Computers & Industrial Engineering 105 (2017) 338–347

Contents lists available at ScienceDirect

Computers & Industrial Engineering

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

A novel multi-criteria decision framework for evaluating green supply

chain management practices
Kazim Sari ⇑
Beykent University, Industrial Engineering Department, Sariyer, Istanbul 34396, Turkey

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

Article history: This study proposes a novel decision framework to evaluate green supply chain management (GSCM)
Received 1 February 2016 practices. The framework is developed by combining Monte Carlo simulation, AHP (Analytical
Received in revised form 15 November 2016 Hierarchy Process) and VIKOR (VlseKriterijumska Optimizacija I Kompromisno Resenje) methods under
Accepted 13 January 2017
fuzzy environment. The AHP is used to obtain relative weights of evaluation criteria for GSCM practices.
Available online 16 January 2017
The VIKOR is used to generate the rankings according to GSCM performance, and then to recommend the
best performing organizations. This framework offers a new way of dealing with uncertainties resulting
Keywords:
from decision makers’ incomplete and/or vague information by utilizing the Monte Carlo simulation. An
Green supply chain
Multi-criteria decision making
illustrative case is used as well to exemplify the proposed approach.
Monte Carlo simulation Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction (Chan, He, & Wang, 2012). Beyond these, they also seen as a tool to
improve the performance of business enterprises (Jabbour,
In today’s business environment with increasing environmental Frascareli, & Jabbour, 2015; Kumar, Teichman, & Timpernagel,
awareness and sustainability, adaptation of Green Supply Chain 2012; Zhu & Sarkis, 2004). In fact, in all of these cases, it is an inevi-
Management (GSCM) practices has become very popular (Hsu, table reality that the managers need to evaluate their own GSCM
2013; Srivastava, 2007). It is in such a way that various organiza- performance or that of suppliers. In this research, we aim to
tions from both manufacturing and service industries employ develop an evaluation tool for GSCM practices. By this way, it is
some sort of GSCM practices in all over the world (Chien & Shih, also made possible to rank the suppliers of an organization accord-
2007; Kannan, de Jabbour, & Jabbour, 2014; Sarkis, Zhu, & Lai, ing to their performance about GSCM.
2011; Zhu, Sarkis, & Lai, 2007). The main purpose of using GSCM Our review of related literature indicates that there is a wide
is to minimize the environmental damage created by supply chain range of studies performed on GSCM (see e.g. Fahimnia, Sarkis, &
related activities (Beamon, 1999). Hence, GSCM makes it necessary Davarzani, 2015; Sarkis et al., 2011; Srivastava, 2007) concentrat-
for business enterprises to reorganize their purchasing, production, ing on supplier or partner selection according to GSCM criteria
and distribution operations. In addition, GSCM also attributes (Awasthi & Kannan, 2016; Hashemi, Karimi, & Tavana, 2015;
importance for reverse logistics to capture value from used prod- Igarashi, De Boer, & Fet, 2013; Lee, Kang, Hsu, & Hung, 2009). In
ucts and materials or to properly dispose them (e.g. Büyüközkan this stream of research, it is very popular to use multi-criteria
& Çifçi, 2012; Haji Vahabzadeh, Asiaei, & Zailani, 2015). decision making (MCDM) methods under fuzzy environment as
Various motivations are reported for business enterprises to the decision situation is very complex and includes vagueness
deploy GSCM practices (e.g. Diabat & Govindan, 2011; (Kannan, Khodaverdi, Olfat, Jafarian, & Diabat, 2013; Kannan
Mathiyazhagan, Diabat, Al-Refaie, & Xu, 2015; Testa & Iraldo, et al., 2014; Tseng & Chiu, 2013; Wu & Barnes, 2016).
2010; Walker, Di Sisto, & McBain, 2008). From these studies, it is These is another stream of studies focused on finding empirical
revealed that while some companies adopt GSCM practices to sat- evidences about GSCM. At this point, the relationship between
isfy customer expectations, some others perform these practices GSCM practices and organizational performance is investigated in
just for complying with environmental regulations. In addition, a large scale (Azevedo, Carvalho, & Cruz Machado, 2011; Jabbour
GSCM practices can also be utilized as a tool for marketing purpose et al., 2015; Laosirihongthong, Adebanjo, & Tan, 2013), even the
relationship between GSCM initiatives and stock prices is investi-
gated (Bose & Pal, 2012). By this way, it is aimed to justify the
⇑ Address: Beykent University, Ayazaga Campus, Faculty of Engineering and belief that GSCM leads to improvements in environmental, opera-
Architecture, Industrial Engineering Department, 34396 Sariyer, Istanbul, Turkey
tional, and financial performances of a business enterprise. In
E-mail address: [email protected]

http://dx.doi.org/10.1016/j.cie.2017.01.016
0360-8352/Ó 2017 Elsevier Ltd. All rights reserved.
 K. Sari / Computers & Industrial Engineering 105 (2017) 338–347 339

addition, pressures, key drivers, and barriers of GSCM adaptation Uygun and Dede (2016) show regard to five main groups as green
are also explored (Drohomeretski, Da Costa, & De Lima, 2014; operations of design, purchasing, transportation, logistics, and
Dubey, Gunasekaran, Papadopoulos, & Childe, 2015; Govindan, reverse logistics. From a different perspective, Zhu et al. (2008)
Kaliyan, Kannan, & Haq, 2014; Wang, Mathiyazhagan, Xu, & consider internal environmental management, green purchasing,
Diabat, 2015). cooperation with customers, eco-design, and investment recovery
Our study provides two major contributions to the literature. are the main groups of GSCM practices. Lau (2011), on the other
First, a comprehensive list of activities is provided for GSCM hand, employs only a few items. They are the green activities
assessment. They are obtained via a detailed review of literature. related with purchasing, packaging, and transportation. Finally,
Secondly and more importantly, a novel decision framework is Rostamzadeh et al. (2015) consider GSCM practices according to
proposed, which integrates Monte Carlo simulation with fuzzy their relationship with the activities of design, purchasing, produc-
MCDM methods. More specifically, fuzzy Analytical Hierarchy Pro- tion, warehousing, transportation, and recycling.
cess (AHP) and fuzzy VIKOR, which stands for ‘‘VIseKriterijumska While constructing our decision hierarchy for GSCM evaluation,
Optimizacija I Kompromisno Resenje” are used together along with we consider the above-mentioned studies by emphasizing more on
Monte Carlo simulation. the definition of Supply Chain Management (SCM) made by Coun-
In fact, it is well liked to use the methods of AHP and VIKOR cil of Supply Chain Management Professionals (CSCMP) (CSCMP,
together under fuzzy environment (Fu, Chu, Chao, Lee, & Liao, 2015). Accordingly, SCM includes ‘‘. . .all activities involved in
2011; Kaya & Kahraman, 2011; Pourebrahim, Hadipour, Mokhtar, sourcing and procurement, conversion, and all logistics”. Thus, based
& Taghavi, 2014; Rezaie, Ramiyani, Nazari-Shirkouhi, & on this definition, we believe that it is more appropriate to catego-
Badizadeh, 2014). For instance, Rezaie et al. (2014) utilize a rize GSCM practices under four main groups. These are the activi-
combination of VIKOR and fuzzy AHP to evaluate performance of ties related with (1) inbound operations, (2) production operations,
cement firms. Similarly, Pourebrahim et al. (2014) propose a (3) outbound operations, and (4) reverse logistics.
decision model based on this approach for priority assessment in Having established the main categories of GSCM practices, we
coastal areas. In addition, Fu et al. (2011) perform benchmarking can now determine the individual GSCM items under each
analysis in the hospitality industry with a combination of AHP category. As it is known, inbound operations include purchasing
and VIKOR. Moreover, Kaya and Kahraman (2011) employ the and arranging the inbound movement of raw materials, parts,
same approach to make a selection among the alternative and components to a manufacturing company. Thus, it is possible
forestation areas. However, in our proposed decision model, we to think of four green practices under this category. These are:
make an improvement over this popular approach by adding a
Monte Carlo simulation analysis. By this way, it is provided a good C1.1. Choosing suppliers by environmental criteria (Azevedo
opportunity for the managers to make a detailed analysis et al., 2011; Mohanty & Prakash, 2014; Rao & Holt, 2005;
of the decision situation. For instance, it is made very simple to Rostamzadeh et al., 2015; Zhu et al., 2008).
analyze the uncertainties about the parameters of a decision C1.2. Guiding suppliers to establish their own environmental
model (i.e. weights of evaluation criteria) with Monte Carlo programs (Azevedo et al., 2011; Rao & Holt, 2005;
simulation. Rostamzadeh et al., 2015; Zhu et al., 2008).
The remainder of this study is organized as follows. Section 2 C1.3. Urging/pressuring suppliers to take environmental actions
clarifies the proposed criteria for evaluating GSCM. The proposed (Azevedo et al., 2011; Mohanty & Prakash, 2014; Rao & Holt,
decision tool is identified in Section 3. In Section 4, a numerical 2005; Rostamzadeh et al., 2015; Zhu et al., 2008).
application along with a discussion is provided. Finally, conclusion C1.4. Purchasing environment friendly items (Awasthi &
and directions for future development are presented in Section 5. Kannan, 2016; Azevedo et al., 2011; Rao & Holt, 2005;
Rostamzadeh et al., 2015; Shen et al., 2013).
2. Proposed criteria for evaluating GSCM
In a similar way, it is possible to consider four green practices
There are various practices that can be performed in an organi- under the category of production operations. These activities are
zation with regard to GSCM. In our attempt to extract criteria for closely related with manufacturing processes of items. At this
GSCM evaluation, research studies by Awasthi and Kannan point, product design can also be considered under this category
(2016), Uygun and Dede (2016), Mohanty and Prakash (2014), as it has great impacts on production operations. They are:
Yang, Lu, Haider, and Marlow (2013), Büyüközkan & Çifçi (2012),
Azevedo et al. (2011), Ubeda, Arcelus, and Faulin (2011), Lau C2.1. Design products for recycling (Awasthi & Kannan, 2016;
(2011), Rostamzadeh, Govindan, Esmaeili, and Sabaghi (2015), Shen et al., 2013; Srivastava, 2007; Zhu et al., 2008).
Rao and Holt (2005), Shen, Olfat, Govindan, Khodaverdi, and C2.2. Using cleaner technology (Awasthi & Kannan, 2016;
Diabat (2013) and Zhu, Sarkis, and Lai (2008) provide very useful Mohanty & Prakash, 2014; Rao & Holt, 2005; Rostamzadeh
resources for us. In addition, excellent GSCM review studies of et al., 2015; Shen et al., 2013; Zhu et al., 2008).
Govindan, Rajendran, Sarkis, and Murugesan (2015) and C2.3. Improving capacity utilization (Rostamzadeh et al., 2015;
Srivastava (2007) as well as the book by Mangan, Lalwani, Uygun & Dede, 2016; Zhu et al., 2008).
Butcher, and Javadpour (2012, p. 334) ensure detailed list of crite- C2.4. Promoting remanufacturing (Awasthi & Kannan, 2016;
ria that can be used for our purpose. Mohanty & Prakash, 2014; Srivastava, 2007).
In multi-criteria decision analysis with AHP, it is very important
to build the decision hierarchy in a proper way. We observe that For outbound operations, on the other hand, six green activities can
GSCM practices are classified differently among researchers. For be listed. These are mainly related with distribution of end items.
example, Rao and Holt (2005) classify them in three groups as They are:
inbound, production, and outbound operations. Büyüközkan & C3.1. Enhancing vehicle operating efficiency (Laari, Töyli,
Çifçi (2012), on the other hand, categorize them under five main Solakivi, & Ojala, 2016; Mangan et al., 2012, p. 334; Yang
groups by considering reverse logistics and packaging as additional et al., 2013).
groups. Apart from this, product design is regarded as the sixth C3.2. Encouraging eco-driving (Laari et al., 2016; Rostamzadeh
main group by Awasthi and Kannan (2016). In a similar way, et al., 2015; Yang et al., 2013).
340 K. Sari / Computers & Industrial Engineering 105 (2017) 338–347

C3.3. Using environmental friendly packaging (Büyüközkan & As a result of this categorization, our hierarchical decision
Çifçi, 2012; Rao & Holt, 2005; Rostamzadeh et al., 2015). framework to evaluate GSCM practices can be formed as shown
C3.4. Reducing empty running (Laari et al., 2016; Mangan et al., in Fig. 1.
2012, p.334; Ubeda et al., 2011).
C3.5. Improving vehicle routing using GPS (Global Positioning
System) and other systems (Laari et al., 2016; Lau, 2011; 3. The proposed decision model
Mangan et al., 2012, p. 334).
C3.6. Increasing vehicle payload capacity (Laari et al., 2016; In this research, a hybrid fuzzy MCDM model with Monte Carlo
Mangan et al., 2012, p.334; Ubeda et al., 2011). simulation is proposed. In the proposed model, relative weights of
each criterion shown in Fig. 1 are obtained by fuzzy AHP. In reality,
Finally, four activities can be considered as green operation fuzzy AHP is a very popular method employed for this purpose
under reverse logistics category. As it is known, reverse logistics (Kwong & Bai, 2002; Sari, 2013; Torfi, Farahani, & Rezapour,
is concerned with the flow of products and materials on the oppo- 2010). Herein, the weights of the evaluation criteria obtained from
site direction of the flow in forward logistics. These are: fuzzy AHP are also fuzzy numbers. In our proposed model, these
fuzzy numbers are converted to random numbers that come from
C4.1. Re-use of products and components (Awasthi & Kannan, a triangular probability distribution function. By this way, it is
2016; Büyüközkan & Çifçi, 2012; Uygun & Dede, 2016). made possible to conduct the Monte Carlo simulation in perform-
C4.2. Recycling of materials (Büyüközkan & Çifçi, 2012; Rao & ing fuzzy VIKOR. At this point, fuzzy VIKOR is used to evaluate
Holt, 2005; Srivastava, 2007; Uygun & Dede, 2016). GSCM practices of each supplier (Sanayei, Farid Mousavi, &
C4.3. Waste management (Azevedo et al., 2011; Mohanty & Yazdankhah, 2010). Indeed, by integrating the Monte Carlo simula-
Prakash, 2014; Rao & Holt, 2005; Srivastava, 2007). tion into this hybrid MCDM model, the uncertainty or vagueness in
C4.4. Taking back packaging (Azevedo et al., 2011; Büyüközkan the weights of evaluation criteria is analyzed in a better way. In
& Çifçi, 2012; Rao & Holt, 2005; Srivastava, 2007). addition, performing sensitivity analysis is also made possible.

Fig. 1. The framework of evaluating the GSCM practices.

 K. Sari / Computers & Industrial Engineering 105 (2017) 338–347 341

Step 1: Construct fuzzy pairwise comparison matrices. Each deci-

sion maker is asked to assign a linguistic term to the pairwise
comparisons among all criteria in the dimensions of the hierar-
chy system. The linguistic terms used in this study and their
numeric conversions are provided in Table 1.

The results of the comparisons are constructed as fuzzy pair-

wise comparison matrices shown in Eq. (1), where A e k is the fuzzy
pairwise comparison matrix of k th decision maker.
2 3 2 3
1 a ~k12    a
~k1n 1 ~k12
a  a ~k1n
6 k 7 6 7
6a~ 1 ... a ~k2n 7 6 1=a
~k12 1 ... a ~k2n 7
e 6 21 7 6 7
A ¼6 .
k
.. .. 7¼6 . .. .. .. 7; k ¼ 1; 2; . . . ; K
6 .. . . ... 7 6 . . 7
4 5 4 . . . 5
~kn1 a
a ~kn2    1 ~k1n 1=a
1=a ~k2n . . . 1
ð1Þ

Step 2: Compute the synthetic pairwise comparison matrix. After

collecting all evaluations of the experts, the next step is to
combine them together. For this purpose, geometric mean tech-
~ij is the aggregated fuzzy
nique shown in Eq. (2) is used, where a
comparison value of dimension i to criterion j, and K is the total
number of experts.
 1=K
~ij ¼ a
a ~1ij  a
~2ij  . . .  a
~Kij ð2Þ

Step 3: Compute the fuzzy weight of each criterion. At this step,

fuzzy weight of each criterion is obtained. To do this, the geo-
metric mean method suggested by Buckley (1985) is utilized.
The geometric mean of fuzzy comparison value of criterion i
to each criterion (~ri Þ can be found by Eq. (3).
 
~ri ¼ a~i1  . . .  a ~in 1=n
~ij  . . .  a ð3Þ

Then, the fuzzy weight of the ith criterion (w ~ i ) indicated by a TFN,

~ i ¼ ðlwi ; mwi ; uwi Þ can be obtained by Eq. (4).
w

~ i ¼ ~r i  ð~r 1  . . .  ~r i  . . .  ~r n Þ1
w ð4Þ

3.2. Fuzzy VIKOR with Monte Carlo simulation

Given that there are m alternative suppliers ðA1 ; A2 ; . . . ; Am Þ, n

decision criteria ðC 1 ; C 2 ; . . . ; C n Þ and K experts, the procedure of
fuzzy VIKOR with Monte Carlo simulation proposed in this study
is explained below:
Fig. 2. The flow chart of the proposed methodology.

Step 1: Determine the weightings of the evaluation criteria. In this

research, fuzzy AHP is employed to find the fuzzy weights of the
Interested readers can see the study by Raychaudhuri (2008) to evaluation criteria. As it is shown in Section 3.1, the fuzzy
have knowledge of Monte Carlo simulation. weights of each evaluation criterion (w ~ i ) is expressed by a
In the previous studies where AHP and VIKOR are used together, TFN, w~ i ¼ ðlwi ; mwi ; uwi Þ.
fuzzy weights obtained from fuzzy AHP are usually converted to
crisp numbers to be used as an input for VIKOR (Kaya & In our proposed model, we convert these TFNs (e.g.
Kahraman, 2011; Rezaie et al., 2014). Although this operation sim- ~ i ¼ ðlwi ; mwi ; uwi Þ) to random numbers ðti Þ that come from a tri-
w
plifies the required computations, it results in loss of information angular probability distribution function with parameters
about the weights of evaluation criteria. In fact, these weights rep-
resent experiences of experts and their subjective judgments about
GSCM. Thus, it is very natural for them to include some sort of Table 1
Triangular fuzzy conversion table.
vagueness, uncertainty, and subjectivity. Therefore, our decision
model deals with this vagueness and uncertainty in a better way Linguistic scale for importance degrees Triangular fuzzy scale
than the previous studies. The details of our proposed decision Absolutely more important (AMI) (5/2, 3, 7/2)
framework are presented in Fig. 2. Very strongly more important (VSMI) (2, 5/2, 3)
Strongly more important (SMI) (3/2, 2, 5/2)
Weakly more important (WMI) (1, 3/2, 2)
3.1. Fuzzy AHP to obtain the weights of GSCM evaluation criteria Equally important (EI) (1, 1, 1)
Weakly less important (WLI) (1/2, 2/3, 1)
Given that there are n decision criteria ðC 1 ; C 2 ; . . . ; C n Þ and K Strongly less important (SLI) (2/5, 1/2, 2/3)
experts for evaluation of GSCM criteria, the procedure of fuzzy Very strongly less important (VSLI) (1/3, 2/5, 1/2)
Absolutely less important (ALI) (2/7, 1/3, 2/5)
AHP can be explained as follows:
342 K. Sari / Computers & Industrial Engineering 105 (2017) 338–347

ðlwi ; mwi ; uwi Þ where ðlwi ; uwi Þ is the range and mwi is the most The index eS  ¼ minj e e  ¼ minj R
S j and R e j represent maximum
likely value. While making this transformation, the guidelines sug- majority rule, and a minimum individual regret of an opponent
gested by Dubois, Foulloy, Mauris, and Prade (2004) is employed. strategy, respectively. In addition, v is the weight of the strategy
In their study, Dubois et al. (2004) discuss the possible ways of of the maximum group utility. It is usually assumed to be 0.5.
transformation of a possibility distribution into a probability
distribution, and vice versa.
Step 8: Defuzzify the e ~ j . In this study, graded mean
e j ; and Q
Sj, R
integration approach is used. According to the graded mean
Step 2: Choose the appropriate linguistic variables for the alterna-
integration approach, a TFN C ~ ¼ ðc1 ; c2 ; c3 Þ can be transformed
tives with respect to criteria. The linguistic variables are
described by TFNs. They are provided in Table 2. into a crisp number by using the Eq. (14).
Step 3: Construct the fuzzy performance rating matrix. To obtain
the fuzzy performance rating matrix shown in Eq. (5), the C ¼ ðc1 þ 4c2 þ c3 Þ=6 ð14Þ
expert ratings for each alternative is aggregated with respect Step 9: Rank the alternatives, sorting the values of S, R, and Q in
to each criterion as shown in Eq. (6). ascending order.
Step 10: Propose a compromise solution. The alternative Að1Þ that
is best ranked by the measure Q (minimum) is proposed as the
compromise solution, if the following two conditions are
ð5Þ
satisfied.
C1. Acceptable advantage

1 1 
~xij ¼ ~xij  ~x2ij  . . .  ~xkij ð6Þ QðAð2Þ Þ  Q ðAð1Þ Þ P DQ ð15Þ
K
where ~xkij is the performance rating of alternative Ai with respect to where Að2Þ is the alternative with second position in the ranking list
of Q, DQ ¼ 1=ðn  1Þ, and n indicates the number of feasible
criterion C j evaluated by the kth expert, and ~xkij ¼
k
ðlij ; mkij ; ukij Þ.
alternatives.
Step 4: Determine the fuzzy best value ~f  and fuzzy worst value ~f 
i i C2. Acceptable stability in decision making
of all criteria.
The alternative Að1Þ must be the best ranked by S or/and R. If one
~f  ¼ max ~x ; ~f  ¼ min ~x ð7Þ
i j ij i j ij of these conditions is not satisfied, then a set of compromise
~ij . It is computed by Eq. (8). solution is proposed, which consists of:
Step 5: Compute the fuzzy difference d
~ ¼ ð~f   ~x Þ=ð~f   ~f  Þ
d ð8Þ
ij ij
i i i
1. Alternatives Að1Þ and Að2Þ if only the condition C2 is not satisfied,
Step 6: Compute the value of e e j . Here e
S j and R S j refers to the sepa- or
ration measure of alternative Aj from the fuzzy best value and R ej 2. Alternatives Að1Þ , Að2Þ , . . ., AðMÞ if the condition C1 is not satisfied;
refers to the separation measure of alternativeAj from the fuzzy AðMÞ is determined by the relation Q ðAðMÞ Þ  Q ðAð1Þ Þ < DQ for the
worst value. There are computed by Eqs. (9) and (10), respec- maximum M.
tively.
4. A numerical application and discussion
X
n
e
Sj ¼ ~
ti  d ð9Þ Consider a manufacturer ABC that is employing green practices
ij
i¼1 in its supply chain operations. Currently, ABC needs to evaluate
  four suppliers according to their performance in GSCM. In making
R ~
e j ¼ maxi t i  d ð10Þ this evaluation, top manager of ABC wants to use our proposed
ij
decision framework.
where ti represent the preference weight of ith criteria. In reality, it Firstly, the fuzzy AHP is employed in order to obtain the relative
is a random number obtained in Step 1. Thus, for each run of the weight of each evaluation criterion shown in Fig. 1. By using lin-
Monte Carlo simulation, there will be a different preference weight. guistic scale provided in Table 1, pairwise comparisons are made
Step 7: Compute the value of Q~ j . It is computed by using the by the manager of ABC. In this regard, first pairwise comparisons
Eqs. (11)–(13). are made for main evaluation criteria. They are presented in
  Table 3.
~ j ¼ v ðe
Q Sj  e
S  Þ=ðe S  Þ  ð1  v Þð R
S  e ej  R
e  Þ= Re  R
e ð11Þ Later, pairwise comparisons for sub-criteria of inbound opera-
tions, production operations, outbound operations, and reverse
e logistics operations are performed in a similar way. Table 4 shows
S  ¼ minj e
Sj ; e
S  ¼ maxj e
Sj ð12Þ the pairwise comparisons formed for inbound operations. Like-
wise, pairwise comparisons for production operations are shown
e  ¼ minj R
R ej; e  ¼ maxj R
R ej ð13Þ in Table 5. In Tables 6 and 7, on the other hand, pairwise compar-
isons for outbound operations and reverse logistics operations are
presented, respectively.
Table 2 Having obtained the pairwise comparisons for the main evalu-
Linguistic scale for evaluating GSCM performance of suppliers. ation criteria and sub-criteria that are shown in Tables 3–7,
Linguistic scale for evaluating suppliers Triangular fuzzy scale Eqs. (3) and (4) are used to obtain the fuzzy weight of each
Very Poor (VP) (0, 1, 3)
criterion in the decision hierarchy. For example, to obtain the fuzzy
Poor (P) (1, 3, 5) weights of the main factors, Table 3 is used along with Eqs. (3) and
Fair (F) (3, 5, 7) (4). As a result, fuzzy weights of (0.16, 0.19, 0.23), (0.18, 0.23, 0.30),
Good (G) (5, 7, 9) (0.30, 0.42, 0.56), and (0.13, 0.17, 0.22) are computed for
Very Good (VG) (7, 9, 10)
inbound, production, outbound, and reverse logistics, respectively.
 K. Sari / Computers & Industrial Engineering 105 (2017) 338–347 343

Table 3 Table 8
Pairwise comparison matrix for the main evaluation criteria. Weights of evaluation criteria.

C1 C2 C3 C4 Fuzzy weight
C1 – EI VSLI EI C1.1 (0.03, 0.05, 0.08)
C2 – SLI SMI C1.2 (0.03, 0.05, 0.08)
C3 – SMI C1.3 (0.04, 0.07, 0.12)
C4 – C1.4 (0.01, 0.02, 0.04)
C2.1 (0.03, 0.05, 0.09)
C2.2 (0.03, 0.06, 0.10)
C2.3 (0.05, 0.09, 0.16)
Table 4 C2.4 (0.02, 0.04, 0.07)
Pairwise comparison matrix for inbound operations. C3.1 (0.03, 0.06, 0.10)
C3.2 (0.04, 0.07, 0.11)
C1.1 C1.2 C1.3 C1.4 C3.3 (0.05, 0.11, 0.20)
C1.1 – EI WLI WSMI C3.4 (0.03, 0.05, 0.09)
C1.2 – SLI AMI C3.5 (0.04, 0.06, 0.10)
C1.3 – SMI C3.6 (0.04, 0.07, 0.12)
C1.4 – C4.1 (0.02, 0.03, 0.06)
C4.2 (0.02, 0.04, 0.08)
C4.3 (0.03, 0.06, 0.11)
C4.4 (0.02, 0.03, 0.05)

Table 5
Pairwise comparison matrix for production operations.
Table 9
C2.1 C2.2 C2.3 C2.4
Linguistic assessments of suppliers according to each criterion.
C2.1 – EI WLI EI
C2.2 – SLI SMI Supplier A Supplier B Supplier C Supplier D
C2.3 – SMI C1.1 VG, G, G P, F, F F, G, G G, VG, F
C2.4 – C1.2 G, F, P VG, G, G G, F, F VG, VG, G
C1.3 P, F, F F, P, G P, F, VP G, F, VG
C1.4 F, G, G F, G, P G, F, VG F, VG, F
C2.1 F, P, P F, G, F F, G, F VG, G, G
Table 6 C2.2 F, G, F G, VG, VG F, F, G VG, G, VG
Pairwise comparison matrix for outbound operations. C2.3 VG, G, VG P, F, P F, G, G VG, G, VG
C2.4 G, VG, VG P, P, F F, P, P P, G, F
C3.1 C3.2 C3.3 C3.4 C3.5 C3.6 C3.1 VG, G, G G, P, G F, G, F VG, F, G
C3.1 – EI WLI EI EI SLI C3.2 VG, G, G F, P, G G, VG, VG F, G, G
C3.2 – SLI SMI EI EI C3.3 VG, G, VG G, F, P F, G, F VG, G, G
C3.3 – SMI SMI WMI C3.4 G, F, VG P, F, F F, G, G VG, VG, VG
C3.4 – EI EI C3.5 G, F, F P, VP, VP F, G, F P, VP, F
C3.5 – EI C3.6 VG, G, G F, P, F P, VP, VP G, VG, G
C3.6 – C4.1 G, VG, VG F, G, G G, VG, F F, P, G
C4.2 VG, G, VG P, F, VP F, F, G VG, G, G
C4.3 F, G, G F, P, G G, VG, F F, P, F
C4.4 VP, P, P F, F, G F, G, G VG, G, G
Table 7
Pairwise comparison matrix for reverse logistics.

C4.1 C4.2 C4.3 C4.4 As it is shown in Fig. 2, the next step in our proposed model is to
C4.1 – EI SLI EI evaluate the four suppliers according to their GSCM performance
C4.2 – WLI SMI for each evaluation criterion. This is achieved by employing Monte
C4.3 – SMI Carlo simulation integrated fuzzy VIKOR explained in Section 3.2.
C4.4 –
For this purpose, linguistic assessments of three experts for each
supplier are obtained. These are provided in Table 9. At this point,
linguistic scales provided in Table 2 are used to make these
evaluations.
In addition, local fuzzy weights of 18 sub-criteria are computed in
After gathering expert evaluations for each supplier, Eq. (6) is
a similar fashion. For example, local fuzzy weights of sub-criteria
used to obtain the fuzzy performance matrix shown. The resulting
for inbound operations (C1.1, C1.2, C1.3, C1.4) are calculated as
fuzzy performance matrix is presented in Table 10.
(0.20, 0.26, 0.36), (0.20, 0.26, 0.34), (0.24, 0.36, 0.51), and (0.09,
0.12, 0.16), respectively. Global fuzzy weights of sub-criteria are The next step is to determine the fuzzy best value ~f  and fuzzy i

calculated by multiplying the respective fuzzy main criteria weight worst value ~f 
i of all criteria. They are obtained by employing Eq.
and local fuzzy weights of 18 sub-criteria. As a result, global fuzzy (7) and provided in Table 11 after normalizing the fuzzy perfor-
weights shown in Table 8 are obtained. mance matrix. Later, by using Eq. (8), fuzzy differences are com-
At this point, instead of converting the fuzzy weights to crisp puted. They are shown in Table 12.
numbers as it is done in existing studies, they are transformed to At this stage, Monte Carlo simulation analysis is required to
random numbers here. We assume that each fuzzy weight is a ran- employ Eqs. (9)–(15) since the weights obtained from fuzzy AHP
dom number that comes from a triangular probability distribution are now random numbers. For performing the Monte Carlo
with respective parameters shown in Table 8. A triangular proba- simulation analysis, Oracle Crystal Ball for Enterprise Performance
bility distribution is defined by three values: the minimum value Management software is used. Crystal Ball is a popular spreadsheet
a, the maximum value b, and the peak value c. For our case, the based risk analysis and simulation tool. In Monte Carlo simulation
weight of the first criterion (C1.1), for example, is a triangular ran- analysis, 10,000 simulation runs are conducted to eliminate
dom number with parameters of a = 0.03, b = 0.08, and c = 0.05. the impact of random variations. As a result of employing
344 K. Sari / Computers & Industrial Engineering 105 (2017) 338–347

Table 10
Fuzzy performance matrix.

Supplier A Supplier B Supplier C Supplier D

C1.1 5.67 7.67 9.33 2.33 4.33 6.33 4.33 6.33 8.33 5.00 7.00 8.67
C1.2 3.00 5.00 7.00 5.67 7.67 9.33 3.67 5.67 7.67 6.33 8.33 9.67
C1.3 2.33 4.33 6.33 3.00 5.00 7.00 1.33 3.00 5.00 5.00 7.00 8.67
C1.4 4.33 6.33 8.33 3.00 5.00 7.00 5.00 7.00 8.67 4.33 6.33 8.00
C2.1 1.67 3.67 5.67 3.67 5.67 7.67 3.67 5.67 7.67 5.67 7.67 9.33
C2.2 3.67 5.67 7.67 6.33 8.33 9.67 3.67 5.67 7.67 6.33 8.33 9.67
C2.3 6.33 8.33 9.67 1.67 3.67 5.67 4.33 6.33 8.33 6.33 8.33 9.67
C2.4 5.00 7.00 9.00 1.67 3.67 5.67 1.67 3.67 5.67 3.00 5.00 7.00
C3.1 5.67 7.67 9.33 3.67 5.67 7.67 3.67 5.67 7.67 5.00 7.00 8.67
C3.2 5.67 7.67 9.33 3.00 5.00 7.00 6.33 8.33 9.67 4.33 6.33 8.33
C3.3 6.33 8.33 9.67 3.00 5.00 7.00 3.67 5.67 7.67 5.67 7.67 9.33
C3.4 5.00 7.00 8.67 2.33 4.33 6.33 4.33 6.33 8.33 7.00 9.00 10.00
C3.5 3.67 5.67 7.67 0.33 1.67 3.67 3.67 5.67 7.67 1.33 3.00 5.00
C3.6 5.67 7.67 9.33 2.33 4.33 6.33 0.33 1.67 3.67 5.67 7.67 9.33
C4.1 6.33 8.33 9.67 4.33 6.33 8.33 5.00 7.00 8.67 3.00 5.00 7.00
C4.2 6.33 8.33 9.67 1.33 3.00 5.00 3.67 5.67 7.67 5.67 7.67 9.33
C4.3 4.33 6.33 8.33 3.00 5.00 7.00 5.00 7.00 8.67 2.33 4.33 6.33
C4.4 0.67 2.33 4.33 3.67 5.67 7.67 4.33 6.33 8.33 5.67 7.67 9.33

Table 11
Fuzzy best and worst values. Table 13
Average value of S, R, and Q.
Best value Worst value
S R Q
C1.1 0.57 0.77 0.93 0.23 0.43 0.63
C1.2 0.63 0.83 0.97 0.30 0.50 0.70 Supplier A 0.31 0.07 0.47
C1.3 0.50 0.70 0.87 0.13 0.30 0.50 Supplier B 0.80 0.13 1.53
C1.4 0.50 0.70 0.87 0.30 0.50 0.70 Supplier C 0.59 0.11 1.15
C2.1 0.57 0.77 0.93 0.17 0.37 0.57 Supplier D 0.29 0.08 0.51
C2.2 0.63 0.83 0.97 0.37 0.57 0.77
C2.3 0.63 0.83 0.97 0.17 0.37 0.57
C2.4 0.50 0.70 0.90 0.17 0.37 0.57
C3.1 0.57 0.77 0.93 0.37 0.57 0.77 Table 14
C3.2 0.63 0.83 0.97 0.30 0.50 0.70 Rank of alternatives.
C3.3 0.63 0.83 0.97 0.30 0.50 0.70
C3.4 0.70 0.90 1.00 0.23 0.43 0.63 Rank
C3.5 0.37 0.57 0.77 0.03 0.17 0.37 Supplier A 1
C3.6 0.57 0.77 0.93 0.03 0.17 0.37 Supplier B 3
C4.1 0.63 0.83 0.97 0.30 0.50 0.70 Supplier C 2
C4.2 0.63 0.83 0.97 0.13 0.30 0.50 Supplier D 1
C4.3 0.50 0.70 0.87 0.23 0.43 0.63
C4.4 0.57 0.77 0.93 0.07 0.23 0.43

By using the average values of these indexes, alternative suppli-

Eq. (9)–(15) along with Monte Carlo simulation analysis, three ers can be ranked according to Step 10 of fuzzy VIKOR with Monte
indexes (S, R, and Q) are produced for each alternative supplier at Carlo simulation presented in Section 3.2. The final rank list based
the end of each simulation run. The average values for these on average values is shown in Table 14. The results in Table 14
indexes are presented in Table 13. indicate that there is a set of compromise solution for this case

Table 12
Fuzzy differences.

Supplier A Supplier B Supplier C Supplier D

C1.1 1.10 0.00 1.22 0.20 1.00 2.33 0.80 0.40 1.67 0.90 0.20 1.44
C1.2 0.20 1.00 2.50 0.90 0.20 1.50 0.40 0.80 2.25 1.00 0.00 1.25
C1.3 0.36 0.67 1.73 0.55 0.50 1.55 0.00 1.00 2.00 1.00 0.00 1.00
C1.4 1.67 0.33 2.60 1.00 1.00 3.40 1.83 0.00 2.20 1.50 0.33 2.60
C2.1 0.00 1.00 2.09 0.50 0.50 1.55 0.50 0.50 1.55 0.92 0.00 1.00
C2.2 0.50 1.00 3.00 1.25 0.00 1.67 0.50 1.00 3.00 1.25 0.00 1.67
C2.3 0.71 0.00 0.83 0.14 1.00 2.00 0.43 0.43 1.33 0.71 0.00 0.83
C2.4 1.20 0.00 1.20 0.20 1.00 2.20 0.20 1.00 2.20 0.60 0.60 1.80
C3.1 1.83 0.00 2.20 1.00 1.00 3.40 1.00 1.00 3.40 1.50 0.33 2.60
C3.2 0.90 0.20 1.50 0.20 1.00 2.50 1.00 0.00 1.25 0.60 0.60 2.00
C3.3 1.00 0.00 1.25 0.20 1.00 2.50 0.40 0.80 2.25 0.90 0.20 1.50
C3.4 0.36 0.43 1.36 0.14 1.00 2.09 0.29 0.57 1.55 0.64 0.00 0.82
C3.5 1.20 0.00 1.00 0.00 1.00 1.83 1.20 0.00 1.00 0.40 0.67 1.58
C3.6 0.69 0.00 0.65 0.13 0.56 1.24 0.38 1.00 1.59 0.69 0.00 0.65
C4.1 1.00 0.00 1.25 0.60 0.60 2.00 0.70 0.40 1.75 0.20 1.00 2.50
C4.2 0.67 0.00 0.71 0.27 1.00 1.79 0.27 0.50 1.29 0.60 0.13 0.86
C4.3 1.25 0.25 1.86 0.75 0.75 2.43 1.38 0.00 1.57 0.50 1.00 2.71
C4.4 0.27 1.00 1.73 0.40 0.38 1.13 0.53 0.25 1.00 0.73 0.00 0.73
 K. Sari / Computers & Industrial Engineering 105 (2017) 338–347 345

Fig. 3. Overlay chart for Q index of each supplier.

Table 15 and Supplier B. As shown in Table 15, it is in such a way that there
95% certainty ranges for Q. is no overlap between them within 95 percent certainty range. This
Lower bound Upper bound can be interpreted as Supplier A and Supplier B are in different rank
Supplier A 0.13 0.84
positions. In addition, it is also observed from Fig. 3 that Q index of
Supplier B 0.92 2.10 Supplier C has a medium level overlap with those of Supplier A and
Supplier C 0.62 1.67 Supplier D. In Table 15, on the other hand, it is shown that Q
Supplier D 0.10 0.93 indexes of Supplier A, Supplier D, and Supplier C are not different
from each other from statistical point view. As a result, it can be
concluded that although average values of the Q indexes produce
Table 16 the rank list shown in Table 14, it should be kept in mind that
Revised rank of alternatives. Supplier C is also within the set of best alternatives together with
Rank Supplier A and Supplier D with 95 percent certainty range.
Supplier A 1
Therefore, the rank of alternatives should be revised accordingly
Supplier B 2 as presented in Table 16.
Supplier C 1
Supplier D 1
5. Conclusion and directions for future research

Deployment of GSCM practices is widespread in today’s busi-

ness environment among various industries. This popularity makes
and Supplier A and Supplier D are the best alternatives in terms of it necessary to develop and to adopt adequate tools for perfor-
their GSCM performances. mance evaluation of GSCM activities. In this regard, two questions
For comparative purpose, the approach of fuzzy AHP and fuzzy become prominent among researchers and practitioners of SCM.
VIKOR without Monte Carlo simulation is also performed. As a These are (1) extraction of a list of criteria for GSCM performance
result, the same rank of alternatives shown in Table 14 is obtained. evaluation and (2) development of a decision framework to be
This means that, our proposed framework with Monte Carlo simu- used for this purpose. In this perspective, this research study aims
lation and the framework without Monte Carlo simulation lead to to provide answers to these two questions. To do this, firstly, an
same ranking of alternatives. However, with our proposed model, extensive review of literature is performed from SCM point of view.
it is possible to perform one additional analysis by using the output Accordingly, a hierarchical decision structure with 18 items is
data of Monte Carlo simulation. To do this, we concentrate on obtained under the main headings of inbound, production, out-
index Q as it is the most important index for ranking alternatives. bound, and reverse logistics. Secondly, a novel decision framework
Namely, an alternative supplier with the best (minimum) Q value that integrates Monte Carlo simulation into the methods of fuzzy
is either the best supplier or one of the best suppliers. For this pur- AHP and fuzzy VIKOR is proposed.
pose, the overlay chart for Q is produced in Fig. 3 by using the data The strength of our proposed approach lies on its practical
obtained from Monte Carlo simulation. In addition, 95% certainty applicability and feasibility. In this regard, a numerical case is
ranges for index Q of each supplier are also provided in Table 15. employed to exemplify our proposed decision framework. At the
In fact, with the help of Fig. 3 and Table 15, it is possible to same time, a comparison is made with our proposed approach
obtain new insights about the rank of alternatives, which cannot and the approach without Monte Carlo simulation. By this way, it
be obtained otherwise. Fig. 3 reveals that there exist overlaps is validated that a better evaluation of GSCM performance is
between Q indexes of alternative suppliers. This overlap between obtained with our proposed framework. In reality, integration of
Supplier D and Supplier A is very strong, which means that they Monte Carlo simulation is not limited to fuzzy AHP and VIKOR
are in the same rank position with nearly same Q indexes. How- methods. Thus, our proposed approach of integrating Monte Carlo
ever, the overlap of Q indexes is very weak between Supplier A simulation can also be used as well for other MCDM methods.
346 K. Sari / Computers & Industrial Engineering 105 (2017) 338–347

For further research, methods of TOPSIS, ELECTRE or PRO- company. European Journal of Operational Research, 233(2), 432–447. http://dx.
doi.org/10.1016/j.ejor.2013.07.023.
METHEE under fuzzy environment can also be employed to
Kannan, D., Khodaverdi, R., Olfat, L., Jafarian, A., & Diabat, A. (2013). Integrated fuzzy
compare the results. multi criteria decision making method and multiobjective programming
approach for supplier selection and order allocation in a green supply chain.
Journal of Cleaner Production, 47, 355–367. http://dx.doi.org/10.1016/j.
References jclepro.2013.02.010.
Kaya, T., & Kahraman, C. (2011). Fuzzy multiple criteria forestry decision making
based on an integrated VIKOR and AHP approach. Expert Systems with
Awasthi, A., & Kannan, G. (2016). Green supplier development program selection
Applications, 38(6), 7326–7333. http://dx.doi.org/10.1016/j.eswa.2010.12.003.
using NGT and VIKOR under fuzzy environment. Computers & Industrial
Kumar, S., Teichman, S., & Timpernagel, T. (2012). A green supply chain is a
Engineering, 91, 100–108. http://dx.doi.org/10.1016/j.cie.2015.11.011.
requirement for profitability. International Journal of Production Research, 50(5),
Azevedo, S. G., Carvalho, H., & Cruz Machado, V. (2011). The influence of green
1278–1296. http://dx.doi.org/10.1080/00207543.2011.571924.
practices on supply chain performance: A case study approach. Transportation
Kwong, C. K., & Bai, H. (2002). A fuzzy AHP approach to the determination of
Research Part E: Logistics and Transportation Review, 47(6), 850–871. http://dx.
importance weights of customer requirements in quality function deployment.
doi.org/10.1016/j.tre.2011.05.017.
Journal of Intelligent Manufacturing, 13(5), 367–377. http://dx.doi.org/10.1023/
Beamon, B. M. (1999). Designing the green supply chain. Logistics Information
A:1019984626631.
Management, 12(4), 332–342. http://dx.doi.org/10.1108/09576059910284159.
Laari, S., Töyli, J., Solakivi, T., & Ojala, L. (2016). Firm performance and customer-
Bose, I., & Pal, R. (2012). Do green supply chain management initiatives impact stock
driven green supply chain management. Journal of Cleaner Production, 112,
prices of firms? Decision Support Systems, 52(3), 624–634. http://dx.doi.org/
1960–1970. http://dx.doi.org/10.1016/j.jclepro.2015.06.150.
10.1016/j.dss.2011.10.020.
Laosirihongthong, T., Adebanjo, D., & Tan, K. C. (2013). Green supply chain
Buckley, J. J. (1985). Fuzzy hierarchical analysis. Fuzzy Sets and Systems, 17(3),
management practices and performance. Industrial Management and Data
233–247. http://dx.doi.org/10.1016/0165-0114(85)90090-9.
Systems, 113(8), 1088–1109. http://dx.doi.org/10.1108/IMDS-04-2013-0164.
Büyüközkan, G., & Çifçi, G. (2012). Evaluation of the green supply chain
Lau, K. H. (2011). Benchmarking green logistics performance with a composite
management practices: A fuzzy ANP approach. Production Planning & Control,
index. Benchmarking: An International Journal, 18(6), 873–896. http://dx.doi.org/
23(6), 405–418. http://dx.doi.org/10.1080/09537287.2011.561814.
10.1108/14635771111180743.
Chan, H. K., He, H. W., & Wang, W. Y. C. (2012). Green marketing and its impact on
Lee, A. H. I., Kang, H.-Y., Hsu, C.-F., & Hung, H.-C. (2009). A green supplier selection
supply chain management in industrial markets. Industrial Marketing
model for high-tech industry. Expert Systems with Applications, 36(4),
Management, 41, 557–562. http://dx.doi.org/10.1016/j.indmarman.2012.04.002.
7917–7927. http://dx.doi.org/10.1016/j.eswa.2008.11.052.
Chien, M. K., & Shih, L. H. (2007). An empirical study of the implementation of green
Mangan, J., Lalwani, C., Butcher, T., & Javadpour, R. (2012). Global logistics & supply
supply chain management practices in the electrical and electronic industry
chain management (2nd ed.). West Sussex: John Wiley & Sons Ltd..
and their relation to organizational performances. International Journal of
Mathiyazhagan, K., Diabat, A., Al-Refaie, A., & Xu, L. (2015). Application of analytical
Environmental Science and Technology, 4(3), 383–394. Retrieved from http://
hierarchy process to evaluate pressures to implement green supply chain
search.proquest.com/docview/14831611?accountid=12217.
management. Journal of Cleaner Production, 107, 229–236. http://dx.doi.org/
CSCMP (2015). Definition of Supply Chain Management Retrieved from https://
10.1016/j.jclepro.2015.04.110.
cscmp.org/about-us/supply-chain-management-definitions.
Mohanty, R. P., & Prakash, A. (2014). Green supply chain management practices in
Diabat, A., & Govindan, K. (2011). An analysis of the drivers affecting the
India: An empirical study. Production Planning & Control, 25(16), 1322–1337.
implementation of green supply chain management. Resources, Conservation
http://dx.doi.org/10.1080/09537287.2013.832822.
and Recycling, 55(6), 659–667. http://dx.doi.org/10.1016/j.resconrec.2010.
Pourebrahim, S., Hadipour, M., Mokhtar, M. Bin., & Taghavi, S. (2014). Application of
12.002.
VIKOR and fuzzy AHP for conservation priority assessment in coastal areas:
Drohomeretski, E., Da Costa, S. G., & De Lima, E. P. (2014). Green supply chain
Case of Khuzestan district, Iran. Ocean & Coastal Management, 98, 20–26. http://
management: Drivers, barriers and practices within the Brazilian automotive
dx.doi.org/10.1016/j.ocecoaman.2014.05.009.
industry. Journal of Manufacturing Technology Management, 25(8), 1105–1134.
Rao, P., & Holt, D. (2005). Do green supply chains lead to competitiveness and
http://dx.doi.org/10.1108/JMTM-06-2014-00840.
economic performance? International Journal of Operations & Production
Dubey, R., Gunasekaran, A., Papadopoulos, T., & Childe, S. J. (2015). Green supply
Management, 25(9), 898–916. http://dx.doi.org/10.1108/01443570510613956.
chain management enablers: Mixed methods research. Sustainable Production
Raychaudhuri, S. (2008). Introduction to monte carlo simulation. In Simulation
and Consumption, 4, 72–88. http://dx.doi.org/10.1016/j.spc.2015.07.001.
Conference, 2008. WSC 2008 (pp. 91–100). Austin, TX: IEEE. http://dx.doi.org/10.
Dubois, D., Foulloy, L., Mauris, G., & Prade, H. (2004). Probability-possibility
1109/WSC.2008.4736059.
transformations, triangular fuzzy sets, and probabilistic inequalities. Reliable
Rezaie, K., Ramiyani, S. S., Nazari-Shirkouhi, S., & Badizadeh, A. (2014). Evaluating
Computing, 10, 273–297. http://dx.doi.org/10.1023/B:REOM.0000032115.
performance of Iranian cement firms using an integrated fuzzy AHP–VIKOR
22510.b5.
method. Applied Mathematical Modelling, 38(21–22), 5033–5046. http://dx.doi.
Fahimnia, B., Sarkis, J., & Davarzani, H. (2015). Green supply chain management: A
org/10.1016/j.apm.2014.04.003.
review and bibliometric analysis. International Journal of Production Economics,
Rostamzadeh, R., Govindan, K., Esmaeili, A., & Sabaghi, M. (2015). Application of
162, 101–114. http://dx.doi.org/10.1016/j.ijpe.2015.01.003.
fuzzy VIKOR for evaluation of green supply chain management practices.
Fu, H. P., Chu, K. K., Chao, P., Lee, H. H., & Liao, Y. C. (2011). Using fuzzy AHP and
Ecological Indicators, 49, 188–203. http://dx.doi.org/10.1016/j.ecolind.2014.
VIKOR for benchmarking analysis in the hotel industry. Service Industries
09.045.
Journal, 31(14), 2373–2389. http://dx.doi.org/10.1080/02642069.2010.503874.
Sanayei, A., Farid Mousavi, S., & Yazdankhah, A. (2010). Group decision making
Govindan, K., Kaliyan, M., Kannan, D., & Haq, A. N. (2014). Barriers analysis for green
process for supplier selection with VIKOR under fuzzy environment. Expert
supply chain management implementation in Indian industries using analytic
Systems with Applications, 37(1), 24–30. http://dx.doi.org/10.1016/j.eswa.2009.
hierarchy process. International Journal of Production Economics, 147, 555–568.
04.063.
http://dx.doi.org/10.1016/j.ijpe.2013.08.018.
Sari, K. (2013). Selection of RFID solution provider: A fuzzy multi-criteria decision
Govindan, K., Rajendran, S., Sarkis, J., & Murugesan, P. (2015). Multi criteria decision
model with Monte Carlo simulation. Kybernetes, 42(3), 448–465. http://dx.doi.
making approaches for green supplier evaluation and selection: A literature
org/10.1108/03684921311323680.
review. Journal of Cleaner Production, 98, 66–83. http://dx.doi.org/10.1016/j.
Sarkis, J., Zhu, Q. H., & Lai, K. H. (2011). An organizational theoretic review of green
jclepro.2013.06.046.
supply chain management literature. International Journal of Production
Haji Vahabzadeh, A., Asiaei, A., & Zailani, S. (2015). Green decision-making model in
Economics, 130(1), 1–15. http://dx.doi.org/10.1016/j.ijpe.2010.11.010.
reverse logistics using FUZZY-VIKOR method. Resources, Conservation and
Shen, L., Olfat, L., Govindan, K., Khodaverdi, R., & Diabat, A. (2013). A fuzzy multi
Recycling, 103, 125–138. http://dx.doi.org/10.1016/j.resconrec.2015.05.023.
criteria approach for evaluating green supplier’s performance in green supply
Hashemi, S. H., Karimi, A., & Tavana, M. (2015). An integrated green supplier
chain with linguistic preferences. Resources, Conservation and Recycling, 74,
selection approach with analytic network process and improved Grey relational
170–179. http://dx.doi.org/10.1016/j.resconrec.2012.09.006.
analysis. International Journal of Production Economics, 159, 178–191. http://dx.
Srivastava, S. K. (2007). Green supply-chain management: A state-of-the-art
doi.org/10.1016/j.ijpe.2014.09.027.
literature review. International Journal of Management Reviews, 9(1), 53–80.
Hsu, C.-C. (2013). Supply chain drivers that foster the development of green
http://dx.doi.org/10.1111/j.1468-2370.2007.00202.x.
initiatives in an emerging economy. International Journal of Operations &
Testa, F., & Iraldo, F. (2010). Shadows and lights of GSCM (Green Supply Chain
Production Management, 33(6), 656–688. http://dx.doi.org/10.1108/IJOPM-10-
Management): Determinants and effects of these practices based on a multi-
2011-0401.
national study. Journal of Cleaner Production, 18(10–11), 953–962. http://dx.doi.
Igarashi, M., De Boer, L., & Fet, A. M. (2013). What is required for greener supplier
org/10.1016/j.jclepro.2010.03.005.
selection? A literature review and conceptual model development. Journal of
Torfi, F., Farahani, R. Z., & Rezapour, S. (2010). Fuzzy AHP to determine the relative
Purchasing and Supply Management, 19(4), 1–17. http://dx.doi.org/10.1016/j.
weights of evaluation criteria and Fuzzy TOPSIS to rank the alternatives. Applied
pursup.2013.06.001.
Soft Computing, 10(2), 520–528. http://dx.doi.org/10.1016/j.asoc.2009.08.021.
Jabbour, A. B. L. de. S., Frascareli, F. C. de. O., & Jabbour, C. J. (2015). Green supply
Tseng, M. L., & Chiu, A. S. F. (2013). Evaluating firm’s green supply chain
chain management and firms’ performance: Understanding potential
management in linguistic preferences. Journal of Cleaner Production, 40,
relationships and the role of green sourcing and some other green practices.
22–31. http://dx.doi.org/10.1016/j.jclepro.2010.08.007.
Resources, Conservation and Recycling, 104, 366–374. http://dx.doi.org/10.1016/j.
Ubeda, S., Arcelus, F. J., & Faulin, J. (2011). Green logistics at Eroski: A case study.
resconrec.2015.07.017.
International Journal of Production Economics, 131(1), 44–51. http://dx.doi.org/
Kannan, D., de Jabbour, A. B. L. de S., & Jabbour, C. J. (2014). Selecting green suppliers
10.1016/j.ijpe.2010.04.041.
based on GSCM practices: Using fuzzy TOPSIS applied to a Brazilian electronics
 K. Sari / Computers & Industrial Engineering 105 (2017) 338–347 347

Uygun, Ö., & Dede, A. (2016). Performance evaluation of green supply chain practices in Chinese manufacturing enterprises. Journal of Operations
management using integrated fuzzy multi-criteria decision making techniques. Management, 22(3), 265–289. http://dx.doi.org/10.1016/j.jom.2004.01.005.
Computers & Industrial Engineering, 102, 502–511. Zhu, Q., Sarkis, J., & Lai, K. (2007). Green supply chain management: Pressures,
Walker, H., Di Sisto, L., & McBain, D. (2008). Drivers and barriers to environmental practices and performance within the Chinese automobile industry. Journal of
supply chain management practices: Lessons from the public and private Cleaner Production, 15(11), 1041–1052. http://dx.doi.org/10.1016/j.jclepro.
sectors. Journal of Purchasing and Supply Management, 14(1), 69–85. http://dx. 2006.05.021.
doi.org/10.1016/j.pursup.2008.01.007. Zhu, Q., Sarkis, J., & Lai, K. (2008). Confirmation of a measurement model for green
Wang, Z., Mathiyazhagan, K., Xu, L., & Diabat, A. (2015). A decision making trial and supply chain management practices implementation. International Journal of
evaluation laboratory approach to analyze the barriers to Green supply chain Production Economics, 111(2), 261–273. http://dx.doi.org/10.1016/j.ijpe.2006.
management adoption in a food packaging company. Journal of Cleaner 11.029.
Production. http://dx.doi.org/10.1016/j.jclepro.2015.09.142.
Wu, C., & Barnes, D. (2016). An integrated model for green partner selection and Kazim Sari earned his Ph.D. in Industrial Engineering from Istanbul Technical
supply chain construction. Journal of Cleaner Production, 112(3), 2114–2132. University, Istanbul, Turkey. He currently serves as Associate Professor and Chair-
http://dx.doi.org/10.1016/j.jclepro.2015.02.023. man of Industrial Engineering Department at Beykent University, Istanbul, Turkey.
Yang, C.-S., Lu, C.-S., Haider, J. J., & Marlow, P. B. (2013). The effect of green supply His principle research areas include analysis and design of supply chains and
chain management on green performance and firm competitiveness in the logistics systems through optimization and simulation modelling. His work has
context of container shipping in Taiwan. Transportation Research Part E: Logistics been published in European Journal of Operational Research, International Journal
and Transportation Review, 55, 55–73. http://dx.doi.org/10.1016/j.tre.2013. of Production Economics, International Journal of Physical Distribution and Logis-
03.005. tics Management, Industrial Management and Data Systems and Kybernetes.
Zhu, Q., & Sarkis, J. (2004). Relationships between operational practices and
performance among early adopters of green supply chain management