Expert Systems with Applications 36 (2009) 8639–8648

Contents lists available at ScienceDirect

Expert Systems with Applications

journal homepage: www.elsevier.com/locate/eswa

An integrated fuzzy model for supplier management: A case study of ISP

selection and evaluation
Saman Hassanzadeh Amin a,*, Jafar Razmi b
a
Department of Industrial Engineering, Islamic Azad University, South Tehran Branch, First Floor, No. 209, North Iranshahr Str., Tehran 11365/4435, Iran
b
Department of Industrial Engineering, Faculty of Engineering, University of Tehran, Iran

a r t i c l e i n f o a b s t r a c t

Keywords:
Supplier selection and evaluation
Internet service provider (ISP)
Quality function deployment (QFD)
Fuzzy set theory
Supplier selection is a multi-criteria decision-making problem which consists of both qualitative and
quantitative metrics. A lot of investigations have been published in the supplier selection area and it
has been notified that in the majority of these publications supplier selection and evaluation and devel-
opment have the same meaning. However, one needs integrated models to cover all of these stages. In
addition, most of the proposed models focused on manufacturing environments and a few papers have
been allocated for service industries. To our knowledge, no Internet service provider (ISP) selection
and evaluation has been published up to now.
In this paper, we propose a new framework on the basis of company’s strategy for supplier manage-
ment including supplier selection, evaluation, and development. In the first phase, quality function
deployment (QFD) is utilized to rank the best ISPs based on qualitative criteria. Then, a quantitative
model is adopted to consider quantitative metrics. Finally, we compose two models and select the best
ISPs. In the next phase, we propose a novel algorithm to evaluate selected ISPs from three perspectives:
customer, performance, and competition. Meanwhile, the fuzzy logic and triangular fuzzy numbers are
utilized to deal with vagueness of human thought. Furthermore, a case study is conducted to illustrate
the stages of ISP selection and evaluation. The implementation of the proposed model is easy and do
not need optimization background.
Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction The objective of supplier selection is to identify suppliers with

Supply chain management (SCM) has generated a substantial
amount of interest both among managers and researchers. SCM
is now distinguished as a governing element in strategy and as
an effective way of creating value for customers. A structure of
supply chain is composed of potential suppliers, producers, distrib-
utors, retailers and customers, etc. (Fiala, 2005). Therefore, suppli-
ers play an important role in achieving the objective of supply
management. They enhance customer satisfaction in a value chain.
Hence, strategic partnership with better performing suppliers
should be integrated within the supply chain for improving the
performance in many directions including reducing costs by elim-
inating wastages, continuously improving quality to achieve zero
defects, improving flexibility to meet the needs of the end-custom-
ers, reducing lead time at different stages of the supply chain, etc.
(Kumar, Vrat, & Shankar, 2004). Manufacturing cannot forecast de-
mand precisely, and they prefer to manage suppliers via different
methods such as supplier development, supplier evaluation, sup-
plier selection, supplier coordination, etc. (Chou & Chang, 2008).
* Corresponding author. Tel.: +98 2188713737; fax: +98 2188013102.
E-mail address: s_hassanzadeh_a@yahoo.com (S.H. Amin).
the highest potential for meeting a firm’s needs consistently and
at an acceptable cost (Kahraman, Cebeci, & Ulukan, 2003). In the
supplier selection process a pool of suppliers is chosen for procure-
ment according to a predefined set of criteria (Aissaoui, Haouari, &
Hassini, 2007). While several methods have been proposed and uti-
lized for evaluation and selection of suppliers, they have limita-
tions including: evaluation solely based on operational metrics,
simple weighted scoring methods based on subjective assess-
ments, and lack of relative evaluation across various suppliers (Tal-
luri & Narasimhan, 2004). In addition, the majority of previous
investigations have focused on supplier selection or supplier eval-
uation or supplier development separately. We aim to develop an
integrated model for supplier management and fill the gap in the
literature. Besides, the new model combines both qualitative and
quantitative criteria simultaneously. Moreover, unlike the case of
parts purchasing, there is only a limited number of studies that
model the service supplier selection problem (Aissaoui et al.,
2007; De Boer, Labro, & Morlacchi, 2001; Degraeve, Labro, & Roo-
dhoofi, 2004). In this article, we propose a novel model for Internet
service provider (ISP) selection and evaluation which can be ap-
plied for most of the service sectors. In addition, most published
models in this area focus only on the customer perspective or

0957-4174/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.eswa.2008.10.012
8640 S.H. Amin, J. Razmi / Expert Systems with Applications 36 (2009) 8639–8648
supplier’s performance perspective, or competition situation, and
hence these models pay no attention to all of these factors simul-
taneously. In order to solve the above shortages we propose a novel
method that considers competition position, current performance
and customers’ viewpoint concurrently.
This paper proposes a decision model that consists of two
phases. In the first phase, QFD model will be integrated with a
quantitative model to select the best ISPs. In the second phase,
the selected ISPs will be evaluated based on a novel algorithm. Fur-
thermore, special attention is paid to the various subjective assess-
ments in the selection and evaluation process by using triangular
fuzzy numbers. Meanwhile, a case study is conducted to illustrate
the proposed model. The model can be easily implemented with
spreadsheet packages.
The organization of this paper is as follows. Section 2 discusses
the literature review. Quality function deployment is introduced in
Section 3. Section 4 describes fuzzy set theory. Next, case review
(case of suppliers’ selection in ISP service sector) is presented in
Section 5. Section 6 introduces ISP selection approach. Subse-
quently, Section 7 presents ISP evaluation process. Furthermore,
sensitivity analysis is presented in Section 8. Finally, conclusions
are presented in Section 9.
2. Literature review
Supplier selection is one of the most widely researched areas in
purchasing with methodologies ranging from conceptual to empir-
ical and modeling streams (Talluri & Narasimhan, 2004). The exis-
tent literature surveys focuses on papers starting from the mid
1960s. The earliest review is by Moore and Fearon (1973) where
they focused on industry applications of computer-assisted sup-
plier selection models. Kingsman (1986) highlighted the inade-
quacy of classical inventory management models for tracking
purchasing decisions and described some of the major models that
were used for the different purchasing decision-making stages.
Holt (1998) reviewed and compared several decisional methods
applied in contractor evaluation and selection. De Boer et al.
(2001) did not restrict the review to the final choice models. They
recognized the prior steps to the ultimate stage and presented the
main published works that deal with all the supplier selection pro-
cess. These stages are problem formulation, formulation of criteria,
qualification and final selection. Recently, Aissaoui et al. (2007)
have presented a literature review that covers the entire purchas-
ing process. They proposed different classifications of the pub-
lished models based on single and multiple periods and Items.
2.1. Criteria selection
Determining the criteria is the first step in supplier selection
process. Dickson (1966) identified 23 different criteria based on a
questionnaire sent to 273 purchasing agent and managers from
United States and Canada. The most prominent ones were quality,
delivery, performance history, warrant and claim policy, produc-
tion facilities and capacity, net price, and technical capabilities. Ell-
ram (1990) applied hierarchy framework including financial,
performance, technology, organizational culture and strategy, and
other factors. Weber, Current, and Benton (1991) selected price,
delivery, quality, facilities and capacity, geographic location, and
technology capability as the most important factors in supplier
selection. Mandal and Deshmukh (1994) proposed interpretive
structural modeling (ISM) as a technique based on group judgment
to identify and summarize relationships between supplier choice
criteria through a graphical model. They suggested it aids the pur-
chaser by separating dependent criteria from independent criteria.
Vokurka, Choobineh, and Vadi (1996) developed an expert system
that covers multiple phases in the supplier selection process,
among which the formulation of supplier selection criteria. Other
(non-experts) users may consult the system to obtain suggestions
as to which criteria to use in a particular situation. Ghodsypour and
O’Brien (1998) stated that cost, quality, and service have consider-
able effects on supplier selection parameters. Kahraman et al.
(2003) mentioned that selection criteria typically fall into one of
four categories: supplier criteria, product performance criteria, ser-
vice performance criteria, and cost criteria. Humphreysa, Wong,
and Chan (2003) integrated environmental criteria into the sup-
plier selection process. Bottani and Rizzi (2006) presented a mul-
ti-attribute approach for selection and ranking of the most
suitable 3PL service provider. They applied service criteria such
as breath of service, business experience, characterization of ser-
vice, compatibility, financial stability, flexibility of service, perfor-
mance, price, physical equipment and information, quality,
strategic attitude, trust and fairness. Recently, Huang and Keskar
(2007) presented basic structure with seven categories for select-
ing the criteria based on the firms’ strategy. In addition, they orga-
nized these categories into three tracks such as product related,
supplier related, and society related. Furthermore, they deter-
mined sub criteria for each one. Demirtas and Ustun (2008) defined
14 different criteria under Benefits, Opportunities, Costs, and Risks
(BOCR) merits.
2.2. Technique selection
Table 1 depicts the supplier selection techniques including ana-
lytic hierarchy process (AHP), multi-objective programming
(MOP), expert systems, data envelopment analysis (DEA), mixed
integer programming (MIP), goal programming (GP), mathematical
programming (MP), linear weighting (LW), total cost of ownership
(TCO), genetic algorithm (GA), analytic hierarchy process (ANP),
case-based reasoning (CBR), data mining (DM), cluster analysis
(CA), activity based costing (ABC), TOPSIS, rough sets theory
(RST), grey approach, neural network (NN), and quality function
deployment (QFD). It can be inferred from Table 1 that in recent
publications, the majority of authors have combined at least two
techniques together. In addition, fuzzy logic has been used because
supplier selection involves subjective, vague and imprecise assess-
ments, which are by nature fuzzy. Fuzzy assessment expressed in
linguistic terms is often intuitive and effective for DMs during
the assessment process (Chou & Chang, 2008). However, in the
most of these papers, the supplier selection and supplier evalua-
tion have similar meaning, and therefore studies become unrealis-
tic for practical decision-makers. In other words, practitioners may
need integrated models that can take into account three stages:
supplier selection, supplier evaluation, and supplier development.
Application of quality function deployment (QFD) as a tool for
supplier selection has been limited. To date there have been few
works that have applied this tool for supplier evaluation purposes.
Ansari and Modarress (1994) discussed the roles of suppliers in the
various phases of QFD. Holmen and Kristensen (1998) illustrated
how the house of quality (HOQ) can be used in the pre-interactive
stage of a single product development project, and how the identi-
fied correlations and non-correlations between the characteristics
of the planned product can be used by a customer as a practical ap-
proach for discrimination between the three types of suppliers.
Temponi, Yen, and Tiao (1999) developed a fuzzy logic-based
extension to HOQ for capturing imprecise requirements to both
facilitate communication of team members and have a formal rep-
resentation of requirements. Recently, Bevilacqua, Ciarapica, and
Giacchetta (2006) suggested a new method that transfers the
house of quality approach typical of QFD problems to the supplier
selection process by using triangular fuzzy numbers. However,
their work only considered qualitative criteria and did not address
 S.H. Amin, J. Razmi / Expert Systems with Applications 36 (2009) 8639–8648 8641

Table 1
Supplier selection techniques.

Techniques Authors Techniques Authors

AHP, MP Ghodsypour and O’Brien (1998) LW Dulmin and Mininno (2003)
AHP, MOP Kokangul and Susuz (in press) Ng (2008)
AHP, fuzzy Kahraman et al. (2003) Fuzzy, ABC Dogan and Sahin (2003)
Chan and Kumar (2007) Fuzzy, TOPSIS Bottani and Rizzi (2006)
Lee (in press) Fuzzy, MOP Amid, Ghodsypour, and O’Brien (2006)
AHP, grey approach Yang and Chen (2005) Amid, Ghodsypour, and O’Brien (in press)
AHP, DEA, TCO Ramanathan (2007) Fuzzy, GP Kumar et al. (2004)
AHP, MOP, MIP, RST Xia and Wu (2007) Kumar, Vrat, and Shankar (2006)
AHP, DEA, NN Ha and Krishnan (2008) Araz, Ozfirat, and Ozkarahan (2007)
DEA Braglia and Petroni (2000) Fuzzy, QFD Bevilacqua et al. (2006)
Talluri and Narasimhan (2004) Fuzzy, GA wang (2008)
Saen (2007) Fuzzy, CA Bottani and Rizzi (2008)
DEA, MOP Weber, Current, and Desai (2000) Grey approach Li, Yamaguchi, and Nagai (2007)
DEA, TCO Garfamy (2006) ANP Gencer and Gürpinar (2007)
MP, TCO Degraeve et al. (2004) ANP, MOP, MIP Demirtas and Ustun (2008)
MIP Ghodsypour and O’Brien (2001) Expert systems Kwong, Ip, and Chan (2002)
Kheljani, Ghodsypour, and O’Brien (in press) Jain, Tiwari, and Chan (2004)
MIP, CBR Faez, Ghodsypour, and O’Brien (in press) Ohdar and Ray (2004)
MIP, DM Feng, Wang, and Wang (2001) Chen, Lin, and Huang (2006)
Hong, Park, Jang, and Rho (2005) Lopez (2007)
MOP Dahel (2003) Chou and Chang (2008)
MOP, GA Liu and Hai (2005) MOP, GP Wadhwa and Ravindran (2007)
Liao and Rittscher (2007)

quantitative metrics. In addition, they supposed equal weights for
decision-makers. But, expertise, experience, authority, and the
responsibilities of different decision-makers (DMs) regarding spe-
cific attributes and alternative rating are not equal and uniform.
2.3. Supplier selection in service industries
The vast majority of the publications have been written in the
context of selecting a supplier for the purchase of a product to be
used in a manufacturing environment. From a point of view of
reflecting purchasing significance in sectors other than manufac-
turing, e.g. service industry, it would be worthwhile to investigate
and illustrate the specific of using decision methods in supplier
selection in those areas as well (Aissaoui et al., 2007; De Boer
et al., 2001). The major difference between parts and services pur-
chasing is that services cannot be ‘stored’ and so there are no
inventory costs associated with service purchasing (Aissaoui
et al., 2007). Degraeve and Roodhooft (2000) proposed an effective
methodology to the service sector in developing an airline selec-
tion model for the procurement of business travel. Oliveira and
Lourenço (2002) discussed the problem of selecting suppliers for
the construction of pipeline networks for gas distribution. They
developed a multi-source and multi-period model that allocated
construction orders to a pool of pre-qualified set of suppliers.
Degraeve et al. (2004) used the concept of a total cost of ownership
to select airlines for a major company. They developed a large com-
plex mixed integer program that accounts for several airline fare
discounting schemed. Klundert, Kuipers, Spieksma, and Winkels
(2005) reported on a model for selecting international telecommu-
nication carriers for a major telecommunication service provider.
They accounted for volume discounts and showed that a special
case of their model results in a min-cost flow model. Bottani and
Rizzi (2006) presented a multi-attribute approach for the selection
and ranking of the most suitable 3PL service provider. To our
knowledge, no supplier selection model for ISP selection and eval-
uation has been published.
3. Quality function deployment
Quality function deployment is a planning tool used to fulfill
customer expectations. It is disciplined approach to product de-
sign, engineering, and production and provides in-depth evalua-
tion of a product. An organization that correctly implements QFD
can improve engineering, productivity, and quality and reduce
costs, product development time, and engineering changes. QFD
focuses on customer expectations or requirements, often referred
to as the voice of customer. It is employed to translate customer
expectations, in terms of specific requirements, into directions
and actions, in terms of engineering or technical characteristics,
that can be deployed through: product planning, part develop-
ment, process planning, production planning, service industries
(Besterfield, Michna, Besterfield, & Sacre, 2003). Further details
on the benefits of using QFD can be found in (Griffin, 1992). The
primary planning tool used in QFD is the House of Quality. The
HOQ translates the voice of customer into design requirements
that meet specific target values and matches those against how
an organization will meet those requirements. The structure of
HOQ can be thought of as a framework of a house, as shown in
Fig. 1 (Besterfield et al., 2003).
4. Fuzzy set theory
Fuzzy sets are sets whose elements have degrees of member-
ship. Fuzzy sets have been introduced by Zadeh (1965) as an exten-
sion of the classical notion of set. FST permits the gradual
assessment of the membership of elements in a set; this is de-
scribed with the aid of a membership function l(x) valued in the
real unit interval [0, 1]. A major contribution of fuzzy set theory
is its capability of representing vague data. FST resembles human
reasoning in its use of approximate information and uncertainty
to generate decisions. It was specifically designed to mathemati-
cally represent uncertainty and vagueness and provide formalized
tools for dealing with the imprecision intrinsic to many problems
(Kahraman et al., 2003).
In this study, fuzzy opinions are represented as triangular fuzzy
numbers (TFNs) because they are intuitive and easy to use. A TFN Ã
can be defined as (a, b, c) which is shown in Fig. 2. Its membership
function lÃ(x) is defined as the following. In addition, the primary
operations for two TFNs are shown in the Appendix (Klir & Yuan,
1995),
8 xa
< ba ; a  x  b
>
leA ðxÞ ¼ cb
cx
; bxc
>
:
0; otherwise
8642 S.H. Amin, J. Razmi / Expert Systems with Applications 36 (2009) 8639–8648

(E)

(C) (A) Customer requirements (voice of the customer), (WHATs)

(B) Prioritized customer requirements
(C) Technical descriptors (Voice of the organization), (HOWs)
(A) (D) (B) (D) Relationship between requirements and descriptors
(E) Interrelationship between technical descriptors
(F) (F) Prioritized technical descriptors

Fig. 1. House of quality.

The majority of authors have proposed the supplier selection

µ (x)
1 grated models for supplier management. Thus, we propose a novel
0 a b c
Fig. 2. A triangular fuzzy number.
5. Case review
Nosaz is a service company in Iran that has been funded in
1969. The investment of this company has been estimated 3 mil-
lion dollars in 2003. Totally, 141 personnel work in Nosaz in three
branches. The central office is located in Tehran. Other personnel
work in two hotels which are located in Noor and Tabriz. The com-
pany purchases its Internet services from a famous ISP. An Internet
service provider (ISP) is a business or organization that provides
consumers or business access to the Internet and related services
(Rao, 2000). In the past, the most ISPs were run by the phone com-
panies. Now, ISPs can be started by just about any individual or
group with sufficient money and expertise. As a result, ISP selec-
tion is an important issue. Dial up technology and asymmetric dig-
ital subscriber line (ADSL) are two important types of Internet
services in Iran. If you want to connect to your office network,
download and exchange music and video files, you’ll want a speedy
broadband connection, such as ADSL (Barroso & Martinez, 2007).
Recently, new problems in connections have revealed. According
to the importance of Internet quality on the customer satisfaction
especially in hotels, the manager of company has decided to select
multiple ISPs, and design a performance assessment process. The
project is conducted by systems manager (DM1) in central office.
Furthermore, purchasing manager (DM2) and marketing manager
(DM3) contribute in decision-making process.
The main goal of this project is to answer two questions: ‘‘What
criteria should be used to measure the performance of ISPs?” and
‘‘Which techniques should be selected to consider both of qualita-
tive and quantitative criteria in ISP management process?”
and evaluation separately. However, in real cases we need inte-
framework for supplier management that includes three stages
which are shown in Fig. 3. Moreover, Ghodsypour and O’Brien
(1998) identified important criteria in supplier selection according
to one of five levels of buyer-supplier integration and strategy
including no integration, logistical integration, operational integra-
tion, integrating process and products with supplier, and business
partnership. Based on this category, we propose buyer–supplier
integration strategies which have been shown in Table 2. Accord-
ing to this table, the systems manager chooses the level 3. There-
fore, he selects supplier selection and evaluation strategy.
Moreover, there are seven eligible suppliers who sell ADSL services
in Tehran, Noor and Tabriz. The members of committee organize
meeting to determine the required criteria. They organize criteria
in two groups: service related and supplier related. Besides, sup-
plier related group is categorized to qualitative and quantitative
ones which have been shown in Fig. 4.
6. Phase 1: ISP selection
This section describes the ISP selection process which consists
of qualitative and quantitative criteria. The company buys trial
Internet connections from seven ISPs to collect the critical informa-
tion for selection process. Each of three DMs uses trial ADSLs to as-
sess the quality of connections. These ADSLs have two common
features. Traffic limit and usage time are unlimited, and the con-
nection speed is up to 512 kbps.
6.1. QFD model
In this section, the QFD model is applied to consider qualitative
criteria in supplier selection process. The QFD enables us to take
into account the customers’ judgments.
Step 1: Let U = {VL, L, ML, M, MH, H, VH} be the linguistic set used
to express opinions on the group of criteria. The linguistic vari-
ables of U can be quantified using triangular fuzzy numbers
(please refer to Fig. 5). Each of the three decision-makers estab-
lishes the level of importance or weight of each of WHATs (ser-

Supplier Management

Supplier Selection Supplier Evaluation Supplier Development

- Select the suppliers on the - Evaluate the suppliers based on - Assist suppliers by supplier
basis of performance and customer, performance, competition. development programs.
customer view point. - Provide feedback to suppliers
about their weaknesses.

Fig. 3. Supplier management framework.

 S.H. Amin, J. Razmi / Expert Systems with Applications 36 (2009) 8639–8648 8643

Table 2
Buyer–supplier integration strategies.

Levels of integration Description Strategy

1. No integration
2. Logistical integration
3. Operational integration
4. Integrating process and products
with supplier
5. Business partnership
Price and quality are important Supplier selection
Besides of price and quality, the operational logistics elements such as reliability and Supplier selection and evaluation
flexibility are important
The supplier’s process capability also should be considered Supplier selection and evaluation
Besides the above criteria the supplier’s human resource such as management ability Supplier selection and evaluation
should be considered
Besides technological and human resource view points, the strategic directions of the Supplier selection and evaluation and
supplier should be considered development

ISP Selection & Evaluation Criteria

Service Related Supplier Related

Qualitative Criteria Quantitative Criteria

WHATs HOWs

Effective marketing & Promotion (C5)

Accessibility (C1) Experience (C6) Monthly fee (C11)
Reliability (C2) Financial strength (C7) Supply variety b (C12)
Security (C3) Management stability (C8) Installation fee (C13)
Speed (C4) Strategic alliances (C9)
Support resource a (C10)

ISP1 ISP2 ISP3 ISP4 ISP ISP6 ISP7

a
An ISP must have technical support staffs that are available to respond to your calls, and solve Internet connection problems
in a short time.
b
Supply variety is the number of Internet services supplied by the ISP.

Fig. 4. ISP selection and evaluation hierarchy.

1 VL L ML M MH H VH VL Very Low (0, 0, 1)

L Low (0, 1, 3)
ML Medium Low (1, 3, 5)
M Medium (3, 5, 7)
MH Medium High (5, 7, 9)
H High (7, 9, 10)
0 1 2 3 4 5 6 7 8 9 10 VH Very High (9, 10, 10)

Fig. 5. A linguistic scale.

vice related criteria) by means of a linguistic variable. The
results are shown in Table 3.
Step 2: The experience, authority and responsiveness of DMs are
not equal in practice (Chou & Chang, 2008). Therefore, it is nec-
essary to determine the weights of DMs. Suppose the weight of
DMn is rn. This parameter can be calculated by linguistic vari-
ables. In this case study, experience index has been considered,
and it is supposed that a DM with more experience is more reli-
able than others. Experience index can be presented as linguis-
tic variables (Fig. 6): Poor = (0, 0, 10); Normal = (5, 10, 15);
Good = (10, 20, 20). In our case, r1 = (10, 20, 20), r2 = (5, 10, 15),
r3 = (5, 10, 15).

Poor Normal Good

Table 3
Weights of WHATs.

WHATs DM1 DM2 DM3

Accessibility (C1) H M H
Reliability (C2) M MH H
Security (C3) VH H H 0 5 10 15 20 years
Speed (C4) MH H MH
Fig. 6. A linguistic scale for experience.
8644 S.H. Amin, J. Razmi / Expert Systems with Applications 36 (2009) 8639–8648

Step 3: The weights assigned by decision-makers for customer Step 6: Now we can complete the HOQ, calculating the weights
requirement should be aggregated. Aggregated weight (wi), is of the HOWs (fj), averaging the aggregated weight for WHATs
calculated by the following equation, where I is the number of (wi), with the aggregated weight between WHATs and HOWs
WHATs (i = 1, 2, . . ., I), N is the number of decision-makers (aij), according to the Eq. (3). Again, these variables are triangu-
(n = 1, 2, . . ., N), and J is the number of HOWs (j = 1, 2, . . ., J). By lar fuzzy numbers. The fuzzy values are shown in the matrix F
using this equation, the aggregated weights are calculated in of Fig. 7,
Table 4. In our case, I = 4, N = 3, and J = 6.
1
fj ¼  ½ðw1  a1j Þ      ðwI  aIj Þ ð3Þ
wi ¼ ðr 1  wi1 Þ  ðr 2  wi2 Þ      ðr N  wiN Þ ð1Þ I

Step 4: Each DM was asked to express an opinion, using the lin-
guistic variables, on the impact of each HOWs on each WHATs.
Here again, triangular fuzzy numbers are used to quantify the
linguistic variables. The opinion expressed by the three deci-
sion-makers, on the impact of each HOWs on each WHATs are
shown in Table 5.
Step 5: Aggregated weight (aij) is calculated by the Eq. (2)
between WHATs and HOWs. Here again, the aij elements
are triangular fuzzy numbers (please refer to matrix D of
Fig. 7),
aij ¼ ðr 1  aij1 Þ  ðr 2  aij2 Þ      ðrN  aijN Þ
Step 7: It is necessary to assess each ISP based on the criterion in
question and combine said assessments with the weight of each
criterion in order to establish a final ranking. In the same way as
before, the linguistic variables are used to quantify by means of
triangular fuzzy numbers, then the SR (SR = supplier rating) are
aggregated based on the following equation:
SRhj ¼ ðr 1  sr hj1 Þ  ðr2  srhj2 Þ      ðr N  sr hjN Þ ð4Þ
where j = 1, 2, . . ., J, and h = 1, 2, . . ., H, which H is the number of
ISPs. Table 6 shows decision-makers opinions on the various
ISPs in relation to each HOW. By using Eq. (4), supplier rating
ð2Þ
is calculated.

Table 4
Aggregated weights.

WHATs r1  wi1 r2  wi2 r3  wi3 Aggregated weight

Accessibility (C1) (70, 180, 200) (15, 50, 105) (35, 90, 150) (120, 320, 455)
Reliability (C2) (30, 100, 140) (25, 70, 135) (35, 90, 150) (90, 260, 425)
Security (C3) (90, 200, 200) (35, 90, 150) (35, 90, 150) (160, 380, 500)
Speed (C4) (50, 140, 180) (35, 90, 150) (25, 70, 135) (110, 300, 465)

Table 5
The Impact of each HOW on each WHAT.

HOWs WHATs Effective marketing (C5) Experience (C6) Financial strength (C7)
DM1 DM2 DM3 DM1 DM2 DM3 DM1 DM2 DM3
Accessibility (C1) VH H MH H MH H H H VH
Reliability (C2) M M H M H VH MH M M
Security (C3) M ML M VH H H M MH H
Speed (C4) MH M M H MH H M ML MH
HOWs WHATs Management stability (C8) Strategic alliances (C9) Support resource (C10)
DM1 DM2 DM3 DM1 DM2 DM3 DM1 DM2 DM3
Accessibility (C1) ML L ML VH MH H VH VH H
Reliability (C2) H H H M MH M VH H H
Security (C3) H MH H M ML H ML M M
Speed (C4) M ML M H M H M L M

C5 C6 C7 C8 C9 C10

C1 (150, 360, 485) (130, 340, 485) (150, 370, 500) (15, 100, 220) (150, 360, 485) (170, 390, 500) (120, 320, 455)
C2 (80, 240, 395) (110, 290, 440) (80, 240, 390) (140, 360, 500) (70, 220, 380) (160, 380, 500) (90, 260, 425)
C3 (50, 180, 320) (160, 380, 500) (90, 260, 425) (130, 340, 485) (70, 220, 365) (40, 160, 310) (160, 380, 500)
C4 (80, 240, 390) (130, 340, 485) (60, 200, 350) (50, 180, 320) (120, 320, 455) (45, 160, 290) (110, 300, 465)

f1 f2 f3 f4 f5 f6

(10500, 79500, 182475) (11550, 84900, 192125) (12175, 88000, 194063)

(16350, 107650, 220800) (10175, 77200, 175975) (11538, 83100, 182463)

Fig. 7. The completed fuzzy-HOQ.

 S.H. Amin, J. Razmi / Expert Systems with Applications 36 (2009) 8639–8648 8645

Table 6
The impact of each potential ISP on the HOWs.

HOWs Effective marketing (C5) Experience (C6) Financial strength (C7)

DM1 DM2 DM3 DM1 DM2 DM3 DM1 DM2 DM3
ISP1 ML ML L H MH H H H MH
ISP2 M M MH M H M ML M M
ISP3 M ML M M MH ML VL VL L
ISP4 MH MH M M ML H ML M L
ISP5 M M M L ML L MH ML M
ISP6 H VH VH M MH M H MH H
ISP7 M M M MH M M M ML MH
HOWs Management stability (C8) Strategic alliances (C9) Support resource (C10)
DM1 DM2 DM3 DM1 DM2 DM3 DM1 DM2 DM3
ISP1 H H M VL L ML MH MH H
ISP2 MH M H H H VH ML M M
ISP3 L ML VL M M M VL VL L
ISP4 MH MH M L M ML M ML ML
ISP5 M MH MH M M M H H H
ISP6 L VL L H H MH M ML MH
ISP7 VH H H MH H H M MH M

Table 7 Table 8
Ranking of ISPs. Measures of suppliers under quantitative criteria.
FSI Defuzzification Score Rank Monthly fee a
(C11) Supply variety (C12) Installation fee (C13)
a b c ISP1 2,499,500 8 500,000
ISP1 1,055,926 21,095,167 72,082,644 31,411,246 0.8493 4 ISP2 2,250,000 5 550,000
ISP2 968,378 20,730,333 74,376,568 32,025,093 0.8878 3 ISP3 2,310,000 5 450,000
ISP3 381,229 9,830,000 43,326,223 17,845,818 0.0000 7 ISP4 2,400,000 3 500,000
ISP4 666,503 16,118,167 63,110,273 26,631,648 0.5501 6 ISP5 2,000,000 6 450,000
ISP5 780,012 17,846,833 67,346,425 28,657,757 0.6769 5 ISP6 2,415,000 4 400,000
ISP6 1,100,172 21,935,833 74,570,288 32,535,431 0.9197 2 ISP7 2,100,000 3 550,000
ISP7 1,087,943 22,415,667 77,949,280 33,817,630 1.0000 1 a
Iran rial (IRR).

Step 8: The FSI index expresses the degree to which each ISP sat-
isfies a given requirement. The FSI index is a triangular fuzzy
number that is obtained from the previously calculated scores
by Eq. (5). Table 7 shows the results,
1
FSIh ¼  ½ðSRh1  f1 Þ      ðSRhJ  fJ Þ ð5Þ
J
Step 9: There are a lot of methods for ranking fuzzy numbers
such as a-cut, defuzzification, hamming distance. The simple
and popular method, centroid method, is adopted to defuzzify
FSI index (Chou & Chang, 2008). This method is simple and easy
to use for practitioners. A defuzzified triangular fuzzy number
à = (a, b, c) is calculated by Eq. (6). In addition, the scores are
normalized in a 0–1 scale by Eq. (7). Finally, the scores can be
ranked. Ultimate ranks and scores based on qualitative criteria
are given in Table 7,
1 xhk  minh¼1;2;...;H fxhk g
DF ¼ ða þ b þ cÞ ð6Þ
3 maxh¼1;2;...;H fxhk g  minh¼1;2;...;H fxhk g
DF h  minh¼1;2;...;H fDF h g
vh ¼ ð7Þ
maxh¼1;2;...;H fDF h g  minh¼1;2;...;H fDF h g
this model for quantitative metrics. Then, the QFD model and this
model will be composed to take into account both of qualitative
and quantitative criteria in supplier selection process. The model
consists of the following steps:
Step1: List all measures in the same sequence as importance of
criteria. Table 8 shows the results.
Step 2: Transform measures so that all measures are positively
related to scores and normalized in a 0–1 scale: we consider a
situation in which a set of H ISPs are available and we want
to select them according to K quantitative criteria. The measure
of ISPh under criterion k is denoted as xhk (h = 1, 2, . . .,
H, k = 1, 2, . . ., K). We normalize all measures xhk into a 0–1 scale
by Eq. (8) and denote them as yhk. For negative criteria, first you
must take reciprocal conversions. Table 9 lists the results,
yhk ¼ ð8Þ

Table 9
Normalized measures of ISPs under quantitative criteria.
6.2. Quantitative model Normalized measures
Monthly fee (C11) Supply variety (C12) Installation fee (C13)
It is undeniable that quantitative metrics such as price have tre-
ISP1 0.0000 1.0000 0.2667
mendous effects on decisions. Ng (2008) proposed a weighted lin-
ISP2 0.4440 0.4000 0.0000
ear program for the multi-criteria supplier selection problem that ISP3 0.3285 0.4000 0.5926
can be easily implemented with a spreadsheet package. Besides, ISP4 0.1660 0.0000 0.2667
the model can be applied to practical situation and does not re- ISP5 1.0000 0.6000 0.5926
quire the user with any optimization background. However, this ISP6 0.1401 0.2000 1.0000
ISP7 0.7617 0.0000 0.0000
model only considers quantitative criteria. In this section, we use
8646 S.H. Amin, J. Razmi / Expert Systems with Applications 36 (2009) 8639–8648

Step 3: Calculate all partial averages by Eq. (9). main purpose of this stage is to compare the performance of each
X
k ISP with other competitors. After a meeting, the members of com-
1
yhg ; k ¼ 1; 2; . . . ; K ð9Þ mittee came to conclusion to apply fuzzy set theory for selected
k g¼1 ISPs. The steps of the proposed algorithm are as follows:

Step 1: Collect customers’ opinions: first, a customer survey is

Step 4: Compare and locate the maximum among these partial
carried to find out the performance of each ISP. In this survey,
averages. The corresponding value is the score of each ISP.
customers are asked to devote linguistic variables for each
Step 5: Rank the scores and identify important ISPs. Table 10
WHATs. In our case, ten staffs have been selected as customers.
shows the scores.
Table 12 shows their opinions about the ISP7. In the same way,
the employees have stated their opinions about ISP5 and ISP6.
Step 2: Calculate aggregated weight: Aggregated weight (zhi) is
6.3. Combination of models calculated by Eq. (11) where H is the number of ISPs
(h = 1, 2, . . ., H), I is the number of WHATs (i = 1, 2, . . ., I), and E
In two previous sections, the scores of ISPs based on qualitative is the number of customers (e = 1, 2, . . ., E). Table 13 shows the
and quantitative criteria were calculated. Now, we should combine results,
them to obtain the final scores. Suppose that vh is the score of ISPh
that is obtained from QFD model (Table 7), and mh is the score of 1
zhi ¼  ðzhi1  zhi2      zhiE Þ ð11Þ
ISPh which is calculated by quantitative model (Table 10). Further- E
more, a1 and a2 are the weights of the QFD model and the quanti-
tative model respectively which a1 + a2 = 1. These parameters are Step 3: Defuzzify TFNs: Eq. (6) is applied to defuzzify triangular
determined by DMs. After organizing meeting the members of fuzzy numbers. Table 13 illustrates the results.
committee decides a1 = 0.5 and a2 = 0.5. The final score of ISPh is Step 4: Identify the weaknesses: in a competitive environment,
calculated by Eq. (10). Table 11 shows the results. According to this the success of a product depends not only on its own perfor-
table, ISP7, ISP5, and ISP6 are selected, mance, but also its competitors. For example, it may perform
a1 v h þ a2 mh ¼ Sh ; h ¼ 1; 2; . . . ; H ð10Þ quite well in meeting a different customer requirement; how-
ever, if its competitors are better, the product can be viewed
to be relatively poorer for the particular customer require-
7. Phase 2: ISP evaluation ment (Lai, Xie, Tan, & Yang, 2008). Therefore, the members
of committee define three ranges to specify the weaknesses
In this section, we propose a new model to evaluate ISPs from for each ISP. This category consists of desirable, tolerable,
three perspectives: customer, performance, and competition. The and undesirable condition. Table 13 shows the results. Now,
the company can provide feedback to ISPs about their
Table 10 weaknesses.
Partial averages and scores of ISPs.
Step 5: Calculate the final scores: not only weaknesses of ISPs
Partial averages with the first k criteria Score Rank are important, but also it is valuable to evaluate ISPs based on
k=1 k=2 k=3 their total performance. In this step, we use the priority
ISP1 0.0000 0.5000 0.4222 0.5000 3
between WHATs that was calculated in Table 4. This variable
ISP2 0.4440 0.4220 0.2813 0.4440 5 is TFN, thus it is necessary to be defuzzified. Then, the final
ISP3 0.3285 0.3642 0.4404 0.4404 6 score for each ISP is calculated by the following equation that
ISP4 0.1660 0.0830 0.1442 0.1660 7 o is the defuzzified priority and p is the defuzzified TFN. Table
ISP5 1.0000 0.8000 0.7309 1.0000 1
14 shows the results. Although ISP7 had a highest score in ISP
ISP6 0.1401 0.1700 0.4467 0.4467 4
ISP7 0.7617 0.3809 0.2539 0.7617 2 selection process, the score of ISP6 dominated in evaluation
process,

Table 11 Final scoreh ¼ ðo1 p1h Þ þ ðo2 p2h Þ þ    þ ðoI pIh Þ ð12Þ
Final scores and ranking.

Score of QFD Score of quantitative Final score Final

model model (Sh) rank
Table 12
ISP1 0.8493 0.5000 0.6747 4 Customers’ opinions for ISP7.
ISP2 0.8878 0.4440 0.6659 5
ISP3 0.0000 0.4404 0.2202 7 WHATs E1 E2 E3 E4 E5 E6 E7 E8 E9 E10
ISP4 0.5501 0.1660 0.3581 6 Accessibility (C1) H L VH M M VH L H H M
ISP5 0.6769 1.0000 0.8385 2 Reliability (C2) M ML H M MH H M MH H M
ISP6 0.9197 0.4467 0.6832 3 Security (C3) VH VL H MH ML H VL M H M
ISP7 1.0000 0.7617 0.8809 1 Speed (C4) ML M MH M H MH M L MH MH

Table 13
Fuzzy matrix.

WHATs TFN Defuzzified TFN Competition situation

ISP7 ISP5 ISP6 ISP1 ISP2 ISP3 ISP7 ISP5 ISP6
Accessibility (C1) (4.8, 6.4, 7.7) (4.5, 5.7, 7.0) (3.6, 5.9, 6.9) 6.3 5.7 5.5 Desirable Tolerable Undesirable
Reliability (C2) (4.4, 6.4, 8.1) (4.2, 5.9, 7.3) (7.1, 8.9, 9.9) 6.3 5.8 8.6 Tolerable Undesirable Desirable
Security (C3) (4.2, 5.7, 7.0) (3.9, 5.5, 6.9) (3.9, 5.9, 7.3) 5.6 5.4 5.7 Tolerable Undesirable Desirable
Speed (C4) (3.7, 5.6, 7.5) (5.1, 6.9, 9.1) (4.5, 6.6, 8.0) 5.6 7.0 6.4 Undesirable Desirable Tolerable
 S.H. Amin, J. Razmi / Expert Systems with Applications 36 (2009) 8639–8648 8647

Table 14 mance simultaneously. Fifth, the proposed approach can ade-

Calculating final scores. quately deal with the inherent uncertainty and imprecision of
WHATs Priority Defuzzified Defuzzified TFN human decision-making. Finally, this research investigated the dif-
priority
ISP7 ISP5 ISP6 ferences among decision-makers by devoting unequal weights.
There are many opportunities for future works in this field. For
Accessibility (120, 320, 455) 298.3 6.3 5.7 5.5
(C1)
example, it is valuable to add ISP development stage to the pro-
Reliability (C2) (90, 260, 425) 258.3 6.3 5.8 8.6 posed model and examine the results in practical environments.
Security (C3) (160, 380, 500) 346.7 5.6 5.4 5.7 Furthermore, some ISPs may have proposed different discount pol-
Speed (C4) (110, 300, 465) 291.7 5.6 7.0 6.4 icy to vend Internet services. Therefore, it is worthwhile to analyze
Final score 7081.6 7112.5 7705.1
the effects of discounts and modify the model. Besides, comparing
Rank 3 2 1
the results of proposed method with fuzzy DEA will be another re-
search topic. It is expected that in the near future, this integrated
model will be applied effectively to various issues such as perfor-
Table 15 mance assessment, business strategies, policy making, personnel
Comparison of crisp and fuzzy values in QFD model.
and employee selection.
Crisp Fuzzy
Score Rank Score Rank
ISP1 0.8951 3 0.8493 4 Appendix A. Fuzzy operations for TFNs
ISP2 0.8661 4 0.8878 3
ISP3 0.0000 7 0.0000 7 If we want to use fuzzy sets in applications, we will have to deal
ISP4 0.4996 6 0.5501 6
with fuzzy numbers operations. Let à = (a, b, c), e
E ¼ ðd; e; f Þ, denote
ISP5 0.6370 5 0.6769 5
ISP6 0.9619 2 0.9197 2
triangular fuzzy numbers. Then (Chou & Chang, 2008; Klir & Yuan,
ISP7 1.0000 1 1.0000 1 1995):

(1) Addition of two fuzzy numbers 

Table 16 ee
A E ¼ ða þ d; b þ e; c þ f Þ
The impact of a1 (a1 + a2 = 1).

Selected ISPs Area

(2) Multiplication of two fuzzy numbers 
1, 5, 7 0 6 a1 < 0.43
5, 6, 7 0.43 6 a1 < 0.73 ee
A E ¼ ða d; b e; c f Þ
2, 6, 7 0.73 6 a1 6 1

(3) Multiplication of any real number k and a fuzzy number

8. Sensitivity analysis
ke
E ¼ ðk d; k e; k f Þ
To identify the differences between the crisp and fuzzy values,
the DMs solve the QFD model again, and they compare the results.
(4) Division of two fuzzy numbers D
This comparison is shown in Table 15. As can be seen from this ta-  
ble, not only the scores will be changed, but also the ranks will be e De a b c
A E¼ ; ;
altered. On the other hand, the decision-makers are interested to d e f
examine the impact of changes in the parameter a1 which was ap-
plied in Section 6.3. Table 16 shows the selected ISPs based on each
(5) Division by any real number k
area. According to this table, there are three areas. Moreover, ISP7  
exists in all ranges. e Dk ¼ a b c
A ; ;
k k k
9. Conclusions
(6) Subtraction of two fuzzy numbers r
In this paper, we have proposed a new model for ISP selection
and evaluation problem. After problem definition in a case of ISP e re
A E ¼ ða  d; b  e; c  f Þ
sector, we proposed the most important criteria based on ADSL’s
attributes. We utilized a combination of fuzzy QFD model and a
(7) Subtraction of any real k and a fuzzy number r
quantitative model for ISP selection. Then, we proposed a novel
algorithm for ISP evaluation which enabled us to identify the e rk ¼ ða  k; b  k; c  kÞ
A
weaknesses and assess the performance of each ISP. Triangular fuz-
zy numbers have been utilized to deal with imprecision in human
thought. The model can be applied to practical situations and can References
be easily implemented by spread sheet packages.
Aissaoui, N., Haouari, M., & Hassini, E. (2007). Supplier selection and order lot sizing
The principal advantages of our model are in sixfold: First, we modeling: A review. Computers and Operations Research, 34(12), 3516–3540.
combined supplier selection, evaluation and development together Amid, A., Ghodsypour, S. H., & O’Brien, C. (2006). Fuzzy multi objective linear model
and proposed a new framework for supplier management. Second, for supplier selection in a supply chain. International Journal of Production
Economics, 104(2), 394–407.
we focused on supplier selection and evaluation in service indus- Amid, A., Ghodsypour, S. H., & O’Brien, C. (in press). A weighted additive fuzzy multi
tries specially Internet Service Provider selection and evaluation objective model for the supplier selection problem under price breaks in a
problem. Third, the proposed model can rate both qualitative and supply Chain. International Journal of Production Economics.
Ansari, A., & Modarress, B. (1994). Quality function deployment: The role of
quantitative criteria and select suitable suppliers effectively. suppliers. International Journal of Purchasing and Materials Management, 30(4),
Fourth, the model considers customer, competition, and perfor- 28–35.
8648 S.H. Amin, J. Razmi / Expert Systems with Applications 36 (2009) 8639–8648
Araz, C., Ozfirat, P. M., & Ozkarahan, I. (2007). An integrated multi criteria decision-
making methodology for outsourcing management. Computers and Operations
Research, 34(12), 3738–3756.
Barroso, J. G. G., & Martinez, J. P. (2007). ADSL deployment in the Community of
Madrid: Investigating the geographical factors of the digital divide. Telematics
and Informatics, 24(2), 101–114.
Besterfield, D. H., Michna, C. B., Besterfield, G. H., & Sacre, M. B. (2003). Total quality
management (3rd ed.). New Jersey: Pearson Education.
Bevilacqua, M., Ciarapica, F. E., & Giacchetta, G. (2006). A fuzzy-QFD approach to
supplier selection. Journal of Purchasing and Supply Management, 12(1), 14–27.
Bottani, E., & Rizzi, A. (2006). A fuzzy TOPSIS methodology to support outsourcing of
logistics services. Supply Chain Management: An International Journal, 11(4), 294–308.
Bottani, E., & Rizzi, A. (2008). An adapted multi-criteria approach to suppliers and
products selection – An application oriented to lead-time reduction.
International Journal of Production Economics, 111(2), 763–781.
Braglia, M., & Petroni, A. (2000). A quality assurance-oriented methodology for
handling trade-offs in supplier selection. International Journal of Physical
Distribution and Logistics Management, 30(2), 96–111.
Chan, F. T. S., & Kumar, N. (2007). Global supplier development considering risk
factors using fuzzy extended AHP-based approach. Omega, 35(4), 417–431.
Chen, C. T., Lin, C. T., & Huang, S. F. (2006). A fuzzy approach for supplier evaluation
and selection in supply chain management. International Journal of Production
Economics, 102(2), 289–301.
Chou, S. Y., & Chang, Y. H. (2008). A decision support system for supplier selection
based on a strategy-aligned fuzzy SMART approach. Expert Systems with
Applications, 34(4), 2241–2253.
Dahel, N. E. (2003). Vendor selection and order quantity allocation in volume
discount environments. Supply Chain Management: An International Journal, 8(4),
335–342.
De Boer, L., Labro, E., & Morlacchi, P. (2001). A review of methods supporting
supplier selection. European Journal of Purchasing and Supply Management, 7(2),
75–89.
Degraeve, Z., Labro, E., & Roodhoofi, F. (2004). Total cost of ownership purchasing of
a service: The case of airline selection at Alcatel Bell. European Journal of
Operational Research, 156(1), 23–40.
Degraeve, Z., & Roodhooft, F. (2000). A mathematical programming approach for
procurement using activity based costing. Journal of Business Finance and
Accounting, 27(1–2), 69–98.
Demirtas, E. A., & Ustun, O. (2008). An integrated multi objective decision making
process for supplier selection and order allocation. Omega, 36(1), 76–90.
Dickson, G. W. (1966). An analysis of vendor selection system and decisions. Journal
of Purchasing, 2(1), 28–41.
Dogan, I., & Sahin, U. (2003). Supplier selection using activity-based costing and
fuzzy present worth techniques. Logistics Information Management, 16(6),
420–426.
Dulmin, R., & Mininno, V. (2003). Supplier selection using a multi-criteria decision
aid method. Journal of Purchasing and Supply Management, 9(4), 177–187.
Ellram, L. M. (1990). The supplier selection decision in strategic partnerships.
Journal of Purchasing and Material Management, 26(4), 8–12.
Fiala, P. (2005). Information sharing in supply chains. Omega, 33(5), 419–423.
Faez, F., Ghodsypour, S. H., & O’Brien, C. (in press). Vendor selection and order
allocation using an integrated fuzzy case-based reasoning and mathematical
programming model. International Journal of Production Economics.
Feng, C. X., Wang, J., & Wang, J. S. (2001). An optimization model for concurrent
selection of tolerances and suppliers. Computers and Industrial Engineering,
40(1–2), 15–33.
Garfamy, R. M. (2006). A data envelopment analysis approach based on total cost of
ownership for supplier selection. Journal of Enterprise Information Management,
19(6), 662–678.
Gencer, C., & Gürpinar, D. (2007). Analytic network process in supplier selection: A
case study in an electronic firm. Applied Mathematical Modeling, 31(11),
2475–2486.
Ghodsypour, S. H., & O’Brien, C. (1998). A decision support system for supplier
selection using an integrated analytic hierarchy process and linear programming.
International Journal of Production Economics, 56–57(1–3), 199–212.
Ghodsypour, S. H., & O’Brien, C. (2001). The total cost of logistics in supplier
selection, under conditions of multiple sourcing, multiple criteria and capacity
constraint. International Journal of Production Economics, 73(1), 15–27.
Griffin, A. (1992). Evaluating QFD’s use in US firms as a process for developing
products. The Journal of Product Innovation Management, 9(3), 171–187.
Ha, S. H., & Krishnan, R. (2008). A hybrid approach to supplier selection for the
maintenance of a competitive supply chain. Expert Systems with Applications,
34(2), 1303–1311.
Holmen, E., & Kristensen, P. S. (1998). Supplier roles in product development:
International versus task partitioning. European Journal of Purchasing and Supply
Management, 4(2–3), 185–193.
Holt, G. D. (1998). Which contractor selection methodology? International Journal of
Project Management, 16(3), 153–164.
Hong, G. H., Park, S. C., Jang, D. S., & Rho, H. M. (2005). An effective supplier selection
method for constructing a competitive supply-relationship. Expert Systems with
Applications, 28(4), 629–639.
Huang, S., & Keskar, H. (2007). Comprehensive and configurable metrics for supplier
selection. International Journal of Production Economics, 105(2), 510–523.
Humphreysa, P. K., Wong, Y. K., & Chan, F. T. S. (2003). Integrating environmental
criteria into the supplier selection process. Journal of Materials Processing
Technology, 138(1–3), 349–356.
Jain, V., Tiwari, M. K., & Chan, F. T. S. (2004). Evaluation of the supplier performance
using an evolutionary fuzzy-based approach. Journal of Manufacturing
Technology Management, 15(8), 735–744.
Kahraman, C., Cebeci, U., & Ulukan, Z. (2003). Multi-criteria supplier selection using
fuzzy AHP. Logistics Information Management, 16(6), 382–394.
Kheljani, J. G., Ghodsypour, S. H., & O’Brien, C. (in press). Optimizing whole supply
chain benefit versus buyer’s benefit through supplier selection. International
Journal of Production Economics.
Kingsman, B. G. (1986). Purchasing raw materials with uncertain fluctuating prices.
European Journal of Operational Research, 25(2), 358–372.
Klir, G. J., & Yuan, B. (1995). Fuzzy sets and fuzzy logic: Theory and applications.
Englewood Cliffs, NJ: Prentice-Hall.
Klundert, J. M., Kuipers, J., Spieksma, F. C. R., & Winkels, M. (2005). Selecting
telecommunication carriers to obtain volume discounts. Interfaces, 35(2),
124–132.
Kokangul, A., & Susuz, Z. (in press). Integrated analytical hierarch process and
mathematical programming to supplier selection problem with quantity
discount. Applied Mathematical Modeling.
Kumar, M., Vrat, P., & Shankar, R. (2004). A fuzzy goal programming approach for
vendor selection problem in a supply chain. Computers and Industrial
Engineering, 46(1), 69–85.
Kumar, M., Vrat, P., & Shankar, R. (2006). A fuzzy goal programming approach for
vendor selection problem in a supply chain. International Journal of Production
Economics, 101(2), 273–285.
Kwong, C. K., Ip, W. H., & Chan, J. W. K. (2002). Combining scoring method and fuzzy
expert systems approach to supplier assessment: A case study. Integrated
Manufacturing Systems, 13(7), 512–519.
Lai, X., Xie, M., Tan, K. C., & Yang, B. (2008). Ranking of customer requirements in a
competitive environment. Computers and Industrial Engineering, 54(2), 202–214.
Lee, A. H. I. (in press). A fuzzy supplier selection model with the consideration of
benefits, opportunities, costs and risks. Expert Systems with Applications,
doi:10.1016/j.eswa.2008.01.045.
Li, G. D., Yamaguchi, D., & Nagai, M. (2007). A grey-based decision-making approach
to the supplier selection problem. Mathematical and Computer Modeling, 46(3–
4), 573–581.
Liao, Z., & Rittscher, J. (2007). A multi-objective supplier selection model under
stochastic demand conditions. International Journal of Production Economics,
105(1), 150–159.
Liu, F. H. F., & Hai, H. L. (2005). The voting analytic hierarchy process method for
selecting supplier. International Journal of Production Economics, 97(3), 308–317.
Lopez, R. F. (2007). Strategic supplier selection in the added-value perspective: A CI
approach. Information Sciences, 177(5), 1169–1179.
Mandal, A., & Deshmukh, S. G. (1994). Vendor selection using interpretive structural
modeling (ISM). International Journal of Operations and Production Management,
14(6), 52–59.
Moore, D. L., & Fearon, H. E. (1973). Computer-assisted decision-making in
purchasing. Journal of Purchasing, 9(1), 5–25.
Ng, W. L. (2008). An efficient and simple model for multiple criteria supplier
selection problem. European Journal of Operational Research, 186(3),
1059–1067.
Ohdar, R., & Ray, P. K. (2004). Performance measurement and evaluation of
suppliers in supply chain: An evolutionary fuzzy-based approach. Journal of
Manufacturing Technology Management, 15(8), 723–734.
Oliveira, R. C., & Lourenço, J. C. (2002). A multi-criteria model for assigning new
orders to service suppliers. European Journal of Operational Research, 139(2),
390–399.
Ramanathan, R. (2007). Supplier selection problem: integrating DEA with the
approaches of total cost of ownership and AHP. Supply Chain Management: An
International Journal, 12(4), 258–261.
Rao, S. S. (2000). Internet service providers in India. Work Study, 49(5), 178–186.
Saen, R. F. (2007). Supplier selection in the presence of both cardinal and ordinal
data. European Journal of Operational Research, 183(2), 741–747.
Talluri, S., & Narasimhan, R. (2004). A methodology for strategic sourcing. European
Journal of Operational Research, 154(1), 236–250.
Temponi, C., Yen, J., & Tiao, W. A. (1999). House of quality: A fuzzy logic based
requirements analysis. European Journal of Operational Research, 117(2),
340–354.
Vokurka, R. J., Choobineh, J., & Vadi, L. (1996). A prototype expert system for the
evaluation and selection of potential suppliers. International Journal of
Operations and Production Management, 16(12), 106–127.
Wadhwa, V., & Ravindran, R. (2007). Vendor selection in outsourcing. Computers and
Operations Research, 34(12), 3725–3737.
Wang, H. S. (2008). Configuration change assessment: Genetic optimization
approach with fuzzy multiple criteria for part supplier selection decisions.
Expert Systems with Applications, 34(2), 1541–1555.
Weber, C. A., Current, J. R., & Benton, W. C. (1991). Vendor selection criteria and
methods. European Journal of Operational Research, 50(1), 2–18.
Weber, C. A., Current, J. R., & Desai, A. (2000). An optimization approach to
determining the number of vendors to employ. Supply Chain Management: An
International Journal, 5(2), 90–98.
Xia, W., & Wu, Z. (2007). Supplier selection with multiple criteria in volume
discount environments. Omega, 35(5), 494–504.
Yang, C. C., & Chen, B. S. (2005). Supplier selection using combined analytical
hierarchy process and grey relational analysis. Journal of Manufacturing
Technology Management, 17(7), 926–941.
Zadeh, L. A. (1965). Fuzzy sets. Information and Control, 8(3), 338–353.