See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/276541887

Application of AHP/EVAMIX Method for Decision Making in the Industrial

Environment

Article in American Journal of Operations Research · January 2013

DOI: 10.4236/ajor.2013.36053

CITATIONS READS

26 1,735

2 authors:

Veera Darji Ravipudi Venkata Rao

C. U. Shah University Sardar Vallabhbhai National Institute of Technology
5 PUBLICATIONS 101 CITATIONS 319 PUBLICATIONS 21,774 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Ravipudi Venkata Rao on 09 October 2016.

The user has requested enhancement of the downloaded file.

American Journal of Operations Research, 2013, 3, 542-569
Published Online November 2013 (http://www.scirp.org/journal/ajor)
http://dx.doi.org/10.4236/ajor.2013.36053

Application of AHP/EVAMIX Method for Decision

Making in the Industrial Environment
Veera P. Darji, Ravipudi V. Rao
Department of Mechanical Engineering, S.V. National Institute of Technology, Surat, India
Email: veera.jani@rediffmail.com, ravipudirao@gmail.com

Received October 1, 2013; revised November 1, 2013; accepted November 8, 2013

Copyright © 2013 Veera P. Darji, Ravipudi V. Rao. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

ABSTRACT
Selection of best alternative among multiple alternatives is a tough task for decision makers in many industrial situa-
tions. This paper explores the applicability and capability of an outranking method known as Evaluation of Mixed Data
(EVAMIX) method combined with Analytical Hierarchy Process (AHP) for selection of right alternative. The novelty
of the proposed methodology is its capability of dealing with both ordinal and cardinal information. The integrated ap-
proach is a significant tool of the decision making process in industrial environments. Five examples are illustrated to
show the effectiveness of method.

Keywords: Industrial Environment; Decision Making; AHP; EVAMIX

1. Introduction the rapid strides in the development of new products and

the emergence of an open economy leading to global
In a competitive industrial environment customers as
competition. Manufacturing industries are compelled to
well as manufacturers are becoming more conscious and
inclined to demand a particular number of customised move away from traditional set-ups to more responsive
products at a particular speed. Therefore it becomes im- and dynamic ones. There is a need for simple, systematic,
portant for the manufacturers to keep pace with dynamic and logical methods or mathematical tools to guide deci-
conditions and rapid changes, be innovative, and adapt to sion makers in considering a number of selection attrib-
new systems, techniques and methodologies. Because of utes and their interrelations. The objective of any selec-
all these factors, the industries and manufacturers have tion procedure is to identify appropriate selection attrib-
begun to make radical changes in their system and struc- utes, and obtain the most appropriate combination of
tures through cost reduction, by achieving higher eco- attributes in conjunction with the real requirements. Thus,
nomic benefit plus environmental benefit to maintain effort needs to be extended to identify those attributes
their position in the global market. The measurement of that influence an alternative selection for a given prob-
the efficiency level of a manufacturing system is a very lem, using simple and logical methods to eliminate un-
critical challenge. Material selection, site selection, Flex- suitable alternatives and to select the most appropriate
ible Manufacturing System selection, Robot selection, alternatives to strengthen the existing selection proce-
computer-integrated manufacturing system selection, dures. The selection of right alternative makes a signifi-
AGV selection, facility layout selection, robot selection, cant change in the productivity and profitability of the
process selection, machine tool selection, cutting fluid manufacturing industries [1].
selection etc. are some of the major issues of industrial In order to evaluate the overall effectiveness of the
environment. candidate alternatives and to select the best option, the
The advancements in the manufacturing technologies multicriteria decision making method requires decom-
have brought about a metamorphism in the world Indus- posing the problem into step such as: defining a set of
trial scene. The advancements include CNC, CAD/CAM, attributes which mostly influence the alternative, prepar-
FMS, robotics, rapid prototyping, environmentally sus- ing a decision matrix, weighing the criteria based on past
tainable technologies, etc., which have become an inte- experience or using an appropriate method, evaluating
gral part of manufacturing. What are parallel to this are the alternatives and ranking them from best to worst.

Open Access AJOR

 V. P. DARJI, R.V. RAO 543

Thus, it is required to extend the decision making method The weighted sum of the dominance scores is then as-
in simple, effective and logical way. This paper presents signed to each alternative.
a very useful outranking method known as Evaluation of Step I: First a set of objective is identified. Then, vari-
Mixed Data (EVAMIX) combined with Analytical Hier- ous attributes and alternatives are short listed for the
archy Process (AHP). given application. Using this information construct a data
The literature shows some applications in the field of matrix of  m  n  size. Where n is number of alterna-
multi criteria analysis in material selection by Chatterjee tives and m is the number of relative attributes chosen
et al. [2], environmental planning by Voogd [3,4], state- for selection problem. Next step is to distinguish the or-
of-art survey for MCDA proposed by Martel and Mata- dinal and cardinal criteria out of decision matrix. Attrib-
razoo [5], multi criteria analysis in physical planning dis- utes are given the linguistic preference, can be converted
cussed in Nijkamp et al. [6], comparison of various al- into its corresponding crisp number as suggested by
ternatives in water resource management carried by Ha- Chen and Hwang [11].
jkowicz and Higgins [7], spatial ranking of hydrological Step II: Normalising the data set is done in the range
vulnerability proposed by Chung and Lee [8], analysis of of 0 - 1 using linear normalization procedure. The bene-
investments in construction by Ustinovichius et al. [9], ficial and non-beneficial attributes are weighted by dif-
and multi criteria analysis used for small-scale forestry ferent equations. For beneficial attributes, normalize the
by Jeffreys [10]. It can be made very clear from the decision matrix using the following equation:
above listed references that, there exist few applications  For beneficial attributes normalize the decision matrix
in science and ecological, financial and non-financial using following equation:
units, and rare applications in the related fields of Indus-
trial environment. rij   xij  min  xij    max  xij   min  xij  
(1)
The Evaluation of Mixed Data (EVAMIX) method  i  1, 2, , m : j  1, 2, , n 
discussed in the present paper is a straightforward and
unequivocal way to tackle the mixed (qualitative and  For non-beneficial attributes the above equation can
quantitative) data for scoring and ranking the available be rewritten as:
data set combined with analytical hierarchy process. The
rij   max  xij    xij    max  xij   min  xij  
next section presents combined EVAMIX methodology (2)
for decision making in the industrial environment.  i  1, 2, , m : j  1, 2, , n 
2. Integrated EVAMIX Approach According to (1) and (2) in the normalised decision
matrix maximum value will always 1 and minimum
The purpose of integrating AHP is done in the present value equal to 0.
paper to provide a vector of weights expressing the rela- Step III: Calculate the evaluative differences of i th
tive importance of alternatives. AHP method helps in alternative on each ordinal and cardinal criterion with
structuring the hierarchy of attributes and alternatives for respect to other alternatives. This step involves the cal-
evaluation. It provides assessment of decision makers’ culation of differences in criteria values between differ-
evaluation by pair-wise comparisons. The priorities for ent alternatives pair-wise. Pair-wise is done based on
attributes and alternatives are calculated by maximum Analytic Hierarchy Process (AHP) Saaty [12-15] and Xu
eigenvector. Last not but the least synthesis of priorities [16]. It provides a way of breaking down the general data
of the alternatives by criteria is carried out into compos- into a hierarch of sub-data, which are easier to evaluate.
ite measures to arrive at a set of ratings for the alterna- These comparisons may be taken from actual measure-
tives. ments or from a fundamental scale which reflects the
Evaluation of Mixed Data (EVAMIX) method was ini- relative strength of preferences introduced by Fechner
tially established by Voogd [3,4], and later advocated by [17] and further advocated by Turstone [18].
Martel and Matarazzo [5]. The novelty of EVAMIX me- In the pair-wise comparison method, attributes and al-
thod is that it deals with mixed (qualitative and quan- ternatives are presented in pairs. It is necessary to evalu-
titative) data. ate individual alternatives. An attribute compared with it
From a procedural point of view, EVAMIX method is always assigned the value 1, so the main diagonal en-
consists of the seven steps discussed in next section. It tries of the pair-wise comparison matrix are all 1. The
commences by identifying unique pairs (criterion-to- numbers 3, 5, 7, and 9 correspond to the verbal judge-
criterion) of alternatives. The degree of pair-wise domi- ments “moderate importance”, “strong importance”, “very
nance for each pair of alternatives is calculated, as the strong importance”, and “absolute importance” (with 2, 4,
difference in score received by the higher performing 6, and 8 for compromise between these values). The
alternative compared to the poorer performing alternative. judgments are given using fundamental scale of AHP.

Open Access AJOR

544 V. P. DARJI, R.V. RAO

 Let A   aij  for all, i, j  1, 2, , n ( ai Vs a j ) direction, i.e., a “higher” score should imply a “large”
denote a square pair-wise comparison matrix. Each preference. It should be noted that the scores  ii  of the
entry in the matrix A is positive  aij  0  and re- quantitative criteria also have to represent “the higher,
the better”.
ciprocal  aij  1 a ji , i, j  1, 2, , n  . Using geo-
Step V: Since  ii  and  ii  will have different meas-
metric mean method; weights are calculated by fol-
urement units, a standardization into the same unit is
lowing steps.
necessary. The standardized dominance scores can be
 Find the relative normalized weight  wi  of each
written as:
attributes by geometric means of rows in matrix
 ii  h  ii   and dii   h   ii 
A   aij  and represent by A1 . (3)
where h represents a standardization function. The
 Calculate matrices A2 and A3 . where, standardized dominance scores can be obtained using
A2  A  A1 and A3  A2 A1 , (4) three different approaches, i.e., (a) subtractive summa-
T tion technique, (b) subtracted shifted interval technique,
where, A1   w1 , w2 , , w j  and (c) additive interval technique. The standardized
 Determine the maximum Eigen value max that is ordinal score  ii   and cardinal dominance score
the average of matrix A3 .
 Calculate the consistency index
 dii  for the alternative pair,  i, i  using additive in-
terval technique is calculated by following equations:
CI   max  m   m  1 . Standardized ordinal dominance score
The smaller the value of CI, the smaller is the devia-   ii     
tion form the consistency.
 ii     
(7)
   


 Obtain the random index (RI) for the number of at-

tributes used in decision making by [12-16].  
where     is the highest (lowest) ordinal domi-
 Calculate the consistency ratio CR = CI/RI. Usually, a nance score for the alternative pair,  i, i   .
CR of 0.1 or less is considered as acceptable, and it Standardized cardinal dominance score
reflects an informed judgment attributable to the   ii     
knowledge of the analyst regarding the problem under  dii     
(8)
   

study.
Step IV: Compute the dominance scores of each al-  
where     is the highest (lowest) cardinal domi-
ternative pair,  i, i   for all the ordinal and cardinal cri- nance score for the alternative pair,  i, i   .
teria using the following equations: Step VI: Let us assume that weights w j have quan-
1C titative properties. The overall dominance measure Dii 
 c
 
 ii     W j sgn  rij  ri j   (5) for each pair of alternatives  i, i   is:
 jO  Dii  wO ii   wC dii  (9)
where where wO is the sum of the weights for the ordinal cri-
1 if rij  rij

 
teria wO   jO w j and wC is the sum of the
sgn  rij  rij   0 if rij  rij

weights for the cardinal criteria w
C 
  jC w j . This
1 if rij  rij (6)
overall dominance score reflects the degree to which
1C
  alternative ai dominates alternative ai  for the given
 
 ii    W j sgn  rij  rij  
c
set of attribute and the weights. In general the measure
 jC  Dii  may be considered as function K of the constitu-
The symbol c denotes an arbitrary scaling parameter, ent appraisal scores: Dii   k  si , si   . This expression
for which any arbitrary positive odd number, like 1, 3, represents a well-known Pairwise comparison problem.
5… may be chosen, O and C are the sets of ordinal and Here for each pair Dii   Di i  1 .
cardinal criteria respectively, and  ii  and  ii are the Step VII: Calculate the appraisal score. The appraisal
dominance scores for alternative pair,  i, i   with re- score for i th alternative  Si  is computed which gives
spect to ordinal and cardinal criteria respectively. In or- the final preference of the candidate alternatives. Higher
der to be consistence, the same value of scaling parame- the appraisal score better is the performance of the alter-
ter c is used in (5) and (6). It is assumed that the value of natives. The best alternative is one which has the highest
c for qualitative evaluation  ii  is taken equal to 1. value of the appraisal score.
Evidently, all standardized scores should have the same Appraisal score

Open Access AJOR

 V. P. DARJI, R.V. RAO 545

 Di i 
1 Attempts are made by Chatterjee et al. [25] to solve the
 Si      (10) material selection problems using TOPSIS and ELEC-
i  Dii   TRE II methods. Shanian and Savadogo [26], applied a
The methodology proposed in this paper enables the non-compensatory compromised approach (ELECTRE
decision maker to rank the alternatives from best to the IV) for material selection. Weighting factor approach is
worst. The method is able to deal with any number at- developed for material selection with combination of
tributes and alternatives by effective mathematical steps. non-linear linearization and modified digital logic me-
In order to demonstrate and validate the applications of thod by [27]. The life cycle analysis (LCA) based meth-
the combined EVAMIX method for industrial environ- odology is proposed by Riberio et al. [28] is lagged by
ment, following five examples are illustrated. deep need of engineering knowledge of all the material
and/or technological alternatives for the analysis. A sig-
3. Examples nificant effort is needed to identify the “best alternative”.
The methodology used by Pecas et al. in [29] is a mix
3.1. Example 1: Automobile Front between a product development process and a classic
Fender Material Selection material selection process, in a way that combines typical
Proper selection of material is a critical issue for the information generated in both processes. In doing so, the
success of manufacturing aids and competitiveness of the selection of materials becomes fully blended in the proc-
industrial environment. Choice of most appropriate ma- ess evolved with the development of the product, from
terial for a particular engineering application is consid- the concept generation to the industrialization.
ered as time consuming and expensive process in past The proposed EVAMIX method is quite effective
years. Material selection is a step in the process of de- compared with other approaches in terms of mathemati-
signing any engineering component. In the context of cal calculations, less time consuming and on hand ap-
product design, the main goal of material selection is to proach.
minimize cost while meeting product performance goals. Step I: A technically good set of candidate materials is
Systematic selection of the best material for a given ap- identified and their important properties are identified.
plication begins with properties and costs of candidate As per Riberio [28] and Pecas [29], ordinal (qualitative)
materials. There exists an ample amount of work being and cardinal (quantitative) attributes are distinguished. A
already done on the problem of material selection. pre-set of candidate materials are listed in Table 1.
EVAMIX is used for material selection of material using This set is composed of the current material used (St 1)
weights employed by Chatterjee et al. [2]. Edwards [19] for the fender construction, two high strength steels, and
dealt with the decision making for material selection is three Aluminium alloys for which all the engineering
some important engineering applications. A method is properties are known in advance.
developed by Chen [20] to solve the tool steel material Using the set of material from above Table 1, the de-
selection problem under fuzzy environment. Jee and cision matrix is formed.
Kang [21] utilized two different decision making theories,
 220 207 7.85 7.85 32 1.50 124 
which are very popular methods. The weighted perform- 350 207 7.85 7.85 16 1.60
ance index values for material selection of rocket motors  197 
are done by Rajan and Narasimhan [22]. Ermolaeva et al.  700 207 7.85 7.85 5 1.60 394 
D 
[23] applied structural optimization method for the opti- 170 69 2.71 2.71 24 0.70 88 
mal choice of foams as a core material for sandwiches 190 71 2.75 2.75 24 0.70 107 
with aluminium alloy faces. Material selection carried  
out by using logical procedure by Rao and Devim [24]. 155 70 2.74 2.74 30 0.70 98 

Table 1. Properties of the candidate material of an automobile front fender [29].

Yield strength Ductility (strain at Young’s Coefficient Hardness Strain hardening Density
Material
(MPa) rupture) (kNm/kg) modulus (GPa) of anisotropy (HV) exponent (Nm/kg) (ton/m3)
HX220YD (St 1) 220 32 207 1.5 124 0.17 7.85
DOCOL600P (St 2) 350 16 207 1.6 197 0.15 7.85
DOCOL1000P (St 3) 700 5 207 1.6 394 0.14 7.85
AL6010 (Al 1) 170 24 69 0.7 88 0.22 2.71
AL2036 (Al 2) 190 24 71 0.7 107 0.23 2.75
GZ45/30 (Al 3) 155 30 70 0.7 98 0.29 2.74

Open Access AJOR

546 V. P. DARJI, R.V. RAO

A quick analysis of the results obtained by Pecas et al. 3),···, (1, 6) etc. for each  i, i   pair are computed for
[29], resulted that the most influent material properties ordinal (C, sets of Cardinal attributes) and cardinal (O,
are the yield strength, young’s modulus and strain hard- sets of Ordinal attributes) attributes using (5) and (6)
ening exponent. Hence, in the present paper these three respectively, and given in Table 3. While calculating the
attributes are chosen to be beneficial attributes. For the dominance scores, the value of c is taken equal to 1.
material to be better in performance if the yield strength Step V: Now the standardized dominance is calculated
is required of higher values than ductility will definitely based on the additive interval technique for all the pairs
require being of higher values. The requirement is to of alternative fender materials using (7) and (8) for the
select a material with high yield strength, which in other ordinal and cardinal criteria respectively and given in
turn says that hardness value must be lower. For the bet- Table 4. It is very clear from the values given in Table 4,
ter formability of material into required shape in auto- that the summation of standardized dominance scores of
mobile fender application requires lower coefficient of each  i, i   and  i , i  is equal to 1.
anisotropy. Thus, it is now clear that hardness and den- Step VI: The overall dominance score for each alter-
sity are non-beneficial attributes. native fender materials pair  i, i   , is calculated using (9)
The importance to each attribute is assigned by Pecas which shows the degree by which fender material i
et al. [29]. They also carried pair-wise comparison in or- dominates material i  . These overall dominance score
der to obtain the engineering property weight (EPWj). To for all the pairs of alternative fender materials are shown
validate the applicability of the method, same weights are in Table 5.
taken in the present calculations. The importance or en- Step VII: Finally the appraisal score for i th alterna-
gineering property weights assigned by Pecas et al. are: tive fender material pair  Si  is computed using (10).
Yield strength—0.231, Young’s modulus—0.23, Strain Higher the appraisal score better is the performance of
hardening exponent—0.165, Density—0.14, Ductility the alternatives fender material. The appraisal score of
(strain at rupture)—0.10, Coefficient of anisotropy— each alternative pair in descending order gives the selec-
0.075 and Hardness—0.06. Out of seven attributes only tion of material from best to the worst performing cate-
yield strength is chosen as ordinal (qualitative) attribute. gory. The scores and ranking obtained for the best alter-
Step II: Normalisation is carried using (1) and (2) for native fender material is HX220YD (St 1).
beneficial and beneficial attributes respectively. Yield In the present perception of the material ranking, it is
strength, ductility, coefficient of anisotropy, young’s mo- clear that the technically the current material i.e.,
dulus, and strain hardening exponent are considered as HX220YD (St 1) fulfills the expected quality for the
beneficial attributes. Whereas hardness and density are manufacturing of front fender of automobile. The rank-
assumed to be non-beneficial attributes. Table 2 shows ing is carried out to eliminate materials with lower tech-
the normalized values of listed engineering properties in nical potential, meaning that any potential advantages of
non-dimensional pattern. Here the value of normalized the other materials might permit better economic and
matrix rang from 0 - 1. environmental performance. The materials Al 1 and Al 3
Step III: After normalizing the decision matrix, the will be excluded avoiding additional effort on subse-
evaluation differences of the i th fender material alterna- quent steps of any further methodology. The other steels
tive for each ordinal and cardinal criterion with respect to (St 2 & St 3) achieve the highest score after St 1. The St
all other material alternatives are calculated. The purpose 1 is the alternative with lower manufacturing costs de-
of method is to show the effectiveness of the scoring and rived from the lower material and fixed costs. St 3 for
ranking procedure of EVAMIX. both thickness (estimated and minimum commercially
Step IV: The dominance scores for each (1, 2), (1, available) having higher manufacturing costs, closely

Table 2. Normalized data of candidate materials.

YS Ductility YM COA Hardness SHE Density

St 1 0.1193 1 1 0.8889 0.8824 0.2 0

St 2 0.3578 0.3251 1 1 0.6438 0.0667 0

St 3 1 0 1 1 0 0 0

Al 1 0.0275 0.7037 0 0 1 0.5333 1

Al 2 0.0642 0.7037 0.0145 0 0.9379 0.6 0.9922

Al 3 0 0.9259 0.0072 0 0.9673 1 0.9922

Open Access AJOR

 V. P. DARJI, R.V. RAO 547

Table 3. Dominance score of each alternative pair.

Pair  ii  ii Pair  ii  ii Pair  ii  ii

(1,2) −0.231 0.6812 (3,1) 0.231 0.0888 (5,1) −0.231 0.3376

(1,3) −0.231 0.6812 (3,2) 0.231 0 (5,2) −0.231 0.5745

(1,4) 0.231 0.4324 (3,4) 0.231 0.1955 (5,3) −0.231 0.5745

(1,5) 0.231 0.4324 (3,5) 0.231 0.1955 (5,4) 0.231 0.616

(1,6) 0.231 0.4324 (3,6) 0.231 0.1955 (5,6) 0.231 0.2725

(2,1) 0.231 0.0888 (4,1) −0.231 0.3376 (6,1) −0.231 0.3376

(2,3) −0.231 0.7700 (4,2) −0.231 0.5745 (6,2) −0.231 0.5745

(2,4) 0.231 0.1955 (4,3) −0.231 0.5745 (6,3) −0.231 0.5745

(2,5) 0.231 0.1955 (4,5) −0.231 0.1540 (6,4) −0.231 0.7345

(2,6) 0.231 0.1955 (4,6) 0.231 0.0355 (6,5) −0.231 0.4975

Table 4. Standardized dominance score of each alternative pair.

Pair  ii d ii Pair  ii d ii Pair  ii d ii

(1,2) 0 0.8846 (3,1) 1 0.1154 (5,1) 0 0.4385

(1,3) 0 0.8846 (3,2) 1 0 (5,2) 0 0.7462

(1,4) 1 0.5615 (3,4) 1 0.2538 (5,3) 0 0.7462

(1,5) 1 0.5615 (3,5) 1 0.2538 (5,4) 1 0.8000

(1,6) 1 0.5615 (3,6) 1 0.2538 (5,6) 1 0.3538

(2,1) 1 0.1154 (4,1) 0 0.4385 (6,1) 0 0.4385

(2,3) 0 1 (4,2) 0 0.7462 (6,2) 0 0.7462

(2,4) 1 0.2538 (4,3) 0 0.7462 (6,3) 0 0.7462

(2,5) 1 0.2538 (4,5) 0 0.2000 (6,4) 0 0.9538

(2,6) 1 0.2538 (4,6) 1 0.0462 (6,5) 0 0.6462

Table 5. Overall dominance scores of each alternative pair.

Pair Dii Pair Dii Pair Dii

(1,2) 0.6812 (3,1) 0.3198 (5,1) 0.3376

(1,3) 0.6812 (3,2) 0.2310 (5,2) 0.5745

(1,4) 0.6634 (3,4) 0.4265 (5,3) 0.5745

(1,5) 0.6634 (3,5) 0.4265 (5,4) 0.8470

(1,6) 0.6634 (3,6) 0.4265 (5,6) 0.5035

(2,1) 0.3198 (4,1) 0.3376 (6,1) 0.3376

(2,3) 0.7700 (4,2) 0.5745 (6,2) 0.5745

(2,4) 0.4265 (4,3) 0.5745 (6,3) 0.5745

(2,5) 0.4265 (4,5) 0.1540 (6,4) 0.7345

(2,6) 0.4265 (4,6) 0.2665 (6,5) 0.4975

Open Access AJOR

548 V. P. DARJI, R.V. RAO

same to that of the material Al 2. If cost is to be con- A thorough literature survey reveals that there have been
sidered for the selection of material than St 1 is doubled a number of studies associated with variety of composite
than the Al 2. products manufactured by one of the liquid-molding
From all above analysis and from the real application techniques like resin transfer molding, vacuum-assisted
of fender it can be made clear that fender material selec- resin transfer molding by various researchers in [32-34].
tion is exclusively based on the manufacturing parame- For all the processes proper selection of subsystems is
ters. For the steel alternatives the use phase importance is very important to get the final desired composite struc-
higher than the production phase. tures. Other few papers also give detailed information
The currently used steel St 1, has the required level of about various composite product and its interactions be-
performance on cost for the material and fender pro- tween them [35-38].
duction, while Al 2 has the value for fender use and dis- Composite product selection is carried out by [39] us-
mantling. The detailed ranking is tabulated in Table 6. ing a technique for order preference by similarity to ideal
Pecas et al. [29] have given further analysis about the solution (TOPSIS) and graphical methods (line graph and
selection of best material as per the requirement. spider diagram) by considering reasonable attributes in
totality. The methodology is a computer assisted MADM
3.2. Example 2: Selection of Composite method. Ranking and scoring of such composite product
Reinforcement system selection should validate by using some other
MADM methods. Hence, combined EVAMIX method is
In order to satisfy the needs of the global economics used very effectively in the present paper.
market, optimized production strategies, expected quality To illustrative the ease and logical approach of the
within shorter lead-time and lower life-cycle cost etc. proposed methodology an example is discussed for the
many industries related to composites, are in need for manufacturing of composite screw rotors for air com-
special mathematical tools and methods for improving pressors by resin transfer molding (RTM) process. Fol-
their product performance as well as to improve competi- lowing steps will explain the scoring and ranking in detail.
tiveness of the final products. A new methodology is Step I: The peer study resulted into manageable num-
proposed to composites industry, designer and manu- ber of candidate reinforcements and their attributes
facturer in deciding the subsystems as well as composite which are listed in Table 7.
product system selection. It helps in achieving expected The possible candidate reinforcements are chopped
quality and properties of final composite product with the glass fibers, chopped carbon fibers and chopped aramid
help of integrated systems approach and also a deci- fibers. The minimum required attributes that is responsi-
sion-making approach based on the attributes identifica- ble for selection of these types of fibers are tensile
tion. Polymer matrix composite products are generally strength, tensile modulus, volume fraction (%) and elon-
manufactured using any one of the liquid molding tech- gation (%). Decision matrix (D) is formed and ordinal
niques. In any liquid molding processes a preform of (qualitative) and cardinal (quantitative) criteria are dis-
reinforced fibers is placed in a closed or open mold and tinguished. Volume fraction can be taken as most influ-
then a liquid polymer resin is injected into the mold to encing ordinal attribute for the selection problem. Using
impregnate the preform. When the mold is full, the the data from Table 7, decision matrix is formulated as:
polymer is cured by a cross-linking reaction to become a
rigid solid. Liquid molding processes offer a way to pro- 1600 35 50 4 
duce high-performance composite products using a rapid D   3528 98 40 1.5 
process with low labor requirement Rudd et al. [30],
Potter [31] gave the material for Resin Transfer molding.  2900 18 30 4.4 

Table 6. Appraisal score of alternative material.

Fender Mat. Si additive interval technique Rank Rank using MSE

St 1 0.4055 1 1

St 2 0.1545 4 2

St 3 0.1052 5 3

Al 1 0.0854 6 6

Al 2 0.2165 2 4

Al 3 0.2073 3 5

Open Access AJOR

 V. P. DARJI, R.V. RAO 549

Durai et al. [39] carried out the analysis through to are strong enough when pulled in the direction of fiber
obtain the weight vector using maximum eigenvlaue. For orientation. Furthermore, in composites, higher loads can
sack of validating the effectiveness of the proposed be safely sustained by the high elongating fibers. Per-
method we have selected the same weights for calcula- centage elongation also plays a role in reducing weight of
tions. The weights assigned are: Tensile strength—0.2662, the product. So, compared with other attributes we can
Tensile modulus—0.2517, Volume fraction—0.2842 and say that elongation is the influencing attribute and cate-
Elongation—0.1979. gorized as ordinal attribute. Table 9 summarized the
Step II: The increase in volumetric fraction of fiber is dominance scores; the value of c is taken equal to 1. The
significant to improve the specific strength and modulus dominance scores for candidate reinforcement pair
and hereby reducing the weight of the product. The resin  i, i  are computed for ordinal and cardinal attributes
system with high performance provides optimum com- using (5) and (6) respectively.
posite strength and sufficient shear modulus to optimize Step V: The standardized dominance score is calcu-
buckling under compression loads. Normalisation is car- lated based on the additive interval technique for all the
ried using (1), for beneficial attributes. Tensile strength, pairs of alternative reinforcement candidate alternatives
tensile modulus, volume fraction and elongation are con- using (7) and (8) for the ordinal and cardinal criteria re-
sidered as beneficial attributes. Three stage analysis car- spectively and given in Table 10. It is very clear from
ried by [39] gave that the attributes selected are all bene- the values given in Table 10, that the summation of stan-
ficial for the final outcoming product. Thus once the dardized dominance scores of each  i, i   and  i , i  is
identification is done properly the selection of best alter- equal to 1.
native will become easy. Table 8 shows the normalized Step VI: The overall dominance score for each alter-
values of listed candidate reinforcements for manufac- native of candidate reinforcement pair  i, i   , is calcu-
turing of engineering properties in non-dimensional pat- lated using (9) which shows the degree by which rein-
tern. Here the value of normalized matrix rang from 0 - forcement i dominates reinforcement i  . These overall
1. dominance score for all the pairs of alternative is shown
Step III: After normalizing the decision matrix, the in Table 11.
evaluation differences of the i th candidate reinforce- Step VII: Finally the appraisal score for i th alterna-
ment for each ordinal and cardinal criterion with respect tive pair ( S i ) is computed using (10). Higher the ap-
to all other reinforcement alternatives are calculated. The praisal score better is the performance of the alternatives
method shows that ranking is still simple using EVAMIX reinforcement candidate. The appraisal score of each
method. Calculations of attributes weights are done using alternative pair in descending order gives the selection of
maximum eigenvlaue analysis in [36]. For the sack of reinforcement from best to the worst performing category.
comparison weights assigned by the previous researchers The scores and ranking is given in Table 12.
are taken to use in the proposed methodology. The systematic mathematical steps have suggested that
Step IV: Resin selected for manufacturing of screw the Chopped carbon fiber is the best alternative for and
rotor of air compressor must be good to absorb energy Chopped aramid fiber is the least preferred alternative for
and reduce stress concentration. This in turn gives better the manufacturing of screw rotor of air compressor. Ta-
fracture toughness and ductility to maximize damage ble 13 shows the comparison of ranking and scoring ob-
tolerance and long-term durability. In composites fibers tained through various MADM methods. Any number of

Table 7. Short listed candidates of reinforcement and their properties [39].

Candidate reinforcements Tensile strength (MPa) Tensile modulus (GPa) Volume fraction (%) Elongation (%)

Chopped glass fiber 1600 35 50 4

Chopped carbon fiber 3528 98 40 1.5

Chopped aramid fiber 2900 18 30 4.4

Table 8. Normalized data for candidates of reinforcement.

Candidate reinforcements Tensile strength (MPa) Tensile modulus (GPa) Volume fraction (%) Elongation (%)

Chopped glass fiber 0 0.2125 1 0.8621

Chopped carbon fiber 1 1 0.5 0

Chopped aramid fiber 0.6743 0 0 1

Open Access AJOR

550 V. P. DARJI, R.V. RAO

Table 9. Dominance score of each alternative pair. posite product system in an industrial environment. As
shown in Table 13 the ranking obtained by various
Composite Reinforcements Pair  ii  ii
methods are same. Hence, the validity of the EVAMIX
(1,2) 0.1979 −0.2337 method is justified by comparing the performance scores
(1,3) −0.1979 0.2697
of all the alternative reinforcements.

(2,1) −0.1979 0.2337 3.3. Example 3: Thermal Power Plant Site

(2,3) −0.1979 0.8021 Selection
(3,1) 0.1979 −0.2697 One of the basic inputs of vital importance which has an
(3,2) 0.1979 −0.8021
impact and influenced industrial development and quality
is energy. The blooming power generation industry is a
sign of growing gross nation. Thermal power generation
Table 10. Standard dominance score of each alternative
pair.
can be considered as one of the main source of electricity
generation. The selection of site for thermal power plant
Composite Reinforcement Pair  ii d ii compared with other is difficult as it involves number of
(1,2) 1 0.3543
factors to be considered for its economic justification;
like availability of materials, disposal facilities, space
(1,3) 0 0.6681 requirements, type of land of site, transportation facility
(2,1) 0 0.6457 etc.
On the industrial font, emphasis must be placed on the
(2,3) 0 1
increased with constant effort to reduce energy consump-
(3,1) 1 0.3319 tion. Fundamental changes in the process, production and
(3,2) 1 0 services can affect considerable energy saving without
affecting the overall economy. The power plant itself
must be useful economically, socially, technically and
Table 11. Overall dominance scores of each alternative pair.
environmental friendly to the society. Ting and Bing [40]
Composite Reinforcement pair Dii introduced TOPSIS based evaluation of thermal power
plant sites. The sitting selection for thermal power plant
(1,2) 0.4821
is supported by few factors;
(1,3) 0.5359 1) Technical conditions—topography and geology
(2,1) 0.5179 conditions, transportation conditions, load conditions and
the quality of resources and environmental conditions;
(2,3) 0.8021 2) Economical conditions—total investment, annual
(3,1) 0.4641 operating cost and maximum voltage
3) Social conditions—The reasonable site of the ther-
(3,2) 0.1979
mal power plant selection requires the view of local gov-
ernment and the living conditions of works.
Table 12. Appraisal score of composite reinforcement. If the site selection is done wrong, it will increase the
Composite Si additive interval technique Rank
cost of electricity, and more critical, it will reduce com-
reinforcements petitiveness in the market to the plant. The potential in-
Chopped glass fiber 0.5154 2 fluence of the irrationality site of the plant to the society
Chopped carbon fiber 0.8492 1 and the environment is imponderable. Therefore, the
comprehensive evaluation using TOPSIS is carried out
Chopped aramid fiber 0.1920 3 for optimizing site selection of thermal power plant site.
Few other researchers, such as Yan [41] proposed com-
materials having any number of conflicting attributes can prehensive evaluation theory, Jing and Dong [42] used
be ranked and scored using EVAMIX method. triangular fuzzy numbers for the sitting of thermal power
The final decision of selecting a set of suitable com- plant, Dong [43] gave application of entropy based fuzzy
posite product system after ranking is basis on various comprehensive evaluation in sitting of thermal power
factors like environmental aspects, cost, technological plant, Dong [44] used grey level analysis in decision
aspects etc. The solutions proposed by Durai et al. pro- making of thermal power plant site, Hai [45] proposed
posed the TOPSIS and graphical representation in the manual application of economic evaluation and Lu et al.
selection of manufacturing process for developing com- [46] used AHP method in the optimal power plant site

Open Access AJOR

 V. P. DARJI, R.V. RAO 551

selection. From the literature it is very clear that the solution. Six alternative sites and seven attributes are
method used in the field of evaluation of sitting for the selected for the present example. The example takes six
thermal power plant is not analysed with strong mathe- alternatives and seven attributes. The raw data about al-
matical steps. The technological conditions for alterna- ternatives are shown in Table 14.
tives are given subjective preference. Ting and Bing [40] Using these data, decision matrix is formed. The
have taken some appropriate objective value. If it is re- weights are taken same for the comparison purpose. The
quired using fuzzy scale we can convert the subjective topography and geology condition is given 0 weights,
value into appropriate objective value. and this tends to remove this attribute from further ana-
There is a need of some better multi criteria decision lysis.
making method, to elaborate the behaviour of alterna- Step II: By analysis of the normalization carried in
tives with respect to the nature of attributes selected. So, Ting and Bing [40], it seems to be clear that only re-
in the present paper EVAMIX is used for the ranking of sources and pre-environment conditions is taken as non
available sites having combined qualitative and quantita- beneficial attribute.
tive attribute values. Next section represents the steps to The technology condition is an influencing criterion
rank the sites for the thermal power plant. for the sitting of power plant. Normalisation is carried
Step I: The feasibility stage of the thermal power plant using (1) and (2) for beneficial and non-beneficial attrib-
for sustainable development evaluation must consider the utes. Table 15 shows the normalized values of listed
combination of the social, technical and economical be- sites for thermal power plant. Here always non-dimen-
nefits. The proposed method enables the combined (or- sional normalization ranges from 0 - 1.
dinal & cardinal) attributes to be analysed for optimal Step III: After normalizing the decision matrix, the

Table 13. Comparison and ranking of candidates of reinforcement using various methods.
Composite Score using Score using Score using Score using
Rank Rank Rank Rank
reinforcements EVAMIX method TOPSIS method line graph spider diagram
Chopped glass fiber 0.5154 2 0.3967 2 0.6522 2 0.4143 2

Chopped carbon fiber 0.8492 1 0.6844 1 0.8553 1 0.6737 1

Chopped aramid fiber 0.1920 3 0.3604 3 0.5218 3 0.3498 3

Table 14. Alternate sites and their evaluation factors [40].

Topography Resources and Area/ten Review of
Transportation Technology Total investment/
Site and geology pro-environment thousand local
conditions conditions ten million
conditions conditions cubic meters Government
1 5 3.0 2 43.5 Bad (2) 6 6

2 5 5.7 6 429.2 Very good (6) 2 5

3 5 4.75 4 392 Very good (6) 2 4

4 5 1.2 6 200 Good (5) 2 3

5 5 3.9 4 580 Very good (6) 2 5

6 5 4.3 2 322 Very good (6) 2 4

Table 15. Normalized data for the site selection.

Resources and Topography
Technology Transportation Area/10,000 m3 Total Review of
Site Pre-Environment and geology
conditions conditions Investment local govt.
Conditions conditions
1 0 0.6 0 0 1 1 6

2 1 0 1 0.7189 1 0 0.6667

3 1 0.2111 0.5 0.6496 1 0 0.3333

4 0.75 1 1 0.2917 1 0 0

5 1 0.4 0.5 1 1 0 0.6667

6 1 0.3111 0 0.5191 1 0 0.3333

Open Access AJOR

552 V. P. DARJI, R.V. RAO

evaluation differences of the i th site ordinal and cardi- calculated and given in Table 16.
nal criterion with respect to all other site alternatives are Step V: The standardized dominance is calculated
calculated. The mathematical steps of EVAMIX in the based on the additive interval technique for all the pairs
present paper are clearer than the method used by [40]. of alternative site using (7) and (8) for the ordinal and
The method shows that ranking is still simple using cardinal criteria respectively and given in Table 17.
EVAMIX method. Weights are taken from [40] for Step VI: The overall dominance score for each alter-
comparison of results and listed as, Topology and geol- native of site for thermal power plant pair  i, i   , is cal-
ogy conditions—0, resources and pre-environment con- culated using (9) which shows the degree by which rein-
ditions—0.32, transportation conditions—0.41, area— forcement i dominates reinforcement i  . These scores
0.32, technology conditions—0.37, total investment— obtained using (9) show the degree by which reinforce-
0.37 and review of local government—0.32. The weights ment i dominates reinforcement i  . These overall do-
assigned are taken same as that of the previous research- minance score for all the pairs of alternative is shown in
ers. It is also clear that the effect of topology and geology Table 18.
conditions is assumed to null. Step VII: Finally the appraisal score for i th alterna-
Step IV: Sitting selection for thermal power plant is tive pair  Si  is computed using (10).
done by taking first attribute to be null and technology Higher the appraisal score better is the site for the
conditions as influencing attribute. Using (5) and (6) re- thermal power plant location. The scores and ranking is
spectively, dominance scores for all candidate sites are given in Table 19.

Table 16. Dominance score of each alternative pair.

Pair  ii  ii Pair  ii  ii Pair  ii  ii

(1,2) −0.37 0.28 (3,1) 0.37 −0.28 (5,1) −0.37 −0.28

(1,3) −0.37 0.28 (3,2) 0.37 −0.73 (5,2) −0.37 0.23

(1,4) 0.37 −0.36 (3,4) 0.37 −0.09 (5,3) −0.37 0.96

(1,5) 0.37 0.28 (3,5) 0.37 −0.96 (5,4) 0.37 −0.09

(1,6) 0.37 0.69 (3,6) 0.37 0.41 (5,6) 0.37 1.37

(2,1) 0.37 −0.28 (4,1) −0.37 0.36 (6,1) −0.37 −0.69

(2,3) −0.37 0.73 (4,2) −0.37 −0.32 (6,2) −0.37 −0.73

(2,4) 0.37 0.32 (4,3) −0.37 0.09 (6,3) −0.37 −0.41

(2,5) 0.37 −0.23 (4,5) −0.37 0.09 (6,4) −0.37 −0.09

(2,6) 0.37 0.73 (4,6) 0.37 0.09 (6,5) −0.37 −1.37

Table 17. Standardized dominance score of each alternative pair.

Pair  ii d ii Pair  ii d ii Pair  ii d ii

(1,2) 0 0.6022 (3,1) 1 0.3978 (5,1) 0 0.3978

(1,3) 0 0.6022 (3,2) 1 0.2336 (5,2) 0 0.5839

(1,4) 1 0.3686 (3,4) 1 0.4672 (5,3) 0 0.8504

(1,5) 1 0.6022 (3,5) 1 0.1496 (5,4) 1 0.4671

(1,6) 1 0.7518 (3,6) 1 0.6496 (5,6) 1 1

(2,1) 1 0.3978 (4,1) 0 0.6314 (6,1) 0 0.2482

(2,3) 0 0.7664 (4,2) 0 0.3832 (6,2) 0 0.2336

(2,4) 1 0.6168 (4,3) 0 0.5328 (6,3) 0 0.3504

(2,5) 1 0.4161 (4,5) 0 0.5328 (6,4) 0 0.4672

(2,6) 1 0.7664 (4,6) 1 0.5328 (6,5) 0 0

Open Access AJOR

 V. P. DARJI, R.V. RAO 553

Table 18. Overall dominance scores of each alternative pair.

Pair Dii Pair Dii Pair Dii

(1,2) 1.0478 (3,1) 1.0622 (5,1) 1.062

(1,3) 1.0478 (3,2) 0.5914 (5,2) 1.2011
(1,4) 0.6414 (3,4) 1.1828 (5,3) 1.6646
(1,5) 1.0478 (3,5) 0.4454 (5,4) 1.1823
(1,6) 1.3082 (3,6) 1.3154 (5,6) 1.9250
(2,1) 1.0622 (4,1) 1.4686 (6,1) 0.8018
(2,3) 1.5185 (4,2) 0.6668 (6,2) 0.5914
(2,4) 1.4432 (4,3) 0.9271 (6,3) 0.7946
(2,5) 0.9089 (4,5) 0.9272 (6,4) 1.1828
(2,6) 1.5186 (4,6) 0.9272 (6,5) 0.1850

Table 19. Appraisal score of each alternative site for ther- tainer handling at port container terminals, arrivals of
mal power plant. container ships cannot be precisely defined in time and
Sites Si additive interval technique Rank
they represent a group of approximately known values
mentioned by Danijela et al. [47]. One of the ways to
1 0.1682 3
improve the total productivity of a container terminal (to
2 0.2817 2 reduce ship turnaround times) is to increase the automa-
3 0.1152 5 tion of the cargo handling and transport, and to use the
most efficient container handling scenario. Advantages in
4 0.1555 4
the fields of electronic, automation, information tech-
5 0.3459 1 nologies and sensors are enabling a development of fully
6 0.0586 6 automated container terminals discussed by loannou [48].
In the studied container terminal the quay cranes were
As per the evaluation process followed by Ting and unloading containers from the ship; the containers were
Bing [40], site no. 2 and 5 are equally good based on the further transported to the storage areas by automated
values obtained for relative adjacent degree. The pro- guided vehicles (AGVs).
posed methodology (EVAMIX) cleared the tie between Ports are an important link in the logistics chain so the
selections of site no. 2 and site no. 5. The appraisal level of port efficiency affects to a large extent the coun-
scores for six sites given in Table 19 reveal that site no. try’s competitiveness, since port efficiency results in
2 is the first choice for thermal power plant. Site no. 5 lower tariffs for exports which, in turn, favour the com-
become second choice according as some difference re- petitiveness of country products in international markets.
flects from resource conditions, transportation conditions Therefore, in order to keep a competitive position in
and area available for the sitting. The comparison of dif- those markets, the countries need to work on the factors
ferent sites about the location of power plant by the that affect the efficiency of their ports and draw con-
method of EVAMIX finds the optimal solution and helps tinuous comparisons on the degree of efficiency among
decision maker in selecting the best site. them and with the ports of other regions [49].
From the literature it is found that the DEA and its
various models are only applied for the efficiency eva-
3.4. Example 4: Evaluation of AGV Fleet
luation in the field of port container terminals and similar
Operation at Port Container Terminal
situations. Efficiency measurement of seaport with DEA
Port container terminal provides a service of container is proposed by Barrows [50]. Cullinane and Song [51]
handling, more precisely, unloading containers from used DEA windows analysis to container port production
ships onto inland vehicles (typically trucks and trains), efficiency. Cullinane and Wang [52] also gave the appli-
and vice versa, with the purpose of distributing contain- cation of DEA. Kaisar [53] measured efficiency of US
ers to the end users. Terminals are equipped with corre- port is carried out using DEA. Analysis of AGV systems
sponding reloading machinery and with opened storage for two container yard system is done in Liu and Jula
areas dedicated for shorter or longer storage of loaded or [54]. Evaluation of international container terminals us-
empty containers. In the process of planning the con- ing DEA is proposed by Min and Park in [55]. Pjevcevic

Open Access AJOR

554 V. P. DARJI, R.V. RAO

and Vukadinovic in [56] characterization of the AGV nal criterion with respect to all other scenarios are calcu-
dispatching rules is discussed. lated. The mathematical steps of combined EVAMIX
It is clear from evaluation of port container systems given in the present paper have maximum impact on se-
required some simple but still logical MCDM method to lection procedure compared with any other methods used
be applied. The proposed combined EVAMIX method- by the previous researchers. The method shows that
ology, deals with evaluation efficiencies of handling of ranking is still simple using EVAMIX method. The re-
containers that arrive at port. This analysis evaluated 12 ciprocal value of the number of containers in the queue is
scenarios for containers dispatching from Berth to stor- taken as ordinal attributes which always shows the ef-
age areas. The purpose of evaluation is to rank these 12 fecttiveness of port terminal. As the length of queue
scenarios. Four attributes selected are number of em- should be low enough to show better dispatching through
ployed AGVs, AGVs active rate, reciprocal value of the the port it is non-beneficial attribute for the alternative
number of containers in the queue and number of served port terminals.
containers. Next sections describe the sequential proce- Pair-wise comparison is done by considering the rela-
dure for ranking of scenarios. tive importance of each attribute compared to other.
Step I: In the illustrative example the aim is to show Matrix A is relative importance matrix.
the effectiveness of number of employed AGVs and dis-
patching rule on container handling i.e., number of con- 1 3 2 2
tainers in queue, AGV utilization etc. using AHP and 1 3 1 2 1 
EVAMIX methods. Only reciprocal value of the number A
1 2 1 2 1 1
of containers in the queue is an ordinal attribute. Other  
three are categorised under cardinal attributes. 12 alter- 1 2 1 1 1
native scenarios and 4 are analysed to calculate the Number of AGVs is moderately more important at-
weights using AHP method. Table 20 shows the raw tribute than the AGV active rate. So, a relative impor-
data for scenario selection. tance value of 3 is assigned to number of AGVs over
Step II: To improve the total productivity of a con- AGV active rate and assigned the value 1/3 to AGV ac-
tainer terminal, the first requirement is to reduce the ship tive rate over number of AGVs. In this way a complete
turnaround times. The entire process is depending on the matrix is A formed. As explained in section 2, A1 ,
effectiveness of unloading containers in the queue wait- A2 and A3 matrices are calculated:
ing to be served by AGVs. Normalization is carried out
using (1) and (2) for beneficial and non-beneficial attrib-  0.4316  1.7780   4.1199 
utes respectively. Table 21 shows the normalized values  0.2095   0.8763  
A1    A   A   4.1823
of scenarios under selection problem.  0.1640  2  0.6795  3  4.1442 
Step III: After normalizing the decision matrix, the      
evaluation differences of the i th site ordinal and cardi-  0.1950  0.7842   4.0221

Table 20. Data of scenario of AGV fleet operation [47].

Number of efficiency AGVs active Reciprocal value of the number Number of served
Scenario
employed AGVs rate of containers in the queue containers

Scenario 1 18 0.94 0.0068 1554

Scenario 2 20 0.94 0.0095 1613
Scenario 3 23 0.92 0.0132 1712
Scenario 4 18 0.97 0.0075 1575
Scenario 5 20 0.96 0.0097 1625
Scenario 6 23 0.95 0.0161 1711
Scenario 7 18 0.93 0.0127 1629
Scenario 8 20 0.92 0.0204 1685
Scenario 9 23 0.91 0.0714 1774
Scenario 10 18 1.00 0.0061 1531
Scenario 11 20 1.00 0.0078 1607
Scenario 12 23 1.00 0.0095 1662

Open Access AJOR

 V. P. DARJI, R.V. RAO 555

The value of maximum eigenvector is calculated, scores and ranking is given in Table 25. The results gave
max  16.4686 4  4.1171 . Consistency index is equal Scenario 7, Scenario 8, and Scenario 9 at first three posi-
to CI   max  4  3  0.0390 . The random index value tions in selection process and Scenario 12 with lowest
is 0.90, which turns to the Consistency ratio = score at last position in selection. The ranking obtained
CR  CI 0.90  0.0434 , which is much less than the comprehends with the results obtained by previous re-
allowable CR value of 0.1. Hence, we can conclude that searches.
a good consistency in the relative importance matrix The analysis of the simulation results dispatching rule
formation is maintained. followed in 7, 8 and 9 alternatives indicated that the
Step IV: Number of AGVs employed and the AGV smallest distance rule provided the best scenario regard-
active rate, the time of AGV in use for both empty and ing the number of served containers and the number of
full move are considered to be non-beneficial attributes. containers in the queue, under the assumption about the
The number of containers in the queue to be unloaded number of employed AGVs. Analysis of AGVs in [47]
from the ship by quay cranes, waiting for AGVs to be idle rate and AGVs active rates showed that the values of
transferred to one of the storage areas and number of these rates were very close among scenario, thus, it was
served containers which are moved from Berth to storage not possible to make a decision of which scenario could
areas are taken as beneficial attributes. The dominance be considered the best.
scores of each alternative scenario pair are calculated The proposed integrated method gave the scenario to
using (5) and (6) respectively. The values of dominance be best with smallest dispatching distance with clearly
scores for all alternative scenarios are tabulated in Ta- distinguishing scores. The scenario with largest distance
bles 22(a) and (b). i.e., 10, 11 and 12 are ranked to be last in the selection.
Step V: The standardized dominance is calculated In simple terms the smallest distance with minimum
based on the additive interval technique for all the pairs number of AGVs required to be employed is the best
of alternative scenarios using (7) and (8) for the ordinal scenario for this application. Whereas the largest distance
and cardinal criteria respectively and given in Tables with maximum number of AGVs employed is ranked to
23(a) and (b). be the last.
Step VI: The overall dominance score for each alter-
native of scenarios for all pair  i, i   , is calculated using
3.5. Example 5: Selection of Facility Layout
(9) which shows the degree by which scenario i domi-
Design
nates scenario i  . These overall dominance score for all
the pairs of alternative is shown in Table 24. In this section, an example is illustrated for the efficiency
Step VII: Finally the appraisal score for i th alterna- and effectiveness measurement of the facility layout de-
tive pair  Si  is computed using (10). Higher the ap- sign for an anonymous leading IC packaging company
praisal score better is the scenario for port terminal. The located in Kaohsiung as discussed by Yang and Kuo [57].

Table 21. Normalized data of scenario of AGV fleet operation.

Number of efficiency AGV active Reciprocal value of the number of Number of served
Scenario
employed AGV rate containers in the queue containers
Scenario 1 1 0.6667 0.0107 0.0947

Scenario 2 0.6 0.6667 0.0521 0.3374

Scenario 3 0 0.8889 0.1087 0.7449

Scenario 4 1 0.3333 0.0214 0.1811

Scenario 5 0.6 0.4444 0.0551 0.3868

Scenario 6 0 0.5556 0.1531 0.7407

Scenario 7 1 0.7778 0.1011 0.4033

Scenario 8 0.6 0.8889 0.2190 0.6337

Scenario 9 0 1 1 1

Scenario 10 1 0 0 0

Scenario 11 0.6 0 0.0260 0.3128

Scenario 12 0 0 0.0521 0.5391

Open Access AJOR

556 V. P. DARJI, R.V. RAO

Table 22. (a) Dominance score of each alternative pair of scenario of AGV fleet operation; (b) Dominance score of each al-
ternative pair of scenario of AGV fleet operation.
(a)
Pair  ii  ii Pair  ii  ii Pair  ii  ii
(1,2) −0.1640 0.2366 (3,1) 0.1640 −0.0271 (5,1) 0.1640 −0.4461
(1,3) −0.1640 0.0271 (3,2) 0.1640 −0.0271 (5,2) 0.1640 −0.0145
(1,4) −0.1640 0.0145 (3,4) 0.1640 −0.0271 (5,3) −0.1640 0.0271
(1,5) −0.1640 0.4461 (3,5) 0.1640 −0.0271 (5,4) 0.1640 −0.0271
(1,6) −0.1640 0.4461 (3,6) −0.1640 0.4045 (5,6) −0.1640 0.0271
(1,7) −0.1640 −0.4045 (3,7) 0.1640 −0.0271 (5,7) −0.1640 −0.8361
(1,8) −0.1640 0.0271 (3,8) −0.1640 −0.2366 (5,8) −0.1640 −0.4045
(1,9) −0.1640 0.0271 (3,9) −0.1640 −0.4045 (5,9) −0.1640 0.0271
(1,10) 0.1640 0.4045 (3,10) 0.1640 −0.0271 (5,10) 0.1640 −0.0271
(1,11) −0.1640 0.4461 (3,11) 0.1640 −0.0271 (5,11) 0.1640 0.4045
(1,12) −0.1640 0.4461 (3,12) 0.1640 0.4045 (5,12) 0.1640 0.4461
(2,1) +0.1640 −0.2366 (4,1) 0.1640 −0.0145 (6,1) 0.1640 −0.4461
(2,3) −0.1640 0.0271 (4,2) −0.1640 0.0271 (6,2) 0.1640 −0.4461
(2,4) 0.1640 −0.0271 (4,3) −0.1640 0.0271 (6,3) 0.1640 −0.4045
(2,5) −0.1640 0.0145 (4,5) −0.1640 0.0271 (6,4) 0.1640 −0.0271
(2,6) −0.1640 0.4461 (4,6) −0.1640 0.0271 (6,5) 0.1640 −0.0271
(2,7) −0.1640 −0.8361 (4,7) −0.1640 −0.4045 (6,7) 0.1640 −0.4461
(2,8) −0.1640 −0.4045 (4,8) −0.1640 0.0271 (6,8) −0.1640 −0.4461
(2,9) −0.1640 0.0271 (4,9) −0.1640 0.0271 (6,9) −0.1640 −0.4045
(2,10) −0.1640 −0.0271 (4,10) −0.1640 0.4045 (6,10) 0.1640 −0.0271
(2,11) 0.1640 0.4045 (4,11) −0.1640 0.4461 (6,11) 0.1640 −0.0271
(2,12) 0 0.4461 (4,12) −0.1640 0.4461 (6,12) 0.1640 0.4045

(b)
Pair  ii  ii Pair  ii  ii Pair  ii  ii
(7,1) 0.1640 0.4045 (9,1) 0.1640 −0.0271 (11,1) 0.1640 −0.4461
(7,2) 0.1640 0.8361 (9,2) 0.1640 −0.0271 (11,2) −0.1640 −0.4045
(7,3) −0.1640 0.0271 (9,3) 0.1640 0.4045 (11,3) −0.1640 0.0271
(7,4) 0.1640 0.4045 (9,4) 0.1640 −0.0271 (11,4) 0.1640 −0.4461
(7,5) 0.1640 0.8361 (9,5) 0.1640 −0.0271 (11,5) 0.1640 −0.4045
(7,6) −0.1640 0.4461 (9,6) 0.1640 0.4045 (11,6) 0.1640 0.0271
(7,8) −0.1640 0.0271 (9,7) 0.1640 −0.0271 (11,7) −0.1640 −0.8361
(7,9) −0.1640 0.0271 (9,8) 0.1640 −0.0271 (11,8) −0.1640 −0.4045
(7,10) 0.1640 0.4045 (9,10) 0.1640 −0.0271 (11,9) −0.1640 0.0271
(7,11) 0.1640 0.8361 (9,11) 0.1640 −0.0271 (11,10) 0.1640 −0.2366
(7,12) 0.1640 0.4461 (9,12) 0.1640 0.4045 (11,12) −0.1640 0.2366
(8,1) 0.1640 −0.0271 (10,1) −0.1640 −0.4045 (12,1) 0.1640 −0.4461
(8,2) 0.1640 0.4045 (10,2) −0.1640 0.0271 (12,2) 0 −0.4461
(8,3) 0.1640 0.2366 (10,3) −0.1640 0.0271 (12,3) −0.1640 −0.4045
(8,4) 0.1640 −0.0271 (10,4) 0.1640 −0.4045 (12,4) 0.1640 −0.4461
(8,5) 0.1640 0.4045 (10,5) −0.1640 0.0271 (12,5) −0.1640 −0.4461
(8,7) 0.1640 0.4461 (10,6) −0.1640 0.0271 (12,6) −0.1640 −0.4045
(8,8) 0.1640 −0.0271 (10,7) −0.1640 −0.4045 (12,7) 0.1640 −0.4461
(8,9) −0.1640 0.0271 (10,8) −0.1640 0.0271 (12,8) 0.1640 −0.8361
(8,10) 0.1640 −0.0271 (10,9) −0.1640 0.0271 (12,9) 0.1640 −0.4045
(8,11) 0.1640 0.4045 (10,11) 0.1640 0.2366 (12,10) 0.1640 −0.2366
(8,12) 0.1640 0.8361 (10,12) −0.1640 0.2366 (12,11) 0.1640 −0.2366

Open Access AJOR

 V. P. DARJI, R.V. RAO 557

Table 23. (a) Standard dominance score of each alternative pair of scenario of AGV fleet operation; (b) Standard dominance
score of each alternative pair of scenario of AGV fleet operation.
(a)
Pair  ii d ii Pair  ii d ii Pair  ii d ii
(1,2) 0 0.6415 (3,1) 1 0.4838 (5,1) 1 0.2332
(1,3) 0 0.5162 (3,2) 1 0.4838 (5,2) 1 0.4913
(1,4) 0 0.5087 (3,4) 1 0.4838 (5,3) 0 0.5162
(1,5) 0 0.7668 (3,5) 1 0.4838 (5,4) 1 0.4838
(1,6) 0 0.7668 (3,6) 0 0.7419 (5,6) 0 0.5162
(1,7) 0 0.2581 (3,7) 1 0.4838 (5,7) 0 0.0000
(1,8) 0 0.5162 (3,8) 0 0.3585 (5,8) 0 0.2581
(1,9) 0 0.5162 (3,9) 0 0.2581 (5,9) 0 0.5162
(1,10) 1 0.7419 (3,10) 1 0.4838 (5,10) 1 0.4838
(1,11) 0 0.7668 (3,11) 1 0.4838 (5,11) 1 0.7419
(1,12) 0 0.7668 (3,12) 1 0.7419 (5,12) 1 0.7668
(2,1) 1 0.3585 (4,1) 1 0.4913 (6,1) 1 0.2332
(2,3) 0 0.5162 (4,2) 0 0.5162 (6,2) 1 0.2332
(2,4) 1 0.4838 (4,3) 0 0.5162 (6,3) 1 0.2581
(2,5) 0 0.5087 (4,5) 0 0.5162 (6,4) 1 0.4838
(2,6) 1 0.7668 (4,6) 0 0.5162 (6,5) 1 0.4838
(2,7) 1 0.0000 (4,7) 0 0.2581 (6,7) 1 0.2332
(2,8) 1 0.2581 (4,8) 0 0.5162 (6,8) 0 0.2332
(2,9) 1 0.5162 (4,9) 0 0.5162 (6,9) 0 0.2581
(2,10) 0 0.4838 (4,10) 1 0.7419 (6,10) 1 0.4838
(2,11) 1 0.7419 (4,11) 0 0.7668 (6,11) 1 0.4838
(2,12) 0.5 0.7668 (4,12) 0 0.7668 (6,12) 1 0.7419

(b)
Pair  ii d ii Pair  ii d ii Pair  ii d ii
(7,1) 1 0.7419 (9,1) 1 0.4838 (11,1) 1 0.2332
(7,2) 1 1 (9,2) 1 0.4838 (11,2) 0 0.2581
(7,3) 0 0.5162 (9,3) 1 0.7419 (11,3) 0 0.5162
(7,4) 1 0.7419 (9,4) 1 0.4838 (11,4) 1 0.2332
(7,5) 1 1 (9,5) 1 0.4838 (11,5) 0 0.2581
(7,6) 0 0.7668 (9,6) 1 0.7419 (11,6) 0 0.5162
(7,8) 0 0.5162 (9,7) 1 0.4838 (11,7) 0 0
(7,9) 0 0.5162 (9,8) 1 0.4838 (11,8) 0 0.2581
(7,10) 1 0.7419 (9,10) 1 0.4838 (11,9) 0 0.5162
(7,11) 1 1 (9,11) 1 0.4838 (11,10) 1 0.3585
(7,12) 1 0.7668 (9,12) 1 0.7419 (11,12) 1 0.6415
(8,1) 1 0.4838 (10,1) 0 0.2581 (12,1) 1 0.2332
(8,2) 1 0.7419 (10,2) 0 0.5162 (12,2) 0.5 0.2332
(8,3) 1 0.6415 (10,3) 0 0.5162 (12,3) 0 0.2581
(8,4) 1 0.4838 (10,4) 0 0.2581 (12,4) 1 0.2332
(8,5) 1 0.7419 (10,5) 0 0.5162 (12,5) 0 0.2332
(8,6) 1 0.7668 (10,6) 0 0.5162 (12,6) 0 0.2581
(8,7) 1 0.4838 (10,7) 0 0.2581 (12,7) 0 0.2332
(8,9) 0 0.5162 (10,8) 0 0.5162 (12,8) 0 0
(8,10) 1 0.4838 (10,9) 0 0.5162 (12,9) 0 0.2581
(8,11) 1 0.7419 (10,11) 0 0.6415 (12,10) 1 0.3585
(8,12) 1 1 (10,12) 0 0.6415 (12,11) 1 0.3585

Open Access AJOR

558 V. P. DARJI, R.V. RAO

Table 24. Overall dominance scores of each alternative pair of scenario of AGV fleet operation.

Pair Dii Pair Dii Pair Dii Pair Dii Pair Dii Pair Dii

(1,2) 0.5364 (3,1) 0.5685 (5,1) 0.3590 (7,1) 0.7843 (9,1) 0.5685 (11,1) 0.3590

(1,3) 0.4316 (3,2) 0.5685 (5,2) 0.5748 (7,2) 1.0001 (9,2) 0.5685 (11,2) 0.2158

(1,4) 0.4253 (3,4) 0.5685 (5,3) 0.4316 (7,3) 0.4316 (9,3) 0.7843 (11,3) 0.4316

(1,5) 0.6411 (3,5) 0.5685 (5,4) 0.5685 (7,4) 0.7843 (9,4) 0.5685 (11,4) 0.3590

(1,6) 0.6411 (3,6) 0.6203 (5,6) 0.4316 (7,5) 1.0001 (9,5) 0.5685 (11,5) 0.2158

(1,7) 0.2158 (3,7) 0.5685 (5,7) 0.0000 (7,6) 0.6411 (9,6) 0.7843 (11,6) 0.4316

(1,8) 0.4316 (3,8) 0.2998 (5,8) 0.2158 (7,8) 0.4316 (9,7) 0.5685 (11,7) 0.0000

(1,9) 0.4316 (3,9) 0.2158 (5,9) 0.4316 (7,9) 0.4316 (9,8) 0.5685 (11,8) 0.2158

(1,10) 0.7843 (3,10) 0.5685 (5,10) 0.5685 (7,10) 0.7843 (9,10) 0.5685 (11,9) 0.4316

(1,11) 0.6411 (3,11) 0.5685 (5,11) 0.7843 (7,11) 1.0001 (9,11) 0.5685 (11,10) 0.4638

(1,12) 0.6411 (3,12) 0.7843 (5,12) 0.8051 (7,12) 0.8051 (9,12) 0.7843 (11,12) 0.7004

(2,1) 0.4638 (4,1) 0.5748 (6,1) 0.3590 (6,1) 0.5685 (10,1) 0.2158 (12,1) 0.3590

(2,3) 0.4316 (4,2) 0.4316 (6,2) 0.3590 (6,2) 0.7843 (10,2) 0.4316 (12,2) 0.2770

(2,4) 0.5685 (4,3) 0.4316 (6,3) 0.3798 (6,3) 0.7004 (10,3) 0.4316 (12,3) 0.2158

(2,5) 0.4253 (4,5) 0.4316 (6,4) 0.5685 (6,4) 0.5685 (10,4) 0.2158 (12,4) 0.3590

(2,6) 0.6411 (4,6) 0.4316 (6,5) 0.5685 (6,5) 0.7843 (10,5) 0.4316 (12,5) 0.1950

(2,7) 0.0000 (4,7) 0.2158 (6,7) 0.3590 (6,7) 0.8051 (10,6) 0.4316 (12,6) 0.2158

(2,8) 0.2158 (4,8) 0.4316 (6,8) 0.1950 (6,8) 0.5685 (10,7) 0.2158 (12,7) 0.1950

(2,9) 0.4316 (4,9) 0.4316 (6,9) 0.2158 (6,9) 0.4316 (10,8) 0.4316 (12,8) 0.0000

(2,10) 0.5685 (4,10) 0.7843 (6,10) 0.5685 (6,10) 0.5685 (10,9) 0.4316 (12,9) 0.2158

(2,11) 0.7843 (4,11) 0.6411 (6,11) 0.5685 (6,11) 0.7843 (10,11) 0.5364 (12,10) 0.4638

(2,12) 0.7231 (4,12) 0.6411 (6,12) 0.7843 (6,12) 1.0001 (10,12) 0.5364 (12,11) 0.4638

Table 25. Appraisal score of each alternative pair. For the industry front layout selection is challenging
and sometime tedious because of its inherent multiple
Scenarios Si additive interval technique Rank
objective nature and its data collection process proposed
Scenario 1 0.0812 7 by Lin and Sharp [58,59]. Past experience and database
Scenario 2 0.0869 4 can be used for the selection but this can not be justified
by the objective values. Hence, sometime the evaluation
Scenario 3 0.0822 6
may create conflict decision for the selection of best
Scenario 4 0.0731 8 layout design. Facility layout design is a systematic stra-
Scenario 5 0.0823 5
tegic approach. The treatment of facilities planning re-
quires highly sophisticated mathematical approach taking
Scenario 6 0.0554 9 advantage of empirical and analytical approaches. The
Scenario 7 0.1792 1 various industrial sectors like manufacturing organiza-
Scenario 8 0.1710 2
tions, mining, railroads etc. include the notion of con-
tinuous improvement in the layout design. It seems very
Scenario 9 0.1450 3
reasonable to suggest some adequate and significant op-
Scenario 10 0.0487 11 portunity to improve planning process as practiced today
Scenario 11 0.0494 10 by Tomkins et al. [60].
The literature for layout design problem falls into two
Scenario 12 0.0357 12
categories; algorithmic and procedural approaches. Muther

Open Access AJOR

 V. P. DARJI, R.V. RAO 559

[61] gave procedural approaches to incorporate both (2) for beneficial and non-beneficial attributes and given
qualitative and quantitative objectives in the design. AHP/ in Table 27.
DEA is used in [62] for facilities layout design, [63] Step III: The attributes considered in the example are
proposed TOPSIS and [64] proposed grey relational all affecting the selection process to some extent but the
analysis in solving the problem of facility layout design maintenance planning involves the required space for
selection and than [65] extended the same problem of engineers and tool movement.
facility layout selection by TOPSIS-DoE analysis. Step IV: The dominance scores for each alternative
In order to suggest some simple and feasible solution pairs of layout designs are calculated and tabulated in
the problem of facility selection is analysed by combined Tables 28(a) and (b) using (5) and (6).
EVAMIX method. In this example of the facility layout Step V: The standardized dominance is calculated
selection problem, 18 alternative layouts and 6 perform- based on the additive interval technique for all the pairs
ance attributes are considered. of alternative layout design using (7) and (8) for the or-
Table 26 shows the attribute values for 18 alternative dinal and cardinal criteria respectively and given in Ta-
plant layout, where flow distance and shape ratio and bles 29(a) and (b).
maintenance would be non-beneficial attributes, whereas Step VI: The overall dominance score for each alter-
adjacency scores, flexibility and accessibility are the native pair  i, i   , is calculated using (9) which shows
beneficial attributes for ranking these alternative plant the degree by which layout design i dominates the de-
layouts. sign i  . These overall dominance score for all the pairs
Step I: Total 18 alternative along with their most ap- of alternative is shown in Tables 30(a) and (b).
plicable 6 attributes are tabulated and decision matrix is Step VII: Finally the appraisal score for i th alterna-
formed. This is carried out to clearly distinguish the tive pair  Si  is computed using (10). Higher the ap-
beneficial and non-beneficial attributes, as well as to find praisal score better is layout design. The scores and rank-
the most influencing attribute to selection process. ing is given in Table 31.
Step II: The normalisation is carried out using (1) and The scores and ranking is given in the performance

Table 26. Data table for facility layout selection [57].

Layout Adjacency Flexibility Accessibility Shape ratio Distance (m) Maintenance

Layout 1 8 0.0494 0.0294 8.280 185.95 0.0130

Layout 2 9 0.0494 0.0147 3.750 207.37 0.0519

Layout 3 8 0.0370 0.0147 7.850 206.38 0.0519

Layout 4 8 0.0370 0.0147 8.280 189.66 0.0519

Layout 5 8 0.0617 0.0147 7.710 211.46 0.0390

Layout 6 5 0.0494 0.0147 2.070 264.07 0.0519

Layout 7 8 0.0247 0.0735 14.000 228.00 0.0649

Layout 8 9 0.0370 0.0441 6.250 185.59 0.0390

Layout 9 9 0.0741 0.0441 7.850 185.85 0.0519

Layout 10 8 0.0741 0.0588 7.850 236.15 0.0649

Layout 11 8 0.0864 0.1029 2.000 183.18 0.0909

Layout 12 8 0.0370 0.0588 13.300 204.18 0.0260

Layout 13 8 0.0247 0.0735 8.140 225.26 0.0519

Layout 14 8 0.0247 0.0588 8.000 205.82 0.0519

Layout 15 9 0.0864 0.1176 8.280 170.14 0.1169

Layout 16 9 0.0741 0.0735 7.710 216.38 0.0519

Layout 17 8 0.0988 0.1324 10.300 179.80 0.0909

Layout 18 10 0.0741 0.0588 10.160 185.75 0.0390

Open Access AJOR

560 V. P. DARJI, R.V. RAO

Table 27. Normalized data table for facility layout selection.

Layout Flexibility Accessibility Maintenance Distance (m) Adjacency Shape ratio
Layout 1 0.333333 0.1249 1 0.8317 0.6 0.4767
Layout 2 0.333333 0.0000 0.625602 0.6036 0.8 0.8542
Layout 3 0.165992 0.0000 0.625602 0.6142 0.6 0.5125
Layout 4 0.165992 0.0000 0.625602 0.7922 0.6 0.4767
Layout 5 0.499325 0.0000 0.749759 0.5601 0.6 0.5242
Layout 6 0.333333 0.0000 0.625602 0.0000 0 0.9942
Layout 7 0 0.4996 0.500481 0.3840 0.6 0.0000
Layout 8 0.165992 0.2498 0.749759 0.8355 0.8 0.6458
Layout 9 0.666667 0.2498 0.625602 0.8327 0.8 0.5125
Layout 10 0.666667 0.3747 0.500481 0.2972 0.6 0.5125
Layout 11 0.832659 0.7494 0.250241 0.8612 0.6 1.0000
Layout 12 0.165992 0.3747 0.87488 0.6376 0.6 0.0583
Layout 13 0 0.4996 0.625602 0.4132 0.6 0.4883
Layout 14 0 0.3747 0.625602 0.6521 0.6 0.5000
Layout 15 0.832659 0.8743 0 1.0000 0.8 0.4767
Layout 16 0.666667 0.4996 0.625602 0.5077 0.8 0.5242
Layout 17 1 1.0000 0.250241 0.8972 0.6 0.3083
Layout 18 0.666667 0.3747 0.749759 0.8338 1 0.3200

Table 28. (a) Dominance score of each alternative pair; (b) Dominance score of each alternative pair.
(a)
Pair  ii  ii Pair  ii  ii Pair  ii  ii Pair  ii  ii
(1,2) 0.2917 0.8264 (3,1) −0.4468 0 (5,1) −0.1366 0.3471 (7,1) −0.0128 0.4857
(1,3) 0.4468 1 (3,2) −0.1551 0.3264 (5,2) 0.2298 0.7572 (7,2) −0.0128 0.4857
(1,4) 0.4468 1 (3,4) 0 0.5000 (5,3) 0.2298 0.7572 (7,3) −0.0128 0.4857
(1,5) 0.1366 0.6529 (3,5) −0.2298 0.2428 (5,4) 0.2298 0.7572 (7,4) −0.0128 0.4857
(1,6) 0.2917 0.8264 (3,6) −0.1551 0.3264 (5,6) 0.2298 0.7572 (7,5) −0.0128 0.4857
(1,7) 0.0128 0.5143 (3,7) 0.0128 0.5143 (5,7) 0.0128 0.5143 (7,6) −0.0128 0.4857
(1,8) 0.0128 0.5143 (3,8) −0.2917 0.1736 (5,8) −0.0619 0.4307 (7,8) −0.0128 0.4857
(1,9) −0.2974 0.1672 (3,9) −0.3721 0.0836 (5,9) −0.2974 0.1672 (7,9) −0.0128 0.4857
(1,10) −0.2974 0.1672 (3,10) −0.2974 0.1672 (5,10) −0.2974 0.1672 (7,10) 0.0619 0.5693
(1,11) −0.2974 0.1672 (3,11) −0.2974 0.1672 (5,11) −0.2974 0.1672 (7,11) −0.2974 0.1672
(1,12) 0.0128 0.5143 (3,12) −0.2917 0.1736 (5,12) −0.1366 0.3471 (7,12) −0.0128 0.4857
(1,13) 0.0128 0.5143 (3,13) −0.0619 0.4307 (5,13) 0.0128 0.5143 (7,13) −0.0747 0.4164
(1,14) 0.0128 0.5143 (3,14) −0.0619 0.4307 (5,14) 0.0128 0.5143 (7,14) 0.1423 0.6592
(1,15) −0.2974 0.1672 (3,15) −0.2974 0.1672 (5,15) −0.2974 0.1672 (7,15) −0.2974 0.1672
(1,16) −0.2974 0.1672 (3,16) −0.3721 0.0836 (5,16) −0.2974 0.1672 (7,16) −0.2298 0.2428
(1,17) −0.2974 0.1672 (3,17) −0.2974 0.1672 (5,17) −0.2974 0.1672 (7,17) −0.2974 0.1672
(1,18) −0.2974 0.1672 (3,18) −0.4468 0 (5,18) −0.3721 0.0836 (7,18) −0.0128 0.4857
(2,1) −0.2917 0.1736 (4,1) −0.4468 0 (6,1) −0.2917 0.1736 (8,1) −0.0128 0.4857
(2,3) 0.1551 0.6736 (4,2) −0.1551 0.3264 (6,2) 0 0.5000 (8,2) 0.1366 0.6529
(2,4) 0.1551 0.6736 (4,3) 0 0.5000 (6,3) 0.1551 0.6736 (8,3) 0.2917 0.8264
(2,5) −0.2298 0.2428 (4,5) −0.2298 0.2428 (6,4) 0.1551 0.6736 (8,4) 0.2917 0.8264
(2,6) 0 0.5000 (4,6) −0.1551 0.3264 (6,5) −0.2298 0.2428 (8,5) 0.0619 0.5698
(2,7) 0.0128 0.5143 (4,7) 0.0128 0.5143 (6,7) 0.0128 0.5143 (8,7) 0.1366 0.6529
(2,8) −0.1366 0.3471 (4,8) −0.2917 0.1736 (6,8) −0.1366 0.3471 (8,8) 0.0128 0.5143
(2,9) −0.3721 0.0836 (4,9) −0.3721 0.0836 (6,9) −0.3721 0.0836 (8,9) −0.0804 0.4100
(2,10) −0.2974 0.1672 (4,10) −0.2974 0.1672 (6,10) −0.2974 0.1672 (8,10) −0.2974 0.1672
(2,11) −0.2974 0.1672 (4,11) −0.2974 0.1672 (6,11) −0.2974 0.1672 (8,11) −0.2974 0.1672
(2,12) −0.1366 0.3471 (4,12) −0.2917 0.1736 (6,12) −0.1366 0.3471 (8,12) −0.2917 0.1736
(2,13) −0.0619 0.4307 (4,13) −0.0619 0.4307 (6,13) −0.0619 0.4307 (8,13) 0.0128 0.5143
(2,14) −0.0619 0.4307 (4,14) −0.0619 0.4307 (6,14) −0.0619 0.4307 (8,14) 0.0128 0.5143
(2,15) −0.2974 0.1672 (4,15) −0.2974 0.1672 (6,15) −0.2974 0.1672 (8,15) −0.2974 0.1672
(2,16) −0.3721 0.0836 (4,16) −0.3721 0.0836 (6,16) −0.3721 0.0836 (8,16) −0.2974 0.1672
(2,17) −0.2974 0.1672 (4,17) −0.2974 0.1672 (6,17) −0.2974 0.1672 (8,17) −0.2974 0.1672
(2,18) −0.4468 0 (4,18) −0.4468 0 (6,18) −0.4468 0 (8,18) −0.3721 0.0836

Open Access AJOR

 V. P. DARJI, R.V. RAO 561

(b)

Pair  ii  ii Pair  ii  ii Pair  ii  ii Pair  ii  ii

(9,1) 0.2974 0.8328 (11,1) 0.2974 0.8328 (13,1) −0.0128 0.4857 (15,1) 0.2974 0.8328

(9,2) 0.3721 0.9164 (11,2) 0.2974 0.8328 (13,2) 0.0619 0.5693 (15,2) 0.2974 0.8328

(9,3) 0.3721 0.9164 (11,3) 0.2974 0.8328 (13,3) 0.0619 0.5693 (15,3) 0.2974 0.8328

(9,4) 0.3721 0.9164 (11,4) 0.2974 0.8328 (13,4) 0.0619 0.5693 (15,4) 0.2974 0.8328

(9,5) 0.2974 0.8328 (11,5) 0.2974 0.8328 (13,5) −0.0128 0.4857 (15,5) 0.2974 0.8328

(9,6) 0.3721 0.9164 (11,6) 0.2974 0.8328 (13,6) 0.0619 0.5693 (15,6) 0.2974 0.8328

(9,7) 0.0804 0.5900 (11,7) 0.2974 0.8328 (13,7) 0.0747 0.5836 (15,7) 0.2974 0.8328

(9,8) −0.1423 0.3408 (11,8) 0.2974 0.8328 (13,8) −0.0128 0.4857 (15,8) 0.2974 0.8328

(9,10) −0.1423 0.3408 (11,9) 0.2974 0.8328 (13,9) 0.0619 0.5693 (15,9) 0.2974 0.8328

(9,11) −0.2974 0.1672 (11,10) 0.2974 0.8328 (13,10) 0.1366 0.6529 (15,10) 0.2974 0.8328

(9,12) −0.1366 0.3471 (11,12) 0.2974 0.8328 (13,11) −0.2974 0.1672 (15,11) 0.1423 0.6592

(9,13) −0.0619 0.4307 (11,13) 0.2974 0.8328 (13,12) −0.0128 0.4857 (15,12) 0.2974 0.8328

(9,14) −0.0619 0.4307 (11,14) 0.2974 0.8328 (13,14) 0.217 0.7428 (15,13) 0.2974 0.8328

(9,15) −0.2974 0.1672 (11,15) −0.1423 0.3408 (13,15) −0.2974 0.1672 (15,14) 0.2974 0.8328

(9,16) −0.217 0.2572 (11,16) 0.2974 0.8328 (13,16) −0.1551 0.3264 (15,16) 0.2974 0.8328

(9,17) −0.2974 0.1672 (11,17) −0.3721 0.0836 (13,17) −0.2974 0.1672 (15,17) −0.4468 0.0000

(9,18) −0.2917 0.1736 (11,18) 0.2974 0.8328 (13,18) −0.0128 0.4857 (15,18) 0.2974 0.8328

(10,1) 0.2974 0.8328 (12,1) −0.0128 0.4857 (14,1) −0.0128 0.4857 (16,1) 0.2974 0.8328

(10,2) 0.2974 0.8328 (12,2) 0.1366 0.6529 (14,2) 0.0619 0.5693 (16,2) 0.3721 0.9164

(10,3) 0.2974 0.8328 (12,3) 0.2917 0.8264 (14,3) 0.0619 0.5693 (16,3) 0.3721 0.9164

(10,4) 0.2974 0.8328 (12,4) 0.2917 0.8264 (14,4) 0.0619 0.5693 (16,4) 0.3721 0.9164

(10,5) 0.2974 0.8328 (12,5) 0.1366 0.6529 (14,5) −0.0128 0.4857 (16,5) 0.2974 0.8328

(10,6) 0.2974 0.8328 (12,6) 0.1366 0.6529 (14,6) 0.0619 0.5693 (16,6) 0.3721 0.9164

(10,7) −0.0619 0.4307 (12,7) 0.0128 0.5143 (14,7) −0.1423 0.3408 (16,7) 0.2298 0.7572

(10,8) 0.2974 0.8328 (12,8) 0.2917 0.8264 (14,8) −0.0128 0.4857 (16,8) 0.2974 0.8328

(10,9) 0.1423 0.6592 (12,9) 0.1366 0.6529 (14,9) 0.0619 0.5693 (16,9) 0.217 0.7428

(10,11) −0.2974 0.1672 (12,10) −0.0804 0.4100 (14,10) −0.0804 0.4100 (16,10) 0.2917 0.8264

(10,12) 0.0804 0.5900 (12,11) −0.2974 0.1672 (14,11) −0.2974 0.1672 (16,11) −0.2974 0.1672

(10,13) −0.1366 0.3471 (12,13) 0.0128 0.5143 (14,12) −0.2298 0.2428 (16,12) 0.2974 0.8328

(10,14) 0.0804 0.5900 (12,14) 0.2298 0.7572 (14,13) −0.217 0.2572 (16,13) 0.1551 0.6736

(10,15) −0.2974 0.1672 (12,15) −0.2974 0.1672 (14,15) −0.2974 0.1672 (16,14) 0.3721 0.9164

(10,16) −0.2917 0.1736 (12,16) −0.2974 0.1672 (14,16) −0.3721 0.0836 (16,15) −0.2974 0.1672

(10,17) −0.2974 0.1672 (12,17) −0.2974 0.1672 (14,17) −0.2974 0.1672 (16,17) −0.2974 0.1672

(10,18) −0.0747 0.4164 (12,18) −0.0804 0.4100 (14,18) −0.2298 0.2428 (16,18) 0.1423 0.6592

Open Access AJOR

562 V. P. DARJI, R.V. RAO

Table 29. (a) Standardised dominance score of each alternative pair; (b) Standardised dominance score of each alternative
pair.
(a)

Pair  ii d ii Pair  ii d ii Pair  ii d ii Pair  ii d ii

(1,2) −0.2121 0.3083 (3,1) −0.1705 0.3459 (5,1) −0.1705 0.3459 (7,1) 0.0109 0.5099

(1,3) 0.1705 0.6541 (3,2) −0.2121 0.3083 (5,2) −0.3717 0.1640 (7,2) −0.3717 0.1640

(1,4) 0.0798 0.5721 (3,4) −0.1705 0.3459 (5,3) −0.1705 0.3459 (7,3) 0.0109 0.5099

(1,5) 0.1705 0.6541 (3,5) 0.1705 0.6541 (5,4) −0.1705 0.3459 (7,4) 0.0109 0.5099

(1,6) 0.5531 1 (3,6) 0.5531 1 (5,6) 0.5531 1 (7,5) 0.0109 0.5099

(1,7) −0.0109 0.4901 (3,7) −0.0109 0.4901 (5,7) −0.0109 0.4901 (7,6) 0.5531 1

(1,8) −0.3717 0.1640 (3,8) −0.3717 0.1640 (5,8) −0.3717 0.1640 (7,8) −0.3717 0.1640

(1,9) −0.3717 0.1640 (3,9) −0.4624 0.0820 (5,9) −0.5531 0 (7,9) −0.3717 0.1640

(1,10) 0.1705 0.6541 (3,10) 0.0798 0.5721 (5,10) −0.0109 0.4901 (7,10) 0.1705 0.6541

(1,11) 0.0109 0.5099 (3,11) 0.0109 0.5099 (5,11) 0.0109 0.5099 (7,11) 0.0109 0.5099

(1,12) −0.0109 0.4901 (3,12) −0.1705 0.3459 (5,12) −0.1705 0.3459 (7,12) 0.0109 0.5099

(1,13) 0.1705 0.6541 (3,13) −0.0109 0.4901 (5,13) −0.0109 0.4901 (7,13) 0.0109 0.5099

(1,14) 0.1705 0.6541 (3,14) −0.1705 0.3459 (5,14) −0.1705 0.3459 (7,14) 0.0109 0.5099

(1,15) −0.4624 0.0820 (3,15) −0.5531 0 (5,15) −0.5531 0.0000 (7,15) −0.3717 0.1640

(1,16) −0.2121 0.3083 (3,16) −0.2121 0.3083 (5,16) −0.3028 0.2263 (7,16) −0.3717 0.1640

(1,17) −0.1705 0.3459 (3,17) −0.1705 0.3459 (5,17) −0.1705 0.3459 (7,17) 0.0109 0.5099

(1,18) −0.5531 0.0000 (3,18) −0.5531 0 (5,18) −0.5531 0 (7,18) −0.3717 0.1640

(2,1) 0.2121 0.6917 (4,1) −0.0798 0.4279 (6,1) −0.5531 0 (8,1) 0.3717 0.8360

(2,3) 0.2121 0.6917 (4,2) −0.2121 0.3083 (6,2) −0.5531 0 (8,2) 0.1705 0.6541

(2,4) 0.2121 0.6917 (4,3) 0.1705 0.6541 (6,3) −0.5531 0 (8,3) 0.3717 0.8360

(2,5) 0.3717 0.8360 (4,5) 0.1705 0.6541 (6,4) −0.5531 0 (8,4) 0.3717 0.8360

(2,6) 0.5531 1 (4,6) 0.5531 1 (6,5) −0.5531 0 (8,5) 0.3717 0.8360

(2,7) 0.3717 0.8360 (4,7) −0.0109 0.4901 (6,7) −0.5531 0 (8,7) 0.5531 1

(2,8) −0.1705 0.3459 (4,8) −0.3717 0.1640 (6,8) −0.5531 0 (8,8) 0.3717 0.8360

(2,9) −0.1705 0.3459 (4,9) −0.3717 0.1640 (6,9) −0.5531 0 (8,9) −0.0109 0.4901

(2,10) 0.3717 0.8360 (4,10) 0.1705 0.6541 (6,10) −0.5531 0 (8,10) 0.3717 0.8360

(2,11) 0.3935 0.8557 (4,11) 0.0109 0.5099 (6,11) −0.3717 0.1640 (8,11) 0.3935 0.8557

(2,12) 0.2121 0.6917 (4,12) −0.0109 0.4901 (6,12) −0.5531 0 (8,12) 0.3717 0.8360

(2,13) 0.3717 0.8360 (4,13) 0.1705 0.6541 (6,13) −0.5531 0 (8,13) 0.3717 0.8360

(2,14) 0.2121 0.6917 (4,14) 0.1705 0.6541 (6,14) −0.5531 0 (8,14) 0.3717 0.8360

(2,15) −0.1705 0.3459 (4,15) −0.4624 0.0820 (6,15) −0.5531 0 (8,15) −0.1705 0.3459

(2,16) −0.0109 0.4901 (4,16) −0.2121 0.3083 (6,16) −0.5531 0 (8,16) −0.0109 0.4901

(2,17) 0.2121 0.6917 (4,17) −0.1705 0.3459 (6,17) −0.5531 0 (8,17) 0.2121 0.6917

(2,18) −0.5531 0 (4,18) −0.5531 0 (6,18) −0.5531 0 (8,18) −0.3935 0.1443

Open Access AJOR

 V. P. DARJI, R.V. RAO 563

(b)

Pair  ii d ii Pair  ii d ii Pair  ii d ii Pair  ii d ii Pair  ii d ii

(9,1) 0.3717 0.8360 (11,1) −0.0109 0.4901 (13,1) −0.1705 0.3459 (15,1) 0.4624 0.9180 (17,1) 0.1705 0.6541

(9,2) 0.1705 0.6541 (11,2) −0.3935 0.1443 (13,2) −0.3717 0.1640 (15,2) 0.1705 0.6541 (17,2) −0.2121 0.3083

(9,3) 0.4624 0.9180 (11,3) −0.0109 0.4901 (13,3) 0.0109 0.5099 (15,13) 0.5531 1 (17,3) 0.1705 0.6541

(9,4) 0.3717 0.8360 (11,4) −0.0109 0.4901 (13,4) −0.1705 0.3459 (15,4) 0.4624 0.9180 (17,4) 0.1705 0.6541

(9,5) 0.5531 1 (11,5) −0.0109 0.4901 (13,5) 0.0109 0.5099 (15,5) 0.5531 1 (17,5) 0.1705 0.6541

(9,6) 0.5531 1 (11,6) 0.3717 0.8360 (13,6) 0.5531 1 (15,6) 0.5531 1 (17,6) 0.5531 1

(9,7) 0.0109 0.5099 (11,7) −0.0109 0.4901 (13,7) −0.0109 0.4901 (15,7) 0.3717 0.8360 (17,7) −0.0109 0.4901

(9,8) 0.4624 0.9180 (11,8) −0.3935 0.1443 (13,8) −0.3717 0.1640 (15,8) 0.1705 0.6541 (17,8) −0.2121 0.3083

(9,10) 0.4624 0.9180 (11,9) −0.3935 0.1443 (13,9) −0.3717 0.1640 (15,9) 0.1705 0.6541 (17,9) −0.2121 0.3083

(9,11) 0.3935 0.8557 (11,10) −0.0109 0.4901 (13,10) 0.1705 0.6541 (15,10) 0.5531 1 (17,10) 0.1705 0.6541

(9,12) 0.3717 0.8360 (11,12) −0.0109 0.4901 (13,11) 0.0109 0.5099 (15,11) 0.5531 1 (17,11) 0.1705 0.6541

(9,13) 0.3717 0.8360 (11,13) −0.0109 0.4901 (13,12) −0.1705 0.3459 (15,12) 0.3717 0.8360 (17,12) −0.0109 0.4901

(9,14) 0.3717 0.8360 (11,14) −0.0109 0.4901 (13,14) 0.0109 0.5099 (15,13) 0.5531 1 (17,13) 0.1705 0.6541

(9,15) −0.1705 0.3459 (11,15) −0.5531 0 (13,15) −0.5531 0 (15,14) 0.5531 1 (17,14) 0.1705 0.6541

(9,16) 0.1705 0.6541 (11,16) −0.3935 0.1443 (13,16) −0.3717 0.1640 (15,16) 0.1705 0.6541 (17,15) −0.3717 0.1640

(9,17) 0.2121 0.6917 (11,17) −0.1705 0.3459 (13,17) −0.1705 0.3459 (15,17) 0.3717 0.8360 (17,16) −0.2121 0.3083

(9,18) −0.5531 0 (11,18) −0.3935 0.1443 (13,18) −0.5531 0 (15,18) −0.3935 0.1443 (17,18) −0.2121 0.3083

(10,1) −0.1705 0.3459 (12,1) 0.0109 0.5099 (14,1) −0.1705 0.3459 (16,1) 0.2121 0.6917 (18,1) 0.5531 1

(10,2) −0.3717 0.1640 (12,2) −0.2121 0.3083 (14,2) −0.2121 0.3083 (16,2) 0.0109 0.5099 (18,2) 0.5531 1

(10,3) −0.0798 0.4279 (12,3) 0.1705 0.6541 (14,3) 0.1705 0.6541 (16,3) 0.2121 0.6917 (18,3) 0.5531 1

(10,4) −0.1705 0.3459 (12,4) 0.0109 0.5099 (14,4) −0.1705 0.3459 (16,4) 0.2121 0.6917 (18,4) 0.5531 1

(10,5) 0.0109 0.5099 (12,5) 0.1705 0.6541 (14,5) 0.1705 0.6541 (16,5) 0.3028 0.7737 (18,5) 0.5531 1

(10,6) 0.5531 1 (12,6) 0.5531 1 (14,6) 0.5531 1 (16,6) 0.5531 1 (18,6) 0.5531 1

(10,7) −0.1705 0.3459 (12,7) −0.0109 0.4901 (14,7) −0.0109 0.4901 (16,7) 0.3717 0.8360 (18,7) 0.3717 0.8360

(10,8) −0.3717 0.1640 (12,8) −0.3717 0.1640 (14,8) −0.3717 0.1640 (16,8) 0.0109 0.5099 (18,8) 0.3935 0.8557

(10,9) −0.4624 0.0820 (12,9) −0.3717 0.1640 (14,9) −0.3717 0.1640 (16,9) −0.1705 0.3459 (18,9) 0.5531 1

(10,11) 0.0109 0.5099 (12,10) 0.1705 0.6541 (14,10) 0.1705 0.6541 (16,10) 0.3717 0.8360 (18,10) 0.5531 1

(10,12) −0.1705 0.3459 (12,11) 0.0109 0.5099 (14,11) 0.0109 0.5099 (16,11) 0.3935 0.8557 (18,11) 0.3935 0.8557

(10,13) −0.1705 0.3459 (12,13) 0.1705 0.6541 (14,12) −0.0109 0.4901 (16,12) 0.2121 0.6917 (18,12) 0.3717 0.8360

(10,14) −0.1705 0.3459 (12,14) 0.0109 0.5099 (14,13) −0.0109 0.4901 (16,13) 0.3717 0.8360 (18,13) 0.5531 1

(10,15) −0.5531 0 (12,15) −0.3717 0.1640 (14,15) −0.5531 0 (16,14) 0.2121 0.6917 (18,14) 0.5531 1

(10,16) −0.3717 0.1640 (12,16) −0.2121 0.3083 (14,16) −0.2121 0.3083 (16,15) −0.1705 0.3459 (18,15) 0.3935 0.8557

(10,17) −0.1705 0.3459 (12,17) 0.0109 0.5099 (14,17) −0.1705 0.3459 (16,17) 0.2121 0.6917 (18,16) 0.5531 1

(10,18) −0.5531 0 (12,18) −0.3717 0.1640 (14,18) −0.5531 0 (16,18) −0.5531 0 (18,17) 0.2121 0.6917

Open Access AJOR

564 V. P. DARJI, R.V. RAO

Table 30. (a) Overall dominance score of each alternative pair; (b) Overall dominance score of each pair.
(a)

Pair Dii Pair Dii Pair Dii Pair Dii Pair Dii Pair Dii

(1,2) 0.5398 (3,1) 0.1913 (5,1) 0.3464 (7,1) 0.4990 (9,1) 0.8345 (11,1) 0.6432

(1,3) 0.8086 (3,2) 0.3164 (5,2) 0.4290 (7,2) 0.3077 (9,2) 0.7713 (11,2) 0.4519

(1,4) 0.7633 (3,4) 0.4147 (5,3) 0.5296 (7,3) 0.4990 (9,3) 0.9172 (11,3) 0.6432

(1,5) 0.6535 (3,5) 0.4703 (5,4) 0.5296 (7,4) 0.4990 (9,4) 0.8719 (11,4) 0.6432

(1,6) 0.9224 (3,6) 0.6990 (5,6) 0.8914 (7,5) 0.4990 (9,5) 0.9252 (11,5) 0.6432

(1,7) 0.5009 (3,7) 0.5009 (5,7) 0.5009 (7,6) 0.7701 (9,6) 0.9626 (11,6) 0.8345

(1,8) 0.3205 (3,8) 0.1683 (5,8) 0.2832 (7,8) 0.3077 (9,7) 0.5456 (11,7) 0.6432

(1,9) 0.1654 (3,9) 0.0827 (5,9) 0.0747 (7,9) 0.3077 (9,8) 0.6600 (11,8) 0.4519

(1,10) 0.4365 (3,10) 0.3912 (5,10) 0.3458 (7,10) 0.6162 (9,10) 0.6600 (11,9) 0.4519

(1,11) 0.3567 (3,11) 0.3567 (5,11) 0.3567 (7,11) 0.3567 (9,11) 0.5480 (11,10) 0.6432

(1,12) 0.5009 (3,12) 0.2689 (5,12) 0.3464 (7,12) 0.4990 (9,12) 0.6175 (11,12) 0.6432

(1,13) 0.5916 (3,13) 0.4636 (5,13) 0.5009 (7,13) 0.4681 (9,13) 0.6549 (11,13) 0.6432

(1,14) 0.5916 (3,14) 0.3838 (5,14) 0.4211 (7,14) 0.5766 (9,14) 0.6549 (11,14) 0.6432

(1,15) 0.1201 (3,15) 0.0747 (5,15) 0.0747 (7,15) 0.1654 (9,15) 0.2660 (11,15) 0.1523

(1,16) 0.2452 (3,16) 0.2079 (5,16) 0.1999 (7,16) 0.1992 (9,16) 0.4767 (11,16) 0.4519

(1,17) 0.2660 (3,17) 0.2660 (5,17) 0.2660 (7,17) 0.3567 (9,17) 0.4573 (11,17) 0.2287

(1,18) 0.0747 (3,18) 0.0000 (5,18) 0.0374 (7,18) 0.3077 (9,18) 0.0776 (11,18) 0.4519

(2,1) 0.4602 (4,1) 0.2367 (6,1) 0.0776 (8,1) 0.6794 (10,1) 0.5634 (12,1) 0.4990

(2,3) 0.6836 (4,2) 0.3164 (6,2) 0.2234 (8,2) 0.6535 (10,2) 0.4628 (12,2) 0.4622

(2,4) 0.6836 (4,3) 0.5852 (6,3) 0.3010 (8,3) 0.8317 (10,3) 0.6088 (12,3) 0.7311

(2,5) 0.5709 (4,5) 0.4703 (6,4) 0.3010 (8,4) 0.8317 (10,4) 0.5634 (12,4) 0.6513

(2,6) 0.7765 (4,6) 0.6990 (6,5) 0.1085 (8,5) 0.7168 (10,5) 0.6541 (12,5) 0.6535

(2,7) 0.6922 (4,7) 0.5009 (6,7) 0.2298 (8,7) 0.8448 (10,6) 0.9252 (12,6) 0.8448

(2,8) 0.3464 (4,8) 0.1683 (6,8) 0.1551 (8,8) 0.6922 (10,7) 0.3838 (12,7) 0.5009

(2,9) 0.2287 (4,9) 0.1281 (6,9) 0.0374 (8,9) 0.4543 (10,8) 0.4628 (12,8) 0.4600

(2,10) 0.5371 (4,10) 0.4365 (6,10) 0.0747 (8,10) 0.5371 (10,9) 0.3399 (12,9) 0.3824

(2,11) 0.5480 (4,11) 0.3567 (6,11) 0.1654 (8,11) 0.5480 (10,11) 0.3567 (12,10) 0.5450

(2,12) 0.5377 (4,12) 0.3487 (6,12) 0.1551 (8,12) 0.5400 (10,12) 0.4549 (12,11) 0.3567

(2,13) 0.6549 (4,13) 0.5543 (6,13) 0.1925 (8,13) 0.6922 (10,13) 0.3464 (12,13) 0.5916

(2,14) 0.5751 (4,14) 0.5543 (6,14) 0.1925 (8,14) 0.6922 (10,14) 0.4549 (12,14) 0.6203

(2,15) 0.2660 (4,15) 0.1201 (6,15) 0.0747 (8,15) 0.2660 (10,15) 0.0747 (12,15) 0.1654

(2,16) 0.3085 (4,16) 0.2079 (6,16) 0.0374 (8,16) 0.3458 (10,16) 0.1683 (12,16) 0.2452

(2,17) 0.4573 (4,17) 0.2660 (6,17) 0.0747 (8,17) 0.4573 (10,17) 0.2660 (12,17) 0.3567

(2,18) 0.0000 (4,18) 0.0000 (6,18) 0.0000 (8,18) 0.1172 (10,18) 0.1861 (12,18) 0.2739

Open Access AJOR

 V. P. DARJI, R.V. RAO 565

(b)

Pair Dii Pair Dii Pair Dii

(13,1) 0.4083 (15,1) 0.8799 (17,1) 0.7339

(13,2) 0.3451 (15,2) 0.7339 (17,2) 0.5426

(13,3) 0.5364 (15,3) 0.9252 (17,3) 0.7339

(13,4) 0.4457 (15,4) 0.8799 (17,4) 0.7339

(13,5) 0.4990 (15,5) 0.9252 (17,5) 0.7339

(13,6) 0.8075 (15,6) 0.9252 (17,6) 0.9252

(13,7) 0.5319 (15,7) 0.8345 (17,7) 0.6432

(13,8) 0.3077 (15,8) 0.7339 (17,8) 0.5426

(13,9) 0.3451 (15,9) 0.7339 (17,9) 0.5426

(13,10) 0.6535 (15,10) 0.9252 (17,10) 0.7339

(13,11) 0.3567 (15,11) 0.8477 (17,11) 0.7713

(13,12) 0.4083 (15,12) 0.8345 (17,12) 0.6432

(13,14) 0.6139 (15,13) 0.9252 (17,13) 0.7339

(13,15) 0.0747 (15,14) 0.9252 (17,14) 0.7339

(13,16) 0.2366 (15,16) 0.7339 (17,15) 0.5375

(13,17) 0.2660 (15,17) 0.4624 (17,16) 0.5426

(13,18) 0.2170 (15,18) 0.4519 (17,18) 0.5426

(14,1) 0.4083 (16,1) 0.7547 (18,1) 0.9252

(14,2) 0.4249 (16,2) 0.6915 (18,2) 0.9999

(14,3) 0.6162 (16,3) 0.7921 (18,3) 0.9999

(14,4) 0.4457 (16,4) 0.7921 (18,4) 0.9999

(14,5) 0.5788 (16,5) 0.8001 (18,5) 0.9626

(14,6) 0.8075 (16,6) 0.9626 (18,6) 0.9999

(14,7) 0.4234 (16,7) 0.8007 (18,7) 0.6922

(14,8) 0.3077 (16,8) 0.6541 (18,8) 0.8828

(14,9) 0.3451 (16,9) 0.5232 (18,9) 0.9224

(14,10) 0.5450 (16,10) 0.8317 (18,10) 0.8139

(14,11) 0.3567 (16,11) 0.5480 (18,11) 0.5480

(14,12) 0.3796 (16,12) 0.7547 (18,12) 0.7260

(14,13) 0.3860 (16,13) 0.7634 (18,13) 0.7829

(14,15) 0.0747 (16,14) 0.7921 (18,14) 0.8914

(14,16) 0.2079 (16,15) 0.2660 (18,15) 0.5480

(14,17) 0.2660 (16,17) 0.4573 (18,16) 0.7054

(14,18) 0.1085 (16,18) 0.2946 (18,17) 0.4573

Open Access AJOR

566 V. P. DARJI, R.V. RAO

Table 31. Appraisal score of facility layouts.

Layouts Si additive interval technique Rank

Layout 1 0.0243 14
Layout 2 0.0531 5
Layout 3 0.0185 16
Layout 4 0.0245 13
Layout 5 0.0135 17
Layout 6 0.0069 18
Layout 7 0.0345 10
Layout 8 0.0483 7
Layout 9 0.0453 8
Layout 10 0.0255 12
Layout 11 0.0495 6
Layout 12 0.0415 9
Layout 13 0.0262 11
Layout 14 0.0226 15
Layout 15 0.1939 2
Layout 16 0.0867 4
Layout 17 0.1099 3
Layout 18 0.1985 1

Table 32. Comparison and ranking of facility layout designs using various methods.
Score using Score using Score using Rank using Rank using Rank using
EVAMIX Rank AHP/DEA Rank TOPSIS Rank GRA TOPSIS SAW
method method meta model method method method
Layout 1 0.0243 14 91.69 7 57.857 18 10 17 14

Layout 2 0.0531 5 98.45 2 59.858 14 8 8 8

Layout 3 0.0185 16 86.39 12 58.653 17 15 15 15

Layout 4 0.0245 13 89.94 8 58.666 16 11 16 12

Layout 5 0.0135 17 86.44 11 59.871 13 13 14 11

Layout 6 0.0069 18 96.62 3 59.789 15 16 9 18

Layout 7 0.0345 10 80.77 15 64.760 8 17 13 16

Layout 8 0.0483 7 96.51 4 60.518 12 7 12 9

Layout 9 0.0453 8 95.69 5 65.223 7 5 6 5

Layout 10 0.0255 12 87.44 9 67.824 5 9 5 7

Layout 11 0.0495 6 100 1 76.219 3 3 1 3

Layout 12 0.0415 9 85.72 14 60.722 11 18 18 17

Layout 13 0.0262 11 86.29 13 63.536 9 14 10 13

Layout 14 0.0226 15 86.66 10 62.039 10 12 11 10

Layout 15 0.1939 2 100 1 80.379 2 1 2 1

Layout 16 0.0867 4 96.74 2 68.175 4 6 4 6

Layout 17 0.1099 3 94.63 6 80.428 1 2 3 2

Layout 18 0.1985 1 100 1 65.524 6 4 7 4

Open Access AJOR

 V. P. DARJI, R.V. RAO 567

frontier proposed by [62] shows 11, 15 and 18 as leading proaches,” In: F. J. Salvatore and G. M. Ehrgott, Eds.,
candidate of facility layout designs. Than the modified Multiple Criteria Decision Analysis: State of the Art Sur-
veys, Springer, New York, 2005, pp. 197-262.
DEA frontier could suggest the best among these three.
The proposed methodology gave final solution by con- [6] P. Nijkamp, P. Rietveld and H. Voogd, “Multicriteria
sidering a large number of design alternatives by avoid- Evaluation in Physical Planning,” North Holland, Am-
sterdam, 1990, pp. 65-100.
ing any problem of same score values. Design alternative
18 is the present alternative and which come out with [7] S. Hajkowicz and A. Higgins, “A Comparison of Multiple
Criteria Analysis Techniques for Water Resource Man-
maximum score in the present approach. The top ranking agement,” European Journal of Operation Research, Vol.
of the design layouts using integrated EVAMIX method 184, No. 1, 2008, pp. 255-265.
are matching with the results obtained by previous re- http://dx.doi.org/10.1016/j.ejor.2006.10.045
searchers and given in Table 32. The alternative 18 is the [8] E. S. Chung and K. S. Lee, “Identification of Spatial
existing facility layout and fulfils all the requirements of Ranking of Hydrological Vulnerability Using Multi-Cri-
manufacturing environment of application. It is clear teria Decision Making Techniques: Case Study of Korea,”
from the Table 32, that alternative 18, 17, 15 and 11 are Water Resource Management, Vol. 23, No. 12, 2009, pp.
found to be more efficient compared any other layout 2395-2416. http://dx.doi.org/10.1007/s11269-008-9387-9
design. [9] L. Ustinovichius, E. K. Zavadskas and V. Podvezko,
“Application of a Quantitative Multiple Criteria Decision
Making (MCDM-1) Approach to the Analysis of Invest-
4. Conclusions ments in Construction,” Control and Cybernetics, Vol. 36,
The integrated EVAMIX methodology is suggested for No. 1, 2007, pp. 251-268.
decision making in the industrial environment to solve [10] I. Jeffreys, “The Use of Compensatory and Non-Compensa-
the decision making problems from a large number of tory Multi-Criteria Analysis for Small-Scale Forestry,”
available alternatives for a given problem. Five examples Small-Scale Forest Ecol Manage Policy, Vol. 3, No. 1,
2004, pp. 99-117.
are included to illustrate the method. The proposed me-
thod incorporates the cardinal and ordinal attributes in a [11] S. J. Chen and C. L. Hwang, “Fuzzy Multiple Attribute
Decision Making-Methods and Applications,” Lecture
separate pattern combined together to provide a better Notes in Economics and Mathematical Systems, Springer,
evaluation of the alternatives. The use of AHP in the de- New York, 1992.
cision making allows the decision maker to assign the [12] T. L. Saaty, “The Analytic Hierarchy Process,” McGraw-
values of relative importance to the attribute based on Hill, New York, 1980.
his/her preferences. The Appraisal scores calculated by
[13] T. L. Saaty and L. G. Vrgas, “The Logic of Priorities, Ap-
the method evaluate and rank the alternatives and lead to plications in Business, Energy, Health, Transportation,”
selection of a suitable alternative. Kluwer-Nijhoff, The Hague, 1981.
The integrated EVAMIX method is a general outrank- [14] T. L. Saaty, “Absolute and Relative Measurement with
ing method and can deal with any number of cardinal and the AHP. The Most Livable Cities in the United States,”
ordinal attributes simultaneously. The method is logical Socio-Economic Planning Sciences, Vol. 20, No. 6, 1986,
and provides good elaboration of the ranking method. pp. 327-331.
The suggested methodology can be used for any type of http://dx.doi.org/10.1016/0038-0121(86)90043-1
selection problem having any number of attributes. [15] T. L. Saaty, “Fundamentals of Decision Making and Pri-
ority Theory with AHP,” RWS Publications, Pittsburg,
2000.
REFERENCES [16] S. Xu, “References on the Analytic Hierarchy Process,”
[1] R. V. Rao, “Decision Making in Manufacturing Envi- Institute of Systems Engineering, Tianjin University,
ronment Using Graph Theory and Fuzzy Multiple Attrib- Tianjin, 1986.
ute Decision Making Methods,” Springer Series in Ad- [17] G. T. Fechner, “Elements of Psychophysics,” Holt, Rine-
vanced Manufacturing, London, Vol. 2, 2013, pp. 1-5. hart and Winston, New York, 1860.
[2] P. Chatterjee, V. M. Athawale and S. Chakraborty, “Ma- [18] L. L. Turstone, “A Law of Comparative Judgements,”
terial Selection Using Complex Proportional Assessment Psychological Reviews, Vol. 34, No. 4, 1927, pp. 272-
and Evaluation of Mixed Data Methods,” Material and 286.
Design, Vol. 32, No. 2, 2011, pp. 851-860.
[19] K. L. Edwards, “Selecting Materials for Optimum Use in
[3] H. Voogd, “Multicriteria Evaluation with Mixed Qualita- Engineering Components,” Mater & Design, Vol. 26, No.
tive and Quantitative Data,” Environment and Planning 5, 2005, pp. 469-473.
Bulletin, Vol. 9, No. 2, 1982, pp. 221-236. http://dx.doi.org/10.1016/j.matdes.2004.07.004
[4] H. Voogd, “Multicriteria Evaluation for Urban and Re- [20] M. S. Chen, “A New Method for Tool Steel Materials
gional Planning,” Pion, London, 1983. under Fuzzy Environment,” Fuzzy Sets and System, Vol.
[5] J. M. Martel and B. Matarazzo, “Other Outranking Ap- 92, No. 3, 1997, pp. 265-274.

Open Access AJOR

568 V. P. DARJI, R.V. RAO

http://dx.doi.org/10.1016/S0165-0114(96)00189-3 http://dx.doi.org/10.1016/S0924-0136(01)00655-0
[21] H. D. Jee and K. J. Kang, “A Method for Optimal Mate- [33] W. D. Brouwer, E. C. F. C. Van Herpt and M. Labordus,
rial Selection Aided with Decision Making Theory,” Ma- “Vacuum Injection Molding for Large Structural Appli-
terials & Design, Vol. 21, No. 3, 2000, pp. 199-206. cations,” Composites Part-A: Applied Science and Manu-
http://dx.doi.org/10.1016/S0261-3069(99)00066-7 facturing, Vol. 34, No. 6, 2003, pp. 551-558.
[22] K. M. Rajan and K. Narasimhan, “An Approach to Selec- http://dx.doi.org/10.1016/S1359-835X(03)00060-5
tion of Material and Manufacturing Processes for Rocket [34] K. Han, S. Jiang, C. Zhang and B. Wang, “Flow Model-
Motor Cases Using Weighted Performance Index,” Jour- ling and Simulation of SCRIMP for Composites Manu-
nal of Materials Engineering and Performance, Vol. 11, facturing,” Composites Part-A: Applied Science and Manu-
No. 4, 2002, pp. 444-449. facturing, Vol. 31, No. 1, 2004, pp. 79-86.
http://dx.doi.org/10.1361/105994902770343980 http://dx.doi.org/10.1016/S1359-835X(99)00053-6
[23] N. S. Ermolaeva, M. B. G. Castro and P. V. Kandachar, [35] R. T. Durai Prabhakaran, B. J. C. Babu and V. P. Agra-
“Materials Selection for an Automotive Structure by In- wal, ”Structural Modeling and Analysis of Composite
tegrating Structural Optimization with Environmental Product System: A Graph Theoretic Approach,” Journal
Impact Assessment,” Materials & Design, Vol. 25, No. 8, of Composite Materials, Vol. 40, No. 22, 2006, pp. 1987-
2004, pp. 689-698. 2007. http://dx.doi.org/10.1177/0021998306061318
http://dx.doi.org/10.1016/j.matdes.2004.02.021 [36] K. L. Reifsnider, “Modeling of the Interphase in Poly-
[24] R. V. Rao and J. P. Davim, “A Decision-Making Frame- mer-Matrix Composite Material Systems,” Composites,
work Model for Material Selection Using a Combined Vol. 25, No. 7, 1994, pp. 461-469.
Multiple Attribute Decision-Making Method,” The Inter- http://dx.doi.org/10.1016/0010-4361(94)90170-8
national Journal of Advanced Manufacturing Technology, [37] H. Liu, A. Uhlherr and M. K. Bannister, “Quantitative
Vol. 35, No. 7-8, 2008, pp. 751-760. Structure—Property Relationships for Composites: Pre-
http://dx.doi.org/10.1007/s00170-006-0752-7 diction of Glass Transition Temperatures for Epoxy Res-
[25] P. Chatterjee, V. M. Athawale and S. Chakraborty, “Se- ins,” Polymer, Vol. 45, No. 6, 2004, pp. 2051-2060.
lection of Materials Using Compromise Ranking and Out- http://dx.doi.org/10.1016/j.polymer.2004.01.008
ranking Methods,” Materials & Design, Vol. 30, No. 10, [38] J. C. Lin, M. H. Nien and F. M. Yu, “Morphological
2009, pp. 4043-4053. Structure, Processing and Properties of Propylene Poly-
http://dx.doi.org/10.1016/j.matdes.2009.05.016 mer Matrix Nanocomposites,” Composite Structures, Vol.
[26] A. Shanian and O. Savadogo, “A Material Selection 71, No. 1, 2005, pp. 78-82.
Model Based on the Concept of Multiple Attribute Deci- http://dx.doi.org/10.1016/j.compstruct.2004.09.021
sion Making,” Materials & Design, Vol. 27, No. 4, 2006, [39] R. T. Durai Prabhakaran, B. J. C. Babu and V. P. Agrawal,
pp. 329-337. “Optimum Selection of a Composite Product System Us-
http://dx.doi.org/10.1016/j.matdes.2004.10.027 ing AMDM Approach,” Material and Manufacturing
[27] B. D. Manshadi, H. Mahmudi, A. Abedian and R. Mah- Processes, Vol. 21, No. 8, 2006, pp. 883-891.
mudi, “A Novel Method for Materials Selection in Me- http://dx.doi.org/10.1080/10426910600773472
chanical Design: Combination of Non-Linear Lineariza- [40] W. Ting and C. X. Bing, “Thermal Power Plant Sitting
tion and a Modified Digital Logic Method,” Materials & Based on TOPSIS Method,” Procedia Engineering, Vol.
Design, Vol. 28, No. 1, 2007, pp. 8-15. 15, 2011, pp. 5384-5388.
http://dx.doi.org/10.1016/j.matdes.2005.06.023 http://dx.doi.org/10.1016/j.proeng.2011.08.998
[28] I. Ribeiro, P. Peças, A. Silva and E. Henriques, “Life [41] J. H. Yan, “Comprehensive Evaluation Theory and Me-
Cycle Engineering Methodology Applied to Material Se- thods,” Science Press, Beijing, 2002.
lection, a Fender Case Study,” Journal of Cleaner Pro-
[42] Y. L. Jing, X. N. Dong and Y. Z. Shi, “The Sitting of
duction, Vol. 16, No. 17, 2008, pp. 1887-1899.
Thermal Power Plant by the Method of Triangular Fuzzy
http://dx.doi.org/10.1016/j.jclepro.2008.01.002
Numbers,” Quantitative & Technical Economics, 2003.
[29] P. Pecas, I. Ribeiro, A. Silva and E. Henriques, “Compre-
[43] Q. L. Dong, “The Application of Entropy-Based Fuzzy
hensive Approach for Informed Life Cycle-Based Mate-
Comprehensive Evaluation Method in the Sitting of
rial Selection,” Mater Design, Vol. 43, 2013, pp. 220-
Thermal Power Plant,” Experimental Investigation, 2010.
232. http://dx.doi.org/10.1016/j.matdes.2012.06.064
[44] X. N. Dong, “Gray-Level Analysis Used in the Optimal
[30] C. D. Rudd, A. C. Long, K. N. Kendall and C. G. E.
Decision-Making of Thermal Power Plant Site,” Journal
Mangin, “Liquid Molding Technologies,” Woodhead Pub-
of Power System Technology, Vol. 11, 1994.
lishing Ltd., Cambridge, 1997.
http://dx.doi.org/10.1533/9781845695446 [45] M. Y. Hai, “Application Manual of Economic Evaluation
and Technological Transformation of Electric Power In-
[31] K. Potter, “Materials for RTM. Resin Transfer. Resin
dustry,” Jilin Photography Press, Jilin, 2005.
Transfer Molding,” Chapman and Hall, London, 1997, p.
200. [46] P. Lu, M. Xing and Y. Kang, “AHP Method Used in the
Optimal Planning of Power Plant Site Selection,” Journal
[32] D. S. Jung and G. L. Dai, “Manufacture of Composite
of North China Electric Power University, Vol. 2, No. 4,
Screw Rotors for Air Compressors by RTM Process,”
1994.
Journal of Materials Processing Technology, Vol. 113,
No. 1-3, 2001, pp. 196-201. [47] P. Danijela, V. Ivana, V. Katarina and T. Dusan, “Appli-

Open Access AJOR

 V. P. DARJI, R.V. RAO 569

cation of DEA to the Analysis of AGV Fleet Operations Envelopment Analysis,” Tehnika, Vol. 65, No. 4, 2010,
in a Port Container Terminal,” Procedia Social and Be- pp. 14-19.
havioural Sciences, Vol. 20, 2011, pp. 816-825. [57] T. Yang and C. Kuo, “A Hierarchical AHP/DEA Meth-
[48] P. Ioannou, H. Jula, C. I. Liu, K. Vukadinovic, H. Pour- odology for the Facilities Layout Design Problem,” Euro-
mohammadi and E. Dougherty Jr., “Advanced Material pean Journal of Operational Research, Vol. 147, No. 1,
Handling: Automated Guided Vehicles in Agile Ports,” 2003, pp. 128-136.
CCDoTT Technical Report, Centre for Advanced Trans- http://dx.doi.org/10.1016/S0377-2217(02)00251-5
portation Technologies, University of Southern California, [58] L. C. Lin and G. P. Sharp, “Quantitative and Qualitative
Los Angeles, 2001. Indices for the Plant Layout Evaluation Problem,” Euro-
[49] M. G. Maria and T. Lourdes, “Reforms and Infrastructure pean Journal of Operational Research, Vol. 116, No. 1,
Efficiency in Spain’s Container Ports,” Transportation 1999, pp. 100-117.
Research Part A: Policy and Practice, Vol. 42, No. 1, http://dx.doi.org/10.1016/S0377-2217(98)00046-0
2008, pp. 243-257. [59] L. C. Lin and G. P. Sharp, “Application of the Integrated
[50] C. P. Barros, “Measurement of Efficiency of Portuguese Framework for the Plant Layout Evaluation Problem,”
Seaport Authorities with DEA,” International Journal of European Journal of Operational Research, Vol. 116, No.
Transport Economics, Vol. 30, No. 3, 2003, pp. 335-354. 1, 1999, pp. 118-138.
[51] K. Cullinane, D. W. Song, P. Ji and T. F. Wang, “An http://dx.doi.org/10.1016/S0377-2217(98)00174-X
Application of DEA Windows Analysis to Container Port [60] J. A. Tomkins, J. A. White, Y. Bozer and J. M. Tanchoco,
Production Efficiency,” Review of Network Economics, “Facilities Planning,” John Wiley & Sons, New York,
Vol. 3, No. 2, 2004, pp. 1-23. 2003.
[52] K. Cullinane and T. F. Wang, “The Efficiency of Euro- [61] R. Muther, “Systematic Layout Planning,” 2nd Edition,
pean Container Ports: Across-Sectional Data Envelop- Cahners Books, Boston, 1973.
ment Analysis,” International Journal of Logistics: Re-
[62] Y. Taho and K. Chunwei, “A Hierarchical AHP/DEA
search and Applications, Vol. 9, No. 1, 2006, pp. 19-31.
Methodology for the Facilities Layout Design Problem,”
http://dx.doi.org/10.1080/13675560500322417
European Journal of Operational Research, Vol. 147, No.
[53] I. E. Kaisar, S. Pathomsiri and A. Haghani, “Efficiency 1, 2003, pp. 128-136.
Measurement of US Ports Using Data Envelopment Ana-
[63] T. Yang and C. C Hung, “2007. Multi-Attribute Decision
lysis,” National Urban Freight Conference, Long Beach
Making Methods for Plant Layout Design Problem,” Ro-
California, 1-3 February 2006, p. 16.
botics and Computer Integrated Manufacturing, Vol. 23,
[54] C. I. Liu, H. Jula, K. Vukadinovic and P. Ioannou, No. 1, 2007, pp. 126-137.
“Automated Guided Vehicle System for Two Container http://dx.doi.org/10.1016/j.rcim.2005.12.002
Yard Layouts,” Transportation Research Part C: Emerg-
[64] Y. Y. Kuo, T. H. Yang and G. W. Huang, “The Use of
ing Technologies, Vol. 12, No. 5, 2004, pp. 349-368.
Grey Relational Analysis in Solving Multiple Attribute
http://dx.doi.org/10.1016/j.trc.2004.07.014
Decision-Making Problems,” Computer & Industrial En-
[55] H. Min and B. I. Park, “Evaluating the Inter-Temporal gineering, Vol. 55, No. 1, 2008, pp. 80-93.
Efficiency Trends of International Container Terminals http://dx.doi.org/10.1016/j.cie.2007.12.002
Using Data Envelopment Analysis,” International Jour-
[65] Y. T. İç, “An Experimental Design Approach Using
nal of Integrated Supply Management, Vol. 1, No. 3,
TOPSIS Method for the Selection of Computer-Integrated
2005, pp. 258-277.
Manufacturing Technologies,” Robotics and Computer-
http://dx.doi.org/10.1504/IJISM.2005.005950
Integrated Manufacturing, Vol. 28, No. 2, 2012, pp. 245-
[56] D. Pjevčević and K. Vukadinović, “Efficiency Measure- 256. http://dx.doi.org/10.1016/j.rcim.2011.09.005
ment of Bulk Cargo Handling at River Port Using Data

Open Access AJOR

View publication stats