows eq.4.1:
a11 a1 j ain
an1 anj ann
(4.1)
Then, the geometric mean method for finding the final fuzzy weights of each
criterion can be formulated as follows eq. 4.2, 4.3:
Wi = ri * (r1+r2++rn)-1,
(4.2)
ri = (ai1* ai2*.ain)1/n
(4.3)
Where
After determining the weights for lower, median and upper comparison matrices,
the weighting average is computed to get the final weights.
The Third Stage : Determining The Final Ranking
In the last stage, calculated weights of the factors are approved by decision
making team. Ranking firms are determined by using TOPSIS method in the
third stage. TOPSIS method is one of the well known ranking methods for
MCDM. TOPSIS is firstly proposed by Hwang and Yoon. This technique based
on the concept that rank alternatives, which has the shortest distance from the
ideal (Best) solution and the longest distance from the ideal (worst) solution.
51
Steps of TOPSIS
Step 1: Construct normalized decision matrix. This step transforms various
attribute dimensions into non-dimensional attributes, which allows comparisons
across criteria.
Normalize scores or data as following eq. 4.4:
rij = xij/ (a2ij) for i = 1, , m; j = 1, , n
(4.4)
Step 2: Construct the weighted normalized decision matrix. Assume we have a set of
weights for each criteria wj for j = 1,n.
Multiply each column of the normalized decision matrix by its associated weight by eq.
4.5. An element of the new matrix is:
vij = wj rij
(4.5)
Step 3: Determine the ideal and negative ideal solutions by eq. 4.6.
Ideal solution.
A* = { v1*, , vn*}, where
vj*={ max (vij ) if j J ; min (vij ) if j J' }
(4.6)
(4.7)
Step 4: Calculate the separation measures for each alternative by eq. 4.8. The
separation from the ideal alternative is:
Si * = [
(vij vj*)2 ]
i = 1, , m
(4.8)
Similarly, by eq. 4.9 the separation from the negative ideal alternative is:
S'i = [ (vij vj' )2 ]
i = 1, , m
(4.9)
Step 5: Calculate the relative closeness to the ideal solution Ci* by eq. 4.10
Ci*= S'i / (Si* +S'i ) ,
Ci* 1
(4.10)
Schematic diagram of the proposed model for best design is provided in Figure 4.1.
52
Set up goal
Make criteria
No
Assess whether the
hierarchy is arranged
properly or not
Yes
End
Fig 4.1: Schematic Diagram for the Fuzzy (GMM) -TOPSIS Integrated Approach [87]
53
I,j =1,2,..n.
(4.11)
Approximation Method
The approximation method is used to calculate the minimum change of weights
(changes= 0.001 in this thesis) that can change the ranking of alternatives by
using different ranking methods.
A1
A2
A3
A4
A5
Criteria (P1)
A1 A2 A3 A4 A5
1
2
5
3
2
1/2 1
7
3
3
1/5 1/7 1 1/4 1/5
1/3 1/3 4
1
3
1/2 1/3 5 1/3 1
A1
A2
A3
A4
A5
A6
National Income(P2)
A1 A2 A3 A4 A5 A6
1
5
3
6
7
5
1/5 1 1/7 1/2 2
2
1/3 7
1
7
3
4
1/6 2 1/7 1 1/2 1
1/7 1/2 1/3 2
1
2
1/5 1/2 1/4 1 1/2 1
A1
A2
A3
A4
A5
A6
Foreign Exchange(P3)
A1 A2 A3 A4 A5 A6
1
4
6
7
2
2
1/4 1
2
2
1 1/3
1/6 1/2 1
2 1/6 1
1/7 1/2 1/2 1 1/5 1/7
1/2 1
6
5
1
1
1/2 3
1
7
1
1
A1
A2
A3
A4
A5
A6
A1
A2
A3
A4
A5
A6
A1
A2
A3
A4
A5
A6
A6
3
1/3
1/5
1/6
1/2
1
LGP
0.356
0.356
0.051
0.119
0.119
E
0.112
0.307
0.1
0.297
0.184
F(GMM)*
0.343
0.317
0.043
0.168
0.126
AN method
AN
C1
C2
C3
C4
C5
A1
0.432
0.364
0.403
0.413
0.048
A2
0.082
0.108
0.178
0.151
0.223
EV
C1
C2
C3
C4
C5
A1
0.44
0.365
0.409
0.416
0.046
A2
0.077
0.11
0.18
0.15
0.209
WLS
C1
C2
C3
C4
C5
C1
0.498
0.394
0.403
0.413
0.048
C2
0.06
0.099
0.16
0.131
0.126
E
C1
C2
C3
C4
C5
A1
0.130
0.141
0.136
0.139
0.107
A2
0.180
0.195
0.197
0.204
0.396
A3
0.281
0.079
0.034
0.036
0.192
A4
0.066
0.04
0.107
0.043
0.325
LLS method
A5
0.081
0.206
0.159
0.129
0.166
A6
0.057
0.203
0.119
0.228
0.045
LLS
C1
C2
C3
C4
C5
A1
0.448
0.384
0.432
0.423
0.053
A2
0.073
0.108
0.167
0.147
0.212
A5
0.078
0.211
0.153
0.125
0.162
A6
0.054
0.197
0.12
0.233
0.043
FPP
C1
C2
C3
C4
C5
A1
0.453
0.389
0.442
0.423
0.17
A2
0.121
0.117
0.184
0.165
0.188
C5
0.075
0.205
0.132
0.117
0.177
C6
0.077
0.203
0.112
0.202
0.051
LGP
C1
C2
C3
C4
C5
A1
0.504
0.391
0.462
0.461
0.056
A2
0.101
0.098
0.154
0.154
0.337
A5
0.178
0.163
0.211
0.182
0.136
A6
0.076
0.145
0.153
0.259
0.116
FGMM
C1
C2
C3
C4
C5
A1
0.444
0.382
0.426
0.42
0.054
A2
0.072
0.108
0.165
0.141
0.206
EV method
A3
0.286
0.078
0.032
0.036
0.187
A4
0.065
0.04
0.106
0.04
0.352
C4
0.057
0.043
0.087
0.039
0.369
A4
0.165
0.128
0.230
0.116
0.123
A3
0.201
0.078
0.063
0.059
0.16
A4
0.05
0.067
0.095
0.045
0.279
A3
0.216
0.065
0.038
0.051
0.169
A4
0.054
0.056
0.077
0.066
0.236
A6
0.056
0.192
0.108
0.231
0.049
A5
0.086
0.194
0.104
0.132
0.146
A6
0.088
0.156
0.112
0.176
0.056
A5
0.072
0.195
0.154
0.115
0.169
A6
0.054
0.195
0.115
0.154
0.034
FGMM method
A3
0.281
0.077
0.035
0.038
0.191
A4
0.063
0.044
0.107
0.039
0.327
A5
0.079
0.204
0.154
0.125
0.167
LGP method
E method
A3
0.271
0.227
0.072
0.099
0.122
A4
0.063
0.041
0.106
0.039
0.326
FPP method
WLS method
C3
0.233
0.056
0.04
0.046
0.22
A3
0.282
0.071
0.034
0.036
0.193
P1
P2
P3
P4
ED
AN
4.583 6.209
5.305
6.797
EV
4.961 6.255
5.359
7.382
WLS
5.508 6.937
5.204
7.114
LLS
4.813 6.119
5.289
7.327
FPP
5.440 8.027
6.904
7.318
LGP
4.550 8.227
5.607
7.005
E
8.573 13.298 12.991 10.146
F(GMM)* 4.504 5.907
5.350
6.930
F(GMM)* Our proposed model
56
P5
P6
6.107
6.451
7.054
6.627
6.634
7.643
13.705
6.270
4.786
5.055
5.331
4.642
8.740
8.162
11.11
4.590
A5
A6
0.079 0.059
0.199 0.19
0.154 0.113
0.131 0.23
0.171 0.049
Comulative ED
69.823
41.194
43.063
33.551
Fuzzy
(GMM)
TOPSIS
Entropy
LGP
37.148
37.63
35.463
33.787
LLS
WLS
EV
AN
FPP
Different Methods
The results of this case, reservoir storage allocation problem Figures 4.2 shows
that, Fuzzy (GMM), AN (AHP) prioritization methods produce the smaller or
close to zero the value of ED respictivily (as comparing criteria i.e. P1, P2 and
P6). Entropy weighting method produce the highest value.
1
1/4
1/3
1
1/3
1/4
4
1
1/7
1/3
5
1
3
7
1
5
5
6
1
3
1/5
1
1
3
57
3
1/5
1/5
1
1
1/3
4
1
1/6
1/3
3
1
LLSM
GLSM
GEM
FPM
CCMA
DEAHP
LP-GFW
F(GMM)*
0.3114 0.1396 0.0367 0.1272 0.2362 0.1487
F(GMM)* Our proposed model
9.24
9.28
9.37
9.28
9.31
9.36
9.86
9.05
8.79
It is shown that, AHP method gives the lowest value of ED, but can not be relied
upon in higher consistency (CR= 0.229). Fuzzy (GMM) prioritization model in
case of high inconsistent matrices produces the second smaller or close to zero
the value of ED as comparing criteria (i.e. Fuzzy model is the best solution).
When compare between Fuzzy (GMM) with different prioritization methods, and
with new approach as integrated model of data envelopment analysis and AHP
(DEAHP), and correlation coefficient maximization approach (CCMA).
58
shows high
GLS
M
.085
.771
GEM FPM
EVM
.771
WLSM
.771
GLSM
.314
GEM
FPM
CCMA
LPGFW
F(GMM)* Our proposed model
CCM
A
1
.314
.085
.771
1
1
.314
.085
.771
DEAH
P
.529
.755
.226
.226
.529
LPGFW
1
.314
.085
.771
1
.529
F(GMM)*
.60
.828
.371
.942
.60
.30
.6
wheelchair design are classified into five aspects; performance (P), safety (S),
cost (C), ergonomic (E) and maintenance (M).
affecting the wheelchair performance: easy to transfer (ETT), easy to use (ETU),
easy to storage (ETS), lightweight (LW) and strong framework (SF). Stability
(ST) and no sharp edge (NSE) are sub-criteria that affect in terms of safety.
While cost of material (CM) and cost of manufacturing process (CMP), easy to
repair (ETR) and easy to dismantle (ETD), are sub-criteria affecting in terms of
cost and maintenance respectively, show (Figure 3). This example is divided into
three sections. Section one presented stages of the Fuzzy (GMM) -TOPSIS
model. Section two presented a comparison between results obtained with AHP
and Fuzzy (GMM) -TOPSIS model by using Euclidean distance. Section three
compares between three different ranking methods SAW, TOPSIS and VIKOR
in terms of methods sensitivity of changing the weights. In this case, three
scenarios are used equal weight, weight from AHP and GMM weight.
Section 1
ETT
ST
Goal level 1
Criteria
level 2
ETR
CM
ETU
CMP
NST
ETD
SupCriteria
level 3
ETS
LW
SF
DC-1
DC-2
DC-3
DC-4
DC-5
DC-6
DC-7
Alternatives level
Initially, the objective or the overall goal of the decision is presented at the top
level of the hierarchy. Specifically, the overall goal of this application is to
select the most suitable wheelchair conceptual design. The second level
represents the main criteria affecting the development of wheelchair design. The
main criteria can be classified into five aspects: performance (P), safety (S), cost
(C), ergonomic (E) and maintenance (M).The sub-criteria is represented at the
third level of the hierarchy. There are five sub-criteria affecting the wheelchair
performance: easy to transfer (ETT), easy to use (ETU), easy to storage (ETS),
lightweight (LW) and strong framework (SF). Stability (ST) and no sharp edge
(NSE) are sub-criteria that affect in terms of safety. While cost of material (CM)
and cost of manufacturing process (CMP), easy to repair (ETR) and easy to
dismantle (ETD), are sub-criteria affecting in terms of cost and maintenance
respectively. Finally, at the lowest level of the hierarchy, the design concept
(DC) alternatives of the wheelchair development are identified.
The Second Stage is Computing The Weights by Fuzzy (GMM)
Step 1: Pair wise comparison matrix
The pair-wise comparisons generate a matrix of relative rankings for each level
of the hierarchy. The number of matrices depends on the number elements at
each level. The order of the matrix at each level depends on the number of
elements at the lower level that it links to Pair-wise comparison begins with
comparing the relative importance of two selected items. There are n (n 1)
judgments required to develop the set of matrices in this step. The decision
makers have to compare or judge each element by using the relative scale pair
wise comparison as shown in Table 4.1. The judgments are decided based on the
decision makers or users experience and knowledge. The scale used for
comparisons in Fuzzy (GMM) enables the decision maker to incorporate
experience and knowledge intuitively. To do pair wise comparison, for instance
as shown in Table 4.8, if performance (P) is strongly more important or essential
over cost (C), then a = 5. Reciprocals are automatically assigned to each pairwise comparison.
61
P
S
C
E
M
[1,1,1]
[2,3,4]
[4,a,6]
[2,3,4]
[4,5,6]
[1/4,1/3,1/2]
[1,1,1]
[2,3,4]
[1,1,1]
[2,3,4]
[1/6,1/5,1/4] [1/4,1/3,1/2]
[1,1,1]
[1/4,1/3,1/2] [2,3,4]
[1/4,1/3,1/2]
[1,1,1]
[2,3,4]
[1,1,1]
[2,3,4]
[1/6,1/5,1/4] [1/4,1/3,1/2] [1/4,1/3,1/2] [1/4,1/3,1/2] [1,1,1]
Step2: Separation
In this step, three matrices (Lower, Medium, and Upper) are separated from the
original matrix. The priority is calculated for each matrix separately.
Step3: Synthesizing the Pair wise Comparison
Normalized vector is computed according to eq.(4.3), where aij the element with i
raw and j column. To calculate the vectors of priorities, sum the each element in
the normalizing column. Then, divide the elements of each column by the sum of
the column as shown in eq. (4.2). The result is priority vector as shown in table
4.9. This process is done on all three matrices (Lower, Medium, and Upper).
Table 4.9: Synthesizing The Pair wise Comparison
Goal
(P)
(S)
(C)
(E)
(M)
P
S
C
1
2
4
1/4 1
2
1/6 1/4 1
1/4 1
2
1/6 1/4 1/4
sum
E
2
1
1/4
1
1/4
M
4
2
2
2
1
2.297
1
0.461
1
0.304
5.062
Priority
0.453
0.197
0.091
0.197
0.060
1
0.453
0.197
0.091
0.197
0.060
2
0.461
0.194
0.090
0.194
0.058
62
3
0.453
0.192
0.096
0.192
0.063
Average
0.456
0.195
0.092
0.195
0.061
This proces is repeated for all levels of hierarchy structure (criteria, sub-criteria
and alternatives).
The priority vectors for criteria, sub-criteria and alternatives are represented in
table 4.11. The overall priority vector can be obtained by multiplying the priority
vector for the design alternatives by the vector of priority of the sub-criteria as
shown in table 4.12.
Table 4.11: Represent Priority Vectors for Criteria, Sub-Criteria and Alternatives.
0.451
ETT
0.265
ETU
0.456
P
0.060
ETS
0.175
0.104
0.140
0.126
0.323
0.080
0.051
0.215
0.081
0.145
0.044
0.201
0.162
0.152
0.093
0.074
0.066
0.070
0.420
0.074
0.204
criteria
Sub-criteria
Alternatives
DC-1
DC-2
DC-3
DC-4
DC-5
DC-6
DC-7
0.140
LW
0.170
0.290
0.140
0.127
0.050
0.167
0.056
0.082
SF
GOAL
0.195
S
0.750 0.261
ST
NSE
0.0928
C
0.750 0.261
CM
CMP
0.195
E
1
E
0.061
M
0.750 0.261
ETR
ETD
0.112
0.061
0.118
0.061
0.311
0.084
0.253
0.144
0.055
0.092
0.062
0.269
0.147
0.230
0.126
0.229
0.126
0.229
0.051
0.169
0.072
0.120
0.066
0.120
0.050
0.322
0.100
0.223
0.262
0.191
0.144
0.093
0.045
0.098
0.068
0.189
0.058
0.174
0.055
0.174
0.174
0.174
0.124
0.227
0.124
0.227
0.047
0.182
0.070
0.195
0.316
0.125
0.101
0.034
0.165
0.065
Table 4.12: Overall Weight Vector for The Alternatives With Respect to The Criteria
ETT
ETU
ETS
LW
SF
ST
NSE
CM
CMP
ETR
ETD
0.206
0.121
0.027
0.064
0.038
0.146
0.051
0.070
0.024
0.195
0.046
0.016
63
ETT
0.401
0.238
0.321
0.289
0.741
0.183
0.117
ETU
0.528
0.199
0.356
0.108
0.494
0.398
0.374
ETS
0.187
0.149
0.133
0.141
0.845
0.149
0.411
LW
0.398
0.679
0.328
0.297
0.117
0.391
0.131
SF
0.249
0.136
0.263
0.136
0.692
0.187
0.563
ST
0.337
0.129
0.215
0.145
0.629
0.344
0.538
NSE
0.468
0.144
0.431
0.136
0.431
0.431
0.431
CM
0.303
0.551
0.303
0.551
0.123
0.406
0.173
CMP
0.297
0.544
0.297
0.544
0.113
0.436
0.168
E
0.269
0.148
0.269
0.112
0.722
0.224
0.500
ETR
0.675
0.492
0.371
0.240
0.116
0.252
0.175
ETD
0.440
0.713
0.282
0.228
0.077
0.372
0.147
Step 2: Construct the Weighted Normalized Decision Matrix by Using eq. (4.5).
In this step multiply each column of the normalized decision matrix by its
associated weight in table 11 as shown in table 4.14
Table 4.14: The Weighted Normalized Decision Matrix
DC-1
DC-2
DC-3
DC-4
DC-5
DC-6
DC-7
ETT
ETU
ETS
LW
SF
ST
NSE
CM
CMP
ETR
ETD
0.082
0.049
0.066
0.059
0.152
0.037
0.024
0.063
0.024
0.043
0.013
0.059
0.048
0.045
0.005
0.004
0.003
0.003
0.022
0.004
0.011
0.025
0.043
0.021
0.019
0.007
0.025
0.008
0.009
0.005
0.010
0.005
0.026
0.007
0.021
0.049
0.018
0.031
0.021
0.091
0.050
0.078
0.023
0.007
0.022
0.006
0.022
0.022
0.022
0.021
0.038
0.021
0.038
0.008
0.028
0.012
0.007
0.013
0.007
0.013
0.002
0.010
0.004
0.052
0.028
0.052
0.021
0.140
0.043
0.097
0.031
0.022
0.017
0.011
0.005
0.011
0.008
0.007
0.011
0.004
0.003
0.001
0.006
0.002
0.152
0.063
0.022
0.043
0.026
0.091
0.023
0.008
0.002
0.140
0.031
0.011
V'
0.024
0.013
0.003
0.007
0.005
0.018
0.006
0.038
0.013
0.021
0.005
0.001
for each element in matrix to find the final separation. The separation from the
negative ideal alternative can be calculated as the separation from the ideal
alternative by eq. (9) as show in table 4.17.
Table 4.16: The Separation from the Ideal Alternative
ETT ETU
ETS
DC-1
4.9
0.0
0.3
DC-2 10.7
1.6
0.4
DC-3
7.5
0.4
0.4
DC-4
8.7
2.6
0.4
DC-5
0.0
0.0
0.0
DC-6 13.2
0.2
0.4
DC-7 16.5
0.4
0.1
Each element in the matrix * 10-3
LW
0.3
0.0
0.5
0.6
1.3
0.3
1.2
SF
0.3
0.4
0.3
0.4
0.0
0.4
0.0
ST
1.8
5.3
3.7
5.0
0.0
1.7
0.2
NSE
0.0
0.3
0.0
0.3
0.0
0.0
0.0
CM
0.2
0.9
0.2
0.9
0.0
0.4
0.0
CMP
0.0
0.1
0.0
0.1
0.0
0.1
0.0
E
7.8
12.5
7.8
14.1
0.0
9.4
1.9
ETR
0.0
0.1
0.2
0.4
0.7
0.4
0.5
ETD
0.0
0.0
0.0
0.1
0.1
0.0
0.1
CMP
0.4
0.0
0.4
0.0
1.3
0.1
1.0
E
0.2
0.0
0.2
0.0
2.6
0.1
1.0
ETR
2.2
1.0
0.4
0.1
0.0
0.1
0.0
ETD
0.9
2.8
0.3
0.2
0.0
0.6
0.0
LW
0.5
2.2
0.3
0.2
0.0
0.5
0.0
SF
0.1
0.0
0.1
0.0
2.1
0.0
1.3
ST
0.3
0.0
0.1
0.0
1.7
0.3
1.2
NSE
0.8
0.0
0.6
0.0
0.6
0.6
0.6
CM
0.4
0.0
0.4
0.0
1.3
0.1
1.0
Step 5: The relative closeness to the ideal solution is calculated by using the eq. (4.10).
In this step each element in row of separation from the negative ideal alternative divides
by the sum of separation ideal and negative ideal alternative. Then, the final ranking is
presented in table 4.18.
Table 4.18: Result of Selection
Ranking
1
2
3
4
5
6
7
65
AHP
DC-5
DC-1
DC-7
DC-6
DC-3
DC-2
DC-4
P
0.456
0.456
0.1779
S
0.191
0.195
0.2675
C
0.099
0.092
0.2079
E
0.191
0.195
0.2675
M
0.061
0.061
0.0793
6.7903
8
6
4
2
0
3.38
3.32
AHP
AHP
Enrtopy
Enrtopy
The results of this case shows that, FGMM produce the smaller value of ED.
Section 3: Comparison between Ranking Methods
ETU
ETS
LW
SF
ST
NSE
CM
CMP
ETR
ETD
0.189
0.114
0.026
0.058
0.069
0.143
0.048
0.074
0.025
0.191
0.047
0.016
Scenario c) Use the weights generated from the Fuzzy GMM as shown in table
4.12. The results of scenario (a),(b),(c) are presented in table 4.21.
66
1
2
3
4
5
6
7
Equal Weights
TOPSIS
VIKOR
DC-5
DC-1
DC-1
DC-5
DC-7
DC-6
DC-2
DC-7
DC-3
DC-3
DC-6
DC-2
DC-4
DC-4
SAW
DC-5
DC-1
DC-7
DC-6
DC-3
DC-2
DC-4
Weights by AHP
TOPSIS
VIKOR
DC-5
DC-5
DC-7
DC-1
DC-1
DC-3
DC-3
DC-7
DC-6
DC-6
DC-2
DC-2
DC-4
DC-4
Table 4.21, while ranking other alternatives, the SAW and VIKOR methods
produce more similar ranks across different scenarios than the TOPSIS method.
Because the TOPSIS method has high sensitivity to the changes in methods for
assigning weights to criteria, it's frequently used as a benchmarking method.
For the wheelchair selection problem, all of the SAW, TOPSIS and VIKOR
methods in each scenario give the top rank to the same alternative (DC-5)except
the VIKOR in scenario "a". Figure 4.6 shows there are small variations in the
Alternatives
Fuzzy VIKOR
4
Rnking
Fuzzy TOPSIS
AHP VIKOR
AHP TOPSIS
AHP SAW
E VIKOR
E TOPSIS
E SAW
Fuzzy SAW
67
E SAW
E TOPSIS
E VIKOR
AHP SAW
AHP TOPSIS
AHP VIKOR
Fuzzy SAW
Fuzzy TOPSIS
Equal
SOPSIS
-.178
Equal
VIKOR
-.178
-.32
AHP
SAW
.2
.428
.392
AHP
TOPSIS
.142
-.75
-.25
-.60
AHP
VIKOR
.928
-.178
.107
.535
0
Fuzzy
SAW
.25
.428
.392
1
-.60
.535
Fuzzy
TOPSIS
.357
.535
-.035
.678
-.285
.428
.678
Fuzzy
VIKOR
.928
-.178
.107
.535
0
1
.535
.428
.027
.031
.064
.038
.060
.042
ranking
DC-5
DC-1
DC-7
DC-3
DC-6
DC-2
DC-4
.027
.030
.064 .038
.067 .041
ranking
DC-5
DC-1
DC-7
DC-6
DC-3
DC-2
DC-4
.027
.037
.064 .038
.054 .048
ranking
DC-5
DC-1
DC-3
DC-7
DC-6
DC-2
DC-4
68
Changes= .003
.146
.051
.07
.024
.142
.051
.07
.024
Results after changing (W)
0.824263
0.434748
0.434599
0.318824
0.269222
0.209023
0.167655
Changes= .004
.146
.051
.07
.024
.149
.051
.07
.024
Results after changing (W)
0.820028
0.599075
0.527373
0.473117
0.472852
0.410121
0.33575
Changes= .01
.146
.051
.07
.024
.156
.051
.08
.034
Results after changing (W)
0.00
0.234395
0.510857
0.604738
0.781943
0.850231
0.930921
.195
.199
.046
.050
ranking
.016
.012
.046
.043
ranking
.016
.013
.046
.036
ranking
.016
.006
DC-5
DC-7
DC-1
DC-3
DC-6
DC-2
DC-1
.195
.198
DC-5
DC-1
DC-7
DC-3
DC-6
DC-2
DC-4
.195
.185
DC-5
DC-7
DC-1
DC-3
DC-6
DC-2
DC-4
The last table explains how changes of weights contribute to the change of raking
and also, shows that the TOPSIS method is more sensitive than others which
changes its alternative's ranking by smallest change in weights of attributes.
69
Material
Fatigue (+)
100
49
78
108
70
165
440
242
616
500
300M
2024T3
7050T73561
Ti6AL4V
E glass epoxy FRP
S glass epoxy FRP
Carbon epoxy FRP
Kevlar 29 epoxy FRP
Kevlar 29 epoxy FRP
Boron epoxy FRP
Toughness (+)
8
13
12
26
10
25
22
28
34
23
Price (-)
4200
2100
2100
10500
2735
4095
35470
11000
25000
3150000
The result of ranking materials for different methods are shown in table 4.25. It is
shown that all methods ranke material number 9 (Kevlar 49-epoxy FRP ) is the
first which has the highest value (.93104). material number 7 (Carbon epoxy
FRP) has the second value (.6884), thus we put 7 in rank 2 of columnaas the
same ranking of column "c" and "d" where column "b" rank material number 8
(Kevlar 29 epoxy FRP) in the second and column "e" rank material number 10
(Boron epoxy FRP).In the same way the material 2 (2024T3) has the lowest
value (.2835) in column a which agree with the ranking of column "d" and "e"
where column "b" rank material number 10 (Boron epoxy FRP) in the last one
and column "c" rank material number 1 (300M) also in the last one .
Thable 4.25: Cumparing Between Defferent Ranking Method .
Ranking
1
2
3
4
5
6
7
8
9
10
a
Topsis
9
7
10
8
6
4
5
1
3
2
b
Jee and kang
9
8
6
7
1
4
3
5
2
10
c
ELECTER
9
7
6
8
10
4
5
3
2
1
d
VIKOR
9
7
10
8
6
4
5
3
1
2
e
L. assignment
9
10
7
8
6
4
1
3
5
2
For the flywheel material selection problem, all of the TOPSIS, Jee and Kang,
ELECTER, VIKOR and the linear assignment methods give the top rank to the
same material (Kevlar 49epoxy FRP). Figure 4.7 shows there are small
variations in the rankings obtained using Jee and Kang, ELECTER, VIKOR and
linear assignment methods.
70
12
10
8
Material
6
4
2
0
1
10
ranking
Linear assignment
Jee and kang
VIKOR
Topsis
ELECTER
71
L. assignment Topsis
0.321
.224
.864
.769
.696
.951
.745
CHAPTER 5
IMPROVEMENT BALANCED SCORECARD INTEGRATED
MODEL
5.1 Introduction
The balanced scorecard is a strategic planning and management system that is
used extensively in business industry, government and nonprofit organizations
worldwide to align business activities to the vision and strategy of the
organization, improve internal and external communications, and monitor
organization performance against strategic goals. This chapter describes a
framework to develop balanced scorecard model step by step. The Balanced
Scorecard perspectives are developed by using new perspective. This model
contains ten steps for establish balanced scorecard in the organization. The
proposed model integrates the origin BSC, perspective "management
commitment" and prioritization technique Fuzzy (GMM) - TOPSIS. This model
attempts to overcome the limitation in balanced scorecard implementation.
The balanced scorecard model helps everyone in an organization understand and
work towards a shared vision. A completed scorecard system aligns the
organizations picture of the future (shared vision), with business strategy,
desired employee behavior and day-to-day operations. Strategic performance
measures are used to better inform decision-making and show progress toward
desired results. The organization can then focus on the most important things that
are needed to achieve its vision and satisfy customers, stakeholders, and
employees. Other benefits include measuring what matters, identifying more
efficient processes focused on customer needs, improving prioritization of
initiatives, improving internal and external communications, improving
alignment of strategy and day-to-day operations and linking budgeting and cost
control processes to strategy.
72
Leadership
Leader is another obvious point concerns performance. Look at any great team or
organization and you invariably see great leadership. So if you want to create a
high performance organization, you must have strong leadership. Leadership is
one of those fundamental building blocks that drive everything else. The
potential of the individual and the organization ultimately boils down to
leadership.
Leadership is about generating movements and currents within the organization.
Leaders produce change through the processes of establishing direction through
vision and strategy, aligning people whose cooperation is needed to achieve the
vision, and motivating and inspiring them to overcome the barriers to change.
Organization Structure
The framework of jobs and departments that make up any organization must be
directed toward achieving the organizations objectives. In other words, the
structure of a lodging business must be consistent with its strategy.
Managers give structure to the organizations through job specialization,
organization, and establishment of patterns of authority and span of control.
One of the major roles of management is to ensure that an organization is
designed to carry out its mission, goals, and strategies. Understanding leadership
requires a fundamental understanding of organizations and the design factors that
must be considered.
The design of an organization is the formal framework for communication and
authority, and is determined by three major components:
Complexity. Is the number of different entities (for example, job titles, reporting
levels, functional departments, and physical work locations) that will exist in the
organization.
Formalization. How much the organization will rely on standard guidelines and
procedures to instruct employee activities.
Centralization. Whether decision making authority is located primarily at upper
management levels or is delegated to lower levels.
Organization Culture
Its important for organizations of different size and level to create the kind of
environment or culture where the positive managerial patterns of listening,
coaching, guiding, involving and problem-solving are actively encouraged and
reinforced. This is where the policy of the Human Resources department is
critical as it reflects and reinforces organizational values and culture.
Organizational culture is important because it promotes healthy competition in
the work place and it also brings all the employees to the same level.
Furthermore, the work culture facilitates the unison of employees who come
from different backgrounds.
Motivation
The setting of goals is important to guide the company and employees toward
defined objectives. Motivation plays a critical role in achieving goals and
business objectives and is equally as important for organizations that work in a
team-based environment or in a workplace comprised of workers who work
independently. There are many reasons why motivation is important. One of the
reasons is because it makes employees to work hard thus leading to high
productivity and improved work quality in organizations and companies.
Financial gain is also another reason why motivation is important.
Empowerment
One of the core components of the management is that of having everyone in the
organization involved in managing and improving quality of the processes for
which they are responsible. This might be done as part of day to day operations
as a member of a natural work team or a self directed team, or in becoming part
of a group that is going to take on a special process design or improvement
project. Regardless of the way it occurs, it involves giving employees greater
responsibility and authority, and is commonly labeled empowerment.
Empowerment is based on the belief that employees have the ability to take on
more responsibility and authority than has traditionally been given them, and that
heightened productivity and a better quality of work life will result. Different
76
words and phrases are used to define empowerment, but must are variations on a
theme: to provide employees with the means for making influential decisions.
Juran define empowerment as" conferring the right to make decisions and take
action.
The cause and effect relationships are the main element in balanced scorecard.
The previous strategy set hypotheses cause and effect relationships between its
perspectives. These relationships between cause and effect can be expressed by a
sequence of if then statements. This relation pointed out by Kaplan and Norton
as shown in Figure 5.1.
Financial
Customer
Vision
and
mission
Internal process
That the direction of the relationships from (financial - customer, internal process
- financial, internal process - learning & growth and learning & growth - internal
process) are illogical. Usually shareholder spends on learning & growth and
internal processes to deliver high quality products or services to meet the
customer requirements. Customer loyalty can be achieved when getting high
quality product or services repeatedly and that returns profit on the organization.
With the addition of the new perspective, "management commitment" we
propose that its role should be central for managing the various aspects of the
vision and mission. These reasons make us think about changing relationships
between the perspectives of the balanced cards as shown in Figure 5.2.
77
Financial
Management
Vision
&
mission
Internal process
Commitment
Customer
5.3
Model Preparation
78
Convey the importance of a new project system for the staff as far as its
ability to increase job fulfillment, empowerment and minimize resistance
to change.
Communicate that these metrics will be developed collectively and are not
to punish individuals but rather to improve the effectiveness of their work
and develop communication mechanisms with the employees so that they
can track and understand the progress and the inherent benefits of the
project.
The proposed implementation action steps.
The proposed timeline of the project.
This action is crucial, since it has been empirically proven that the active
involvement of employees in the design of a performance measurement system
will lead to a significantly larger performance increase, compared to an
introduction of an identical system whereby employees do not participate (telland-sell) in the deployment process. For this reason, at the beginning of the BSC
project, interviews must be scheduled and questionnaires could be distributed to
selected employees from all hierarchical levels, in order to assess change
readiness, establish buy in for the imperative to change and clarify the role of
each employee along with the personal and corporate level benefits from the
implementation of this strategic management tool [44].
79
drivers in the other perspectives, whilst ensuring a cause and effect relation
between them. The selection of the 1015 most appropriate strategic objectives
can be achieved in a project team workshop by ranking each strategic objective
on categories.
(10) Revaluation of
the completed
scorecard done
BSC
Model
(4) Identify strategic
objectives
(8) Developing
strategic programs for
achieving objectives
(7) Balanced
Scorecard Weights
strategy mapping, and thereupon, the findings of this phase must also be
presented and explained to selected stakeholders [57].
This is the first stage; a problem is decomposed into a hierarchical structure that
consists of an objective (i.e., overall goal of the decision making), the general
criteria which impact the goal directly, sub-criteria (objectives), sub-sub-criteria
(measures) etc.
In this stage, to determine the criteria weights, a team of experts formed pair wise
comparison matrices for evaluating the criteria. Each expert of the team has
individual evaluation. Computing the geometric mean of the values obtained
from individual evaluations, a final pair wise comparison matrix on which there
is a consensus is found. The weights of the critical success factors are calculated
based on this final comparison matrix as shown in the next Table 5.1 and
equations.
Table 5.1. The Pair-wise Comparison of Linguistic Variables Using Fuzzy Numbers
Intensity of fuzzy scale
~
1
Fuzzy number
(L,M,U)
User define
= ( ,1, )
~
3
~
5
~
7
~
9
~~~~
2, 4 ,6 ,8
(L,M,U)
= ( ,3, )
(L,M,U)
= ( ,5, )
(L,M,U)
= ( ,7, )
(L,M,U)
= ( ,9, )
Intermediate values
(L,M,U)
= ( , , )
Where: L is the lower limit, M is the medium limit, U is the upper limit.
83
Next, from the information of the pair wise comparison, we can form the fuzzy
positive reciprocal matrix as follows:
a11 a1 j ain
(5.1)
Then, the geometric mean method for finding the final fuzzy weights of each
criterion can be formulated as follows eq.5.2, 5.3:
Wi = ri * (r1+r2++rn)-1,
(5.2)
ri = (ai1* ai2*.ain)1/n
(5.3)
Where
After determining the weights for lower, median and upper comparison matrices,
the weighting average is computed to get the final weights.
In the last stage, calculated weights of the factors are approved by decision
making team. Ranking firms are determined by using TOPSIS method in the
third stage. TOPSIS method is one of the well known ranking methods for
MCDM. TOPSIS is firstly proposed by Hwang and Yoon. This technique based
on the concept that rank alternatives, which has the shortest distance from the
ideal (Best) solution and the longest distance from the ideal (worst) solution.
Steps of TOPSIS
Step 1: Construct normalized decision matrix. This step transforms various
attribute dimensions into non-dimensional attributes, which allows comparisons
across criteria.
Normalize scores or data as follows eq. 5.4:
rij = xij/ (a2ij) for i = 1, , m; j = 1, , n
(5.4)
84
Multiply each column of the normalized decision matrix by its associated weight
by eq. 5.5 an element of the new matrix is:
vij = wj rij
(5.5)
Step 3: Determine the Ideal and Negative Ideal Solutions by eq. 5.6.
Ideal solution
A* = { v1*, , vn*}, where
vj*={ max (vij) if j J ; min (vij ) if j J' }
(5.6)
(5.7)
Step 4: Calculate the Separation Measures for Each Alternative by eq. 5.8. The
separation from the ideal alternative is:
Si * = [
(vij vj*)2 ]
i = 1, , m
(5.8)
Similarly, Calculate the separation from the negative ideal alternative by eq. 5.9 is:
S'i = [
(vij vj' )2 ]
i = 1, , m
(5.9)
Step 5: Calculate the Relative Closeness to the Ideal Solution Ci* eq. 5.10
Ci*= S'i / (Si* +S'i ) ,
Ci* 1
(5.10)
85
86
87
CHAPTER 6
APPLYING THE BALANCED SCORECARD TO FACULTY
OF ENGINEERING
6.1 Introduction
Higher education institutes are facing new challenges in order to improve the
quality of education. There is a pressure for restructuring and reforming higher
education in order to provide quality education and bring up graduates who
become fruitful members of their societies. In higher education as in business
there are acceptable conventions of measuring excellence. As a result, the
implementation of Balanced Scorecard in higher education has been a target of
interest in recent years. However, rather than emphasizing on financial
performance, higher education has emphasized on academic measures in its
Balanced Scorecard.
The proposed BSC model is applied in Faculty of engineering (A). Faculty of
Engineering (A) has five scientific departments, namely:
- Department of Civil Engineering
- Department of Mathematics and Physics Engineering, a division not enrolls
students
- Department of Electrical Engineering
- Department of Architecture
- Department of Industrial Engineering
The Faculty (A) has nearly 182 faculty members and their assistants who teach
for about 2500 students.
several meetings to raise the awareness of quality and actions that should be
implemented to achieve steps toward accreditation. The required data has been
obtained with the help of this team and also through randomly questionnaire.
Data of the unity of quality and accreditation were analyzed. In addition, 110
copies of a questionnaire were distributed among which 86 copies of
questionnaire were returned (return rate of 78.18%). Table 6.1 shows the
characteristics of the study sample society. As part of the verification of the
questionnaire, expert's person reviewed it and acknowledged its validity for
display devoted to measure. In order to test the reliability of the questionnaire,
Cronbachs Alpha was found to be 0.727, which indicated that the questionnaire
has high internal reliability.
Table 6.1: Sample Characteristics
Factor
Position Professor
Assistant Professor
Lecturer
Stuff
Researcher
Graduate
Student
Frequency Percentage
7
8.14%
8
9.30%
12
13.95%
10
11.62%
6
6.97%
13
15.11%
30
34.88%
Professor Assistant
Professor
0.4
89
0.8
0.6
0.7
0.9
0.3
0.9
0.52
0.1
0.0
0.3
0.7
.07
0.3
0.8
0.1
0.73
0.1
-0.1
Ranking
5
4
0
F
IP
L&G
MC
Alternatives
Student
Graduate
Researcher
Stuff
Lecturer
Assistant Professor
Professor
knowledge, skills, good behavior, and the ability to think creatively and continue
learning so that they can compete in the local labor market and the Arab world
and face the challenges of the twenty-second oven. Quality characteristic of each
activities Faculty of engineering, and is keen Faculty to maximize its role in
serving and development environment and actively participate in projects,
programs and plans for economic development and social, a locomotive for
development and enlightenment the local community through educational
programs, research and service centers in the university.
90
Values
The higher values from which the faculty of Engineering (A) following are:
Leadership (Aleightcar, creativity and excellence)
Quality (continuous improvement with professionalism and proficiency)
Ethics
Collective Action
Loyalty
92
Threats
The country's sluggish evolution reflects the absence of partial budget
support.
The uncertainty of future annual budgets leads to poor long-term planning.
The limited budget of the University reflects a kind of difficulties in
spending colleges.
Lack of competitive advantage that will help the Faculty to compete with
other universities.
Migration of faculty members to private universities because of the
financial incentive they receive unlike public universities.
Provide quality educational service after studying the needs of the local
market, regional and global levels of graduates.
94
Objectives
Indicator
A5
A6
B1
B2
B3
B4
C1
C2
C3
C4
C5
C6
C7
C8
C9
C10
C11
A1
A2
A3
Financial
(FL)
Customer
(CT)
Development
the finance
Capacity
Customer
satisfaction
Customer
retention
Customer
acquisition
Reduce
customer
complaints
Quality
educational
service.
Internal
process (IP)
A4
C12
Contribute to
the
development
of integrated
community.
C13
C14
C15
D1
Learning
and growth
(LRG)
Raise the
efficiency of
faculty.
D2
D3
D4
Work on
development
D5
95
educational
process.
D6
Faculty accreditation
Preparation
courses for
continuing
Education and
training.
Leadership
Management
commitment
(MC)
`
Complete the
organizational
structure.
Development
motivation
mechanics.
Development
culture
Development
empowerment
mechanics
E6
E7
E8
E9
E10
96
Increase tuition
Financial
(FL)
Customer
(CT)
Internal
process (IP)
Learning and
growth (L&G)
Management
commitment
(MC)
Increase
external grants
fees
Increase Budget
Increase contracts
with industry
Increase Students
'satisfaction
Improve
university image
Increase
academic stuff
Increase customer
satisfaction
Increase
teaching aids
Increase papers
research
Leadership
Increase of practical
projects
Culture Development
Increase
Empowerment
Increase
environmental
research
Motivation
Organizational
structure.
97
FL
[1,1,1]
[2,3,4]
[4,5,6]
[4,5,6]
[8,9,9]
CT
[1/4,1/3,1/2]
[1,1,1]
[1,1,1]
[1,2,3]
[3,4,5]
IP
[1/6,1/5,1/4]
[1,1,1]
[1,1,1]
[6,7,8]
[2,3,4]
L& G
[1/6,1/5,1/4]
[1/3,1/2,1]
[1/8.1/7,1/6]
[1,1,1]
[1,2,3]
MC
[1/9,1/9,1/8]
[1/5,1/4,1/3]
[1/4,1/3,1/2]
[1/3,1/2,1]
[1,1,1]
Step2: Separations
Three matrices (Lower, Medium, and Upper) are separated from the original
matrix. The priority is calculated for each matrix separately.
Step3: Synthesizing the Pair wise Comparison
Normalized vector is computed according to eq. (4.3), where aij the element with
I raw and j column. To calculate the vectors of priorities, sum the each element in
the normalizing column. Then, divide the elements of each column by the sum of
the column as shown in eq. (4.2). The result is priority vector as shown in table
6.5. This process is done on all three matrices (Lower, Medium, and Upper).
Table 6.5: Synthesizing the Pair wise Comparison
Goal
FL
CT
IP
L&G
MC
FL
1
2
4
4
8
CT
1/4
1
1
1
3
IP
1/6
1
1
6
2
sum
L& G
1/6
1/3
1/8
1
1
MC
1/9
1/5
1/4
1/3
1
NV
0.2385
0.6683
0.6597
1.5156
2.1689
5.2512
Priority
0.0454
0.1272
0.12563
0.2886
0.41303
1
0.045
0.127
0.126
0.289
0.419
2
0.039
0.120
0.110
0.298
.4302
3
0.038
0.124
0.102
0.318
0.414
98
Average weights
.0414
0.124
0.112
0.302
0.4194
FL
0.100
0.199
0.398
0.398
0.796
CT
0.072
0.288
0.288
0.288
0.864
IP
0.026
0.154
0.154
0.926
0.309
L& G
0.114
0.227
0.085
0.681
0.681
MC
0.100
0.181
0.226
0.301
0.903
Step 7: Construct the Weighted Normalized Decision Matrix by Using eq. (4.5).
In this step multiply each column of the normalized decision matrix by its
associated average weight in table 6.5 as shown in table 6.8.
Table 6.8: The Weighted Normalized Decision Matrix
Goal
FL
CT
IP
L&G
MC
FL
0.004
0.008
0.016
0.016
0.033
CT
0.009
0.036
0.036
0.036
0.107
IP
0.003
0.017
0.017
0.104
0.035
L& G
0.034
0.069
0.026
0.206
0.206
MC
0.042
0.076
0.095
0.126
0.378
0.033
0.004
0.107
0.009
0.104
0.003
0.206
0.026
0.378
0.042
sum each element in the row of separate on matrix. Calculate the root of the sum
for each element in matrix to find the final separation. The separation from the
negative ideal alternative can be calculated as the separation from the ideal
alternative by eq. (4.9) as show in table 6.11.
Table 6.10: The Separation from the Ideal Alternative
Goal
FL
CT
IP
L&G
MC
FL
0.001
0.001
0.000
0.000
0.000
CT
0.010
0.005
0.005
0.005
0.000
IP
0.010
0.007
0.007
0.000
0.005
L& G
0.029
0.019
0.032
0.000
0.000
MC
0.113
0.092
0.081
0.064
0.000
FL
0.000
0.000
0.000
0.000
0.001
CT
0.000
0.001
0.001
0.001
0.010
IP
0.000
0.000
0.000
0.010
0.001
L& G
0.000
0.002
0.000
0.032
0.032
MC
0.000
0.001
0.003
0.007
0.113
Step 10: The relative closeness to the ideal solution is calculated by using the
eq.(4.10). In this step each element in row of separation from the negative ideal
alternative divides by the sum of separation ideal and negative ideal alternative.
Then, the final ranking is presented in table 6.12.
Table 6.12: Result of Ranking
Perspective
Rank
FL
5
CT
3
IP
4
L& G
2
MC
1
This proces is repeated for all stratgic objectives and all indecators .[96]
The Balances Scorecard Weighting Results
The results of this research show that "Management Commitment" is the most
important perspective of educational balanced scorecard in faculty of engineering
(A) in the first level. It is noted from table 6.12 that the most important
perspective after management is learning and growth and the least important is
finance. When going to the second level the result show that " Motivation
development"," education development", "customer satisfaction",
"Quality
101
B4 (.085)
Complains Reduction
(CR) (.085)/(4)
B3 (.129)
Customer Acquisition
(CA) (.129)/(2)
B2 (.224)
B1 (.559)
Customer Satisfaction
(CS) (.559)/(1)
Customer (CT)
(.124)/ (3)
D8 (.119)/(2)
D9 (.559)/(1)
D10 (.264)/(4)
C10 (.044)/(10)
C11 (.069)/(5)
C12 (.105)/(3)
D6 (.257)/(2)
D5 (.743)/(1)
D4 (.16)/(4)
D3 (.45)/(1)
D2 (.27)/(2)
D1 (.1)/(3)
Faculty
Efficiency (FE)
D7 (.0563)/(3)
Educational
Development (ED)
Continuing
Education (CE)
C15 (.319)/(2)
C14 (.567)/(1)
C13 (.113)/(3)
Community
Development (CD)
C9 (.039)/(11)
C8 (.085)/(4)
C7 (.029)/(12)
C6 (.136)/(1)
C5 (.095)/(9)
C4 (.074)/(7)
C3 (.145)/(2)
C2 (.095)/(6)
C1 (.083)/(8)
Education
Quality (EQ)
Internal process
(IP) (.112)/ (4)
Fig.6.3: The Weights / Ranking of the Perspectives, Objectives, and Performance Indicators
A6 (.064)/(5)
A5 (.095)/(6)
A4 (.068)/(4)
A3 (.284)/(2)
A2 (.086)/(3)
A1 (.401)/ (1)
Financial Development
(DF) (.41)/(1)
Financial (FL)
(.041)/ (5)
Mission
Vision
E8 (.249)/(2)
E7 (.741)/ (1)
Motivation Developing
(MD)(.321)/(1)
E6 (.5)
E5 (.5)
E9 (.193)
Culture Developing
(CD)(.152)/(4)
E4 (.25)
E3 (.25)
E2 (.25)
E1 (.25)
E10 (.129)
Empowerment
Developing (ED) (.193)/(3)
Leadership (L)
(.204)/(2)
Commitments /Management
(MC) (.419) /(1)
(A). Figure 6.3 shows the rankings of the perspectives, objectives, and
performance indicators.
Weight Chart
Weight chart makes the exact weight for perspectives and strategic objectives in
the entire System apparent (Figure 6.4 and Figure 6.5).
Prespectives Weights
0.4194
0.302
0.124
0.112
CT
IP
0.0414
FL
L&G
MC
0.524
0.41
0.224
0.129
0.085
DF
FL
CS
RR
CA
CT
CR
0.321
0.301
0.25
0.2040.193
0.1510.129
0.173
EQ
CD
FE
IP
ED
CE
OS
L&G
102
CD
MC
ED
Objectives
Development
the finance
Capacity 20 %
at the end of
2017
Contribute to the
development of
integrated
community
within 10% in
2015
Applied the
planned
maintenance for
IT system at the
end of 2014
Increasing the
scientific
research within
20% in 2016
Adoption
scientific journal
of the Faculty
Suggestions
Preparing the plan
of projects that
need to achieve
next years
Review the plan in
university meeting
to get the budget
Accreditation more
Laboratory
The announcement
of the accredited
laboratories
Creating practical
solution for
industrial
companies
Time
Responsible
Six
months
The administrator of
faculty meeting
Six
months
Dean of faculty
One year
The administrator of
especial units
One year
The administrator of
especial units
Two years
Faculty members of
Graduation Project
teams
Increasing the
projects which
developing the
environment
One year
The administrator of
environment
developing
Review the IT
system once per
month
One year
The administrator of
IT system
Six month
Faculty members
One year
Academic members
One year
Faculty members
One year
Dean of faculty
Increasing the
scientific research
with successive
position
directed towards
patents and
international
awards
Reward innovative
research
Accreditation
scientific journal
of the Faculty
103
Status
Adoption the
motivation policy
Awareness
Applied system
Review the system
Preparing to
accreditation
One year
Six month
One year
One year
Six month
Faculty members
Quality assurance
and accreditation
team
Faculty (A)
1584671
625000
5083000
2310000
1116400
1360000
1215000
9538110
86%
1
463000
7632000
71%
13
163
21
1/6
1/9
87%
46
40
1/2
1/14
60%
93%
65
46
90%
15%
75%
37
15
80%
30%
35945
Once per year
5680
Once per year
Indicator
Budget allocated annually.
Annual revenue from tuition fees compared with the
budget allocated annually.
The value of contracts with industry annually pound
compared with the budget allocated annually.
Annual revenue from foreign donations with the
budget allocated annually.
The value of external grants for the allocated budget.
Cost of salaries for the budget annually.
Ratio of Students satisfaction grade.
Universitys position in national and international
rankings.
The number of contracts with industry annually.
Number of complaints made annually.
Ratio of staff to academic staff
Ratio of academic staff to students
Ratio of academic staff satisfaction of education
level.
Ratio of students' satisfaction of education level.
Number of classrooms to the number of students.
Number of laps for faculty.
Ratio of students Satisfaction of teaching aids.
Ratio of things that need to be learned after
graduation.
Number of books in library
Time cycle for up-to-dating the library.
104
90%
Run to failure
76%
Run to failure
100
50
40%
26%
65%
15
60%
15
115
20%
60
11%
81%
55%
59%
32%
52%
33%
30%
10%
77%
59%
90%
75%
98%
95%
93%
90%
Centralization
80%
30%
80%
Centralization
80%
16%
60%
105
Performance Chart
Performance chart is an easy to understand column graph that shows the
performance for the whole system as shown in:
Figure 6.6 illustrates the differences between financial performance of faculties
of engineering (A) and (B). The variance is clear in column (A1) budget
allocated annually and (A3) value of contracts with industry.
Financial prespective
Fayoum
1A
2A
3A
Cairo
4A
5A
6A
The following Figure 6.7 shows the differences between the performance of
Faculties of Engineering (A) and (B) in dealing with customer perspective. The
Figure 6.7 illustrates the ability of faculty (B) to acquire customers, it is
considered a guide to the rest of the indicators customer satisfaction and knowing
the proper handling of their problems.
106
Customer
Cairo
B
1B
Fayoum
A
2B
3B
4B
Fig. 6.7: The Differences between the Performance of Faculties of Engineering (A) and
(B) in Dealing with Customer Perspective
The following Figure 6.8 presents the comparison between the performances of
the internal processes in faculties of engineering (A) and milestone (B). The
difference in the intensity of the number of labs in the faculty of engineering (B)
is more than the faculty of engineering (A).
Internal Process
Cairo
B
1C
2C
3C
4C
5C
6C
Fayoum
A
7C
8C
9C
10C
11C
12C
Fig. 6.8: The Differences between the Performance of Internal Process Perspective of
Faculties of Engineering (A) and the Milestone (B).
107
The following Figure 6.9 presents the comparison between the performances of
the learning and growth perspective in faculty of engineering (A) and milestone
(B). The differences is clear in the large number of the researches carried out by
the faculty of engineering (B), increasing the percentage of faculty members who
are trained and increasing the practical projects carried out by the faculty.
1D
2D
3D
4D
A
Fayoum
5D
6D
7D
8D
9D
10D
Fig. 6.9: The Differences between Learning and Growth Perspective Performance of
Faculties of Engineering (A) and the Milestone (B).
Management
Cairo
B
1E
2E
3E
4E
Fayoum
A
5E
6E
7E
8E
9E
10E
108
109
CHAPTER 7
SAMMARY AND CONCLUSIONS
7.1 Introduction
In the previous chapters, balanced scorecard is discussed circumstantially.
Chapter 1 presented the literature review. Chapter 2 discussed balanced scorecard
benefits and limitations. Chapter 3 compared between MADM methods. Chapter
4 proposed integration Fuzzy (GMM) - TOPSIS model to assign the balanced
scorecard perspective's weights and verified this model by various cases. Chapter
5 presented the new balanced scorecard model which combining of origin BSC,
new perspective "management commitment" and prioritization technique Fuzzy
(GMM) - TOPSIS. Chapter 6 discussed the results of the practical study. This
chapter focuses on summary and conclusions as well as a critical evaluation of
the research.
7.2 Summary
The balanced scorecard makes organizational strategy tangible which makes a
strategy more easily understood by employees, and enables the strategy to
become visible knowledge through organizational communication. The balanced
scorecard also can be a catalyst for sharing strategy with organization. It clarifies
each employee's responsibility, and contributes to integrating the corporate
culture with the strategy.
The strategy that is described well by the balanced scorecard, urges an
organization to create the capability that an organization wants to have or
improve. What the organization wants becomes explicit in the four perspectives
model. The process to connect the capability with the strategy necessary
generates communication between top management and employees. Well
attracted strategy in the strategy map and the scorecard enable people easily to
understand why, what, and how to have such capabilities.
110
The literature review illustrated that there are opportunities to develop balanced
scorecard. These opportunities are to develop the weights of perspectives, causeeffect relationships, implementation, time-dimension, communication within an
organization, feedback about the organizations strategy and management
commitment.
Different MADM methods are used to develop balanced scorecard and
prioritizing its perspectives. In the scope of the study, the integrated Fuzzy
(GMM) - TOPSIS model is proposed to estimate BSC indicator's weights. This
model is verified by comparing the results with other reported research works.
Sensitivity analyses for the model provide the smallest ED among other model
which means more accurate results of priority.
The traditional BSC contains four perspectives (i.e., financial, customer, internal
business processes, and organizational learning and growth activities), this thesis
added new perspective "management commitment" to develop and support the
balanced scorecard implementation process. The concept of management
commitment is the degree of members' recognition of organizational goals and
values that they are willing to work extraordinarily hard to help the organization
complete its goals. The origin BSC, perspective "management commitment" and
prioritization technique Fuzzy (GMM)- TOPSIS are the new integrated BSC
model. The new balanced scorecard model is combining of origin BSC, new
perspective "management commitment" and prioritization technique Fuzzy
(GMM) - TOPSIS. This model contains ten steps for establishing balanced
scorecard in the organization.
Higher-education plays a vital role in countries economic growth and shaping
the future of the nation. This thesis implemented the new model of balanced
scorecard in the Faculty of engineering (A). This model provides a perfect
framework to determine the importance of each of these perspectives, objectives
and indicators. The conclusions of the results are stated in the next section.
The management by utilizing the balanced scorecard encourages an organization
to internalize strategic ways of thinking as a normative mantel model. The
111
7.3 Conclusion
The implementation of the BSCs involves several challenges in embedding the
information systems and the culture that help receive, process, and analyze
performance information, and to implement action plans in time to respond to
unexpected turns of events. A BSC requires time and effort to "get it right",
commitment to implement it, and perseverance to ensure that it is used and
updated.
The following conclusions may be drawn from this research:
-
Fuzzy (GMM) is used for computing weight which is an advanced step for
TOPSIS to finding the final rank, fast, precise and easy.
Sensitivity analysis for the model provide that it has the smallest ED
among other models and its sensitivity to change in alternative weighs is
the best between all other (VIKOR, SAW) which means more accurate
result of priority.
The proposed model which integrates the BSC with Fuzzy (GMM)
TOPSIS method shows to be a feasible and effective assessment model for
faculty of engineering performance evaluation and it could be extended to
other faculties as well, or digging deeply to assesses faculty department
also.
The most significant advantage of the use of the balanced scorecard is that
it provides a wider development of metrics that are closely connected to
the strategic goals of institution (here faculty of engineering).
113
REFERENCES
[1] Javad Dodangeh, Rosnah Bt Mohd Yusuff, Javad Jassbi," Using Topsis Method
with Goal Programming for Best selection of Strategic Plans in BSC Model ", Journal
of American Science , Vol.6, No.3, 2010.
[2] Kaplan, R. and D. Norton," The Balanced Scorecard - Measures that drive
performance ", Harvard Business Review, Vol.70, No.1, pp71-79, 1992.
[3] Kaplan, R. S., and Norton, D. P., "Using the Balanced Scorecard as a Strategic
Management System", Harvard Business Review, Vol.74, No.1, pp.74-85, 1996a.
[4] Malczewski J. , " GIS-based Multi Criteria Decision Analysis", International Journal
of Geographical Information Science, Vol.20, No.7, pp. 703726, 2006 .
[5] Der-JangChi & Hsu-Feng Hung, "Is the balanced score card really helpful for
improving performance? Evidence from software companies in China and Taiwan",
African Journal of Business Management, Vol.5, No., pp.224-239, 2011.
[6] Bernard Marr," Balanced score card explained, examples, templates and case
studies KPI library", International Journal of Electronic Business Management ,Vol.1,
No.2, pp. 21- 35, 2011.
[7] Kaplan R. & Norton D., "The strategy focused organization." New York Harvard
business school press, 2001.
[8] Ali-Rahimi M.," Joint application of balanced scorecard and excellence model using
TQFM, SWOT and MADM tools", Arabian Journal of Business Management and
Review, Vol.2, No.5, pp. 1-7, 2013.
[9] Horngreen C. T., "A Managerial Emphasis, Pearson Education ", International of
Prentice Hall, Cost accounting, Vol.8, No.6, 2009.
[10] Abbas Asosheh, Soroosh Nalchigar, Mona Jamporazmey," Information technology
project evaluation: An integrated data envelopment analysis and balanced scorecard
approach", Expert Systems with Applications , Vol.37 , No.5, pp.59315938, 2010.
[11] Nurul Fadly Habidin," Balanced scorecard strategy in automotive industry",
LASSIB Forum, Vol.30, No.6, 2012.
[12] Amir Manian, Asie Omidian," Performance Evaluating of IT Department Using a
Modified Fuzzy TOPSIS and BSC Methodology (Case study: Tehran Province Gas
Company)", Journal of Management Research, Vol.3, No.2, pp.899-1941, 2011.
[13] Ing-Long Wu, Ching-Hui Chang," Using the balanced scorecard in assessing the
performance of e-SCM diffusion: A multi-stage perspective", Decision Support
Systems, Vol.52, No.5, pp.474485, 2012.
[14] Deok Joo Lee, Sung-Joon Park, Kyung-Taek Kim," A Development of Key
Performance Indicators for the Public R&D of Energy Technology using Balanced
Scorecard Approach", Proceedings of the World Congress on Engineering and
Computer Science (WCECS), Vol. 1, No.2 ,pp. 23-25, 2013.
114
[15] Bhagwat, R., & Sharma, M. K.," Performance measurement of supply chain
management: A balanced scorecard approach", Computers and Industrial Engineering,
Vol.53, No.5, pp.4362, 2007.
[16] Chwan Yi, C., Lin, B. ,"An Integration of Balanced Scorecards and Data
Envelopment Analysis for Firms Benchmarking Management.", Total Quality
Management, Vol.11, No.20, pp.1153-1172, 2009.
[17] Zhang, Y., Li, L.," Balanced Scorecard of Commercial Bank in Performance
Management System ", International Symposium on Web Information Systems and
Applications (WISA), Vol.22, No.24, pp.206-209, 2009.
[18] Ahmed, Z., Bowra, Z., Ahmad, I., Nawaz, M., & Khan, M. ," Performance
Measures Used by the Commercial Banks in Pakistan within the Four Perspectives of
Balanced Scorecard", Journal of Money, Investment and Banking, Vol.21, No.5 pp.1220, 2011.
[19] Al Sawalqa, F., Holloway, D., & Alam, M.," Balanced Scorecard Implementation
in Jordan: An Initial Analysis", International Journal of Electronic Business
Management, Vol.3, No.9, pp.196-210, 2011.
[20] Sabah M. Al-Najjar, Khawla H. Kalaf," Designing a Balanced Scorecard to
Measure a Bank's Performance: A Case Study", International Journal of Business
Administration, Vol.3, No.4, pp.4-9, 2012.
[21] Kettunen, J.," Strategies for the cooperation of educational institutions and
companies in mechanical engineering", International Journal of Educational
Management, Vol. 20, No.1, pp.19-28, 2006.
[22] Greiling, D., "Balanced Scorecard Implementation in German non-profit
Organizations", International Journal of Productivity and Performance Management,
Vol. 6, No.59, pp.534-554, 2010.
[23] Kollberg, B., & Elg, M., "The Practice of the Balanced Scorecard in Health Care
Services", International Journal of Productivity and Performance Management, Vol. 5,
No.60, pp.427-445, 2010.
[24] Sunisa Atiwithayaporn, Wanchai Rivepiboon," A Development of the
Effectiveness Evaluation Model for Agile Software Development using the Balanced
Scorecard", Proceedings of the International Multi Conference of Engineers and
Computer Scientists, Vol.1, No. 2, pp.13 - 15, 2013.
[25] Karathanos, D., Karathanos, P.," Applying the balanced scorecard to education",
Journal of Education for Business, Vol.80, No.4, pp. 222-230, 2005.
[26] Chang, O.H. and Chow, C.W., " The balanced scorecard: a potential tool for
supporting change and continuous improvement in accounting education", Journal
Accounting Education, Vol.14 No.3, pp.395-412, 1999.
115
117
[66] Lohman, C., Fortuin, L. and Wouters, M., "Designing a Performance Measurement
System: A Case Study", European Journal of Operational Research, Vol.156, No.2,
pp.267-286, 2004.
[67] Alexandros, P., George, I., Gregory P. and Soderquist, K., "An Integrated
Methodology for Putting the Balanced Scorecard into Action", European Management
Journal, Vol.23, No.2, pp.214-227, 2005.
[68] Pandey, I.M.," Balanced Scorecard: Myth and reality", Journal for Decision
Makers, Vol.30, No.1, pp.51-64, 2005.
[69] Kennerley, M. & Neely, A.," A Framework of the Factors Affecting Evolution of
Performance Measurement System", International Journal of Operations and Production
Management, Vol.22, No.11, pp.12-22, 2002.
[70] Neely, A., Gregory, M. & Platts, K. ," Performance measurement system design: A
literature review and research agenda", International Journal of Operations and
Production Management, Vol.25, No.12, pp. 12-28, 2005.
[71] Johnson, J.J.," Balanced scorecard for state-owned enterprises", Philippines: Asian
Development Bank, Vol.2, No.1 pp. 6068, 2007.
[72] Hwang, H.N., RAU, H. ," Design and planning of the balanced scorecard: A case
study", Journal of Human System Management, Vol.26, No.1, pp. 2-8, 2007.
[73] Gumbus, A. & Lussier, R.N.," Entrepreneurs use a Balanced Scorecard to translate
strategy into performance measures", Journal of Small Business Management, Vol.44,
No.3, pp.407-408, 2006.
[74] Anthens, G.H., "Balanced Scorecard. In: Computerworld Balanced Scorecard
Institute. Vol.37, No.7, pp.26- 34, 2009.
[75] Noell, Chr. and Lund, M.," The Balanced Scorecard (BSC) for Danish Farms
Vague Framework or Functional Instrument?" vol.11, No.95 pp.101- 109, 2010.
[76] Russell T.westcott," The Certified Manager of Quality Organizational Excellence
handbook'', 2006.
[77] Niven, P.R.," Balanced Scorecard. Step-by-step: Maximizing performance and
maintaining results", New York: John Wiley & Sons, Inc., Vol.4, No.6, 2002.
[78] Gwo-Hshiung, Tzeng and Jih-Jeng Huang," New Frontiers of Multiple Attribute
Decision Making", kainan university, 2007.
[79] Basar oztaysi, Irem UCal, "Comparing MADM Techniques for Use in Performance
Measurement", Proceedings of the International Symposium on the Analytic Hierarchy
Process 2009.
[80] Heracles Polatidisdias a. Haralambopoulos, Giussepe Munda, Ron Vreeker,"
Selecting an Appropriate Multi-Criteria Decision Analysis Technique for Renewable
Energy Planning", Vol.1, No.1, pp. 81-193, 2006.
119
[81] Meysam Borajee, Siamak Haji Yakchali," Using the AHP-ELECTRE III integrated
method in a competitive profile matrix", Vol.11, No.2, pp. 86-93, 2011.
[82] Milan Janic, Aura Reggiani, "An Application of the Multiple Criteria Decision
Making (MCDM) Analysis to the Selection of a New Hub Airport", Vol.2, No.2,
pp.113, 2002.
[83] Rafael De Santiago, Javier Estrada, " Geometric Mean Maximization: Expected,
Observed, and Simulated Performance", 2011.
[84] Alireza Afshari, Majid Mojahed and Rosnah Mohd Yusuff, " Simple Additive
Weighting approach to PersonnelSelection problem, Vol.1, No.5, pp.24- 38, 2010.
[85] Opricovic, S., and Tzeng, G-H., "Compromise solution by MCDM methods: A
comparative analysis of VIKOR and TOPSIS", European Journal of Operational
Research, Vol. 156, pp. 445455, 2004.
[86] Mansour Momeni, Mohammed Reza Fathi, Mohammed Karimi Zarchi and Sirous
Azizollohi," Fuzzy TOPSIS based approach to maintenance strategy selection: case
study", Vol.3, No.2, pp.699-706, 2011.
[87] Mohammed F. Aly , Hazem A. Attia and Ayman M. Mohammed, " Integrated
Fuzzy (GMM) -TOPSIS Model for Best Design Concept and Material Selection
Process", Vol.2, No.11, pp.2319-8753, 2013.
[88] Bojan Srdjevic, "Combining Different Prioritization Methods in the Analytic
Hierarchy Process Synthesis", Al Sevier, Computers and Operations Research, Vol.32,
No.7, pp.1897-1919, 2005.
[89] Ying-Ming, W., Celik P. and Ying L., " A Linear Programming Method for
Generating the Most Favorable Weights from a Pair-Wise Comparison Matrix" ,
Computers and Operations Research", Vol.35, No.12, pp.3918-3930, 2008.
[90] Hambali Ariff, Mohd. Sapuan Salit, Napsiah Ismail& Y. Nukman, "Use of
Analytical Hierarchy Process (AHP) for Selecting The Best Design Concept", Journal
Technology, Vol.49, pp. 118, 2008.
[91] Ali Jahan, Md Yusof Ismail, Faizal Mustapha, Salit Mohd Sapuan," Material
selection based on ordinal data", Vol. 31, pp.31803187, 2010.
[92] Ashu Sharma, "Implementing Balance Scorecard for Performance Measurement",
Faculty, The Icfai Business School, Jaipur, India Vol.15, No.3, 2009.
[93] Kwang, M., Young W., Seung, H., Jae, H. and Kyoung, S., "A Study on
Development of Balanced Scorecard for Management Evaluation Using Multiple
Attribute Decision Making", Journal of Software Engineering and Applications, Vol.3,
No.3, pp.268-272, 2010.
[94] Yee-Ching and Lilian Chanan, " An Analytic Hierarchy Framework for Evaluating
Balanced Scorecards of Healthcare Organizations", Canadian Journal of Administrative
Sciences, Vol.23, No.2, pp.85-104, 2009.
120
[95] Amy, H.I., Lee, Wen-C., Chen and Ching-Jan, C.,"A Fuzzy AHP and BSC
Approach for Evaluating Performance of IT Department in the Manufacturing Industry
in Taiwan", Expert Systems with Applications, Vol.34, No.1, pp.96-107, 2008.
[96] Mohammed F. Aly , Hazem A. Attia and Ayman M. Mohammed, " Establish the
Balanced Scorecard and Prioritizing the Performance Indicators by Using GMMTOPSIS Approach in Faculty of Engineering ", international conference in prospects of
Engineering Solutions and challenges of the times, Faculty of engineering Fayoum
university, 2013.
121
APPENDICES
Appendix A: MATLAB Code for Fuzzy (GMM)-TOPSIS Model
Fuzzy (GMM) Code
%% GMM
Clc
a=input(' Input Judgment matrix = ')
[n1 n2]=size(a)
for i1=1:n1
for i2=1:n2
if a(i1,i2)>1
b(i1,i2)=a(i1,i2)+1;
c(i1,i2)=a(i1,i2)-1;
elseif a(i1,i2)==1
b(i1,i2)=1;
c(i1,i2)=1;
else
b(i1,i2)=1/a(i1,i2)-1;
c(i1,i2)=1/a(i1,i2)+1;
b(i1,i2)=1/b(i1,i2);
c(i1,i2)=1/c(i1,i2);
end
end
end
a1=prod(a')
a3=a1.^(1/n1)
a4=sum(a3)
w1=a3./a4
b1=prod(b')
b3=b1.^(1/n1)
b4=sum(b3)
w2=b3./b4
c1=prod(c')
c3=c1.^(1/n1)
c4=sum(c3)
w3=c3./c4
w=[w1' w2' w3' ]
w= Average(w')
122
TOPSIS Code
%% TOPSIS
Clc
clear all
w=input('weight=')
a=input('matrix a=')
% w=[.1 .4 .3 .2]
% a=[7 9 9 8; 8 7 8 7;9 6 8 9;6 7 8 6]
a1=a.^2
a_sum=sum(a1)
root_sum=sqrt(a_sum)
for i1=1:length(a)
for i2=1:length(a)
a2(i1,i2)=a(i1,i2)/root_sum(i2)
end
end
end
for i1=1:length(a)
for i2=1:length(a)
a3(i1,i2)=a2(i1,i2)*w(i2)
end
end
end
a44=max(a3(:,[1:length(a)-1]))
a444=min(a3(:,length(a)))
a4=[a44 a444]
a55=min(a3(:,[1:length(a)-1]))
a555=max(a3(:,length(a)))
a5=[a55 a555]
for i1=1:length(a)
for i2=1:length(a)
s(i1,i2)=(a3(i1,i2)-a4(i2))^2
end
end
s1=sum(s)
s2=sqrt(s1)
123
........................................... : ....................................................
:
-
-
-
-
-
)( : .
)/( : .
) ( : ) - -
( .
) ( : ) - -
- - - (
)( : ) - -
- - - - - -
- - - (.
) . (
) (
) (
) (
/
/
/
/
- )(
) / (
- )(
) (
- )(
) (
- )(
)(
-)/(
) (
-) /(
) (
-) /(
) (
-) (
) (
-) (
)(
-) (
)(
giving as much background information about their strengths, weaknesses, history of use and future
potential as possible. The authors also pointed that; implementing multi -materials in product design leads
to higher product performance in terms of functionality, manufacturability, costs and aesthetics.
Multi-material selection is considered as one of the design strategies implemented to attain product
efficiency according to Wang [4]. Each product is different, and therefore several products may require
numerous functions that could not be satisfied by utilizing a single material. A design that incorporates
multi-material selection is a feasible alternative in order to achieve the functional requirements of a
product. Novita S., et. al. [5],Presents a MCDM for material selection during the conceptual design phase
and applied on an automotive body assembly. J.C. Albi ana, C. Vila [6] draw up a framework proposal
for integrated material and process selection in product design. Fuzzy multicriteria decision making
(MCDM) approach is proposed to select the best prototype product by Hao-Tien Liu [7]. Kuo- Chen
Hung et. al. [8] presents a fuzzy integrated approach to assess the performance of design concepts. A new
integrated design concept evaluation approach based on vague sets is presented by Xiuli Geng et. al.
[9].Hambali Ariff et. al. [10], presents the methodology of selecting design concepts using analytical
hierarchy process.
I.III. MATERIAL SELECTION
An ever increasing variety of materials is available today, with each having its own characteristics,
applications, advantages, and limitations. When selecting materials for engineering designs, a clear
understanding of the functional requirements for each individual component is required and various
important criteria or attributes need to be considered. Material selection attribute is defined as an attribute
that influences the selection of a material for a given application. These attributes include: physical
properties, electrical properties, magnetic properties, mechanical properties, chemical properties,
manufacturing properties (machinability, formability, weld ability, cast ability, heat treatability, etc.),
material cost, product shape, material impact on environment, performance characteristics, availability,
fashion, market trends, cultural aspects, aesthetics, recycling, target group, etc.
The selection of an optimal material for an engineering design from among two or more alternative
materials on the basis of two or more attributes is a multiple criteria decision making (MCDM) problem.
The selection decisions are complex, as material selection is more challenging today. There is a need for
simple, systematic, and logical methods or mathematical tools to guide decision makers in considering a
number of selection attributes and their interrelations. The objective of any material selection procedure is
to identify appropriate selection attributes, and obtain the most appropriate combination of attributes in
conjunction with the real requirement. Thus, efforts need to be extended to identify those attributes that
influence material selection for a given engineering design to eliminate unsuitable alternatives, and to
select the most appropriate alternative using simple and logical methods, [3].
Chiner [11] proposed five steps for material selection: definition of design, analysis of material
properties, screening of candidate materials, evaluation and decision for optimal solution, and verification
tests. Farag [12] on his handbook for material selection described the different stages of design and the
related activities of the material selection. Farag defined three stages of selection: namely initial
screening, developing and comparing alternatives, and selecting the optimum solution. Moreover, Van
Kesteren et al. [13] suggested basic materials selection activities as follow: formulating material criteria,
making a set of candidate materials, comparing candidate materials and choosing candidate material.
I.IV. MATERIAL SELECTION MODELS
The objectives of performance, cost and environmental sensitivity drive engineering design, and are
generally limited by materials. Selection of the materials that best meet the requirements of the design and
give maximum performance and minimum cost is the goal of optimum product design as Buggy
approach[14]. However, some conflicting situations are generally observed between these objectives and
criteria (i.e. young modulus/cost, or toughness/hardness) and there is a necessity to decide which property
is more important than others. Using simple and logical methods, the criteria that influence material
selection for a given engineering application must be identified to eliminate unsuitable alternatives and to
select the most appropriate one according to Chatterjee and Edwards [15,16].
In order to solve the material selection issue of engineering components and to increase the efficiency in
design process, many materials selection methods have been developed such as Ashby approach [17,18],
TOPSIS (Technique for Order Performance by Similarity to Ideal Solution) [19,20,21,22,23], VIKOR,
which means Multi-criteria Optimization and Compromise Solution) [24-27], ELECTRE stands for:
(ELimination Et Choix Traduisant la REalit ) which means (ELimination and Choice Expressing the
REality), [28-30], PROMETHEE (Preference Ranking Organization Method for Enrichment Evaluation)
[15], COPRAS (complex proportional assessment) [31,32] and COPRAS-G [33] , graph theory and
matrix approach [34], preference selection index (PSI) method [35] and linear assignment method [36].
A variety of quantitative selection methods have been developed to analyse the material selection process,
thus a systematic evaluation for these methods is performed by A. Jahan, et. Al. [37]. The study seeks to
address the following questions: (1): What is the contribution of the literature in the field of screening and
choosing the materials?, (2) What are the methodologies, systems, tools for material selection of
engineering components?. (3) Which approaches were prevalently applied?. (4) Is there any inadequacy
of the approaches?. Interested reader could find many methods and analysis in this study.
TOPSIS method takes attention of many researchers in the field of material selection. Sharma et al. [38]
proposed an expert system based on (TOPSIS) for aid in material selection process. Jee and Kang [22]
introduced hybrid of Entropy and TOPSIS as tool in computer aided engineering (CAE) to help design
engineers for material selection. As an example, the procedure of optimal material selection for a
flywheel has been developed in their work. Milani [24] applied Entropy and TOPSIS in gear material
selection and studied on effect of normalization norms on ranking of materials. Shanian and Savadogo
[19] showed application of TOPSIS as a MADM method for solving the material selection problem of
metallic bipolar plates for polymer electrolyte fuel cell. Since the Entropy method for deciding the
relative importance of attributes does not give scope to designers preferences, in
their study a revised Entropy method was used for calculation of relative importance of each
criterion. They also compared ordinary TOPSIS method to a modified version and showed efficiency of
proposed method. Huang et al. [39] used the possible solutions search algorithm (PSSA) to pre-select the
materials to obtain the feasible solutions, and applied TOPSIS method to acquire the optimal solution.
Rao and Davim [23] offered a decision-making framework model for material selections using a
combined multiple attribute decision-making method. The procedure was based on a combined TOPSIS
and AHP method. According to A. Jahan, et. al. [37], the most popular approach adopted in the literature
of material evaluation and selection are TOPSIS, ELECTRE and AHP have been the most popular state
of the art methods in material choosing. Chart method, Computer-aided materials selection and
knowledge-based systems are the most prevalent approach in material screening. Fuzzy methods
prevalently have been used either individually or with other methods such as Genetic Algorithm [40],
Neural Nnetworks [41], KBS [42], improved compromise ranking method [28], Graph theory [34] and
Fuzzy rating [43].
Critical analyze to the MCDM approaches and try is cited by A. Jahan, et. Al. [36]. The authors list many
advantages and drawbacks for screening material methods. Instead of analyzing every single approach in
material choosing methods, the main focus of the authors are due to TOPSIS, ELECTRE and AHP, which
are the three most popular selection approaches after 2005.
They listed the following drawbacks for these main models:
Although ELECTRE methods have good output, they have a number of limitations: As the
number of alternatives increases, the amount of calculations increases quite rapidly and
computational procedures are quite elaborate.
ELECTRE only determine the rank of each material and do not give numerical value for better
understanding of differences between alternatives.
AHP is a powerful and flexible decision making procedure to help one set priorities and make
the best decision when both tangible and non-tangible aspects of a decision need to be
considered, but it only can compare a very limited number of decision alternatives, which is
usually not more than 15. When there are hundreds or thousands of option to be compared, the
pair wise comparison manner provided by the traditional AHP is obviously infeasible.
TOPSIS is a good choice for material selection because of following reasons:
It is useful for qualitative and quantitative data.
It is relatively easy and fast, with a systematic process.
The output can be a preferential ranking of the candidate materials
With a numerical value that provides a better understanding of differences and similarities
among alternatives.
This is especially useful when dealing with a large number of alternatives and criteria; the
methods are completely suitable for linking with computer databases dealing with material
selection.
However, there are two major drawbacks for TOPSIS method. The first drawback is the operation of
normalized decision matrix in which the normalized scale for each criterion is usually derived a narrow
gap among the performed measures. That is, a narrow gap in the TOPSIS method is not good for ranking
and cannot reflect the true dominance of alternatives. Another drawback is that we never considered the
risk assessment for a decision maker in the TOPSIS method. According to risk propensity, it has been
commonly observed that decision makers differ in that willingness to overestimate the probability of a
gain or a loss, the risk attitudes for a decision maker is usually categorized as risk-seeking, risk-neutral,
and risk-averse. Without considering risk propensity, the subjective propensity associated with different
decision maker preference cannot be determined, Ruey-Chyn [44]
I.V. WORK OBJECTIVES
Although a good amount of research work has already been carried out by the past researchers on design
concept and materials selection applications using different MCDM methods, there is still a need to
employ a simple and systematic mathematical approach to guide the decision maker in taking an
appropriate product and material decisions for a specific engineering application. Although several
techniques have been combined or integrated with the classical TOPSIS, many other techniques have not
been investigated. These techniques make the classical TOPSIS more representative and workable in
handling practical and theoretical problems by providing necessary analysis for original data.
This paper aims to: 1) explodes the possibility to propose an integrated decision-making approach based
on fuzzy linguistic variables and geometric mean method integrated with TOPSIS framework which can
support product development and material selection process under uncertain environments. 2) Apple this
model after verification to a real life problem of design concept evaluation and other for material
selection process.
This paper is divided into six sections. The initial section is the introduction. Section two proposes Fuzzy
(GMM) TOPSIS integrated approach in detail. Section three introduces evaluation criteria. Section four
presents verification examples and section five represent real life applications for practical cases study,
where section six present conclusions about the results.
II. PROPOSED FUZZY (GMM) -TOPSIS INTEGRATED APPROACH
The proposed model integrated approach composed of Fuzzy (GMM) and TOPSIS methods consist of
three basic stages. The first stage data gathering to structure the hierarchy, stage two deals with Fuzzy
computation where the third stage is values determination of the final ranking using TOPSIS method.
II.I. THE FIRST STAGE: STRUCTURING THE HIERARCHY
This is the first stage; a problem is decomposed into a hierarchical structure that consists of an objective
(i.e., overall goal of the decision making), the general criteria which impact the goal directly, sub-criteria
(objectives), sub-subcriteria (measures) etc.
II.II. THE SECOND STAGE: COMPUTING THE WEIGHTS
In this stage, to determine the criteria weights, a team of experts formed pair wise comparison matrices
for evaluating the criteria. Each expert of the team established individual evaluation. Computing the
geometric mean of the values obtained from individual evaluations, a final pair wise comparison matrix
on which there is a consensus is found. The weights of the critical success factors are calculated based on
this final comparison matrixes according to Fuzzy linguistic variables shown in table1and equations.[45]
Table 1. The pair wise comparison of linguistic variables using fuzzy numbers
Intensity of fuzzy scale
~
1
~
3
~
5
~
7
~
9
~~~~
2, 4 ,6 ,8
Fuzzy number
(L,M,U)
(L,M,U)
(L,M,U)
(L,M,U)
(L,M,U)
(L,M,U)
User define
= ( ,1, )
= ( ,3, )
= ( ,5, )
= ( ,7, )
= ( ,9, )
= ( , , )
Where: L is the lower limit, M is the medium limit, U is the upper limit.
The corresponding membership function can be depicted as shown in Figure 1.
SI
MI
II
DI
EI
Scale
Next, from the information of the pair wise comparison, we can form the fuzzy positive reciprocal matrix
as follows:
a11 a1 j ain
an1 anj a nn
Where aij
a ji 1 and aij wi
wj
(1)
Then, the geometric mean method for finding the final fuzzy weights of each criterion can be formulated
as follows:
Wi=ri *(r1+ r2 + . rn)-1,
(2)
where
ri=(ai1*ai2**ain )1/n.
(3)
After determining the weights for lower, median and upper comparison matrices, the weighting average
is computed to get the final weights.[46]
II.III.
In the last stage, calculated weights of the factors are approved by decision making team. Ranking firms
are determined by using TOPSIS method in the This stage. In TOPSIS technique based on the concept
that rank alternatives, which has the shortest distance from the ideal (Best) solution and the longest
distance from the ideal (worst) solution.[47]
Steps of TOPSIS
Step 1: Construct normalized decision matrix. This step transforms various attribute dimensions into nondimensional attributes, which allows comparisons across criteria.
Normalize scores or data as follows:
a2ij) for i = 1, , m; j = 1, , n
rij =xij/ (
(4)
Step 2: Construct the weighted normalized decision matrix. Assume we have a set of weights for each
criteria wj for j = 1,n. Multiply each column of the normalized decision matrix by its associated weight.
An element of the new matrix is:
V ij = wj.rij= 1,2,3,.,
j= 1,2,.
(5)
(6)
(7)
Step 4: Calculate the separation measures for each alternative. The separation from the ideal alternative
is:
Si *= [
(Vij Vj )2 ] 1/2
i = 1, , m
(8)
(Vij Vj ) 2 ] 1/2
i = 1, , m
(9)
Ci* 1
(10)
Step 6: Select an alternative with maximum Ci* or alternative in the descending order based on the value
of Ci*.
Schematic diagram of the proposed model for best design is provided in Figure 2.
Set up goal
Make criteria
No
Buildcomparison matrix
End
Fig 2.Schematic diagram for the Fuzzy (GMM) -TOPSIS integrated approach
III.
EVALUATION CRITERIA
Evaluation criteria are the tool needed to measure the performance of each prioritization methods. These
criteria will compare and determine the best method among all the prioritization methods. In this study,
Euclidean Distance (ED) and Approximation method is applied. For each particular comparison matrix in
the hierarchy evaluation with aid of an error criteria ED will be performed. More precisely, the most
appropriate method for each matrix can be selected by performing the multi criteria analysis of derived
priority vectors across minimizing criteria ED.
III.I.
ED is used to estimate the overall distance between all the judgment elements in the comparison matrix
and associated ratios of the priorities from the derived vector weight. The best method is determined by
the least ED value. The ED is measured in the following way:
ED= (
III.II.
(11)
APPROXIMATION METHOD
The approximation method is used to calculate the minimum change of weights (changes= 0.001 in this
paper) that can change the ranking of alternatives by using different ranking methods.
I.
VERIFICATION
EXAMPLE 1
In this example, a pair wise comparison matrix has been conducted based on data that are taken from
Ying-Ming Wang et al. [48].This example compares between different prioritization methods in terms of
Euclidean distance (ED) as comparing criteria as shown in Figure 3. The results obtained from the YingMings study and the result using Fuzzy (GMM), are list in Table 2.
4
1
1/7
1/3
5
1
3
7
1
5
5
6
1
3
1/5
1
1
3
3
1/5
1/5
1
1
1/3
4
1
1/6
1/3
3
1
W6
0.1391
0.1253
0.1477
0.1305
0.0751
0.1392
0.1615
0.1875
0.2056
0.1487
13.13
9.47 11.73 9.24 9.28
9.28 9.37 9.31 9.36 9.05
EVM
WLSM
GLSM
GEM
FPM
CCMA
LP-GFW
WLSM
0.314
GLSM
.085
.771
GEM
.771
.771
.314
FPM
1
.314
.085
.771
CCMA
1
.314
.085
.771
1
DEAHP
.529
.755
.226
.226
.529
LP-GFW
1
.314
.085
.771
1
.529
FUZZY (TOPSIS)
.60
.828
.371
.942
.60
.30
.6
IV.II
Example 2:
In this example, a case has been conducted based on data that are taken from Bojans study [49]. The
selected case is reservoir storage allocation problem. The analyzed problem is allocating the surface water
reservoir storage to multiple uses. A global economical goal is defined as the most profitable use of
reservoir, and six purposes are considered as decision alternatives: electric power generation (A1);
irrigation (A2); flood protection (A3); water supply (A4); tourism and recreation (A5); and river traffic
(A6). Alternatives are evaluated across five economical criteria of different metrics: gain in national
income (C1); earning foreign exchange (C2); improvement of the balance of payment (C3); import
substitution (self-sufficiency) (C4); and gain in regional income (C5). P1 is the matrix where criteria are
compared by importance with respect to the goal, and matrices containing judgments of alternatives with
respect to criteria C1, C2,., C5 are referred to as P2,., P6, respectively as shown in Table 4. The
priority vectors for criteria is presented in table 5 and the Value of ED for all methods are presented in
table 6 and comulative ED presented in Figure 4.
Table 4: Compression matrices for reservoir storage allocation problem .
A1
A2
A3
A4
A5
A1
1
1/2
1/5
1/3
1/2
Criteria (P1)
A2 A3
A4
2
5
3
1
7
3
1/7
1
1/4
1/3
4
1
1/3
5
1/3
National Income(P2)
A2 A3 A4 A5
5
3
6
7
1 1/7 1/2
2
7
1
7
3
2 1/7
1 1/2
1/2 1/3
2
1
1/2 1/4
1 1/2
A1
A2
A3
A4
A5
A6
A1
1
1/5
1/3
1/6
1/7
1/5
A1
A2
A3
A4
A5
A6
Foreign Exchange(P3)
A1 A2 A3 A4 A5
1
4
6
7
2
1/4
1
2
2
1
1/6 1/2
1
2 1/6
1/7 1/2 1/2 1 1/5
1/2
1
6
5
1
1/2
3
1
7
1
A5
2
3
1/5
3
1
A6
5
2
4
1
2
1
A6
2
1/3
1
1/7
1
1
A1
A2
A3
A4
A5
A6
A6
1
1/3
1/7
1/6
1/3
1/4
4
1/2
1/3
2
2
1
3
1
1/5
1/2
1/3
2
7
5
1
4
7
3
6
2
1/4
1
2
1/2
3
3
1/7
1/2
1
1/2
A1
A2
A3
A4
A5
A6
A1
1
1/3
1/9
1/7
1/4
1/3
A6
3
1/3
1/5
1/6
1/2
1
A1
A2
A3
A4
A5
A6
A1
1
5
3
6
3
1
A6
1
4
3
7
5
1
Criteria
C1
C2
C3
C4
C5
EV
0.358
0.306
0.041
0.171
0.123
WLS
0.411
0.291
0.047
0.14
0.111
LLS
0.356
0.313
0.042
0.166
0.122
FPP
0.391
0.283
0.065
0.152
0.109
LGP
0.356
0.356
0.051
0.119
0.119
Entropy
0.112
0.307
0.1
0.297
0.184
Fuzzy (GMM)
0.343
0.317
0.043
0.168
0.126
P2
P3
4.583
4.961
5.508
4.813
5.440
4.550
8.573
4.504
6.209
6.255
6.937
6.119
8.027
8.227
13.298
5.907
5.305
5.359
5.204
5.289
6.904
5.607
12.991
5.350
Method
P4
P5
P6
6.797
7.382
7.114
7.327
7.318
7.005
10.146
6.930
6.107
6.451
7.054
6.627
6.634
7.643
13.705
6.270
4.786
5.055
5.331
4.642
8.740
8.162
11.11
4.590
TD
AN
EV
WLS
LLS
FPP
LGP
Entropy
Fuzzy (GMM) TOPSIS
Comulative ED
69.823
33.551
Fuzzy Entropy
(GMM)
TOPSIS
41.194
43.063
LGP
FPP
37.148
37.63
35.463
33.787
LLS
WLS
EV
AN
Different Methods
Cases Study
In the previous section verification of the model is presented. In this section two real life applications will
be cited to demonstrate the applicability, simplicity and accuracy of the integrated model. These two
applications have already been solved by the past researches and different ranking of the alternatives have
been obtained. The first case is cited by H. Ariff, et. al. [10]. This case is about the selection of the best
design concept for chair wheel transfer product. The second case is addressed by Ali Jahan et. al.[50]
which is the selection of the most suitable material for the design of a flywheel.
V.I
Problem1:
This proposed model is applied to a real problem in the industry. Inaccurate decision during the design
stage can cause the product to be redesign or remanufactured. A study has been conducted based on data
that are taken from case study used by H. Ariff, M. Sapuan, N. Ismail and Y. Nukman.[10]. This study is
about wheelchair transfer problems. There are seven wheelchair design concepts of wheelchairs.. The
main criteria affecting the development of wheelchair design are classified into five aspects; performance
(P), safety (S), cost (C), ergonomic (E) and maintenance (M). There are five sub-criteria affecting the
wheelchair performance: easy to transfer (ETT), easy to use (ETU), easy to storage (ETS), lightweight
(LW) and strong framework (SF). Stability (ST) and no sharp edge (NSE) are sub-criteria that affect in
terms of safety. While cost of material (CM) and cost of manufacturing process (CMP), easy to repair
(ETR) and easy to dismantle (ETD), are sub-criteria affecting in terms of cost and maintenance
respectively, show (Figure 3). This example is divided into three sections. Section one presented stages of
the Fuzzy (GMM) -TOPSIS model. Section two presented comparison between results obtained by AHP
and Fuzzy (GMM) -TOPSIS model by using Euclidean distance. Section three compares between three
different ranking methods SAW, TOPSIS and VIKOR in terms of methods sensitivity of changing the
weights. In this case, three scenarios are used equal weight, weight from AHP and GMM weight..
Section 1
The first stage is Structuring the hierarchy.
In this section, a hierarchy model for structuring design concept decisions is introduced. A four level
hierarchy decision process displayed in Figure 5 is described below:
Selection of the best wheelchair conceptual design
ETT
ETU
CM
NST
CMP
Criteria
ST
Goal level 1
level 2
ETR
SupCriteria
level 3
ETD
ETS
LW
SF
DC-1
DC-2
DC-3
DC-4
DC-5
DC-6
DC-7
Alternatives level
maintenance respectively. Finally, at the lowest level of the hierarchy, the design concept (DC)
alternatives of the wheelchair development are identified.
The pair-wise comparisons generate a matrix of relative rankings for each level of the hierarchy. The
number of matrices depends on the number elements at each level. The order of the matrix at each level
depends on the number of elements at the lower level that it links to.
Pair-wise comparison begins with comparing the relative importance of two selected items. There are n
(n 1) judgments required to develop the set of matrices in this step. The decision makers have to
compare or judge each element by using the relative scale pair wise comparison as shown in Table 1. The
judgments are decided based on the decision makers or users experience and knowledge. The scale used
for comparisons in Fuzzy (GMM) enables the decision maker to incorporate experience and knowledge
intuitively. To do pair wise comparison, for instance as shown in Table 7, if performance (P) is strongly
more important or essential over cost (C), then a = 5. Reciprocals are automatically assigned to each pairwise comparison.
Table 7:Construct a Pair-wise Comparison Matrix
Goal
Performance (P)
Safety (S)
Cost (C)
Ergonomic (E)
Maintenance (M)
P
[1,1,1]
[1/4,1/3,1/2]
[1/6,1/5,1/4]
[1/4,1/3,1/2]
[1/6,1/5,1/4]
S
[2,3,4]
[1,1,1]
[1/4,1/3,1/2]
[1,1,1]
[1/4,1/3,1/2]
C
[4,a,6]
[2,3,4]
[1,1,1]
[2,3,4]
[1/4,1/3,1/2]
E
[2,3,4]
[1,1,1]
[1/4,1/3,1/2]
[1,1,1]
[1/4,1/3,1/2]
M
[4,5,6]
[2,3,4]
[2,3,4]
[2,3,4]
[1,1,1]
Step 2 : Separation
In this step, three matrices (Lower, Medium, Upper) are separated from the original matrix. The priority
is calculated for each matrix separately.
Step 3 : Synthesizing the Pair wise Comparison
Normalized vector is computed according to eq.(3), where aij the element with I raw and j column. To
calculate the vectors of priorities, sum the each element in the normalizing column. Then, divide the
elements of each column by the sum of the column as shown in eq. (2). The result is priority vector as
shown in table 8. This process is done on all three matrices (Lower, Medium, and Upper).
Table 8: Synthesizing the Pair wise Comparison
Goal
(P)
(S)
(C)
(E)
(M)
sum
P
1
1/4
1/6
1/4
1/6
S
2
1
1/4
1
1/4
C
4
2
1
2
1/4
E
2
1
1/4
1
1/4
M
4
2
2
2
1
2.297
1
0.461
1
0.304
5.062
Priority
0.453
0.197
0.091
0.197
0.060
1
0.453
0.197
0.091
0.197
0.060
2
0.461
0.194
0.090
0.194
0.058
3
0.453
0.192
0.096
0.192
0.063
Average
0.456
0.195
0.092
0.195
0.061
These processes are repeated for all levels of hierarchy structure (criteria, sub-criteria and alternatives).
The priority vectors for criteria, sub-criteria and alternatives are represented in table 10. The overall
priority vector can be obtained by multiplying the priority vector for the design alternatives by the vector
of priority of the sub-criteria as shown in table 11.
Table 10: represent priority vectors for criteria, sub-criteria and alternatives.
0.45
ETT
0.456
P
0.26 0.06
ETU ETS
0.14
LW
0.17
0.10
0.14
0.12
0.32
0.08
0.05
0.21
0.08
0.14
0.04
0.20
0.16
0.15
0.17
0.29
0.14
0.12
0.05
0.16
0.05
criteria
Sub-criteria
Alternatives
DC-1
DC-2
DC-3
DC-4
DC-5
DC-6
DC-7
0.09
0.07
0.06
0.07
0.42
0.07
0.20
0.08
SF
GOAL
0.195
S
0.75 0.26
ST
NSE
0.0928
C
0.75 0.26
CM CMP
0.195
E
1
E
0.061
M
0.75 0.26
ETR ETD
0.11
0.06
0.11
0.06
0.31
0.08
0.25
0.14
0.05
0.09
0.06
0.26
0.14
0.23
0.12
0.22
0.12
0.22
0.05
0.16
0.07
0.12
0.06
0.12
0.05
0.32
0.10
0.22
0.26
0.19
0.14
0.09
0.04
0.09
0.06
0.18
0.05
0.17
0.05
0.17
0.17
0.17
0.12
0.22
0.12
0.22
0.04
0.18
0.07
0.19
0.31
0.12
0.10
0.03
0.16
0.06
Table 11: Overall weight vector for the alternatives with respect to the criteria
ETT
0.206
ETU
0.121
ETS
0.027
LW
0.064
SF
0.038
ST
0.146
NSE
0.051
CM
0.070
CMP
0.024
E
0.195
ETR
0.046
ETD
0.016
This step transforms various attribute dimensions into non-dimensional attributes, by using the eq. (4). To
calculate the normalizing decision matrix, squaring each element of the matrix of alternatives. Then, sum
the squaring elements in each column. After that, calculate the root for the sum in each column. Divide
the elements in alternatives matrix of each column by the root in each column and the resulted normalized
matrix stated in table12.
DC-1
DC-2
DC-3
DC-4
DC-5
DC-6
DC-7
ETT
0.40
0.23
0.32
0.28
0.74
0.18
0.11
ETU
0.52
0.19
0.35
0.10
0.49
0.39
0.37
ETS
0.18
0.14
0.13
0.14
0.84
0.14
0.41
LW
0.39
0.67
0.32
0.29
0.11
0.39
0.13
SF
0.24
0.13
0.26
0.13
0.69
0.18
0.56
ST
0.33
0.12
0.21
0.14
0.62
0.34
0.53
NSE
0.46
0.14
0.43
0.13
0.43
0.43
0.43
CM
0.30
0.55
0.30
0.55
0.12
0.40
0.17
CMP
0.29
0.54
0.29
0.54
0.11
0.43
0.16
E
0.26
0.14
0.26
0.11
0.72
0.22
0.50
ETR
0.67
0.49
0.37
0.24
0.11
0.25
0.17
ETD
0.44
0.71
0.28
0.22
0.07
0.37
0.14
Step 2: Construct the weighted normalized decision matrix by using eq. (5). In this step multiply each
column of the normalized decision matrix by its associated weight in table 11 as shown in table 13.
Table 13: The weighted normalized decision matrix
DC-1
DC-2
DC-3
DC-4
DC-5
DC-6
DC-7
ETT
0.08
0.04
0.06
0.05
0.15
0.03
0.02
ETU
0.06
0.02
0.04
0.01
0.05
0.04
0.04
ETS
0.00
0.00
0.00
0.00
0.02
0.00
0.01
LW
0.02
0.04
0.02
0.01
0.00
0.02
0.00
SF
0.00
0.00
0.01
0.00
0.02
0.00
0.02
ST
0.04
0.01
0.03
0.02
0.09
0.05
0.07
NSE
0.02
0.00
0.02
0.00
0.02
0.02
0.02
CM
0.02
0.03
0.02
0.03
0.00
0.02
0.01
CMP
0.00
0.01
0.00
0.01
0.00
0.01
0.00
E
0.05
0.02
0.05
0.02
0.14
0.04
0.09
ETR
0.03
0.02
0.01
0.01
0.00
0.01
0.00
ETD
0.00
0.01
0.00
0.00
0.00
0.00
0.00
0.15
0.02
0.06
0.01
0.02
0.00
0.04
0.00
0.02
0.00
0.09
0.01
0.02
0.00
0.00
0.03
0.00
0.01
0.14
0.02
0.03
0.00
0.01
0.00
SF
0.3
0.4
0.3
0.4
0.0
0.4
0.0
ST
1.8
5.3
3.7
5.0
0.0
1.7
0.2
NSE
0.0
0.3
0.0
0.3
0.0
0.0
0.0
CM
0.2
0.9
0.2
0.9
0.0
0.4
0.0
CMP
0.0
0.1
0.0
0.1
0.0
0.1
0.0
E
7.8
12.5
7.8
14.1
0.0
9.4
1.9
ETR
0.0
0.1
0.2
0.4
0.7
0.4
0.5
ETD
0.0
0.0
0.0
0.1
0.1
0.0
0.1
SF
0.1
0.0
0.1
0.0
2.1
0.0
1.3
ST
0.3
0.0
0.1
0.0
1.7
0.3
1.2
NSE
0.8
0.0
0.6
0.0
0.6
0.6
0.6
CM
0.4
0.0
0.4
0.0
1.3
0.1
1.0
CMP
0.4
0.0
0.4
0.0
1.3
0.1
1.0
E
0.2
0.0
0.2
0.0
2.6
0.1
1.0
ETR
2.2
1.0
0.4
0.1
0.0
0.1
0.0
ETD
0.9
2.8
0.3
0.2
0.0
0.6
0.0
Step 5: The relative closeness to the ideal solution is calculated by using the eq.(10). In this step each
element in row of separation from the negative ideal alternative divides by the sum of separation ideal and
negative ideal alternative. Then, the final ranking is presented in table17.
Table 17: Result of selection
Ranking
1
2
3
4
5
6
7
AHP
DC-5
DC-1
DC-7
DC-6
DC-3
DC-2
DC-4
P
0.456
0.456
S
0.191
0.195
C
0.099
0.092
E
0.191
0.195
M
0.061
0.061
0.1779
0.2675
0.2079
0.2675
0.0793
6.7903
8
6
4
2
0
3.38
3.32
AHP
AHP
Enrtopy
Enrtopy
ETU
0.114
ETS
0.026
LW
0.058
SF
0.069
ST
0.143
NSE
0.048
CM
0.074
CMP
0.025
E
0.191
ETR
0.047
ETD
0.016
Scenario c) Scenario c) Use the weights generated from the Fuzzy (GMM) as shown in table 11.The
results of scenario (a),(b),(c) are presented in table 20.
Table 20: Comparing between deferent ranking methods in deferent scenario
Ranking
1
2
3
4
5
6
7
Equal Weights
SAW TOPSIS
DC-5 DC-5
DC-1 DC-1
DC-2 DC-7
DC-6 DC-2
DC-7 DC-3
DC-3 DC-6
DC-4 DC-4
VIKOR
DC-1
DC-5
DC-6
DC-7
DC-3
DC-2
DC-4
Weights by AHP
SAW TOPSIS
DC-5 DC-5
DC-1 DC-7
DC-7 DC-1
DC-6 DC-3
DC-3 DC-6
DC-2 DC-2
DC-4 DC-4
VIKOR
DC-5
DC-1
DC-3
DC-7
DC-6
DC-2
DC-4
Table20, while ranking other alternatives, the SAW and VIKOR methods produce more similar ranks
across different scenarios than the TOPSIS method. Because the TOPSIS method has high sensitivity to
the changes in methods for assigning weights to criteria, it's frequently used as a benchmarking method.
For the wheelchair selection problem, all of the SAW, TOPSIS and VIKOR methods in each scenario
give the top rank to the same alternative (DC-5)except the VIKOR in scenario "a". Figure 7 shows there
are small variations in the rankings obtained using SAW, TOPSIS and VIKOR methods.
Alternatives
6
4
2
0
1
Fuzzy VIKOR
AHP VIKOR
E VIKOR
4
Rnking
Fuzzy TOPSIS
AHP TOPSIS
E TOPSIS
Fuzzy SAW
AHP SAW
E SAW
Table 21 represents Spearmans rank correlation coefficient between mentioned approaches. High rank
correlation between Fuzzy TOPSIS and Fuzzy SAW (0.678), Fuzzy TOPSIS and AHP VIKOR (.428) and
Fuzzy TOPSIS and (equal) TOPSIS (0.535) .
Table 21 Spearmans rank correlation coefficient between MCDM methods .
E SAW
E TOPSIS
E VIKOR
AHP SAW
AHP TOPSIS
AHP VIKOR
Fuzzy SAW
Fuzzy TOPSIS
Equal
SOPSIS
-.178
Equal
VIKOR
-.178
-.32
AHP
SAW
.2
.428
.392
AHP
TOPSIS
.142
-.75
-.25
-.60
AHP
VIKOR
.928
-.178
.107
.535
0
Fuzzy
SAW
.25
.428
.392
1
-.60
.535
Fuzzy
TOPSIS
.357
.535
-.035
.678
-.285
.428
.678
Fuzzy
VIKOR
.928
-.178
.107
.535
0
1
.535
.428
This study uses an approximation method to calculate the minimum change of weights (changes= 0.001
in this study) that can change the ranking of alternatives by using different ranking methods TOPSIS,
SAW, and VIKOR. The results for this study is presented in table 22.
Table 22: Compression between different ranking methods with minimum change of weights that can
change the ranking of alternatives.
TOPSIS
.Weights (W) .206
.121
W changes
.210
.121
Results before changing (W)
.027 .064
.031 .060
ranking
.814683
.438619
.427345
.321119
.269081
.217484
.171268
SAW
.Weights (W) .206
.121
W changes
.203
.121
Results before changing (W)
DC-5
DC-1
DC-7
DC-3
DC-6
DC-2
DC-4
.813289
.59916
.525928
.477828
.4732433
.41713
.339316
VIKOR
.Weights (W) .206
.121
W changes
.206
.111
Results before changing (W)
DC-5
DC-1
DC-7
DC-6
DC-3
DC-2
DC-4
0
.502833
.604242
.739321
.768263
.859454
.961268
DC-5
DC-1
DC-3
DC-7
DC-6
DC-2
DC-4
.027 .064
.030 .067
ranking
.027 .064
.037 .054
ranking
.038
.042
.038
.041
.038
.048
Changes= .003
.146 .051 .07
.024
.142 .051 .07
.024
Results after changing
(W)
.824263
.434748
.434599
.318824
.269222
.209023
.167655
Changes= .004
.146 .051 .07
.024
.149 .051 .07
.024
Results after changing
(W)
.820028
.599075
.527373
.473117
.472852
.410121
.33575
Changes= .01
.146 .051 .07 .024
.156 .051 .08 .034
Results after changing
(W)
0
.234395
.510857
.604738
.781943
.850231
.930921
.195 .046
.199 .050
ranking
.016
.012
DC-5
DC-7
DC-1
DC-3
DC-6
DC-2
DC-1
.195 .046
.198 .043
ranking
.016
.013
DC-5
DC-1
DC-7
DC-3
DC-6
DC-2
DC-4
.195 .046
.185 .036
ranking
.016
.006
DC-5
DC-7
DC-1
DC-3
DC-6
DC-2
DC-4
The last table explains how changes of weights contribute to the change of raking and also, shows
that the TOPSIS method is more sensitive than others which changes its alternative's ranking by
smallest change in weights of attributes.
V.II.
Problem2:
This example has been conducted based on data that are taken from Ali Jahan et. al.[50]. This sace deals
with the selection of the most suitable material for design of a flywheel which is a device to store kinetic
energy as used in automobiles, urban subway trains, mass transit buses, wind-power generators, etc. The
most important requirements in a flywheel design are to store the maximum amount of kinetic energy per
unit mass and to ensure against premature failure due to fatigue or brittle fracture. The following
characteristics are required for flywheel: (1) performance index of rlimit/q (where, rlimit is the fatigue
limit of the material and q is the material density). This signifies that the higher the value of rlimit/q, the
lower the weight of the material for a given fatigue strength and consequently, the kinetic energy per unit
mass of the flywheel will be higher. (2) Fracture toughness (KIC) of the material will be the performance
measure for failure due to brittle fracture. (3) The fragmentability of the flywheel material is an essential
property from the safety point of view. If the flywheel breaks into small pieces at final failure, the hazard
will be much reduced. (4) Price per unit mass. Among these four criteria, the beneficial attributes are
fatigue limit, fracture toughness and fragmentability where higher values are desirable, and price/mass is
a non-beneficial attribute where smaller value is always preferable. For more clarification in this example
proposed method illustrated step by step.
The problem which consists of ten alternative materials and four material selection criteria are shown in
Table 23. According to entropy method, [51] the weights of the considered criteria are as follow: wa =
0.4, wb = 0.3, wc = 0.2 and wd = 0.1.
Table23: Candidate materials for a flywheel
No.
1
2
3
4
5
6
7
8
9
10
Material
300M
2024T3
7050T73561
Ti6AL4V
E glass epoxy FRP
S glass epoxy FRP
Carbon epoxy FRP
Kevlar 29 epoxy FRP
Kevlar 29 epoxy FRP
Boron epoxy FRP
Fatigue (+)
100
49
78
108
70
165
440
242
616
500
Toughness (+)
8
13
12
26
10
25
22
28
34
23
Price (-)
4200
2100
2100
10500
2735
4095
35470
11000
25000
3150000
The result of ranking materials for different methods are shown in table 24. It is shown that all methods
ranke material number 9 (Kevlar 49-epoxy FRP ) is the first which has the highest value (.93104).
material number 7 (Carbon epoxy FRP) has the second value (.6884), thus we put 7 in rank 2 of
columnaas the same ranking of column "c" and "d" where column "b" rank material number 8 (Kevlar
29 epoxy FRP) in the second and column "e" rank material number 10 (Boron epoxy FRP).In the same
way the material 2 (2024T3) has the lowest value (.2835) in column a which agree with the ranking of
column "d" and "e" where column "b" rank material number 10 (Boron epoxy FRP) in the last one and
column "c" rank material number 1 (300M) also in the last one .
Table 24: Cumparing between defferent ranking method .
Ranking
1
2
3
4
5
6
7
8
9
10
TOPSIS
9
7
10
8
6
4
5
1
3
2
ELECTER
9
7
6
8
10
4
5
3
2
1
VIKOR
9
7
10
8
6
4
5
3
1
2
Linear assignment
9
10
7
8
6
4
1
3
5
2
For the flywheel material selection problem, all of the TOPSIS, Jee and Kang, ELECTER, VIKOR and
the linear assignment methods give the top rank to the same material (Kevlar 49epoxy FRP). Figure8
shows there are small variations in the rankings obtained using Jee and Kang, ELECTER, VIKOR and
linear assignment methods.
12
10
Ranking
8
6
4
2
0
1
10
Material
Linear assignment
VIKOR
Topsis
ELECTER
I.
ELECTER
-0.042
VIKOR
0.296
.793
L. assignment
0.321
.864
.696
Topsis
.224
.769
.951
.745
Conclusion
In this paper, Fuzzy (GMM) -TOPSIS compound decision making methods have been used in proposed
approach.This method proposed to deal with both qualitative and quantitative criteria and select the
suitable design.The following conclusions could be summarized as follows:
1-TOPSIS framework provides a perfect way to rank the candidate alternatives according to a decision
matrix, while fuzzy-GMM is effective in conducting preliminary analysis of uncertainty in decision
matrix.
2-GMM is used for computing weight which is an advanced step for TOPSIS to finding the final rank,
fast, precise and easy.
3-From the numerical illustration for design concept evaluation of the wheelchair problem the analysis
reveals that the design concept-5 is the most appropriate for further development because it has the
highest value among the other design concepts. Application of Fuzzy-GMM TOPSIS model for selecting
conceptual design at conceptual design stage can improve quality of product and shorten product
development process.
4-Sensitivity analysis for the model provide that it has the smallest TD among other model and its
sensitivity to change in alternative weighs is the best between all other (VIKOR, SAW) which mean more
accurate result of priority.
5-The two cited examples demonstrate the potentiality, applicability and simplicity solving design
concept and material selection decision-making problems and that the model is quite simple to implement
involving a large reduction of mathematics as compared to the other conventional material selection
methods.
6-The results derived using both this model show an excellent correlation with those obtained by the past
researchers which specifically prove the global applicability of this method while solving such type of
complex design or material selection problems.
References
[1] Chen, S. J., & Hwang, C. L., "Fuzzy multiple attribute decision making:Methods and applications", Lecture notes in economics
and mathematical systems. New York: Springer, 1992.
[2] Fodor, J., & Roubens, M.," Fuzzy preference modeling and multicriteria decision support", Dordrecht: Kluwer Academic
Publishers, 1994.
[3] Ashby MF, Brchet Y, Cebon D., "Selection strategies for materials and processes" Adv. Eng. Mater. , Vol.4(6), pp. 32734,
2004.
[4] Cui X, Wang S, Hu SJ., "A method for optimal design of automotive body assembly using multi-material construction", Mater
Des. , Vol. 29(2), pp. 381387, 2008.
[5] Novita Sakundarini, ZahariTaha , SalwaHanim Abdul-Rashid , Raja Ariffin Raja Ghazila, " Optimal multi-material selection for
lightweight design of automotive body assembly incorporating recyclability", Materials and Design Vol.50, pp. 846857, 2013.
[6] J.C. Albiana,
C. Vila, "A framework for concurrent material and process selection during conceptual product design stages",
Materials and Design,Vol.41,pp. 433446, 2012.
[7] Hao-Tien Liu, "Product design and selection using fuzzy QFD and fuzzy MCDM approaches", Applied Mathematical
Modelling, Vol. 35,pp. 482496, 2011.
[8] Kuo-Chen Hung, Peter Julian , Terence Chien , Warren Tsu-huei Jin,"A decision support system for engineering design basedon
an enhanced fuzzy MCDM approach", Expert Systems with Applications,Vol. 37, pp 202213, 2010.
[9] Xiuli Geng, Xuening Chu, Zaifang Zhang "A new integrated design concept evaluation approach based on vague sets, "Expert
Systems with Applications, Vol. 37, pp. 66296638, 2010.
[10] Hambali Ariff, Mohd. Sapuan Salit, Napsiah Ismail& Y. Nukman " Use of Analytical Hierarchy Process (AHP) for Selecting
The Best Design Concept", Journal Technology ,Vol.49 A Dis., pp. 118, 2008.
[11] Chiner M., "Planning of expert systems for materials selection", Mater Des. Vol. 9, pp.195203, 1988.
[12] Farag M.," Quantitative methods of materials selection", In: Kutz M, editor. Handbook of materials selection,2002.
[13] Van Kesteren IEH, Kandachar PV, Stappers PJ. "Activities in selecting materials by product designers", In: Proceedings of the
international conference on advanced design and manufacture, Harbin, China, 2006.
[14] Thakker A, Jarvis J, Buggy M, Sahed A., " A novel approach to materials selection strategy; case study: wave energy extraction
impulse turbine blade", Mater Des., Vol. 29, No.19, pp 7380, 2008.
[15] Chatterjee P, Chakraborty S., "Material selection using preferential ranking methods", Mater Des., Vol.35, No.3, pp. 8493,
2012.
[16] Edwards KL.," Materials influence on design: a decade of development", Mater Des., Vol.32, No.10, pp. 7380, 2011.
[17] Prashant Reddy G, Gupta N.," Material selection for microelectronic heat sinks: an application of the Ashby approach", Mater
Des., Vol.31, No.1, pp 137, 2010.
[18] Rashedi A, Sridhar I, Tseng KJ.," Multi-objective material selection for wind turbine blade and tower: Ashbys approach",
Mater Des , Vol.37, No.5, pp 2132, 2012.
[19] Shanian A, Savadogo O.," TOPSIS multiple-criteria decision support analysis for material selection of metallic bipolar plates
for polymer electrolyte fuel cell", J Power Sources; Vol.159, No.10, pp.95104, 2006.
[20] Gupta N. " Material selection for thin-film solar cells using multiple attribute decision making approach" Mater Des.; Vol. 32,
No.16, pp. 6771, 2011.
[21] Dagdeviren M, Yavuz S, Klnc N.," Weapon selection using the AHP and TOPSIS methods under fuzzy environment",
Expert Syst Appl; Vol.36, No.81, pp. 4351, 2009.
[22] Jee D-H, Kang K-J.," A method for optimal material selection aided with decision making theory",Mater,Vol.21, pp.199206,
Des 2000.
[23] Rao RV, Davim JP.," A decision-making framework model for material selection using a combined multiple attribute decisionmaking method", Int J AdvManuf Technol; Vol.35, No.7, pp.5160,2008.
[24] Milani AS, Shanian A, Madoliat R, NemesJA.,"The effect of normalization norms in multiple attribute decision making
methods: a case study in gear materialSelection", Struct Multidisc Optim, Vol.29,No.31, pp.28, 2005.
[25] Jahan A, Mustapha F, Ismail MY, Sapuan SM, Bahraminasab M. A.," Comprehensive VIKOR method for material selection",
Mater. Des.,Vol. 32 No.12, pp.1521, 2011.
[26] Jeya Girubha R, Vinodh S. ,"Application of fuzzy VIKOR and environmental impact analysis for material selection of an
automotive component" Mater Des., Vol.3, No. 74, pp. 7886, 2012.
[27] Ali Jahan , K.L. Edwards, " VIKOR method for material selection problems with interval numbers and target-based criteria"
Materials and Design,Vol. 47, pp.759765, 2013.
[28] Rao RV. ,"A decision making methodology for material selection using an improved compromise ranking method", Mater
Des., Vol.29, No.19, pp. 4954, 2008.
[29] Shanian A, Savadogo O.,"A material selection model based on the concept of multiple attribute decision making", Mater. Des.,
Vol.27, No.3, pp.2937, 2006.
[30] Shanian A, Savadogo O. "A non-compensatory compromised solution for material selection of bipolar plates for polymer
electrolyte membrane fuel cell (PEMFC) using ELECTRE IV" ElectrochimActa,Vol.51, No.53, pp.715,2006 .
[31] Chatterjee P, Athawale VM, Chakraborty S. "Selection of materials using compromise ranking and outranking methods" Mater
Des., Vol. 30, No. 40, pp.4353, 2009.
[32] Chatterjee P, Athawale VM, Chakraborty S." Materials selection using complex proportional assessment and evaluation of
mixed data methods", Mater Des. , Vol.32, No.8, pp.5160, 2011.
[33] Maity SR, Chatterjee P, Chakraborty S.," Cutting tool material selection using grey complex proportional assessment method",
Mater Des.,Vol.36, No.37, pp.28, 2012.
[34] Rao RV. "A material selection model using graph theory and matrix approach" Mater. Sci. Eng. A, Vol.431, No. 2, pp. 4855,
2006.
[35] Maniya K, Bhatt MG.,"A selection of material using a novel type decision making method: preference selection index method"
Mater.Des.,Vol.31, No.178, pp.59, 2010.
[36] Jahan A, Ismail MY, Mustapha F, Sapuan SM.," Material selection based on ordinal data" Mater Des., Vol.31, No.318, pp. 0
7, 2010.
[37] A. Jahan , M.Y. Ismail , S.M. Sapuan , F. Mustapha "Material screening and choosing methods A review", Materials and
Design, Vol. 31, pp. 696705, 2010.
[38] Sharma PK, Aggarwal A, Gupta R, Suryanarayan D. "Expert system for aid in material selection process", In:IEEE
international engineering management conference Delhi, India. IEEE, pp. 2731, 1993.
[39] Huang H, Liu G, Liu Z, Pan J. "Multi-objective decision-making of materials selection in green design" Jixie Gongcheng
Xuebao/Chin J Mech Eng, Vol. 42:pp. 131136, 2006.
[40] Jarupan L, Kamarthi SV, Gupta SM. ,"Application of combinatorial approach in packaging material selection In:Gupta SM,
editor. Proceedings of SPIE", the international society for optical engineering. Philadelphia, PA, pp. 207 223, 2004.[41] Shi P.
"Neural network approach to material selection for injection moldedParts" Harbin Gongye Daxue Xuebao/J Harbin InstTechnol:37,
2005 .
[42] Zha XF., "A web-based advisory system for process and material selection in concurrent product design for a manufacturing
environment", Int J AdvManuf Technol, Vol. 25, pp. 233243, 2005.
[43] Thurston DL, Carnahan JV. , "Fuzzy ratings and utility analysis in preliminary
design evaluation of multiple attributes" J Mech Des Trans ASME, Vol. 114: PP. 648658, 1992.
[44] Ruey-Chyn, Tsaur "Decision risk analysis for an interval TOPSIS method" Applied Mathematics and Computation, Vol. 218,
pp. 42954304, 2011.
[45] Buckley, J. J., "Fuzzy hierarchical analysis. Fuzzy Sets and Systems " New York: McGraw-Hill, Vol. 17, No (3), ISSN: 233247, 1985.
[46] Zadeh, L. A.," Fuzzy sets. Information and Control" New York: McGraw-Hill, Vol. 8, No (2), ISSN: 338-353, 1965.
[47] Mansour Momeni, Mohammed Reza Fathi, Mohammed Karimi Zarchi and Sirous Azizollohi , " Fuzzy TOPSIS based
approach to maintenance strategy selection: case study" Middle East Journal of Operational Research, vol. 3, pp 699-706, 2011.
[48] Ying-Ming, W., Celik P. and Ying L., A Linear Programming Method for Generating the Most Favorable Weights from a
Pair-Wise Comparison Matrix, Computers and Operations Research", Vol.35, No.12, pp.3918-3930, 2008.
[49] Bojan Srdjevic, Combining Different Prioritization Methods in the Analytic Hierarchy Process Synthesis Al Sevier,
Computers and Operations Research, Vol.32, No.7, pp.1897-1919, 2005.
[50] Ali Jahan, Md Yusof Ismail, Faizal Mustapha, Salit Mohd Sapuan," Material selection based on ordinal data",
31,pp.31803187,2010.
Vol.
Abstract
In todays world of global competition, providing quality service is a key for success, and many experts
concur that the most powerful competitive trend currently shaping marketing and business strategy is
service quality. Institutes of higher education are also focusing on ways to render high quality education
to their educators and have a better performance. Higher education institutes are facing new challenges in
order to improve the quality of education. There is a pressure for restructuring and reforming higher
education in order to provide quality education and bring up graduates who become fruitful members of
their societies. Therefore, these institutes are trying to recognize the dimensions of a quality education
and define strategies to reach their pre-defined standards and goals.
The first objective of this article is proposed a balanced scorecard as a performance evaluation model for
engineering educational systems. The second objective prioritizes performance indicators within
engineering education balanced scorecard using GMM- TOPSIS. This study will collect and arrange
suitable performance evaluation configurations and indices by literature reviews and interviews to
department heads in engineering educational systems.
level of acceptable service as they realize that service will not always reach the desired levels; this is the
adequate service level. Between these two service levels is a zone of tolerance that customers are willing
to accept. Finally, customers have a predicted level of service, which is the level of service they believe
the company will perform.
Karathanos and Karathanos [14] performed a study aimed at showing the performance indicators of the
firstthree winners of the Malcolm Baldrige National Quality Award [7]. The study concentrated on the
need for alignment of performance measures with vision, mission and strategic goals. Chen, Yang and
Shiau (2006) have used the BSC to create a system for evaluating the performance of the Chin Nmin
Institute of Technology in Taiwan.
In another study conducted by Umashankar and Dutta [19], BSC was used to measure the efficiency of
the management at Indian universities. The study found that the BSC could enable these universities to
identify and correct significant deviations and design appropriate strategies. Nayeri, Mashhadi and
Mohajeri [20] developed the BSC model in order to assessthe strategic environment of higher education
in the field of business in Iran. Raghunadhan [21] assessed the institutes of higher education which is
funded by the government of India, and used the BCS to compare institutes surveyed. The results
indicated that the concepts of strategic management are applied in these institutes. Beard [22] argued that
the BSC is suitable for use in higher education, and he has shown many successful applications of the
BSC in this area. Also, Umayal and Suganthi [23] presented a model for measuring performance of an
educational institution based on BSC approach. Measurement criteria were also suggested to assess the
performance according to the four perspectives of the BSC. In addition, Yu, Hamid, Ijab and Soo [24]
discussed the appropriateness of adopting electronic BSC to measure the quality of performance for
academic staff in higher education. The research showed that the electronic BSC is appropriate and
effective for this purpose.A lot of researchers like Munteanu et al [25], Adcroft et al.[26], Mourad et
al.[27], Mazzarol [28], Durkin et al.[29] emphasized on the requirement of higher education
organizations for collecting data based on students expectations. Dhara Jha and Vijay [30] used BSC as a
tool to manage the education academics. Seth. A. and Oyugi L. A. [31] studied the relationship between
the balanced scorecard and organizational performance of higher learning in Kenya as a means of
improving organizational performance. Josua Tarigan and very recently Deborah Christine [32] studied
the relationship between non-financial performance and financial performance using balanced scorecard
in higher education in Indonesia universities .Mohammad H Yarmohammadian et. al. [33] developed An
Integrated strategic quality model and BSC applied on Iranian higher education system. Teresa et. al.
studied the validation of a Balanced Scorecard (BSC) model and a Strategic Map for the University by
studying the relationships of efficiency between its dimensions. This work determines factors of the
performance in this type of institution. These factors are: the participation of teaching staff in innovation
activities; the number of doctorate-level staff; the academic subjects and credits in the Virtual Campus;
and the scores in the surveys of student satisfaction. Amene and Farhad [34] discussed performance
evaluation in higher education institutes with the use of combinative model AHP and BSC.
Vision and
mission
Increase tuition
Financial
(FL)
Customer
(CT)
Internal
process (IP)
Learning and
growth (L&G)
Management
commitment
Increase
external grants
fees
Increase Budget
Increase contracts
with industry
Increase Students
'satisfaction
Improve
university image
Increase
academic stuff
Increase customer
satisfaction
Increase
teaching aids
Increase papers
research
Culture Development
Leadership
Increase
Empowerment
(MC)
Increase
environmental
research
Increase of practical
projects
Motivation
Organizational
structure.
Objectives
Indicator
A5
A6
B1
B2
A1
A2
A3
Financial
(FL)
Customer
Development
the finance
Capacity
Customer
satisfaction
Customer
A4
(CT)
retention
Customer
acquisition
Reduce
customer
complaints
Quality
educational
service.
Internal
process (IP)
Contribute to
the
development
of integrated
community.
Raise the
efficiency of
faculty.
B3
B4
C1
C2
C3
C4
C5
C6
C7
C8
C9
C10
C11
C12
C13
C14
C15
D1
D2
D3
D4
Learning and
growth
(LRG)
Work on
development
educational
process.
D5
D6
D7
Preparation
courses for
continuing
Education and
training.
D8
E5
E6
E2
E3
Complete the
organizational
structure.
E4
E1
Management
commitment
(MC)
`
Faculty accreditation
D9
D10
Leadership
Development
motivation
mechanics.
Development
culture
Development
empowerment
mechanics
E7
E8
E9
E10
~
1
~
3
~
5
~
7
~
9
~~~~
2, 4, 6, 8
Fuzzy number
(L,M,U)
User define
= ( ,1, )
(L,M,U)
= ( ,3, )
(L,M,U)
= ( ,5, )
(L,M,U)
= ( ,7, )
(L,M,U)
= ( ,9, )
Intermediate values
(L,M,U)
= ( , , )
Where: L is the lower limit, M is the medium limit, U is the upper limit.
Next, from the information of the pair wise comparison, we can form the fuzzy positive reciprocal matrix
as follows:
a11 a1 j ain
an1 a nj ann
Here aij
(1)
w
a ji 1 and aij i .
wj
Then, the geometric mean method for finding the final fuzzy weights of each criterion can be formulated
as follows:
w i ri
(r1 r2 rn )1 ,
(2)
ri (ai1
ai 2 ain )1/ n .
(3)
Where
After determining the weights for lower, median and upper comparison matrices, the weighting average
is computed to get the final weights. [35]
Step 1: Construct normalized decision matrix. This step transforms various attribute dimensions into nondimensional attributes, which allows comparisons across criteria.
Normalize scores or data as follows:
rij = xij/ (a2 ij) for i = 1, , m; j = 1, , n
(4)
Step 2: Construct the weighted normalized decision matrix. Assume we have a set of weights for each
criteria wj for j = 1,n.
Multiply each column of the normalized decision matrix by its associated weight. An element of the new
matrix is:
vij = wj rij
(5)
(6)
(vij vj*)2 ]
i = 1, , m
(8)
i = 1, , m
(9)
Ci* 1
(10)
F
[1,1,1]
[2,3,4]
[4,5,6]
[4,5,6]
[8,9,9]
C
[1/4,1/3,1/2]
[1,1,1]
[1,1,1]
[1,2,3]
[3,4,5]
IP
[1/6,1/5,1/4]
[1,1,1]
[1,1,1]
[6,7,8]
[2,3,4]
L& G
[1/6,1/5,1/4]
[1/3,1/2,1]
[1/8.1/7,1/6]
[1,1,1]
[1,2,3]
MC
[1/9,1/9,1/8]
[1/5,1/4,1/3]
[1/4,1/3,1/2]
[1/3,1/2,1]
[1,1,1]
Step 2: Separations
Three matrices (Lower, Medium, and Upper) are separated from the original matrix. The priority is
calculated for each matrix separately.
F
1
2
4
4
8
C
1/4
1
1
1
3
IP
1/6
1
1
6
2
sum
L& G
1/6
1/3
1/8
1
1
MC
1/9
1/5
1/4
1/3
1
0.2385
0.6683
0.6597
1.5156
2.1689
5.2512
Priority
0.0454
0.1272
0.12563
0.2886
0.41303
1
0.045
0.127
0.126
0.289
0.419
2
0.039
0.120
0.110
0.298
.4302
3
0.038
0.124
0.102
0.318
0.414
Average weights
.0414
0.124
0.112
0.302
0.4194
This step transforms various attribute dimensions into non-dimensional attributes, by using eq. (4). To
calculate the normalizing decision matrix, square each element of the matrix of alternatives. Then, sum of
the squares of each element square in each column. After that, calculate the root for the sum in each
column. Divide the elements in alternatives matrix of each column by the root in each column and the
resulted normalized matrix stated in table 6.
Table 6: Normalized decision matrix.
Goal
F
C
IP
L&G
MC
F
0.100
0.199
0.398
0.398
0.796
C
0.072
0.288
0.288
0.288
0.864
IP
0.026
0.154
0.154
0.926
0.309
L& G
0.114
0.227
0.085
0.681
0.681
MC
0.100
0.181
0.226
0.301
0.903
Step 7: Construct the weighted normalized decision matrix by using eq. (5). In this step multiply each
column of the normalized decision matrix by its associated average weight in table 5 as shown in table 7.
Table 7: The weighted normalized decision matrix
Goal
F
C
IP
L&G
MC
F
0.004
0.008
0.016
0.016
0.033
C
0.009
0.036
0.036
0.036
0.107
IP
0.003
0.017
0.017
0.104
0.035
L& G
0.034
0.069
0.026
0.206
0.206
MC
0.042
0.076
0.095
0.126
0.378
0.033
0.004
0.107
0.009
0.104
0.003
0.206
0.026
0.378
0.042
F
0.001
0.001
0.000
0.000
0.000
C
0.010
0.005
0.005
0.005
0.000
IP
0.010
0.007
0.007
0.000
0.005
L& G
0.029
0.019
0.032
0.000
0.000
MC
0.113
0.092
0.081
0.064
0.000
F
0.000
0.000
0.000
0.000
0.001
C
0.000
0.001
0.001
0.001
0.010
IP
0.000
0.000
0.000
0.010
0.001
L& G
0.000
0.002
0.000
0.032
0.032
MC
0.000
0.001
0.003
0.007
0.113
Step 10: The relative closeness to the ideal solution is calculated by using the eq.(10). In this step each
element in row of separation from the negative ideal alternative divides by the sum of separation ideal and
negative ideal alternative. Then, the final ranking is presented in table11.
Table 11: Result of ranking
Perspective
Rank
F
5
C
3
IP
4
L& G
2
MC
1
This proces is repeated for all stratgic objectives and all indecators. The result of the rank for all levels
(perspectives, objectives, and performance indicators) is shown in figure 3.
5-Results
The results of this research show that "Management" is the most important perspective of educational
balanced scorecard in faculty of engineering in the first level. It is noted from table 11 that the most
important perspective after management is learning and growth and the least important is finance. When
going to the second level the result show that " Motivation development"," education development",
"customer satisfaction", "Quality education service" and "Financial development" are considered as the
most important objectives of educational balanced scorecard in faculty of engineering. According to the
result of third level the "Budget allocated annually", "Number of laps found to factor required for each
department."," Number of projects that involved the overall development of the environment", " Number
of international conferences, which was attended by members of the faculty", " Number of international
agreements for the exchange of graduates", " Number of practical projects, which was attended by
members of the faculty in the development of society and the development environment", and " number
of person motivated bar moth" are considered as most important indicators of educational balanced
scorecard in faculty of engineering.
B4 (.085)
Complains Reduction
(CR) (.085)/(4)
B3 (.129)
Customer Acquisition
(CA) (.129)/(2)
B2 (.224)
B1 (.559)
Customer Satisfaction
(CS) (.559)/(1)
Customer (CT)
(.124)/ (3)
D8 (.119)/(2)
D9 (.559)/(1)
D10 (.264)/(4)
C10 (.044)/(10)
C11 (.069)/(5)
C12 (.105)/(3)
D6 (.257)/(2)
D5 (.743)/(1)
D4 (.16)/(4)
D3 (.45)/(1)
D2 (.27)/(2)
D1 (.1)/(3)
Faculty Efficiency
(FE) (.173)/(3)
D7 (.0563)/(3)
Educational Development
(ED) (.301)/(1)
C15 (.319)/(2)
C14 (.567)/(1)
C13 (.113)/(3)
Community Development
(CD) (.25)/(2)
C9 (.039)/(11)
C8 (.085)/(4)
C7 (.029)/(12)
C6 (.136)/(1)
C5 (.095)/(9)
C4 (.074)/(7)
C3 (.145)/(2)
C2 (.095)/(6)
C1 (.083)/(8)
Education Quality
(EQ) (.75)/(1)
Internal process
(IP) (.112)/ (4)
Fig.3: The weights / ranking of the perspectives, objectives, and performance indicators
A6 (.064)/(5)
A5 (.095)/(6)
A4 (.068)/(4)
A3 (.284)/(2)
A2 (.086)/(3)
A1 (.401)/ (1)
Financial Development
(DF) (.41)/(1)
Financial (FL)
(.041)/ (5)
Mission
Vision
E8 (.249)/(2)
E7 (.741)/(1)
Motivation Developing
(MD)(.321)/(1)
E6 (.5)
E5 (.5)
Organization Structure
(OS) (.129)/(5)
E9 (.193)
Culture Developing
(CD)(.152)/(4)
E4 (.25)
E3 (.25)
E2 (.25)
E1 (.25)
E10 (.129)
Empowerment Developing
(ED) (.193)/(3)
Leadership (L)
(.204)/(2)
Commitments /Management
(MC) (.419) /(1)
6- Conclusion
The effectiveness of the higher education sector can be defined generally by, the degree to which the
goals and objectives specified in higher education policies, plans, projects and programs are achieved to
the satisfaction of the stakeholders. The ultimate objective of improving higher education effectiveness is
the overall improvement in specifically the nations human capital and generally, in national development
while making the most efficient use of resources. This study used the BSC as a strategic tool to evaluation
the faculty of engineering performance. Fuzzy (GMM) TOPSIS model is used to prioritize levels of all
BSC perspectives. The following conclusions could be drawing:
1- The weights calculated by Fuzzy (GMM) TOPSIS prioritize the importance of the BSC
evaluation criteria for faculty of engineering performance with respect to the relative weights of
the criteria, it not only revels the ranking order of the faculty performance but it also pinpoints
the gaps to better achieve faculty goal by using the MCDM analytical methods.
2- The proposed which integrate the BSC with MCDM method shows to be a feasible and effective
assessment model for faculty of engineering performance evaluation and it could be extended to
other faculties as well or digging deeply to assesses faculty department also.
3- The result from Fuzzy (GMM)-TOPSIS model found out that management perspective has the
first priority which means that it is the most important component of the five balanced scorecard
perspectives the faculty performance.
4- The most significant advantage of the use of the balanced scorecard is that it provide a wider
development of metrics that are closely connected to the strategic goals of institution (here
faculty of engineering).
5- Organizing an appropriate set of metrics through an academic scorecard provides a useful way to
conceptualize and display the overall education and financial performance of certain units with
the organization.
References
[1] Harvey L. and Green D., Defining quality, Assess Eval High Educ, Vol. 18 No. 1, pp. 934, 1993.
[2] Clewes, A student-centred conceptual model of service quality in higher education, Qual High Educ, Vol. 9, No. 1, pp. 6985,
2003.
[3] O'Neill,M.A.,Pamer, A., ImportancePerformance Analysis: A Useful Tool for Directing Continuous Quality Improvement in
Higher Education, Qual. Assur. Educ., Vol. 12, No. 1, pp. 39-52, 2004.
[4] Guolla, M., Assessing the Teaching Quality to Student Satisfaction Relationship: Applied Customer Satisfaction Research in
the Classroom, J. Mark. Theory Pract., Vol. 7, No. 3, pp. 87-97, 1999.
[5] Zeithaml, V. A, Berry, L. L., Parasuraman, A., The Nature and Determinants of Customer Expectations of Services, J. Acad.
Mark. Sci., Vol. 21, No. 1, pp. 1-12, 1993.
[6] Kettunen, J.," Strategies for the cooperation of educational institutions and companies in mechanical engineering", International
Journal of Educational Management, Vol. 20, No.1, pp.19-28,2006.
[7] Chen, S. H., Yang, C. C., & Shiau, J. Y.," The application of balanced scorecard in the performance evaluation of higher
education", The TQM Magazine, Vol.18, No. 2, 2006.
[8] Kaplan Robert S. & Norton David P., The Strategy Focused Organization: How Balanced Scorecard Companies Thrive in the
New Business Environment, Boston: Harvard Business school press 2001.
[9] Kaplan Robert S. & Norton David P.,"Strategy Maps", Harvard Business School Publishing Corp., Boston, MA. 2004.
[10] Wilson, C., Hagarty, D., Gauthier, J., Result Using the Balanced Scorecard in the Public Sector, Journal of Corporate Real
Estate, Vol. 6, No. 1, pp. 53-63, 2003.
[11] Ricardo Correa Gomes, and Joyce Liddle,"The Balanced Scorecard as a Performance Management Tool for Third Sector
Organizations: the Case of the Arthur Bernardes Foundation, Brazil", Vol. 6, No. 4, ISSN354-366 - 2009.
[12] Nopadol R.,"Why the Balanced Scorecard Fails in SMEs: A case Study", International Journal of Business and Management,
Vol. 6, No.11, 2011.
[13] Panagiotis C., Michael G., George V.," A Proactive Balanced Scorecard ", International Journal of Information Management,
Vol. 31, pp. 460-468,2011.
[14] Karathanos, D., & Karathanos, P.," Applying the balanced scorecard to education",Journal of Education for Business, Vol. 80,
No. 4, ISSN222-230, 2005.
[15]Chang, O.H. and Chow, C.W., The balanced scorecard: a potential tool for supporting change and continuous improvement in
accounting education, Issues in Accounting Education, Vol. 14 No. 3, pp. 395-412, 1999.
[16] Amaratunga, D., and Baldry, D.," Assessment of facilities management performance in higher education properties Facilities",
Vol.18, No. 7/8, pp. 293-301, 2000.
[17] Delker, S. G. B.," Balanced scorecard: An instrument of change for facilities services", A Project Presented to the Faculty of
California State University, San Bernardino, 2003.
[18] Cullen, J., Joyce, J., Hassall, T., & Broadbent, M.," Quality in higher education: From monitoring to management", Quality
Assurance in Education, Vol.11 No.1, pp. 5-14, 2003.
[19]Umashankar, V., & Dutta, K.," Balanced scorecards in managing higher education institutions", An Indian perspective.
International Journal of Educational Management, Vol.21, No.1,pp.54-67, 2007.
[20] Nayeri, M. D., Mashhadi, M. M., & Mohajeri, K. ," Universities strategic evaluation using balanced scorecard", International
Journal of Social Sciences, Vol.2, No.4,pp.231-236, 2007.
[21] Raghunadhan, T. " Strategy: A pedagogy for efficient, accountable and socially responsive higher education", Global Business
and Management Research: An International Journal, Vol.1, No.1, pp.36-49, 2009.
[22] Beard, D. F.,"Successful applications of the balanced scorecard in higher education",Journal of Education for Business, Vol.84,
No.5,pp.275-282, 2009.
[23] Umayal, K. P. L., & Suganthi, L.," A strategic framework for managing higher educational institutions", Advances in
Management, Vol.3, No.10, pp.15-21,2010.
[24] Yu, M. L., Hamid, S., Ijab, M. T., & Pei, S. H. ," The e-balanced scorecard (e-BSC) for measuring academic staff performance
excellence", The International Journal of Higher Education and Educational Planning, Vol.57, No.6, pp. 813-828, 2009.
[25] Munteanu, C., Ceobanu, C., Boblca, C., & Anton, O. "An analysis of customer satisfaction in a higher education context",
International Journal of Public Sector Management, Vol. 23, No.2, pp. 124 140, 2010.
[26] Adcroft, A., Teckman, J., & Willis, R., "Is higher education in the UK becoming more competitive?", International Journal of
Public Sector Management, Vol. 23, No.6, pp.578 588, 2010.
[27] Mourad, M., Ennew, C., & Kortam, W., "Brand equity in higher education", Marketing Intelligence & Planning, Vol.29, No. 4,
pp.403 420, 2012.
[28] Mazzarol, T., & Soutar, G.,"Revisiting the Global Market for Higher Education", Asia Pacific Journal of Marketing and
Logistics, Vol. 24, No.5, pp.153 161, 2012.
[29] Durkin, M., McKenna, S., &Cummins, D.,"Emotional connections in higher education marketing", International Journal of
Educational Management, Vol.26 No.2, pp. 153 161, 2012.
[30] Dhara Jha, and Vijay Kumar Gupta,"Balanced Scorecard: AStrategic Management and Control Tool: Does it Only Deal with
Strategy?, A Perspective from Management Students", International Journal of Marketing, Financial Services & Management
Research, Vol.2, No. 1,pp. 2277- 3622, 2013.
[31] Seth A. and Oyugi L. A,"Influence of Balanced scorecard on Organizational Performance in Institutionsofhigher Learningin
Kenya., A Case Studyof University of Nairobi", International Journal of Education and Research,Vol.1, No.8,2013.
[32] Josua Tarigan and Deborah Christine Widjaj," The Relationship between Non-Financial Performance and Financial
Performance Using Balanced Scorecard Framework: A Research in Education Context", Journal of Economics, Business and
Management, Vol. 2, No.1, 2014.
[33] Mohammad H Yarmohammadian, Maryam Fooladvand, Somayeh Shahtalebi, Ahmad Ali Foroughi Abari, and Badri
Shahtalebi," An Integrated Strategic Quality Model (ISQM) for Non-governmental Universities: Integration between Strategic
Planning and BSC applied Iranian higher education system", Interdisciplinary Journal of Contemporary Researchin Businesscopy
right, Vol.4, No.12, 2013.
[34] Teresa Garca Valderrama, Vanesa Rodrguez Cornejo, Daniel Revuelta Bordoy," Balanced Scorecard and Efficiency: Design
and Empirical Validation of a Strategic Map in the University by Means of DEA", American Journal of Operations Research, Vol.3,
ISSN 30-52, 2013.
[35] Amene Kiarazm, and Farhad Koohkan,"Performance Evaluation in Higher Education Institutes with the Use of Combinative
Model AHP and BSC", Journal of Basic and Applied Scientific Research, Vol.3, No.4, 2013.
[36] Dr. B.C Roy,"Analytic Hierarchy Process & TOPSIS Method to Evaluate Faculty Performance in Engineering Education",
Dipendra Nath Ghosh et al Uniascit, Vol. 1 No.2, pp. 63-70, 2011.
[37] Ramlal Porika and Prashanthpataley,"Performance Appraisal of Faculty in Various Engineering Colleges Using A Conceptual
Model of Performance Appraisal System",Vol.1, No.1, pp.11-23, 2012.
[38]Buckley, J. J., "Fuzzy hierarchical analysis. Fuzzy Sets and Systems " New York: McGraw-Hill, Vol. 17, No (3), ISSN: 233247, 1985.
[39] C. L. Hwang, K. Yoon," Multiple attribute decision making. Springer- verlag, Berlin" New York: McGraw-Hill, 1981.
. .
:
.
.
.
.
. .
.
.
-:
, .
" " .
.
.
.
Fuzzy (GMM)- TOPSIS .
.
.
. .
. " "
Fuzzy (GMM)- TOPSIS .
.
)( .
. - TOPSIS
) Fuzzy(GMM .
.
Fuzzy (GMM) - TOPSIS
.
) (.