Upload
vonhu
View
224
Download
1
Embed Size (px)
Citation preview
APPLICATIONS OF VECTOR OPTIMIZATION
PROBLEMS
BY
VANDANA BAGLA
UNDER THE SUPERVISION OF
DR. ANJANA GUPTA
Delhi Technological University (D.T.U.), DELHI
Submitted
in fulfillment of the requirement of Research Summary
to the
TEERTHANKER MAHAVEER UNIVERSITY
MORADABAD
SUMMARY
OF
RESEARCH WORK
The research work titled “Applications of Vector Optimization Problems”
focuses on various perspectives of formation and solutions of Multi Criteria
Decision Making (MCDM) problems. Over the last few decades, various new
approaches have been developed and the methodologies of decision making
processes have been improving gradually. Decision-making approaches are
broadly employed for the selection of the best compromise solution, as any
alternative cannot usually excel in all the criteria under consideration simul-
taneously. Besides, the real criteria values by which a decision is made, the
selection of the best solution also depends on the decision maker’s individual
preferences. Traditional decision making processes and models are falling be-
hind in the fast pace of expedition. Although mathematical advances have
been made in this area, these research areas are still contemporary and inno-
vative. In this study, integrated solution methodologies for general MCDM
problems are developed based on various MCDM approaches. Model-based
MCDM is investigated, with the focus being on improving and applying ap-
proaches based on interactive multi objective optimization methods. In ad-
dition to the methodological aspects, there has been a focus on applying the
interactive multi-objective optimization to the application areas that contain
complex processes and conflicting targets, which have gathered increasing in-
terest of modeling and optimization during recent years. The contribution of
this thesis lies in a state of decision making from the applications in various
scenarios taken up during the research. In this work, a new approach to de-
cision making has been introduced that facilitates fast and reliable decisions.
It enables pair-wise comparisons with necessary sagacity to obtain a clear
and unambiguous conclusion.
3
Organization Of The Thesis
The thesis consists of eight chapters followed by
References and Bibliography
4
Contents
1 Introduction 8
1.1 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.2 Introduction to MCDM Methods . . . . . . . . . . . . . . . . . . . 9
1.2.1 Analytical Hierarchy Process (AHP) . . . . . . . . . . . . . 10
1.2.2 Preference Ranking Organization Method for Enrichment Eval-
uation (PROMETHEE) . . . . . . . . . . . . . . . . . . . . 10
1.2.3 Technique for Order Preference by Similarity to Ideal Solution
(TOPSIS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.2.4 Rank Order Centroids (ROC) . . . . . . . . . . . . . . . . . 12
1.2.5 Ratio Method . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.2.6 Lexicographic Approach . . . . . . . . . . . . . . . . . . . . 13
1.2.7 Weighted Penalty Method . . . . . . . . . . . . . . . . . . . 13
1.3 Publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2 AHP Based Assignment Model For Allotting Parking Slots To Dif-
ferent Localities 15
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.2 Problem Description and Mathematical formulation . . . . . . . . . 16
2.3 Solution Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3 Site Selection Using Optimization Techniques 18
5
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.2 Problem Description and Mathematical Formulation . . . . . . . . . 19
3.3 Solution Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4 Leader Culling Using AHP - PROMETHEE Methodology 21
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.2 Problem Description . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.3 Solution Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
5 Agronomic Business Promotion 23
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5.2 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5.3 Solution Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
6 A Qualitative Assessment of Educational Software 25
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
6.2 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
6.3 Solution Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
7 Improving Consistency of Comparison Matrices in Analytical Hi-
erarchy Process 27
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
7.2 Research Methodology . . . . . . . . . . . . . . . . . . . . . . . . . 28
8 Conclusion of the Thesis 29
8.1 Summary of the Content and Conclusions . . . . . . . . . . . . . . 29
8.2 Summarizing the Findings and Limitations . . . . . . . . . . . . . . 32
8.3 Suggested Implications (Research Contribution) . . . . . . . . . . . 33
8.4 Recommendations for Further Work (Research) . . . . . . . . . . . 34
6
References 35
Bibliography 43
Applications of Vector Optimization Problems
7
Chapter 1
Introduction
This chapter is introductory and covers review of literature since 20th century till
date. A brief preamble of Multi Criteria Decision Making (MCDM) methodologies
is also included.
1.1 Literature Review
The study of decision problems has a long history and in the last few decades, it
has been one of the major research field in optimization discipline. A preeminent
work has been carried out in 20th century in field of optimization theory. The
mathematical modeling of these decision making problems gained momentum with
monumental contributions of various eminent economists and mathematicians.
In context of MCDM problems, Major advances have been recorded since 1980s.
Saaty(1980) introduced the Analytic Hierarchy Process (AHP), which brought evo-
lutionary advances in MCDM approaches. Saaty being placed in Fortune magazine,
is visibly one of the most successful people in MCDM. AHP, developed by Saaty
is one of the most popular techniques used by the researchers and practitioners.
Another useful technique, Technique for Order Preference by Similarity to Ideal So-
lution (TOPSIS) was developed by Hwang and Yoon(1981) and it is a popular ap-
8
proach to MCDM problems. In early 1980s, Brans(1982) laid fundamentals of Pref-
erence Ranking Organization Method for Enrichment Evaluation (PROMETHEE),
which is a part of outranking methods used as decision-aid in various MCDM prob-
lems. Considerable work was carried out for more than a decade to bring out various
versions of PROMETHEE by Brans et al.(1982; 1984; 1988; 1992; 1994; 1996).
Saaty & Takizawa(1986) revised AHP in their study so that it could handle the
non-linear hierarchies. Edwards & Barron(1994) first propounded Multi Attribute
Global Inference of Quality(MAGIQ) which was based on rank order centroids to
assign attributes’ weights. Carlsson & Fuller(1995) in mid 1990s introduced inter-
dependency concept into MCDM which later on was improved by Ostermark(1997).
Saaty(1996; 1999) developed the Analytical Network Process (ANP) by the end of
20th century, which is a more general form of AHP. Numerous other credible ap-
proaches have been cultivated and fostered till date, by eminent researchers in the
specified field.
1.2 Introduction to MCDM Methods
MCDM models are divided into two groups of Multiple Attribute Decision Making
(MADM) and Multiple Objective Decision Making (MODM). With regard to the
subject under consideration, the problems of decision making in this research cover
both MADM and MODM models. By considering the various techniques of MCDM
Models, some of the MCDM methods used in the present work are: Analytical Hier-
archy Process (AHP), Technique for Order Preference by Similarity to Ideal Solution
(TOPSIS), Preference Ranking Organization Method for Enrichment Evaluation
(PROMETHEE), Multi Attribute Global Inference of Quality (MAGIQ), Simple
Additive Weighting (SAW), Ratio Method, etc.
9
1.2.1 Analytical Hierarchy Process (AHP)
The formulation of AHP given by Saaty(1980) emerged as a paradigm for decision
making scenarios categorized under MADM. It is based on the well-defined mathe-
matical structure of consistent matrices and their associated eigenvector’s ability to
generate true or approximate weights. AHP can be considered to be both a descrip-
tive and prescriptive model of decision making and it is perhaps, the most widely
used decision making approach in the world today. Its validity is based on the
many thousands of actual applications in which the AHP results are accepted and
used by the aspirants. It is an approach to decision making that involves structur-
ing multiple judgment criteria into a hierarchy, assessing the relative importance of
these criteria, comparing alternatives for each criterion, and determining an overall
ranking of the alternatives. It provides a comprehensive and rational framework for
structuring a decision problem for representing and quantifying its elements. The
outcome of AHP is a prioritized weighting of each decision alternative. The AHP
converts these evaluations to numerical values that can be processed and compared
over the entire range of the problem. A numerical weight or priority is derived for
each element of the hierarchy, allowing diverse and often incommensurable elements
to be compared to one another in a rational and consistent way.
1.2.2 Preference Ranking Organization Method for Enrich-
ment Evaluation (PROMETHEE)
PROMETHEE, an outranking decision making technique, was introduced by Pro-
fessor Jean-Pierre Brans(1982) in 1982. At that time he proposed only the ba-
sic PROMETHEE I (partial ranking) and PROMETHEE II (complete ranking)
versions. Soon thereafter he started working with Bertrand Mareschal(1984) and
they proposed Graphical Analysis for Interactive Assistance (GAIA) method as
10
a descriptive extension of PROMETHEE in 1988. The visual interactive module
GAIA provides a marvelous graphical representation supporting the PROMETHEE
methodology and it is one of a very few effective descriptive MADM methods till
date. PROMCALC (later PROMCALC-GAIA) was one of the first truly interac-
tive software with a strong emphasis on user interface, graphical representations
and sensitivity analysis. The PROMETHEE method is based on mutual compar-
ison of each alternative pair with respect to each of the selected criteria. In order
to perform alternative ranking by the PROMETHEE method, it is necessary to
define preference function P (a, b) for alternatives a and b after defining the crite-
ria. Alternatives a and b are evaluated according to the criteria functions. It is
considered that alternative a is better than alternative b according to criterion f , if
f(a) > f(b). The decision maker has the authority to assign the preference to one
of the alternatives on the basis of such comparison. The preference can take values
on the scale from 0 to 1.
1.2.3 Technique for Order Preference by Similarity to Ideal
Solution (TOPSIS)
TOPSIS is a MADM method to identify solutions from a finite set of alternatives.
As suggested by Hwang and Yoon(1981), a MADM problem may be viewed as a
geometric system in which the m alternatives that are evaluated by n attributes
are similar to m points in a n dimensional space. Therefore, the most preferable
alternative should be the point in that space that is nearest to the ideal solution
and farthest from the worst solution.
11
1.2.4 Rank Order Centroids (ROC)
This method is a simple way of giving weights to a number of items ranked according
to their importance. The decision-makers usually can rank items much more easily
than giving weights to them. This method takes those ranks as inputs and converts
them to weights for each of the items. The term ‘Rank Order Centroid’ was coined
by Barron and Barrett(1996) who also argued for its use in MADM problems. The
idea is to convert ranks into values that are normalized on a 0.0 to 1.0 interval scale.
As proposed by James D. McCaffrey(2005), ROC is a MADM technique based on a
hierarchical decomposition of comparison attributes. It has features similar to AHP
and is used to assign a single, overall measure of quality to each member of a system
having arbitrary number of comparison attributes. This technique uses rank order
centroids to assign normalized numeric weights to the comparison attributes as an
overall measure of quality with reference to a set of evaluation criteria. MAGIQ
was originally developed to validate the results of AHP by adopting a comparatively
simplified course of action and the technique has produced highly useful results in
practice.
1.2.5 Ratio Method
The Ratio Method is another simple way of calculating weights for a number of
critical factors. A decision-maker should first rank all the items according to their
importance in the preferred domain. The next step is giving weight to each item
based on its rank in the interval [10, 90]. Here lowest ranked item will be given
a weight of 10 and rests of the items are rated in multiples of 10 based on the
preferences given by decision maker. For example if item II is five times more
important to item I (the lowest ranked item), then item II is provided with a
rating 50. The last step is normalizing these raw weights as proposed by Weber &
Borcherding(1993).
12
1.2.6 Lexicographic Approach
Multi-objective optimization consists of optimizing a number of objectives that
are usually conflicting. One way to tackle MODM problems is the lexicographic
method. Here each objective is optimized one at a time subject to a pre-defined
ordering established by the decision makers. It is important that the decision-maker
must express preferences in order to establish the ordering and the performance of
the method is vulnerable to the priorities given to various objectives. The single ob-
jective problem taking into account the most prioritized objective is solved, subject
to given constraints by usual methods ignoring rest of the objectives. The problem
is reconstituted considering the next prioritized objective with the previous solution
as an added constraint. The process is repeated until all the objectives are dealt one
by one adding the previous solutions in the constraint inequations. The approach
is useful while dealing with few objectives (two or three) but the procedure may be
lengthy when number of objectives exceed.
1.2.7 Weighted Penalty Method
While solving the MODM problems for a variety of parameters, scalarization is
the most practical and feasible approach. Several possible solutions are generated
depending on the priorities provided to various objectives. In the last decades,
the main focus was on finding one optimal solution to such problems by interactive
methods(1999; 1995) but now due to availability of much advanced application soft-
ware, it is possible to represent the whole efficient set without much manual efforts.
The decision maker gets a better insight to the problem structure by visualizing
the whole solution set at a stretch. A wide variety of scalarization methods exist
based on which one can assemble a MODM problem into prioritized single objec-
tive problem. In this work, a Weighted Penalty Cost Approach to put restrictions
13
on assignments, apart from achieving a prioritized efficient solution by exercising
priorities to preemptive factors, has been employed.
1.3 Publications
The subject matter of the thesis is published / under publication in the form of
following research papers worked out by the author.
I. Vandana Bagla & Anjana Gupta (2011). Analytical Hierarchy Process Based
Assignment Model for Allotting Parking Slots to Different Localities, Journal of
Multi-Criteria Decision Analysis, Wiley Online Library, 18, 173 - 185.
II. Vandana Bagla, Anjana Gupta & Deepika Kukreja (2011). A Qualitative As-
sessment of Educational Software, International Journal of Computer Applications,
36(11), 1 - 7.
III. Vandana Bagla, Anjana Gupta & Bhavna Sharma (2012), Leader Culling
Using AHP-PROMETHEE Methodology, presented in Third International Confer-
ence on Computational Intelligence Applications (ICCIA -2012) held on Feb 11 -
12, 2012 in Sandip Institute of Technology & Research Centre, Nashik-4222013 and
proceedings published in International Journal of Computer Application (IJCA).
IV. Vandana Bagla & Anjana Gupta (2012), Agronomic Business Promotion
published in the book titled “Business Intelligence and Data Warehousing”, Narosa
publications, 132 - 143, ISBN 978-81-8487-212-5.
V. Vandana Bagla , Anjana Gupta & Aparna Mehra (2013), Improving Consis-
tency of Comparison Matrices in Analytical Hierarchy Process, accepted for publi-
cation in International Journal of Current Engineering and Technology.
VI. Vandana Bagla & Anjana Gupta (2013), Site Selection Using Optimiza-
tion Techniques, accepted for publication in International Journal of Scientific &
Engineering Research.
14
Chapter 2
AHP Based Assignment Model
For Allotting Parking Slots To
Different Localities
2.1 Introduction
This chapter propounds a multi-criteria decision-making assignment model to allo-
cate appropriate parking slots to various localities. An algorithm is developed that
allows the consumers to choose the most suitable car parking according to their
requirements. Due to growing population and a consequent increase in number of
vehicles, the problem of parking personal vehicles is becoming explosive. Parking
has been widely recognized to be an important transportation policy issue in the
broad context of the transportation-land use relationship. All the aspects of sharing
of parking slots are visualized subject to land availability. An effective resolution
is suggested by framing a bi-objective problem to minimize the weighted cost and
maximum distances traveled by the residents. MADM technique, Analytic hier-
archy process (AHP) has been used to rank various relating attributes. MODM
15
model ‘Weighted Penalty Method’ is used for scalarization purpose.
2.2 Problem Description and Mathematical for-
mulation
Suppose there are m localities and n available parking slots. Each of the locality
is to be allocated with a parking slot subject to the conditions that development
cost of the prescribed parking slot should be minimized, traveling distance should
be minimum and the prescribed parking slot should justify the requirements of the
locality. For m > n , available parking slots can be shared by aspiring localities
and for m ≤ n , no sharing is required .
Let the localities li (i = 1, . . . ,m) are to be allocated with pj (j = 1, . . . , n)
parking slots. Suppose ai be the requirement of a particular locality li (i = 1, . . . ,m)
and bj be the capacity of the parking slot pj. The set up cost and weight of the
parking slot pj (j = 1, . . . , n) be denoted by cj and wj respectively.
Let ρj =cj∑j cj
(1 − wj) denote the weighted cost of parking slot pj . Here we
take the weights (1 − wj) as our purpose is to minimize the weighted set up cost.
The distance between the locality li and parking slot pj is denoted by dij.
Then above described model is formulated as the following two objective non
linear programming problem.
min (C(x), D(x))
C(x) =m∑
i=1
n∑j=1
ρjxij
D(x) = max{dij : xij = 1 (i = 1, . . . ,m ; j = 1, . . . , n)}
(P ) subject to the constraints
16
n∑j=1
xij = 1 (i = 1, . . . ,m) (2.1)
m∑i=1
xij ≤ β 1 ≤ β ≤ m; β ∈ N ; (j = 1, . . . , n) (2.2)
m∑i=1
aixij ≤ bj ( j = 1, . . . , n) (2.3)
xij = 0 or 1 (i = 1, . . . ,m ; j = 1, . . . , n) (2.4)
It is to be noted that β = 1 if m ≤ n whereas 1 < β ≤ m for m > n.
2.3 Solution Procedure
The above formulated problem is a bi-objective integer programming problem. The
prioritized two-objective problem is reduced to an equivalent single-objective inte-
ger programming problem by using scalarization techniques. Surveys are conducted
and corresponding positive reciprocal matrices are constructed at each level of hier-
archy. MADM technique, AHP is used to calculate the weights for different criteria
associated with a parking slot. The quandary is scalarized using MODM model
‘Weighted Penalty Method’ and subsequently solved using Hungarian Method to
assignment problem and 0− 1 integer programming.
17
Chapter 3
Site Selection Using Optimization
Techniques
3.1 Introduction
This chapter extends the study conducted in chapter 2 by analyzing a three objec-
tive problem in relevance to site selection model. Almost all investment decisions
involve multiple, diverse and complex set of social and financial factors which are
quite hard to be overcome by mere intuition. A number of allocation problems
have been solved in the recent past; see Ignizio(1982), Ignizio and Cavalier(1994).
In Multi objective Programming Problems, there may not be a single solution that
simultaneously optimizes each objective to its fullest. In each case an objective
must have reached a point such that, while attempting to optimize it further, other
objectives suffer as a result. Finding such a solution and quantifying how much
better this solution is compared to many other such solutions, is the goal when
setting up and solving a Multi-objective Optimization Problem; see Azarm(1994),
Prakash and Gupta(2006), Serafini(1985) and Steuer(1986). Here we envision a site
selection model for commercial activities which efficiently explores a multi-criteria
18
decision-making model involving three objectives. Maximization of profit and mini-
mization of set-up cost are the major objectives . Third objective is to rank various
attributes such as capacity, neighborhood, connectivity, transport availability and
proximity which are considerably important factors for a flourishing business.
3.2 Problem Description and Mathematical For-
mulation
Suppose a corporate body/M.N.C. deals in m different business/outlets and there
are n available sites (m ≤ n). Problem is to allocate a suitable site out of the
available sites to each business/outlet. While doing this, the two major objectives
of the company are to maximize the overall profit and to minimize the overall cost.
At the same time, the company wants to prioritize the sites carrying more weights
for flourishing business.
Let pij (i = 1, . . . ,m; j = 1, . . . , n) denote the expected profit , when ith business
is set up at jth site. Also let cj be the overall cost and wj be the weight of jth site
(j = 1, . . . , n) calculated using AHP. Let ρj denote the normalized cost of the jth
site. Then (1− ρj) = rj denotes the reversed cost of the jth site. It is to be noted
that we have reversed the cost as our objective is to minimize the cost and on the
contrary, we are dealing with a maximization problem. Then the above described
model is formulated as the following three objective problem.
(A) Maximize Z(X) = (P (x), R(x), W (X))
P (x) =m∑
i=1
n∑j=1
pijxij , R(x) =m∑
i=1
n∑j=1
rjxij , W (x) =m∑
i=1
n∑j=1
wjxij
19
subject to the constraints:
n∑j=1
xij = 1 (i = 1, . . . ,m) (3.1)
m∑i=1
xij ≤ 1 (j = 1, . . . , n) (3.2)
xij = 0 or 1 (i = 1, . . . ,m ; j = 1, . . . , n) (3.3)
3.3 Solution Procedure
The solution procedure consists of two phases. In the first phase, weights are allo-
cated to various available sites based on various important aspects such as neigh-
boring locality, broad or narrow connecting roads, area/capacity of the available
sites, proximity factors such as competitive business rivals in the nearby areas,
transport availability such as metro or other public transports. To accomplish this,
an approach of Analytic Hierarchy Process (AHP) given by Saaty (1980) is used.
In the second phase, the three objective problem seeking to maximize the profit,
minimize the set up cost born by the investors and to maximize qualitative stan-
dards is untangled using Lexicographic approach and thereafter solved using integer
programming. A comparative solution is projected using scalarization method.
20
Chapter 4
Leader Culling Using AHP -
PROMETHEE Methodology
4.1 Introduction
This chapter proposes the use of improved PROMETHEE methodology for candi-
date selection eligibility and has the potential to re-shape and influence the way
the general public perceive social justice for a rational decision. PROMETHEE
is a MCDM method well adapted to problems where a finite number of alterna-
tives are to be ranked considering several, sometimes conflicting, criteria. It has
successively been applied in many fields especially in the investment analysis and
performance evaluation. Albadvi(2007), Babic and Plazibat(1998), Bouri(2000),
Mareschal(1988), Mareschal and Brans(1988; 1991) and Vranegl(1996) , all applied
PROMETHEE as a decision making tool to solve the different problems in the field
of finance.
The developed methodology facilitates the selection procedure in a rational and
transparent way that can be examined and understood by the concerned voters.
The objective quality of this method keeps inconsistency of decision makers’ op-
21
posing views within reasonable limits. The study is substantiated using Simple
Additive Weighting (SAW) technique. The factors considered here explore personal
as well as professional aspects of the aspiring candidates. The employed methodol-
ogy utilizes systematic approach to screen and ultimately select the most suitable
candidate.
4.2 Problem Description
A group of voters is asked to select a candidate among a set of contesting candidates.
Each voter has a personal ranking of the candidates according to his/her preferences.
The development of a ranking procedure requires a consensus on a set of criteria
for evaluation, upon which the selection of candidates will be based. Potentially
large number criteria are to be judged on a valid scale that is acceptable to all the
participants. The most critical phase in designing the selection process model is
to structure the decision problem. The main goal is to select the best candidate,
according to a set of criteria for evaluation. As conflicting views may arise among
different communities in determining the most important criteria of evaluation in a
given decision making setting, a general survey was conducted to develop the main
criteria, sub criteria and categories for each sub criteria for achieving the goal.
4.3 Solution Procedure
Various surveys were conducted to rate each attribute to others in a series of pair
wise comparisons. An approach of AHP is applied to find out the weights of each
criterion at different levels of hierarchy. The evaluation process finally generates
the global weights for each requisite criterion of interest. After having weight allo-
cations to the set of perceived criteria via AHP, the problem of selection of suitable
candidate is submitted to PROMETHEE II methodology for a pertinent ranking.
22
Chapter 5
Agronomic Business Promotion
5.1 Introduction
Present chapter critically analyzes the agronomic market opportunities for the in-
vestors so as to formulate better merchandise programs. Rural markets in India
constitute a wide and untapped market for many products and services which are
being marketed for the urban masses. The urban market reaching almost to a sat-
uration level, demand from rural India has been on a high trajectory. It provides
a promising business as many new products have already made their entry into the
rural consumer basket. Thus, Indian rural markets have caught the attention of
many companies, advertisers and multinational companies. Contemporary desider-
atum is what kind of investments is required to unlock long-term value from rural
markets.
5.2 Problem Statement
A marketer trying to market his product or service in the rural areas is faced by
many challenges; the first is posed by the geographic aspects like location, profit cost
revenue, population density and climatic conditions. The second challenge is from
23
the demographic factors like purchasing power, literacy level and denominations in
the region. Third is psychographic segmentation like activities, interests and ethics
of the residents. And last but not the least are behavioristic influences namely usage
rates, brand loyalty and festivity. One of the key issues, which require research, is
the methodology by which we can design a cost effective, efficient, environment
friendly model which is befitting on the above set of criteria.
5.3 Solution Procedure
To evaluate the hierarchy, decision makers are asked to allot rankings to the leveled
criteria according to their requisite priorities. Numeric weights are provided to
all the criteria using MAGIQ technique. TOPSIS has been used to provide final
rankings to available agronomic locations. Various surveys were conducted to rate
each attribute to others in a series of pair wise comparisons. An approach of AHP is
applied to find out the weights of each criterion at different levels of hierarchy. The
evaluation process finally generates the global weights for each requisite criterion of
interest. After having weight allocations to the set of perceived criteria via AHP,
the problem of selection of suitable candidate is submitted to PROMETHEE II
methodology for a pertinent ranking.
24
Chapter 6
A Qualitative Assessment of
Educational Software
6.1 Introduction
This chapter inspects the prerequisite of children’s software evaluation in the light
of dynamic nature of edutainment perspective. The work provides comparative
ranking of educational software for scholastic programs scouting both technical and
non technical aspects. Work was conducted to help educational institutions gain a
deeper understanding of professional decisions they face and reduce their initial state
of uncertainty about the best course of action. The Internet-related technologies
have helped to lay the blueprint for the futuristic software evaluation information in
the 21st century. More studies need to be accomplished in this field, as that done by
Escobedo and Evans(1997), where the ratings assigned by the published software
methods are compared with actual child selection or Tammen and Brock(1997),
where middle school students are asked to identify issues they feel are important
for the evaluation of software programs.
25
6.2 Problem Statement
The problem of evaluation of alternatives submitted to a multi-criteria decision
analysis is a contentious task. Usually there is no optimal solution as no alternative
is the best one on each criterion. Problem is to evaluate the software according
to their credibility on the set of weighted judgment criteria. The most demanding
phase in designing the evaluation model is to structure the decision problem. The
goal is to reckon the software according to a set of criteria for evaluation. The con-
ventional approach to predictive evaluation is to use a checklist exploring technical
and non-technical features of the requisite software.
6.3 Solution Procedure
Rank Order Centroid (ROC) methodology and Ratio Method are used to accomplish
the result. The key advantage of ROC Methodology is its simplicity in surveying
whereas Ratio method requires quantified rating of prioritized alternatives. Ratio
Method has its own advantages when the decision makers influence their prioritized
specifications regarding significances.
26
Chapter 7
Improving Consistency of
Comparison Matrices in
Analytical Hierarchy Process
7.1 Introduction
This chapter designs a contemporary MADM model to assist the decision making
scenarios and is specifically applicable to multi-criteria decision problems associated
with large number of attributes. Antecedent of inconsistencies are analyzed and a
corrective application based approach is introduced to construct nearly consistent
matrices. The analysis extends the previous researches in many ways. Firstly, it
analyzes the root causes of inconsistencies in pairwise comparison matrices and
suggests the constructive approaches to overcome the discrepancies. Secondly, the
model enables us to visualize the impact of various criteria on the final ranking
and determine the level of importance of each criterion based on survey analysis.
Thirdly, implementing above methodology can serve as a paradigm for MCDM
problems for which similar models can be developed, modified and improved. A
27
further advantage concerns this model’s inherent ability to procure instant reporting
of analysis results, where numerous subtle factors are involved.
7.2 Research Methodology
Decision Makers are more likely to be cardinally inconsistent because they cannot
estimate precise measurement values even from a known scale and worse when
they deal with intangibles (a is preferred to b twice and b to c three times, but a
is preferred to c only five times). Here the first reason of inconsistency appears.
This is referred to as inconsistency of reciprocity. Again, judgments are ordinally
intransitive if (A is preferred to B and B to C but C is preferred to A). One of
the apparent reason is, large number of pairwise comparisons (n(n− 1)/2) required
to workout the attribute weights at a given level of hierarchy. This results in
baffling responses relating to pairwise comparisons. This is termed as inconsistency
of transitivity and it goes on increasing with the size of the matrix.
28
Chapter 8
Conclusion of the Thesis
In this work, an integrated solution methodology for a general discrete Multi Cri-
teria Decision Making (MCDM) problem is developed based on various interactive
approaches. The concept is designed to make better choices when faced with com-
plex decisions involving several dimensions. MCDM tactics are especially helpful
when there is a need to combine hard data with subjective preferences, to make
trade-offs between desired outcomes and to involve multiple decision makers. In the
decision making process, there are typically multiple conflicting criteria that need
to be assessed simultaneously with about same degree of precedence. This craves
the search for an approach which deals the predicament with necessary sagacity
to obtain a clear and unambiguous conclusion. In this chapter, we summarize the
work done in this dissertation in the said field and draw conclusions of this R and D.
We also discuss limitations of this investigation and suggested implications. Future
plan of action for further research is also acknowledged.
8.1 Summary of the Content and Conclusions
Chapter 1 is introductory and covers literature review and introduction to Multi
Criteria Decision Making (MCDM) methodologies.
29
Chapter 2 propounds a multi-criteria decision-making assignment model to al-
locate appropriate parking slots to various localities. The proposed methodology
provides a decision making method for development of parking slots to decision
maker who is interested in minimizing the total cost as well as the distance tra-
versed by the residents. Sets of efficient solutions for allotting parking slots seeking
minimization of cost as the first priority objective and minimization of distance as
the second priority objective has been obtained. Concepts of assignment model and
integer programming have been used to assign parking slots to various localities.
Heuristics have effectively helped in getting precise and practicable solution. The
study suggests sharing of parking slots if number of localities exceeds the available
number of parking slots.
Chapter 3 augments the study conducted in chapter 2 by analyzing a three objec-
tive problem in relevance to site selection model. While strategically managing our
location decisions, we need solutions that address multiple logistics and economic
factors involving real estate and our customers. Maximization of profit and min-
imization of set-up cost are the two major objectives. Third objective is to rank
the sites qualitatively, for a flourishing business. MADM model, AHP has been
used to rank various relating attributes. A comparative study is conducted using
MODM models ‘Weighted Penalty Method’ and ‘Lexicographic Approach’. The
method also suggested that once starting up with a low profit/low cost model, high
futuristic growth may be expected if qualitative ranks are high.
Chapter 4 proposes to combine two MADM tactics for efficient manipulation of
candidate selection procedure. An approach of AHP is applied to find out the
weights of each criterion at different levels of hierarchy. After having weight alloca-
tions to the set of perceived criteria via AHP, the problem of selection of suitable
candidate is submitted to PROMETHEE II methodology for a pertinent ranking.
The present study reveals that above methodology can be utilized as a valuable
30
decision-making process to evaluate the contestants in a logical and consistent fash-
ion for many reasons.
Chapter 5 critically analyzes the agronomic market opportunities for the investors
so as to formulate better merchandise programs. Here also two MADM methodolo-
gies are merged, intending to simplify the ranking procedure, taking into account
illiteracy problem in rural areas. Numeric weights are provided to all the criteria
using MADM technique, Multi Attribute Global Inference of Quality (MAGIQ).
Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) has been
used to provide final rankings to available agronomic locations. There is a promis-
ing business and definitely a lot of money in rural India but the smart thing would
be to weigh crucial attributes strategically for tactical gains.
Chapter 6 inspects the prerequisite of childrens software evaluation in the light of
dynamic nature of edutainment perspective. Work was conducted to help educa-
tional institutions gain a deeper understanding of professional decisions they face
and reduce their initial state of uncertainty about the best course of action. Com-
parative solutions are provided using two MADM techniques Rank Order Centroid
(ROC) methodology and Ratio Method.
Chapter 7 designs a novel MADM model to assist the decision making scenarios
and is specifically applicable to multi-criteria decision problems associated with
large number of attributes. We have described the implementation of a corrective
model, assisting the decision-maker in the construction of a consistent comparison
matrix. Group decision making is becoming increasingly important in decision
scenarios associated with MCDM problems. The proposed methodology provides
an insight into the impediments to effective group processes and on techniques that
can improve group decisions. We hope that the projected approach can refocus the
attention of researchers from the race of finding better judgments for inconsistent
and near consistent matrices.
31
8.2 Summarizing the Findings and Limitations
This study has shown that methodologies adapted from MCDM tactics have ver-
satile applications in almost all decision making layouts. Application feasibility as
well as efficiency of the tactics in solving diversified problems have been described
illustratively in this study. Expert opinions are sought and implemented to scout
empirical perfections. Some limitations to this pilot study need to be acknowledged.
One of the more significant findings that emerge from this study is that, AHP is
a useful decision tool to consolidate evaluation data. It provides a useful mecha-
nism for checking the consistency of the evaluation measures and alternatives. This
method is preferred over other established methodologies as it does not demand
prior knowledge of the utility function; rather it is based on a hierarchy of criteria
and attributes facilitating the understanding of the problem. It allows relative and
absolute comparisons, making this method a very robust tool. This method al-
lows combinations with other techniques as well as scenario analysis and simulation
exercises. The further strength of the AHP is its ability to detect inconsistent judg-
ments. Last but not the least, AHP allows group decision-making and is convenient
in numerical handling. Free on line software support is an added advantage.
The limitation of the AHP is that, it only works for positive reciprocal matri-
ces. The other seeming drawback is the scale used in judgment mechanism which
inculcates inconsistencies in assessments. Large number of pairwise comparisons
required for sizable comparison matrices is another snag which seeks research for
further improvement.
It was also shown that MAGIQ technique using rank order centroids for weight
determination is another useful MADM model. It is far more quicker to other
analogous techniques used for the decision mechanism. The key advantage of the
methodology is its simplicity in surveying, yet it has its own limitations when the
32
decision makers influence their prioritized specifications regarding significances.
This study also found that PROMETHEE and TOPSIS used in the present work are
simple and logical MADM tactics widely used in decision scenarios. Still they have
their own limitations in practical implications. Major of them is, the requirement
of predefined criteria weights which seeks the help of other MADM procedures.
Further complication in PROMETHEE is requirement of preference functions as
proposed in original inception. But these preference functions require the definition
of some preferential parameters, such as the preference and indifference thresholds.
However, in real time applications, it may be difficult for the decision maker to
specify which specific form of preference function is suitable for each criterion and
also to determine the parameters involved. Another apparent constraint is, limited
software support which enhances manual computations.
Another important findings regarding the two MODM techniques used in this work
indicate that Lexicographic Approach is useful only when objectives considered in
the study are few. Also it does not provide alternatives to the aspirants for in-
dividual preferences whereas Weighted Penalty Method provides a set of efficient
solutions which can be evaluated for prioritized preferences. However it was ob-
served that weights alloted for scalarization purpose in Weighted Penalty Method
in the present work may significantly effect the decision layouts.
8.3 Suggested Implications (Research Contribu-
tion)
The present study confirms previous findings and contributes additional evidences
that suggest that pairwise comparison methods can always be used to draw the
final conclusions in a comparatively accessible and intelligible way. However, as the
results suggest that AHP should be used in combination with other decision tools to
33
support because AHP is efficient only when the number of criteria and alternatives
are few.
The methodology of PROMETHEE and its algorithm has been demonstrated in
detailed and simplified way and it is believed that it will be useful for decision
makers to apply such MCDM tools to supplement their decision-making efforts.
A contemporary decision making technique has been brought up in this work in
which antecedent of inconsistencies are analyzed and a corrective application based
approach is introduced, which helps the decision-maker to build consistent matrices
with controlled error by appreciably less number of pairwise comparisons.
8.4 Recommendations for Further Work (Research)
The current findings add substantially to our understanding of incorporating MCDA
tactics in decision layouts. The empirical findings in this study will serve as a base
for future studies in various application scenarios.
34
References
[1] Albadvi, A., Chaharsooghi, S.K. & Esfahanipour, A. (2007). Decision making in
stock trading, An application of PROMETHEE, European Journal of Operational
Research, 177, 673 - 683.
[2] Antonio, A. J. (2006). Consistency in the analytic hierarchy process, a new
approach, International Journal of Uncertainty, Fuzziness and Knowledge-Based
Systems, 14(4), 445 - 459.
[3] Azarm, S. (1994), Reduction Method with System Analysis for Multiobjective
Optimization-Based Design, Structural Optimization, 7, 47 - 54.
[4] Babic, Z. & Plazibat, N. (1998). Ranking of enterprises based on multi criteria
analysis, International Journal of Production Economics, 29 - 35, 56 - 57.
[5] Ball, V. C., Noel, J. & Srinivasan (1994). Using the analytic hierarchy process
in house selection, Journal of Real Estate Finance and Economics, 9, 69 - 85.
[6] Barron, F. H. & Barrett, B. E. (1996). Decision Quality Using Ranked Attribute
Weights, Management Science, 42 (11), 1515 - 1523.
[7] Basak, I. & Saaty, T. L. (1993). Group decision making using the analytic
hierarchy process, Mathematical Computer Modelling, 17, 101 - 109.
[8] Belton, V. & Gear, A. E. (1983). On a Short-Coming of Saaty’s Method of
Analytic Hierarchies, Omega, 11, 228 - 230.
35
[9] Belton, V. & Stewart, T. (2002). Multiple Criteria Decision Analysis, An Inte-
grated Approach, Chapter 5, 119 - 162, Boston: Kluwer Academic Publishers.
[10] Bhushan, N. & Ria, K. (2004). Strategic Decision Making: Applying the Ana-
lytic Hierarchy Process, London: Springer-Verlag London Limited.
[11] Bouri, A. & Martel, J. M. (2000). A multi-criterion approach for selecting
attractive portfolio, Journal of Multi-Criteria Decision Analysis, 11, 269 - 277.
[12] Bozoki, S. & Rapcsak, T. (2007). On Saatys and Koczkodaj’s inconsistencies of
pairwise comparison matrices, Computer and Automation Institute, Hungarian
Academy of Sciences.
[13] Brans, J. P. (1982). The engineering of decision: Elaboration instruments of
decision support method PROMETHEE, Laval University, Quebec, Canada.
[14] Brans, J. P., Mareschal, B. & Vincke, P. (1984). PROMETHEE: A new family
of outranking methods in MCDM, in IFORS’84, North Holland.
[15] Brans, J. P. & Mareschal, B. (1988). Geometrical Representation for MCDM,
the GAIA procedure, European Journal of Operation Research, 34, 69 - 77.
[16] Brans, J. P. & Mareschal, B. (1992). PROMETHEE V : MCDM problems with
additional segmentation constraints, INFOR, 30 (2), 85 - 96.
[17] Brans, J. P. & Mareschal, B. (1994). PROMCALC & GAIA : A new decision
support system for multicriteria decision aid, Decision Support Systems, 12, 297
- 310.
[18] Brans, J. P. (1996). The space of freedom of the decision maker modelling the
human brain. European Journal of Operational Research, 92 (3), 593 - 602.
36
[19] Bryson, N. & Mobolurin, A. (1994). An approach to using the analytic hierarchy
process for solving multiple criteria decision making problems, European Journal
of Operational Research, 76, 440 - 454.
[20] Buckleitner, W. (1985). A Survey of Early Childhood Software,Ypsilanti, Michi-
gan : High/Scope Educational Research Foundation.
[21] Byun, D. H. (2001). The AHP method for selecting an automobile purchase
model, Information and Management, 38, 289 - 297.
[22] Carlsson, C. & Fuller, R. (1995). Multiple criteria decision making: The case
for interdependence, Computers & Operations Research, 22 (3), 251 - 260.
[23] Cogger, K. O. & Yu, P. L. (1985). Eigen weight vectors and least distance
approximation for revealed preference in pairwise weight ratio, Journal of Opti-
mization Theory and Applications, 46, 483 - 491.
[24] Crain, W. F. (2003). Multi-attribute Weight Determination: Elicitation
and Approximation, (Doctoral dissertation), George Mason University, Virginia,
United States.
[25] Davis, H. A. (1963). The Method of Pairwise Comparisons, Griffin, London.
[26] Dong, H. J. & Kang, K.J. (2000). A method for optimal material selection
aided with decision making theory, Materials and Design, 99 (21), 199 - 206.
[27] Dyer, J. S. (1990a). Remarks on the Analytic Hierarchy Process, Management
Science, 36, 249 - 258.
[28] Edwards, W. & Barron, F. H. (1994). SMARTS and SMARTER: Improved
Simple Methods for Multi attribute Utility Measurement, Organizational Behav-
ior and Human Decision Processes, 60 (3), 306 - 325.
37
[29] Escobedo, T. H. & Evans, S. (1997, March 28). A Comparison of Child-Tested
Early Childhood Education Software with Professional Ratings, Paper presented
at the Annual Meeting of the American Education Research Association, Chicago.
[30] Fang, S. & Zhang, Z. Y. (2005). Application of TOPSIS to the evaluation of
hospital performance, Chinese Journal of Health Statistics, 22 (3), 169 - 170.
[31] Forman, E. H. (1990). Random indices for incomplete pairwise comparison
matrices, European Journal of Operational Research , 48, 153 - 155.
[32] Haugland, S. W. & Shade, D. D. (1988). Developmentally appropriate software
for young children. Young Children, 43 (4), 37 - 43.
[33] Hummel, M. (2001). Supporting Medical Technology Development with the An-
alytic Hierarchy Process, Groningen, The Netherlands: Rijksuniversiteit Gronin-
gen.
[34] Hwang, C. L. & Masud, A. S. Md. (1979). Multiple Objective Decision Mak-
ing Methods and Applications (Lecture Notes in Economics and Mathematical
Systems), Berlin, Heidelberg, New York : Springer-Velarg.
[35] Hwang, C.L. & Yoon, K. (1981). Multiple Attribute Decision Making Meth-
ods and Applications (Lecture Notes in Economics and Mathematical Systems),
Berlin, Heidelberg, New York : Springer-Velarg.
[36] Ignizio, J. P. (1982). Linear Programming in Single and Multi-objective Sys-
tems, Englewood Cliffs, New Jersy : Prentice-Hall.
[37] Ignizio, J. P. & Cavalier, T. M. (1994). Linear Programming, Englewood Cliffs,
New Jersy : Prentice-Hall.
[38] Ishizaka, A. & Lusti, M. (2004). An expert module to improve the consistency of
38
AHP matrices, International transactions in operational research (ITOR), Black-
well Publishing 11 (1), 97 - 105.
[39] Jianglong, T., Xiaomin, Z. & Xueyi, S. (2005). Application of ideal point
method to the evaluation of land use planning project. Transactions of the CASE,
21 (2), 56 - 59.
[40] Jones, N. B. & Vaughan, L. (1983). Evaluation of Educational Software: A
Guide to Guides, Austin, TX : Southwest Educational Development Lab, North-
east Regional Exchange Incorporation, Chelmsford.
[41] Koczkodaj, W. W.(1993). A new definition of consistency of pairwise compar-
isons. Mathematical Computational Modeling, 18 (7), 79 - 84.
[42] Koksalan, M. & Sagala, P. (1992). An interactive approach for choosing the
best of a set of alternatives, Journal of the Operational Research Society, 43, 259
- 263.
[43] Lathrop, A. & Goodson, B. (1983). Courseware in the Classroom: Selecting,
Organizing, and Using Educational Software, Menlo Park, CA: Addison-Wesley.
[44] Littman, T. (1995, October 23). Parking requirement impacts on housing affod-
ability, Victoria Transport Policy Institute, Victoria, Canada : British Columbia.
[45] Liu, P. D. (2009). Multi-Attribute Decision-Making Method Research Based on
Interval Vague Set and TOPSIS Method. Technological Development of Economy,
15 (3), 453 - 463.
[46] Lunging, F. (1992). Analytical hierarchy in transportation problems, an appli-
cation for Istanbul, Urban Transportation Congress of Istanbul, 2, 16 - 18.
[47] Macharis, C., Springael, J., De Brucker, K. & Verbeke, A. (2003).
PROMETHEE and AHP: The design of operational synergies in multicriteria
39
analysis, Strengthening PROMETHEE with ideas of AHP, European Journal of
Operational Research, 153, 307 - 317.
[48] Mareschal, B. (1988). Weight stability intervals in multicriteria decision aid,
European Journal of Operation Research, 33, 54 - 64.
[49] Mareschal, B. & Brans, J. P. (1991). Bank Adviser : An industrial evaluation
system, European Journal of Operational Research, 54, 318 - 324.
[50] McCaffrey, J. (2005). Multi-Attribute Global Inference of Quality (MAGIQ),
Software Test and Performance Magazine, 7 (2), 28 - 32.
[51] McCaffrey, J. & Koski, N. (2006). Competitive Analysis Using MAGIQ, MSDN
Magazine, 11 (21), 35 - 39.
[52] Miettinen, K. (1999). Nonlinear multiobjective optimization, Boston : Kluwer
Academic Publishers.
[53] Paolo, F. (2003). A method for choosing from among alternative transportation
projects, European Journal of Operational Research, 150, 194 - 203.
[54] Prakash, S. & Gupta, A. (2006). Efficient solutions for the problem selecting
sites for polling stations and assigning voter-area to them with two objectives,
Proceedings of National Conference on Management Science and Practice (MSP-
2006), 141 - 151.
[55] Saaty, T. L. (1980). The Analytic Hierarchy Process, New York : McGraw Hill.
[56] Saaty, T. L. & Vargas, L.G. (1984). Comparison of eigenvalue, logarithmic least
square and least square methods in estimating ratio, Journal of Mathematical
Modelling, 5, 309 - 324.
40
[57] Saaty, T. L. & Takizawa, M. (1986). Dependence and Independence: from
Linear Hierarchies to Non-Linear Networks, European Journal of Operational Re-
search, 26, 229 - 237.
[58] Saaty, T. L. (1990). Eigenvector and logarithmic least squares, European Jour-
nal of Operational Research, 48, 156 - 160.
[59] Saaty, T. L. (1996). Decision Making with Dependence and Feedback : The
Analytical Network Process, Pittsburgh : RWS Publications.
[60] Saaty, T. L. (1999, Aug. 12-14). Fundamentals of the Analytical Network Pro-
cess, Proceeding of ISAHP 1999, Kobe, Japan.
[61] Serafini, P. (1985). Mathematics of Multi Objective Optimization, New
York : Springer-Verlag.
[62] Shih, H. S., Syur, H. J. & Lee, E. S. (2007). An extension of TOPSIS for group
decision making, Mathematical and Computer Modeling, 45 (7-8), 801 - 813.
[63] Shoup, D. C. (1995). An opportunity to reduce minimum parking requirements,
Journal of American Planning Association, 61, 14 - 28.
[64] Steuer, R. E. (1986). Multiple Criteria Optimization: Theory, Computation,
and Application, New York : John Wiley and Sons.
[65] Tammen, J. & Brock, L (1997). CD-ROM Multimedia: What Do Kids Really
Like, MultiMedia Schools, 4 (3), 54 - 56.
[66] Triantaphyllou, E. (2000). Multi-Criteria Decision Making Methods: A Com-
parative Study, Boston, MA, U.S.A : Kluwer Academic Publishers.
[67] Vrangel, S., Stanojevic, M., Stevanovic, V. & Luein, M. (1996). INVEX, In-
vestment advisory expert system. Expert Systems, 13 (2), 105 - 120.
41
[68] Wang, J. J. & Yang, D. L. (2007). Using a hybrid multi - criteria decision aid
method for information systems outsourcing, Computers & Operations Research,
34, 3691 - 3700.
[69] Weber, M., Borcherding, K. (1993). Behavioral influences on weight judgments
in multiattribute decision making. European Journal of Operations Research, 67,
1 - 12.
[70] Yu, P. L. (1985). Multiple Criteria Decision Making : Concepts, Techniques
and Extensions, New York : Plenum Press.
[71] Zhang, A. (2006). The Application of TOPSIS to the Venture Evaluation Index
System, Popular Science and Technology, 87 (1), 164 - 165.
[72] Zopounidis, C. & Doumpos M. (2000). Intelligent Decision Aiding Sys-
tems Based on Multiple Criteria for Financial Engineering, Boston, MA,
U.S.A. : Kluwer Academic Publishers.
42
Bibliography
[1] Arthur, M. G. (1968). Proper efficiency and the theory of vector maximization,
Journal of Mathematical Analysis and Applications, 22, 618 - 630.
[2] Arthur, M. G. (1970). Theory Series, Management Science, 16 (11), 676 - 691.
[3] Bellman, R. (1957). Dynamic Programming, Princeton University Press, New
York, NY: Dover Publications.
[4] Borda, J. C.(1781). Mmoire sur les lections au scrutin, Histoire de lAcademie
Royale des Sciences.
[5] Cantor, G. (1915), Contributions to the Founding of the Theory of Transfinite
Numbers, Philip, J. (ed.) (1955), New York: Dover, ISBN 978-0-486-60045-1.
[6] Charnes, A., Cooper W. & Ferguson R. (1955). Optimal estimation of executive
compensation by linear programming, Management Science, 1, 138 - 151.
[7] Charnes, A. & Cooper, W.W. (1961). Management Models and Industrial Ap-
plications of Linear Programming, New York : John Wiley and Sons.
[8] Cochrane, J. L. & Zeleny, M.(Eds.) (1973). Multiple Criteria Decision Making,
Proceedings of International Conference on MCDM, Columbia : University of
South Carolina Press.
[9] Condorcet, M. (1785). Essai sur lapplication de lanalyse la probabilit des
dcisions rendues la pluralit des voix, Paris.
43
[10] Dantzig, G. B.(1947). Maximization of a linear function of variables subject to
linear inequalities, Published in Activity Analysis of Production and Allocation,
Koopmans T. C (ed.) (1951), 339 - 347, New York, London : Wiley & Chapman-
Hall.
[11] Dantzig, G. B. (1953). Notes on linear programming, California : RAND Cor-
poration.
[12] DeSanctis, G. & Gallupe, R. A. (1987). Foundation for the study of group
decision support systems, Management Science, 33 (5).
[13] Edwards, W. (1954). The theory of decision making, Psychological Bulletin,
41, 380 - 417.
[14] Edwards, W. (1961). Behavioral decision theory, Annual Review of Psychology,
12, 473 - 498.
[15] Ehrgott, M. & Gandibleux, X. (2002). Multiple Criteria Optimization: State of
the Art Annotated Bibliographic Surveys, Boston : Kluwer Academic Publishers.
[16] Fonseca, C. M. & Fleming, P. J.(1995). An overview of evolutionary algorithms
in multiobjective optimization, Evolutionary Computation, 3 (1), 1 - 16.
[17] Gass, S. I. & Saaty, T. L. (1954). The Parametric Objective Function - Part I,
Operations Research, 2 (3), 316 - 319.
[18] Gass, S. I. & Saaty, T. L. (1955). Parametric Objective Function - Part II:
Generalization, Operations Research, 2 (1), 39 - 45.
[19] Gass, S. I. & Saaty, T.L. (1955). The Computational Algorithm for the Para-
metric Objective Function, Naval Research Logistics Quarterly, 2 (1 - 2), 39 -
45.
44
[20] Gass, S. I. (1985). Decision Making Models and Algorithms: A First Course,
New York, NY: John Wiley.
[21] Hadamard, J. S. (1902). Sur les problemes aux derivees partielles el leur sig-
nification physique, Bull. Univ. Princeton, 13, 49 - 62.
[22] Jelassi, T., Kersten, G. & Zionts, S. (1990). An introduction to group de-
cision and negotiation support, Readings in Multiple Criteria Decision Aid,
C.A.Bana E Costa (Ed.), Berlin : Springer-Velarg.
[23] Jensen, J .L. (1906). Sur les fonctions convexes et les inegualites entre les
valeurs moyennes, Acta Mathematica, 30, 175 - 193.
[24] Kantorovich, L. V. (1940). A new method of solving some classes of extremal
problems, Doklady Akad Sci USSR, 28, 211 - 214.
[25] Keeney, R. & Raiffa, H. (1976). Decisions with Multiple Objectives: Preferences
and Value Tradeoffs, New York: Wiley, U.S.A.
[26] Kirkwood, C. W. (1997). Strategic Decision Making: Multiobjective De-
cision Analysis with Spreadsheets, Belmont, Calif., U.S.A. : Duxbury Press,
ISBN : 0534516920.
[27] Koopmans, T. C. (1951). Analysis of Production as an Efficient Combination
of Activities, Activity Analysis of Production and Allocation, John Wiley and
Sons, 33 - 97.
[28] Korhonen, P., Wallenius, J. & Zionts, S. (1984). Solving the discrete multiple
criteria problem using convex cones, Management Science, 30, 1336 - 1345.
[29] Kuhn, H. W. & Tucker, A. W. (1951). Nonlinear programming, Proceedings
of 2nd Berkeley Symposium, Berkeley, California : University of California Press,
481 -492.
45
[30] Kuhn, H. W. (1955). The Hungarian method for assignment problem , Naval
Research Logistics Quarterly, 2, 83 - 97.
[31] Lai, Y. & Hwang, C. (1996). Fuzzy multiple objective decision making, Methods
and applications. Berlin : Springer-Velarg.
[32] Markowitz, H.M. (1952, March). Portfolio Selection, The Journal of Finance,
7 (1), 77 - 91.
[33] Markowitz, H. M. (1991). Autobiography, The Nobel Prizes 1990, Tore
Frangsmyr (Ed.), Stockholm : Nobel Foundation.
[34] Neumann, J. V. & Morgenstern, O. (1944), Theory of Games and Economic
Behavior, Princeton, N. J. : Princeton University Press.
[35] Neumann V. J. & Morgenstern, O. (1947). Theory of Games and Economic
Behavior, (vol. 2), Princeton, N. J. : Princeton University Press.
[36] Ostermark R. (1997). Temporal interdependence in fuzzy MCDM problems,
Fuzzy Sets and Systems, 88, 69 - 79.
[37] Raiffa, H. & Schlaifer, R. (1961). Applied statistical decision theory, Boston:
Division of Research, Graduate School of Business Administration, Harvard Uni-
versity.
[38] Raiffa, H. (1968). Decision Analysis : Introductory Lectures on Choices Under
Uncertainty, Massachusetts : Addison-Wesley.
[39] Ronald, H. & Kimball, G. E. (1959). Sequential Decision Processes : Notes on
Operations Research, Cambridge, Massachusetts : Technology Press.
[40] Ronald, A. H. & James, E. M. (1990). Principles and Applications of Decision
Analysis (vol. 1), Menlo Park, California : Strategic Decisions Group.
46
[41] Roy, B., Vanderpooten, D. (1996). The European School of MCDA: Emergence,
Basic Features and Current Works, Journal of Multi-criteria Decision Analysis,
5, 22 - 38.
[42] Schwefel, H.P. (1995). Evolution and Optimum Seeking. New York : John Wiley
& Sons Incorporation.
[43] Slowinski, R. & Teghem, J. (1990). Stochastic versus fuzzy approaches to
multiobjective mathematical programming under uncertainty, Boston : Kluwer
Academic Publishers.
[44] Thorndike, E. L. (1920). A constant error in psychological ratings, Journal of
Applied Psychology, 4, 25 - 29.
[45] Thurstone, L. L. (1927). The method of paired comparisons for social values,
Journal of Abnormal and Social Psychology, 21, 384 - 400.
[46] Walker, D. F. (1984). Computers in Schools: Potentials and Limitations and
The Software Problem, San Francisco, California : Far West Laboratory.
[47] Wierzbicki, A. (1980). The Use of Reference Objectives in Multiobjective
Optimization, Lecture Notes in Economics and Mathematical Systems, Springer,
Berlin, 177, 468 - 486.
[48] Zionts, S. & Wallenius, J. (1976). An Interactive Programming Method for
Solving the Multiple Criteria Problem. Management Science, 22 (6), 652 - 663.
[49] Zoutendijk, G. (1960). Methods of feasible directions: a study in linear and
nonlinear programming, University of California : Elsevier Pub. Co.
47