Professional Documents
Culture Documents
Petrillo 2016
Petrillo 2016
Vol. 15 (2016)
°c World Scienti¯c Publishing Company
DOI: 10.1142/S021962201640006X
mortiz1@cuc.edu.co
Fabio de Felice
University of Cassino and Southern Lazio
Department of Civil and Mechanical Engineering
G. Di Biasio 43 Cassino, 03043, Italy
defelice@unicas.it
Antonella Petrillo§
University of Naples Parthenope, Department of Engineering
Isola C4 Centro Direzionale Napoli Napoli, 80143, Italy
antonella.petrillo@uniparthenope.it
Keywords: Analytic hierarchy process (AHP); technique for order of preference by similarity to
ideal solution (TOPSIS); multi-criteria decision making (MCDM); tomography; medical
applications.
§ Corresponding author.
1
2 M. A. O. Barrios et al.
1. Introduction
Multi Criteria Decision-Making (MCDM) methods are a set of methods designed to
ensure that a synthesis of multiple sources of information is documented and directed
toward a stated goal which can help with the comparison of alternatives based on
decision matrices.1,2
MCDM has seen an incredible amount of use over the last several decades.
MCDM is frequently used to solve real-world problems with multiple, con°icting,
and incommensurate attributes.3,4 In particular, MCDM is successfully used in
many di®erent selection processes.5 The aim of the present paper is not to analyze in
by NANYANG TECHNOLOGICAL UNIVERSITY on 07/17/16. For personal use only.
detail the several common MCDM methods, but only to highlight the most wide-
spread methods. In order to identify those methods, a search was conducted for
Int. J. Info. Tech. Dec. Mak. Downloaded from www.worldscientific.com
common MCDM methods by title, abstract, and keywords, while utilizing the fol-
lowing databases: Elsevier, Springer, ScienceDirect, and IEEExplore.6 The following
MCDM methods were identi¯ed: Analytic Hierarchy Process (AHP),7 Analytic
Network Process (ANP),8 Technique for Order of Preference by Similarity to Ideal
Solution (TOPSIS),9 Elimination Et Choix Traduisant la REalite (ELECTRE),10
Vlse Kriterijumska Optimizacija Kompromisno Resenje, means Multicriteria Opti-
mization and Compromise Solution (VIKOR),11 and Preference Ranking Organi-
zation Method for Enrichment Evaluation (PROMETHEE).12 According to the
literature search conducted, it became apparent that the selection process can be
modeled as a MCDM problem. Selection problems can have various di®erent and
con°icting criteria for making a selection among predetermined decision alter-
natives.13 Some relevant applications about the selection problem in medical sector
are: the use of AHP to select medical equipment such as pacemakers and im-
plantable de¯brillators proposed by Balestra et al.14 or the selection of an intensive
care ventilators analyzed by Chatburn and Primiano15 or the selection of medical
products and materials in Korea as developed by Cho and Kim.16 Although di®erent
MCDM exist, all of them follow a number of common steps: problem de¯nition,
identi¯cation of alternatives, criteria selection, preparation of the decision matrix,
and assigning weights to the criteria. The newest trend with respect to MCDM
method use is to combine two or more methods to make up for shortcomings in a
single method.17 For instance Arikan and Kucukce18 applied AHP and PRO-
METHEE II to minimize economic losses from the inadequate assessment of sup-
pliers in medical ¯eld. While, Santos and Garcia19 used AHP and ELECTRE to
acquire hospital medical equipment. In the present paper our aim is to present an
integrated MCDM approach based on the AHP and TOPSIS method applied within
the medical ¯eld. In fact, in the literature several applications exist in which AHP
and TOPSIS are frequently used to solve real-world problems.20 But literature does
not provide studies that directly focus on tomography equipment selection. Thus,
the novelty of this research is to develop a methodological tool, based on AHP-
TOPSIS methods, to the ¯eld of medicine which represents a strategic ¯eld for the
development of the human community. Through the proposed tool is possible to
An AHP-Topsis Integrated Model 3
decision analysis method for the following reasons: the ability to provide complete
ranking results; the suitability to be combined with stochastic analysis; the use of
Int. J. Info. Tech. Dec. Mak. Downloaded from www.worldscientific.com
weights and objective data to calculate the relative distances; the smoothing of
tradeo®s by dealing with nonlinear relationships and the ease at which it can be
converted into a programmable procedure. The primary goal of our research is to
provide a useful decision-making tool to develop a model to consider the decision
makers' (DMs') preferential system and judgments. As stated by Tavana and
Hatami-Marbini23 the selection of an MCDM framework or method should be done
carefully according to the nature of the problem, types of choices, measurement
scales, dependency among the attributes, type of uncertainty, expectations of the
DMs, and quantity and quality of the available data and judgments. For the above
reasons in the present paper an AHP/TOPSIS integrated model is presented. Ad-
ditionally, the AHP is able to reduce the complexity of decision making in a reliable
way, support both a single person or a group decision and usually does not require
the presence of a specialist.24 While, the main objective of TOPSIS is to provide
ranking of alternatives that ¯nd the optimal alternative which should be nearest to
the positive ideal solution and furthest from the negative ideal solution.25 As
highlighted above, in addition to the single MCDM approach, there are some inte-
grated approaches studied in alternative selection problems in the literature. How-
ever, there is no evidence of any applications in the medical ¯eld. An example of an
application in the literature is given by Zaidan et al.,26 they evaluated and selected a
health informatics open-source EMR software package-based MCDM using inte-
grated AHP–TOPSIS in group context of the available options of open-source EMR
software. Other applications in di®erent and various ¯elds include: Misra and Ray27
proposed an integrated AHP–TOPSIS model for software selection and a multi-
criteria decision-making evaluation was de¯ned. The model used AHP and TOPSIS.
Nwokoagbara et al.28 identi¯ed the best micro algae strain for viable biodiesel pro-
duction by using an AHP and TOPIS method. Furthermore, some authors propose a
Fuzzy AHP-TOPSIS method such as Ref. 29. They proposed a Fuzzy Analytic
Hierarchy Process (Fuzzy AHP) and fuzzy technique in order of preference by
similarity to the ideal solution (fuzzy TOPSIS) to successfully determine the priority
weights of multiple criteria and selects the ¯ttest suppliers by taking the vagueness
and imprecision of human assessments into consideration. Beikkhakhian et al.,30
propose the application of a model in evaluating agile supplier selection criteria and
4 M. A. O. Barrios et al.
ranking suppliers using fuzzy TOPSIS–AHP methods. Taylan et al.31 applied Fuzzy
AHP to create appropriate weights for ¯ve main criteria, i.e., time, cost, quality,
safety, and environmental sustainability and then use Fuzzy Technique for Order of
Preference by Similarity to Ideal Solution (Fuzzy TOPSIS) to rank the 30 con-
struction projects on the basis of the opinions of seven DM in various sectors in Saudi
Arabia. Ren and Sovacool32 propose an integrated method based on AHP–TOPSIS
has been developed to rank the low-carbon energy systems from most to least im-
portant, with major implications for Chinese DMs and stakeholders. De¯nitively,
there are applications in several sub¯elds and activities, but healthcare applications
by NANYANG TECHNOLOGICAL UNIVERSITY on 07/17/16. For personal use only.
are not addressed. Thus, the methodological approach presented in this paper also
tries to cover this gap. In detail, in the proposed methodological approach to select
Int. J. Info. Tech. Dec. Mak. Downloaded from www.worldscientific.com
2. Hierarchy De¯nition
The project was discussed with the maintenance boss and the ethics committee of a
clinic. This research was carried out through interviews with all the departments
involved with tomography equipment. In Fig. 1 is shown the tomography equipment,
understudy.
The research did not involve patient participation. The participating depart-
ments, as a decision-making team, identi¯ed a total of three purchase options that
satisfy the requirements of the clinic considering patient pro¯les and internal process
analysis. For the hierarchy design, ¯ve criteria, 17 sub-criteria, and a goal were
identi¯ed in order to take into account the needs of this organization and health care
sector.33–35
The AHP model was checked and discussed during multiple meetings in order
to verify that it was understandable and clear. The AHP model is presented in
Fig. 2.
by NANYANG TECHNOLOGICAL UNIVERSITY on 07/17/16. For personal use only.
The aim of the AHP model is to ensure both an appropriate investment and the
protection of patients. In detail, ¯ve criteria were identi¯ed: PERFORMANCE (C1),
Int. J. Info. Tech. Dec. Mak. Downloaded from www.worldscientific.com
Fig. 2. Hierarchy for selection of the best purchase option at the moment of acquiring tomography
equipment in clinics.
6 M. A. O. Barrios et al.
ASPECTS cover all the aspects in reference to the physical structure of the medical
equipment.
Below is a description for each sub-criteria. In particular, the PERFORMANCE
criterion (C1) is made up of ¯ve sub-criteria: SPATIAL RESOLUTION (S1), RUN
SPEED (S2), PROCESSING SOFTWARE (S3), IMAGE RESOLUTION (S4), and
NUMBER OF CUTS (S5). First, SPATIAL RESOLUTION is described as the
ability of the tomography equipment to discriminate between two adjacent high-
contrast objects. RUN SPEED is de¯ned as the velocity of the equipment at the
moment of scanning; while PROCESSING SOFTWARE deals with the software
by NANYANG TECHNOLOGICAL UNIVERSITY on 07/17/16. For personal use only.
TOPSIS technique is used to evaluate the purchase options with basis of quantitative
information through which criteria and sub-criteria are measured. As a ¯nal point,
Int. J. Info. Tech. Dec. Mak. Downloaded from www.worldscientific.com
the purchase option with the highest closeness coe±cient is chosen as the best al-
ternative for the clinical needs. Figure 3 shows the framework of the proposed
methodology.
After forming the decision hierarchy, criteria and sub-criteria weights are calcu-
lated using the AHP. Technically, by using the AHP one determines the relative
Fig. 3. Proposed methodology for the selection of the most appropriate purchase for tomography
equipment in a clinic.
8 M. A. O. Barrios et al.
. Make pairwise comparison matrices for both the criteria and sub-criteria. In this
study, a three point scale has been used as previous studies37–39 because it has been
demonstrated that most responders do not use more than three judgments (equal,
more much more) and lay users reported puzzled at the moment of using a more
complex scale.40–43
. Determine the relative normalized weight ðW j Þ of each criterion/sub-criterion by
by NANYANG TECHNOLOGICAL UNIVERSITY on 07/17/16. For personal use only.
hY N i1=N
GM j ¼ a
j¼1 ij
; ð3:1Þ
,
X
N
W j ¼ GM j GM j : ð3:2Þ
j¼1
A3 ¼ A1 A2; ð3:3Þ
A3
A4 ¼ : ð3:4Þ
A2
where
A2 ¼ W 1 ; W 2 ; . . . ; W j T ð3:5Þ
. Determine the maximum eigenvalue which is the mean of matrix A4.
. Calculate the consistency index (CI).
. Calculate the consistency ratio (CR).
4. Survey Design
The survey was carried out to allow each participant to determine the relative
importance of each criterion with respect to the rest of the criteria; and for each sub-
criterion with respect to all other sub-criteria. The survey design is illustrated in
Fig. 4.
For each pairwise comparison it was asked: In the selection of tomography
equipment, according to your expertise, how relevant is RADIATION DOSE with
respect to EASE OF PATIENT MONITORING ? Participants responded using the
three-level scale previously mentioned. This process was then repeated, creating
similar surveys to determine the relative importance of each criterion and sub-cri-
terion. This design was taken to reduce confusion while ¯lling it out and to eliminate
intransitive judgments.
An AHP-Topsis Integrated Model 9
. One Electronic engineer, the head of the Maintenance Department, with more
than seven years of experience in the clinic.
. Three Radiologists, with more than 10 years of experience in contact with
tomography equipment.
. One Administrative and Financial Management personnel with signi¯cant expe-
rience in leading these types of companies.
. One Industrial engineer from the University with a master in integrated logistics
and with extensive knowledge and experience in the design of decision-making
models in healthcare systems.
These participants made the comparisons with the support of the designed sur-
vey. Each respondent was required to answer for the department in which they were
working. It is also worth noting that none of the participants from the clinic an
author of this paper.
During the pairwise comparisons process, some loss of interest or distractions may
appear, which will generate inconsistencies in the decision-making process.44 If this
Int. J. Info. Tech. Dec. Mak. Downloaded from www.worldscientific.com
occurs, the comparisons must be made again by the participants because incon-
sistencies reduce the level of reliability in the decision making. Although, some in-
consistency could be present, it is necessary to calculate it in order not to overpass
the limit value. For this, respondent's consistency is determined by consistency index
(CI). This indicator is divided by random index (RI) whose values depend on the
matrix size. The resulting division is named as a consistency ratio (CR) which should
not overpass 0.1 to consider judgments as appropriate.45–49
5. TOPSIS Approach
Int. J. Info. Tech. Dec. Mak. Downloaded from www.worldscientific.com
This technique is based upon the concept that the chosen candidate (alternative)
should be the shortest distance from the positive ideal solution and the farthest from
the negative ideal solution.51
The steps of TOPSIS are given as below:
. Create an assessment matrix by considering \y" purchase options and \z" sub-
criteria with the intersection of each purchase option and sub-criteria given as X ij ;
then the resulting matrix is ðX ij Þ yxz .
2 3
S S S
P1 6 x 1 x 2 x z 7
P2 6
11 12 1z 7
6 x 21 x 22 x 2z 7
6 7
6 7
D ¼ P3 6 7 ð5:1Þ
.. 66 x 31 x 32 x 3z 7
. 6. 7
.. .. 7
4 .. . . 5
Py
x y1 x y2 x yz
. Normalize the judgment matrix (X ij Þ yxz to transform it into R ¼ ðn ij Þ yxz using the
following formula:
x ij
n ij ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Py ffi: ð5:2Þ
x 2
i¼1 ij
where:
J ¼ fj ¼ 1; 2; . . . ; zjj associated with the benefit sub criteriag;
by NANYANG TECHNOLOGICAL UNIVERSITY on 07/17/16. For personal use only.
CC i ¼ 1 if A i ¼ Aþ ;
CC i ¼ 0 if A i ¼ A :
a 3-level scale where (0) represents low quality (1) medium quality, and (2) high
quality. PROCESSING SOFTWARE (S5) is de¯ned according to the existence of
software that permits data management. Thus, it is quali¯ed with (0) if it does not
have processing software and (1) if it has it. RADIATION DOSE (S6) is measured in
MHU (Mega heat units). EASE OF PATIENT MONITORING (S7) is dimensioned
in s/rotation. CONTRAST CONTROL (S8) is de¯ned in mm. APPLICATION
SOFTWARE (S9) is measured taking into account the presence of this software.
Therefore, it is judged with (0) if tomography equipment does not have application
software and (1) if it has it. INTEROPERATIVITY (S10) is evaluated according to
by NANYANG TECHNOLOGICAL UNIVERSITY on 07/17/16. For personal use only.
the capability that tomography equipment has to be connected and exchange data
with an information system. In this way, if the equipment does not have connections
Int. J. Info. Tech. Dec. Mak. Downloaded from www.worldscientific.com
with \IS" is quali¯ed with (0); otherwise with (1). AUTOMATIC HANDLING (S11)
is assessed with (0) if the equipment cannot be managed automatically; otherwise
with (1). By the other side PRICE (S12) and MAINTENANCE COST (S13) are
measured in USD. EASE OF PAYMENT (S14) is de¯ned in terms of advance
payment percentage (%). INFORMATION STORING (S15) is measured in GB
(Gigabytes); ENERGY CONSUMPTION (S16) is de¯ned in KW (kilowatts) and
¯nally, EASE OF MAINTENANCE (S17) is measured in terms of year guarantee. It
is good to remark that all this information was collected from the speci¯cations
established in the webpage of the manufacturers.
In the next step, by using Eq. (5.2), the normalized decision matrix is calculated.
After this, a weighted normalized matrix is determined using Eq. (5.3). Then, by
using Eqs. (5.4) and (5.5), respectively, positive and negative ideal solutions are set.
Next, it is necessary to determine the distance of the purchase options from the
positive ideal solution and the negative ideal solution. These distances are calculated
by employing Eqs. (5.6) and (5.7) correspondingly. Finally, the closeness coe±cients
of the purchase options are de¯ned using Eq. (5.8).
sub-criterion was RADIATION DOSE with 0.38483. However, with a similar score
(0.3843), CONTRAST CONTROL got the second place with respect to the de¯ni-
tion of this criterion. In the TECHNOLOGY cluster APPLICATION SOFTWARE
represented more than 50% of the criterion importance with 0.55669 which shows the
high relevance of these kinds of software in tomography operation. On the other
hand, in FINANCIAL ASPECTS cluster, EASE OF PAYMENT was chosen as
the most signi¯cant sub-criterion with almost 50% of the total criterion relevance
Fig. 5. Sub-criteria weights in a decision-making model designed for the purchase of tomography
equipment.
An AHP-Topsis Integrated Model 15
by NANYANG TECHNOLOGICAL UNIVERSITY on 07/17/16. For personal use only.
Int. J. Info. Tech. Dec. Mak. Downloaded from www.worldscientific.com
Fig. 6. Criteria contributions in a decision-making model designed for the purchase of tomography
equipment.
Criteria 0.01860
Performance 0.05623
Patient safety 0.09989
Technology level 0.00094
Financial aspects 0.00107
Technical aspects 0.00198
16 M. A. O. Barrios et al.
Table 9 presents the TOPSIS decision matrix where purchase options with re-
spect to criteria are shown. Sub-criteria values for each purchase option were entered
according to the description in Sec. 5.1. Positive and negative ideal solutions for each
sub-criterion have been de¯ned. The sub-criteria weights calculated through AHP
are also presented in this table as a link between the two MCDM methods. Table 10
denotes the normalized TOPSIS decision matrix according to Eq. (5.2) Table 11
shows a weighted, normalized decision matrix for TOPSIS. Closeness coe±cients and
a rank of purchase options from TOPSIS are shown in Fig. 7. On the other hand,
Fig. 8 and Table 12 describe the separation of each purchase option from the positive
ideal solution. Speci¯cally, Table 12 indicates the contribution of each sub-criterion
to the total separation from positive ideal solution. Meanwhile, Fig. 9 and Table 13
present the separation from the ideal negative solution. In detail, Table 13 shows the
speci¯c contribution of each sub-criterion to the total separation from negative ideal.
In this way, it can be noted which sub-criteria contribute to each measure.
Figure 7 shows that purchase option \P2" that corresponds to LIGHTSPEED 16
CUTS of the manufacturer \GENERAL ELECTRIC" obtained the ¯rst place of the
ranking with 0.6582 and purchase option \P3" got the lowest score with 0.3872. The
comparative view of closeness coe±cient of purchase options is seen in the ¯gure
Int. J. Info. Tech. Dec. Mak. Downloaded from www.worldscientific.com
by NANYANG TECHNOLOGICAL UNIVERSITY on 07/17/16. For personal use only.
Table 10. Normalized TOPSIS decision matrix for purchase options with respect to criteria.
P1 0.577 0.577 0.577 0.577 0.577 0.633 0.615 0.615 0.577 0.577 0.577 0.523 0.577 0.577 0.669 0.599 0.577
P2 0.577 0.577 0.577 0.577 0.577 0.498 0.492 0.615 0.577 0.577 0.577 0.751 0.577 0.577 0.666 0.531 0.577
P3 0.577 0.577 0.577 0.577 0.577 0.593 0.615 0.492 0.577 0.577 0.577 0.402 0.577 0.577 0.330 0.599 0.577
Aþ 0.577 0.577 0.577 0.577 0.577 0.498 0.492 0.615 0.577 0.577 0.577 0.402 0.577 0.577 0.669 0.531 0.577
A− 0.577 0.577 0.577 0.577 0.577 0.633 0.615 0.492 0.577 0.577 0.577 0.751 0.577 0.577 0.330 0.599 0.577
W 0.03895 0.04192 0.03909 0.04998 0.04343 0.15669 0.0645 0.13098 0.05823 0.03966 0.04211 0.04641 0.04699 0.05549 0.04181 0.0398 0.06394
Table 11. Weighted and Normalized TOPSIS decision matrix for purchase options with respect to criteria.
P1 0.022 0.024 0.023 0.029 0.025 0.099 0.040 0.081 0.034 0.023 0.024 0.024 0.027 0.032 0.028 0.024 0.037
P2 0.022 0.024 0.023 0.029 0.025 0.078 0.032 0.081 0.034 0.023 0.024 0.035 0.027 0.032 0.028 0.021 0.037
P3 0.022 0.024 0.023 0.029 0.025 0.093 0.040 0.064 0.034 0.023 0.024 0.019 0.027 0.032 0.014 0.024 0.037
Aþ 0.022 0.024 0.023 0.029 0.025 0.078 0.032 0.081 0.034 0.023 0.024 0.019 0.027 0.032 0.028 0.021 0.037
A 0.022 0.024 0.023 0.029 0.025 0.099 0.040 0.064 0.034 0.023 0.024 0.035 0.027 0.032 0.014 0.024 0.037
An AHP-Topsis Integrated Model 19
by NANYANG TECHNOLOGICAL UNIVERSITY on 07/17/16. For personal use only.
20
P1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0004 0.0001 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0234
P2 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0003 0.0000 0.0000 0.0000 0.0000 0.0000 0.0162
P3 0.0000 0.0000 0.0000 0.0000 0.0000 0.0002 0.0001 0.0003 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0002 0.0000 0.0000 0.0274
P1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0003 0.0000 0.0000 0.0000 0.0001 0.0000 0.0000 0.0002 0.0000 0.0000 0.0239
P2 0.0000 0.0000 0.0000 0.0000 0.0000 0.0004 0.0001 0.0003 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0002 0.0000 0.0000 0.0312
P3 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0003 0.0000 0.0000 0.0000 0.0000 0.0000 0.0173
An AHP-Topsis Integrated Model 21
by NANYANG TECHNOLOGICAL UNIVERSITY on 07/17/16. For personal use only.
Int. J. Info. Tech. Dec. Mak. Downloaded from www.worldscientific.com
possible solution has a few sub-criteria out of this constraint, it will not be chosen as
the most suitable. In conclusion, upon analyzing both scenarios P2 is considered
as the best because the solution adjusts to the positive ideal solution and separates
from the negative solution.
7. Conclusions
In recent years, because of the ease of use due to advancing technologies, combining
di®erent methods has become commonplace within multi criteria decision analysis.
The combination of multiple methods addresses de¯ciencies that may be seen in
certain methods. This paper utilized two of the more common methods of MCDM to
allow practitioners to choose a method for solving a speci¯c problem. The identi¯-
cation of the most important criteria at the moment of purchasing tomography
equipment is a fundamental part of operational performance, patient safety and
¯nancial management of a clinic. The hybrid method AHP–TOPSIS described in this
paper not only allowed for the identi¯cation of the criteria, but was useful to select
the best purchase option using quantitative and qualitative information which cor-
responded to each option. Additionally, the AHP–TOPSIS method provided an
understandable framework for the decision-making process which is really important
in the healthcare sector where DMs need to explain their selections. Speci¯cally,
TOPSIS involves analysis in which each alternative is compared to a positive and a
negative ideal solution which allows for the identi¯cation of the purchase options
that are nearer to a positive ideal solution and farther from negative scenarios. This
paper has shown that in this case of tomography equipment, department perceptions
di®er signi¯cantly with low or medium correlation coe±cients. These outcomes have
important implications for the manufacturers of tomography equipment in order to
improve this machine and satisfy all departments' needs of a clinic. This paper also
describes a decision-making model that will be more e®ective if key criteria are
22 M. A. O. Barrios et al.
can be applied as a guideline for a variety of future selection problems, and similar
models can be established to address di®erent problems regarding the medical ¯eld.
Based on the results, the proposed model can be widely used for the evaluation of
various policies. For future research, a comparative study will be conducted by using
another MCDM methods such as VIKOR and ELECTRE in order to validate the
present results. Furthermore, a fuzzy ANP approach could be explored to identify
the interactions between criteria and sub-criteria.
References
1. I. Linkov, P. Welle, D. Loney, A. Tkachuk, L. Canis, J. B. Kim and T. Bridges, Use
of multicriteria decision analysis to support Weight of Evidence evaluation, Risk Anal.
31(8) (2011) 1211–1225.
2. H. Çaliskan, B. Kursuncu, C. Kurbanoglu and S. Y. Güven, Material selection for the tool
holder working under hard milling conditions using di®erent multi criteria decision
making methods, Materials & Design 45 (2013) 473–479.
3. J. Figueira, S. Greco and M. Ehrgott, Multiple Attributes Decision Analysis: State of the
Art Surveys (Springer, New York: 2005).
4. C. Kahraman, O. Engin, O. Kabak and I. Kaya, Information systems outsourcing deci-
sions using a group decision-making approach, Engineering Applications of Arti¯cial
Intelligence 22(6) (2009) 832–841.
5. J. M. Sanchez-Lozano, M. S. García-Cascales and M. T. Lamata, Evaluation of suitable
locations for the installation of solar thermoelectric power plants, Computers & Industrial
Engineering 87 (2015) 343–355.
6. M. Velasquez and P. T. Hester, An analysis of multi-criteria decision making methods,
International Journal of Operations Research 10(2) (2013) 56–66.
7. T. L. Saaty, The Analytical Hierarchy Process (McGraw-Hill, New York, 1980).
8. T. L. Saaty, Decision Making with Dependence and Feedback: The Analytic Network
Process (RWS Publications, Pittsburgh, Pennsylvania, 1986).
9. C. L. Hwang and K. Yoon, Multiple Attribute Decision Making: Methods and Applications
(Springer-Verlag, New York, 1981).
10. B. Roy, Classement et choix en presence de points de vue multiples (la methode
ELECTRE), La Revue d'Informatique et de Recherche Operationelle (RIRO) (1968)
57–75 (in French).
11. P. L. Yu, A Class of solutions for group decision problems, Management Science
19(8) (1973) 936–946.
An AHP-Topsis Integrated Model 23
12. J. P. Brans and P. Vincke, A preference ranking organisation method: The PRO-
METHEE method for MCDM, Management Science 31(6) (1985) 647–656.
13. F. De Felice and A. Petrillo, Decision–making analysis to improve public participation in
strategic energy production management, in Studies in Fuzziness and Soft Computing,
eds. A. G. S. Ventre, A. Maturo, S. Hoskova-Mayerova and J. Kacprzyk (Springer Verlag,
2013), pp. 129–142.
14. G. Balestra, M. Kna°itz, R. Massa and M. Sicuro, AHP for the acquisition of biomedical
instrumentation, in Annual Int. Conf. IEEE Engineering in Medicine and Biology Society
January 2007, IEEE Lyon, pp. 3581–3584.
15. R. L. Chatburn and F. P. Primiano, Decision analysis for large capital purchases: how to
buy a ventilator, Respiratory Care 46(10) (2001) 1038–1053.
by NANYANG TECHNOLOGICAL UNIVERSITY on 07/17/16. For personal use only.
16. K. T. Cho and S. M. Kim, Selecting medical devices and materials for development in
Korea: the analytic hierarchy process approach, The International Journal of Health
Int. J. Info. Tech. Dec. Mak. Downloaded from www.worldscientific.com
51. M. Pal and K. Choudhury, Exploring the dimensionality of service quality: An applica-
tion of TOPSIS in the Indian banking industry, Asia-Paci¯c Journal of Operational
Research 26(1) (2009) 115–133.