2
A novel hybrid MCDM approach based on fuzzy DEMATEL, fuzzy ANP and fuzzy
3
TOPSIS to evaluate green suppliers
4 Glin Bykzkan
Q1 , Gizem ifi
5 Industrial Engineering Department, Faculty of Engineering and Technology, Galatasaray University, 34357 Ortaky,
_
Istanbul, Turkey
6
7
9
a r t i c l e i n f o
10 11 Keywords:
12 Green supply chain
13 Supplier selection
14 Fuzzy ANP
15 Fuzzy DEMATEL
16 Fuzzy TOPSIS
17
1 8
a b s t r a c t
19 It is well known that green principles and strategies have become vital for companies as the public
20 awareness increased against their environmental impacts. A companys environmental performance is
21 not only related to the companys inner environmental efforts, but also it is affected by the suppliers
22 environmental performance and image. For industries, environmentally responsible manufacturing,
23 return ows, and related processes require green supply chain (GSC) and accompanying suppliers with
24 environmental/green competencies. During recent years, how to determine suitable and green suppliers
25 in the supply chain has become a key strategic consideration. Therefore this paper examines GSC man-
26 agement (GSCM) and GSCM capability dimensions to propose an evaluation framework for green suppli-
27 ers. However, the nature of supplier selection is a complex multi-criteria problem including both
28 quantitative and qualitative factors which may be in conict and may also be uncertain. The identied
29 components are integrated into a novel hybrid fuzzy multiple criteria decision making (MCDM) model
30 combines the fuzzy Decision Making Trial and Evaluation Laboratory Model (DEMATEL), the Analytical
31 Network Process (ANP), and Technique for Order Performance by Similarity to Ideal Solution (TOPSIS)
32 in a fuzzy context. A case study is proposed for green supplier evaluation in a specic company, namely
33 Ford Otosan.
34 2011 Published by Elsevier Ltd.
35
36
37 1. Introduction
38 Money, components, processes and information ows might
39 establish a supply chain management system but simultaneously,
40 due to government legislation and increasing awareness among
41 the people to protect the environment; rms today cannot ignore
42 environmental issues if they want to survive in the global market.
43 In this sense, green supply chain management (GSCM) has emerged
44 as a way for rms to achieve prot and market share objectives by
45 lowering environmental impacts and increasing ecological ef-
46 ciency (van Hock & Erasmus, 2000). In response to demands, com-
47 panies have to nd ways to incorporate environmental and social
48 aspects into their supply chain management.
49 In order to reap the greatest benets from environmental man-
50 agement, rms must integrate all members in the green supply
51 chain (GSC) (Lee, Kang, Hsu, & Hung, 2009). Among these expecta-
52 tions, increasing attention is devoted to suppliers social responsi-
53 bility with a particular focus on fair and legal use of natural
54 resources. Hence, strategic partnership with environmentally,
55 socially and economically powerful suppliers should be integrated
56 within the GSC for improving the performance in many directions
57 including reducing costs and lead time, eliminating wastages,
58 improving quality and exibility to meet the needs of the custom-
59 ers, etc. For this reason, the aim of this study is to propose an
60 evaluation model to judge the appropriateness of suppliers for an
61 organization which has environmental goals and measure the
62 validity of the model with a real case study.
63 There are various mathematical techniques for evaluation of
64 suppliers, such as data envelopment analysis (DEA) (Wu, 2009),
65 heuristics (He, Chaudhry, Lei, & Baohua, 2009; Sen, Bas ligil, Sen, &
66 Barali, 2007), analytic hierarchy process (AHP) (Sevkli, Koh, Zaim,
67 Demirbag, & Tatoglu, 2007), fuzzy AHP (Chan & Kumar, 2007; Lee
68 et al., 2009; Rao & Holt, 2005), fuzzy goal programming (Kumar,
69 Vrat, & Shankar, 2006; Tsai & Hung, 2009), fuzzy analytic network
70 process (ANP) (Lin, 2009; Tuzkaya & nt, 2008) in literature. For
71 the purpose of evaluating and selecting green suppliers, both qual-
72 itative and quantitative factors must be considered. Thus, green
73 supplier selection is a kind of multiple criteria decision making
74 (MCDM) problem and we need to employ MCDM methods to han-
75 dle it appropriately. Here emphasis is placed on the relationships of
76 factors which can be handled by ANP (Saaty, 1996) effectively. The
77 ANP can deal with the dependence in feedback systematically. In
78 this study also Decision Making Trial and Evaluation Laboratory
79 (DEMATEL) method (Gabus & Fontela, 1972) is used to extract the
80 mutual relationships of interdependencies within criteria and the
81 strength of interdependence. Lastly to choose the alternative for
82 ideal solution of this problem, Technique for Order Performance
0957-4174/$ - see front matter 2011 Published by Elsevier Ltd.
doi:10.1016/j.eswa.2011.08.162
n
j1
u
ij
; then
511 511
512
X s
A: 1 514 514
515 Step 2.3: Acquire fuzzy total-relation matrix. As soon as the normal-
516 ized direct-relation matrix
X is obtained, the total-relation matrix
517
_
_
_
; X
2
0 m
12
. . . m
1n
m
21
0 . . . m
2n
: :
: :
: :
m
n1
m
n2
. . . 0
_
_
_
_
;
X
3
0 u
12
. . . u
1n
u
21
0 . . . u
2n
: :
: :
: :
u
n1
u
n2
. . . 0
_
_
_
_
:
523 523
524 According to the crisp case, we dene the total-relation fuzzy
525 matrix
T through (2):
526
T
XI
X
1
: 2 528 528
529
Let
T
~
t
11
~
t
12
. . .
~
t
1n
~
t
21
~
t
22
. . .
~
t
2n
: :
: :
: :
t
n1
t
n2
. . .
~
t
nn
_
_
_
_
;
531 531
532 where
~
t
ij
l
0
ij
; m
0
ij
; u
0
ij
then
533
Matrix l
0
ij
_ _
X
l
I X
l
1
; 3
535 535
536
Matrix m
0
ij
_ _
X
m
I X
m
1
; 4
538 538
539
Matrix u
0
ij
_ _
X
u
I X
u
1
: 5
541 541
542 By applying these formulas, the total-relation matrix acquired is
543 given in Table 4.
544 Step 2.4: Obtain the inner dependence matrix. In this step, after
545 defuzzication of the total-relation matrix
T by using Eq. (6), the
546 sum of each column in total-relation matrix became equal to 1
547 by the normalization method.
548
F
~
t
ij
1=2
_
1
0
inf
x2R
~
t
a
ij
sup
x2R
~
t
a
ij
_ _
da: 6
550 550
551 Then the inner dependence matrix can be acquired to put in the un-
552 weighted supermatrix of ANP later. Table 5 shows the inner depen-
553 dence matrix of organizational performance dimension and can be
554 seen in Fig. 6 as matrix B of the supermatrix.
555 Step 3: Establish remaining relations using the fuzzy ANP. In
556 ANP, like AHP, pair wise comparisons of the elements in each level
557 are conducted with respect to their relative importance towards
558 their control criterion. By using triangular fuzzy numbers again,
559 the relative strength of each pair of elements and the preferences
560 of the decision maker in the same hierarchy are indicated. Via
561 pair-wise comparison, the fuzzy judgment matrix
A
0
is constructed
562 as:
6 G. Bykzkan, G. ifi / Expert Systems with Applications xxx (2011) xxxxxx
ESWA 6898 No. of Pages 12, Model 5G
13 September 2011
Please cite this article in press as: Bykzkan, G., & ifi, G. A novel hybrid MCDM approach based on fuzzy DEMATEL, fuzzy ANP and fuzzy TOPSIS to
evaluate green suppliers. Expert Systems with Applications (2011), doi:10.1016/j.eswa.2011.08.162
Edited by Foxit Reader
Copyright(C) by Foxit Software Company,2005-2007
For Evaluation Only.
563
A
0
~ a
0
11
~ a
0
12
. . . ~ a
0
1n
~
a
0
21
~
a
0
22
. . .
~
a
0
2n
: :
: :
: :
~
a
0
n1
~
a
0
n2
. . .
~
a
0
nn
_
_
_
_
; 7
565 565
566 where ~ a
0
ij
l
0
ij
; m
0
ij
; u
0
ij
_ _
indicates the importance among the com-
567 pared criteria (importance of i over j) where i = j = 1,2, . . . , n. Table
568 6 gives examples of linguistic and fuzzy evaluations between green
569 logistics dimensions and goal. This evaluation can be seen in Fig. 6
570 as matrix A of supermatrix. Other evaluations are populated in the
571 same way.
572 Step 3.1: Calculate the relative importance weights. The priority
573 vectors for each pairwise comparison matrix will be needed to
574 complete the various supermatrix submatrices. Estimate triangular
575 fuzzy priorities ~ w
k
where k = 1, 2. . . , n from the judgment matrix.
576 The logarithmic least-squares method can be used for calculating
577 these weights (Ont, Kara, & Isik, 2009; Ramik, 2007; Tuzkaya
578 et al., 2009; Tuzkaya & nt, 2008):
579
~ w
k
w
l
k; w
m
k
; w
u
k
k 1; 2; . . . ; n where; 581 581
582
w
s
k
n
i1
a
s
kj
1=n
n
i1
n
i1
a
m
ij
1=n
; s 2 fl; m; ug 8
584 584
585 for 0 < a 6 1 and all i, j, where i = 1, 2, . . . , n, j = 1, 2. . . , n. In order to
586 control the result of the method, the consistency ratio for each of
587 the matrices and the overall inconsistency for the hierarchy are cal-
588 culated. The Consistency Ratio (CR) is used to directly estimate the
589 consistency of the pair-wise comparisons and should be less than
590 0.10. Then it can be said the comparisons are acceptable, otherwise
591 they are not acceptable. In this study, the inconsistency ratios for all
592 the comparison matrices were calculated for the mean values of the
593 fuzzy numbers. Because the lower and upper values provide exi-
594 bility for human judgments, they are not expected to have rigid
595 consistency.
Organizational
Performance Dimension
Improve green supply chain
management activities
Cost
(OP2)
Quality
(OP1)
Time
(OP3)
Flexibility
(OP4)
GOAL
Production
(GL2)
Procurement
(GL1)
Distribution
(GL3)
Packaging
(GL5)
Reverse L.
(GL4)
Green Logistics
Dimension
Recycle
(GOA2)
Reduce
(GOA1)
Remanufacture
(GOA3)
Disposal
(GOA5)
Reuse
(GOA4)
Green Organizational
Activities Dimension
(A)
(B)
(D)
(E)
(F)
(H) ) I ( ) G (
(C)
Organization
(EC1)
Financial Performance
(EC2)
Green Competencies
(EC5)
Service Quality
(EC3)
Technology
(EC4)
Green Supplier
Evaluation Criteria
S1 S2 S3 S4 S5
Fig. 4. Detailed evaluation model.
Table 1
Corresponding linguistic terms for evaluation.
Linguistic term Abbrev. Fuzzy scales
None N (0, 0, 1)
Very Low VL (0, 0.1, 0.2)
Low L (0.1, 0.2, 0.3)
Fairly Low FL (0.2, 0.3, 0.4)
More or less Low ML (0.3, 0.4, 0.5)
Medium M (0.4, 0.5, 0.6)
More or less Good MG (0.5, 0.6, 0.7)
Fairly Good FG (0.6, 0.7, 0.8)
Good G (0.7, 0.8, 0.9)
Very Good VG (0.8, 0.9, 1)
Excellent E (0.9, 1, 1) Fig. 5. Fuzzy membership functions for linguistic values.
G. Bykzkan, G. ifi / Expert Systems with Applications xxx (2011) xxxxxx 7
ESWA 6898 No. of Pages 12, Model 5G
13 September 2011
Please cite this article in press as: Bykzkan, G., & ifi, G. A novel hybrid MCDM approach based on fuzzy DEMATEL, fuzzy ANP and fuzzy TOPSIS to
evaluate green suppliers. Expert Systems with Applications (2011), doi:10.1016/j.eswa.2011.08.162
596 Step 3.2: Defuzzify the weights obtained from fuzzy matrices. In
597 this step, defuzzication of the weights is done in the same way as
598 Eq. (6).
599 Such an example, priority calculation of the production one of
600 the green logistics dimensions with respect to goal (from Table 6)
601 is as follows. By applying Eq. (8), the fuzzy weight is obtained as,
602
603 Then using this fuzzy vector and applying Eq. (6), defuzzied
604 weight 0.4046 is obtained.
605 Step 4: Forma supermatrix. ANP uses the formation of a superm-
606 atrix to allow for the resolution of the effects of the interdepen-
607 dence that exists between the clusters within the decision
608 network hierarchy. The supermatrix is a partitioned matrix, where
609 each submatrix is composed of a set of relationships between two
610 clusters in the graphical model. A generic supermatrix is shown
611 in Fig. 6, with the notation representing the various relationships
612 fromFig. 4; for instance, A is the submatrix representing the inu-
613 ence relationship between green logistics dimension elements and
614 control factor of the goal of selecting a green supplier.
615 By entering the priorities found by fuzzy DEMATEL and fuzzy
616 ANP into the appropriate columns, initial supermatrix can be con-
617 structed. Table 7 presents the initial supermatrix of the study.
618 4.1: Solve the supermatrix. To complete this task, rstly each of
619 the columns may either be normalized by dividing each weight in
620 the column by the sum of that column. Then, the nal step in the
621 process is to obtain a priority ranking for each of the alternatives.
622 To derive the overall priorities of elements, the normalized
623 supermatrix is raised to limiting powers to calculate the overall
624 priorities, and thus the cumulative inuence of each element on
625 every other element with which it interacts is obtained. In this
626 case, the supermatrix is raised to the power 25. This weighted
627 supermatrix is shown in Table 8.
628 According to this weighted supermatrix, weights of the criteria
629 on the objective of green supplier selection are shown in the Goal
630 column to use in fuzzy TOPSIS steps later.
631 Step 5: Evaluate the alternatives by using fuzzy TOPSIS steps.
632 The technique is adapted from Chen (2000) and the steps of the
633 methodology are as follows.
634 Step 5.1: Establish fuzzy decision matrix for evaluation of the
635 green supplier alternatives. With m alternatives and n criteria, fuz-
636 zy MCDM problem can be expressed as:
637
Table 2
Fuzzy direct relation matrix of organizational performance dimension.
Quality (OP1) Cost (OP2) Time (OP3) Flexibility (OP4)
Quality (OP1) (0.8, 0.9, 1) (0.4, 0.5, 0.6) (0, 0.1, 0.2)
Cost (OP2) (0.5, 0.6, 0.7) (0.4, 0.5, 0.6) (0, 0, 0.1)
Time (OP3) (0.4, 0.5, 0.6) (0.6, 0.7, 0.8) (0, 0.1, 0.2)
Flexibility (OP4) (0.3, 0.4, 0.5) (0.8, 0.9, 1) (0.5, 0.6, 0.7)
Table 3
Fuzzy normalized direct relation matrix of organizational performance dimension.
(OP1) (OP2) (OP3) (OP4)
(OP1) (0.36, 0.40, 45) (0.18, 0.22, 0.27) (0, 0.04, 0.09)
(OP2) (0.22, 0.27, 0.31) (0.18, 0.22, 0.27) (0, 0, 0.04)
(OP3) (0.18, 0.22, 0.27) (0.27, 0.31, 0.36) (0, 0.04, 0.09)
(OP4) (0.13, 0.18, 0.22) (0.36, 0.40, 0.45) (0.22, 0.27, 0.31)
Table 4
Fuzzy total direct relation matrix of organizational performance dimension.
(OP1) (OP2) (OP3) (OP4)
(OP1) (0.17, 0.33, 0.70) (0.51, 0.73, 1.23) (0.31, 0.49, 0.89) (0, 0.08, 0.29)
(OP2) (0.32, 0.47, 0.82) (0.19, 0.34, 0.76) (0.28, 0.42, 0.78) (0, 0.04, 0.23)
(OP3) (0.30, 0.48, 0.86) (0.42, 0.63, 1.10) (0.13, 0.27, 0.62) (0, 0.08, 0.28)
(OP4) (0.34, 0.56, 1.03) (0.60, 0.86, 1.43) (0.40, 0.61, 1.07) (0, 0.05, 0.26)
Table 5
Inner dependence matrix of organizational performance dimension.
(OP1) (OP2) (OP3) (OP4)
(OP1) 0.19 0.28 0.27 0.29
(OP2) 0.25 0.14 0.24 0.20
(OP3) 0.26 0.25 0.16 0.28
(OP4) 0.30 0.33 0.33 0.23
Goal GL OP GOA EC
Goal 0 0 0 0 0
Green Logistics Dimensions (GL) A E D 0 0
Organizational Performance (OP) 0 C B 0 0
Green Organizational Activities (GOA) 0 F 0 0 0
Green Supplier Evaluation Criteria (EC) 0 H G I I
Fig. 6. General submatrix notation for supermatrix. Note: I is the identity matrix.
w
l
k
0:20:10:61=11
1=5
10:91=0:41=0:31=0:3
1=5
1=0:911=0:41=0:21=0:2
1=5
0:40:411=0:71=0:7
1=5
0:30:20:711
1=5
0:30:20:71=11
1=5
0:3048:
w
m
k
0:30:20:71=11
1=5
10:91=0:41=0:31=0:3
1=5
1=0:911=0:41=0:21=0:2
1=5
0:40:411=0:71=0:7
1=5
0:30:20:711
1=5
0:30:20:71=11
1=5
0:3828:
w
u
k
0:40:30:81=0:91
1=5
10:91=0:41=0:31=0:3
1=5
1=0:911=0:41=0:21=0:2
1=5
0:40:411=0:71=0:7
1=5
0:30:20:711
1=5
0:30:20:71=11
1=5
0:5478:
8 G. Bykzkan, G. ifi / Expert Systems with Applications xxx (2011) xxxxxx
ESWA 6898 No. of Pages 12, Model 5G
13 September 2011
Please cite this article in press as: Bykzkan, G., & ifi, G. A novel hybrid MCDM approach based on fuzzy DEMATEL, fuzzy ANP and fuzzy TOPSIS to
evaluate green suppliers. Expert Systems with Applications (2011), doi:10.1016/j.eswa.2011.08.162
Table 6
Linguistic and fuzzy evaluation matrices of green logistics with respect to goal.
Linguistic terms Fuzzy terms
GL1 GL2 GL3 GL4 GL5 GL1 GL2 GL3 GL4 GL5
1 VG 1 (0.8, 0.9, 1) (1/0.5, 1/0.4, 1/0, 3) (1/0.4, 1/0.3, 1/0, 2) (1/0.4, 1/0.3, 1/0, 2)
1 (1/1, 1/0.9, 1/0, 8) 1 (1/0.5, 1/0.4, 1/0, 3) (1/0.3, 1/0.2, 1/0, 1) (1/0.3, 1/0.2, 1/0, 1)
ML ML 1 (0.3, 0.4, 0.5) (0.3, 0.4, 0.5) 1 (1/0.8, 1/0.7, 1/0, 6) (1/0.8, 1/0.7, 1/0, 6)
FL L FG 1 E (0.2, 0.3, 0.4) (0.1, 0.2, 0.3) (0.6, 0.7, 0.8) 1 (0.9, 1, 1)
FL L FG 1 (0.2, 0.3, 0.4) (0.1, 0.2, 0.3) (0.6, 0.7, 0.8) (1/1, 1/1, 1/0, 9) 1
Table 7
Initial supermatrix of green supplier selection for the improvement of GSC Q4 .
Goal GL1 GL2 GL3 GL4 GL5 OP1 OP2 OP3 OP4 GOA1 GOA2 GOA3 GOA4 GOA5 C1 C2 C3 C4 C5
Goal 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
GL1 0.32 0.18 0.30 0.22 0.25 0.29 0.43 0.44 0.08 0.06 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
GL2 0.40 0.34 0.21 0.26 0.30 0.28 0.29 0.20 0.34 0.16 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
GL3 0.13 0.13 0.10 0.09 0.14 0.16 0.14 0.20 0.34 0.44 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
GL4 0.08 0.15 0.21 0.17 0.11 0.13 0.06 0.07 0.17 0.26 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
GL5 0.09 0.20 0.18 0.26 0.19 0.14 0.07 0.07 0.09 0.08 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
OP1 0.00 0.61 0.12 0.24 0.13 0.52 0.19 0.28 0.27 0.29 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
OP2 0.00 0.26 0.06 0.09 0.59 0.26 0.25 0.14 0.24 0.20 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
OP3 0.00 0.08 0.25 0.53 0.23 0.16 0.26 0.25 0.16 0.28 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
OP4 0.00 0.06 0.57 0.14 0.06 0.06 0.30 0.33 0.33 0.23 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
GOA1 0.00 0.46 0.46 0.46 0.46 0.46 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
GOA2 0.00 0.10 0.10 0.10 0.10 0.10 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
GOA3 0.00 0.15 0.15 0.15 0.15 0.15 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
GOA4 0.00 0.26 0.26 0.26 0.26 0.26 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
GOA5 0.00 0.05 0.05 0.05 0.05 0.05 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
C1 0.00 0.20 0.15 0.20 0.25 0.10 0.19 0.11 0.20 0.25 0.10 0.10 0.10 0.15 0.20 1.00 0.00 0.00 0.00 0.00
C2 0.00 0.25 0.25 0.25 0.10 0.20 0.20 0.30 0.25 0.25 0.15 0.10 0.15 0.15 0.20 0.00 1.00 0.00 0.00 0.00
C3 0.00 0.25 0.25 0.15 0.20 0.20 0.30 0.19 0.15 0.19 0.15 0.10 0.15 0.15 0.10 0.00 0.00 1.00 0.00 0.00
C4 0.00 0.20 0.25 0.30 0.15 0.20 0.20 0.20 0.30 0.26 0.24 0.35 0.24 0.20 0.20 0.00 0.00 0.00 1.00 0.00
C5 0.00 0.10 0.10 0.10 0.30 0.30 0.11 0.20 0.10 0.05 0.36 0.35 0.36 0.35 0.30 0.00 0.00 0.00 0.00 1.00
Table 8
Weighted supermatrix of green supplier selection for the improvement of GSC.
GOAL GL1 GL2 GL3 GL4 GL5 OP1 OP2 OP3 OP4 GOA1 GOA2 GOA3 GOA4 GOA5
C1 0.16 0.16 0.16 0.16 0.17 0.14 0.18 0.15 0.18 0.20 0.11 0.10 0.11 0.15 0.20
C2 0.20 0.20 0.20 0.20 0.17 0.19 0.21 0.24 0.22 0.22 0.15 0.10 0.15 0.15 0.20
C3 0.19 0.20 0.19 0.17 0.18 0.18 0.23 0.19 0.18 0.19 0.15 0.10 0.15 0.15 0.10
C4 0.23 0.22 0.24 0.25 0.21 0.23 0.22 0.22 0.26 0.25 0.24 0.35 0.24 0.20 0.20
C5 0.22 0.22 0.21 0.22 0.27 0.26 0.16 0.20 0.16 0.14 0.35 0.35 0.35 0.35 0.30
Table 9
Linguistic and fuzzy decision matrix for green supplier alternative evaluation.
Linguistic terms Fuzzy terms
C1 C2 C3 C4 C5 C1 C2 C3 C4 C5
S1 FG G FG MG ML (0.6, 0.7, 0.8) (0.7, 0.8, 0.9) (0.6, 0.7, 0.8) (0.5, 0.6, 0.7) (0.3, 0.4, 0.5)
S2 VG G G VG VG (0.8, 0.9, 1) (0.7, 0.8, 0.9) (0.7, 0.8, 0.9) (0.8, 0.9, 1) (0.8, 0.9, 1)
S3 VG E E G G (0.8, 0.9, 1) (0.9, 1, 1) (0.9, 1, 1) (0.7, 0.8, 0.9) (0.7, 0.8, 0.9)
S4 MG VG G VG MG (0.5, 0.6, 0.7) (0.8, 0.9, 1) (0.7, 0.8, 0.9) (0.8, 0.9, 1) (0.5, 0.6, 0.7)
S5 M MG FG G G (0.4, 0.5, 0.6) (0.5, 0.6, 0.7) (0.6, 0.7, 0.8) (0.7, 0.8, 0.9) (0.7, 0.8, 0.9)
Table 10
Weighted decision matrix for green supplier alternative evaluation.
C1 C2 C3 C4 C5
S1 (0.10, 0.11, 0.13) (0.14, 0.16, 0.18) (0.11, 0.13, 0.15) (0.12, 0.14, 0.16) (0.07, 0.09, 0.11)
S2 (0.13, 0.14, 0.16) (0.14, 0.16, 0.18) (0.13, 0.15, 0.17) (0.18, 0.21, 0.23) (0.18, 0.20, 0.22)
S3 (0.13, 0.14, 0.16) (0.18, 0.20, 0.20) (0.17, 0.19, 0.19) (0.16, 0.18, 0.21) (0.15, 0.18, 0.20)
S4 (0.08, 0.10, 0.11) (0.16, 0.18, 0.20) (0.13, 0.15, 0.17) (0.18, 0.21, 0.23) (0.11, 0.13, 0.15)
S5 (0.06, 0.08, 0.10) (0.10, 0.12, 0.14) (0.11, 0.13, 0.15) (0.16, 0.18, 0.21) (0.15, 0.18, 0.20)
G. Bykzkan, G. ifi / Expert Systems with Applications xxx (2011) xxxxxx 9
ESWA 6898 No. of Pages 12, Model 5G
13 September 2011
Please cite this article in press as: Bykzkan, G., & ifi, G. A novel hybrid MCDM approach based on fuzzy DEMATEL, fuzzy ANP and fuzzy TOPSIS to
evaluate green suppliers. Expert Systems with Applications (2011), doi:10.1016/j.eswa.2011.08.162
639 639
640
D represents the fuzzy decision matrix with alternatives A and cri-
641 teria C, and can be seen with linguistic and fuzzy terms in Table 9.
642 Step 5.2: Normalize the decision matrix. Normalized fuzzy deci-
643 sion matrix
R is calculated as:
644
R ~r
ij
n
; i 1; 2. . . ; m; j 1; 2. . . n; 646 646
647
~r
ij
a
ij
C
j
;
b
ij
C
j
;
c
ij
C
j
_ _
; 9
649 649
650 where C
j
max
i
C
ij
.To avoid the complicated normalization for-
651 mula used in the classical TOPSIS, the linear scale transformation
652 is used to transform the various criteria scales into a comparable
653 scale (Chen, 2000). Linear scale transformation for normalization
654 is also employed by Kuo et al. (2007) and Celik et al. (2009). Here
655 normalized decision matrix remains the same because max C
ij
= 1.
656 Step 5.3: Compute weighted decision matrix. Weighted normal-
657 ized fuzzy decision matrix that is shown in Table 10 is computed
658 by using Eq. (11), where w
j
is the weight for the criterion j obtained
659 from supermatrix
660
~ v
ij
~r
ij
~ w
j
; 10 662 662
663 where ~ v ~ v
ij
n
; i 1; 2; . . . ; m; j 1; 2; . . . ; n.
664 Step 5.4: Calculate the distances from positive and negative
665 ideal points. Since the triangular fuzzy numbers are included in
666 [0, 1] range, positive and negative ideal reference points (FPIRP,
667 FNIRP) are as follows:
668
A
f~ v
1
; ~ v
2
; . . . ~ v
n
g; A
f~ v
1
; ~ v
2
; . . . ~ v
n
g; 11 670 670
671 where ~ v
j
1; 1; 1, ~ v
j
0; 0; 0.
672 The next step is to calculate the distance of alternatives from
673 FPIRP and FNIRP.
674
d
n
j1
d~ v
ij
; ~ v
j
; i 1; 2; . . . ; m; j 1; 2. . . n; 12
676 676
d
n
j1
d~ v
ij
; ~ v
j
; i 1; 2. . . m; j 1; 2; . . . ; n; 13
678 678
679
d
A;
1
3
a
1
b
1
2
a
2
b
2
2
a
3
b
3
2
_ _
_
: 14
681 681
682 Positive and negative distances of the green supplier alternatives
683 can be seen in Table 11.
684 Step 5.5: Rank the alternatives. The performance indices are
685 computed to rank the alternatives. Performance indices are sorted
686 in a decreasing order. Table 12 shows the nal ranking and accord-
687 ing to this hybrid methodology, the best possible green supplier is
688 S3 with a score of 0.1767.
689 5. Conclusion
690 This study suggests a novel hybrid MCDM approach to evaluate
691 green suppliers for the need of improving GSCM initiatives. Based
692 on the literature survey and with the validation of industrial ex-
693 perts, possible green supplier evaluation criteria were dened
694 and a new evaluation model was formulated. The proposed model
695 was implemented in Ford Otosan, one of the pioneering companies
696 about environmental subjects in Turkey.
697 The combined fuzzy ANP and fuzzy DEMATEL approaches used
698 in this study offered a more precise and accurate analysis by inte-
699 grating interdependent relationships within and among a set of cri-
700 teria. Moreover, fuzzy TOPSIS method helped to choose the
701 alternative for ideal solution of this problem efciently.
702 While it is believed that the presented model provides value,
703 there are also further points that can be included. To our knowl-
704 edge, no previous work investigated such a problem by an inte-
705 grated method with DEMATEL, ANP, and TOPSIS in fuzzy
706 environment. As the proposed approach is novel, it might be
707 applied to other MCDM problems.
708 6. Uncited references
709 Fekri et al. (2009) Tseng (2010) and Tseng and Lin (2009). Q2
710 Acknowledgements
711 The authors would like to express their deep gratitude towards
712 the industrial experts of Ford Otosan, especially to Vedat Okyar
713 (Senior Purchasing Manager-Trim Parts in Glck Plant) and Serdar
714 Aydn (New Project Chief in Glck Plant). The authors acknowl-
715 edge Alis an apan for his contribution in realizing the application
716 part and R.Ufuk Bilsel for his help in improving the linguistic qual-
717 ity of the paper.
Table 11
Positivenegative distances and nal performance indices of green supplier alternatives.
Positive Negative
d1 d2 d3 d4 d5 d
TOT
d1 d2 d3 d4 d5 d
TOT
S1 0.89 0.84 0.87 0.86 0.91 4.37 0.11 0.16 0.13 0.14 0.09 0.64
S2 0.86 0.84 0.85 0.79 0.80 4.14 0.14 0.16 0.15 0.21 0.20 0.86
S3 0.86 0.81 0.82 0.82 0.82 4.12 0.14 0.19 0.18 0.18 0.18 0.88
S4 0.90 0.82 0.85 0.79 0.87 4.23 0.10 0.18 0.15 0.21 0.13 0.77
S5 0.48 0.88 0.87 0.82 0.82 3.87 0.08 0.12 0.13 0.18 0.18 0.70
Table 12
Final performance indices of green supplier
alternatives Q5 .
Performance index
S1 0.1272
S2 0.1728
S3 0.1767
S4 0.1541
S5 0.1528
10 G. Bykzkan, G. ifi / Expert Systems with Applications xxx (2011) xxxxxx
ESWA 6898 No. of Pages 12, Model 5G
13 September 2011
Please cite this article in press as: Bykzkan, G., & ifi, G. A novel hybrid MCDM approach based on fuzzy DEMATEL, fuzzy ANP and fuzzy TOPSIS to
evaluate green suppliers. Expert Systems with Applications (2011), doi:10.1016/j.eswa.2011.08.162
718 References
719 Aiello, G. (2009). Clean agent selection approached by fuzzy TOPSIS decision-
720
making method. Fire Technology, 45, 405418.
721
Amiri, M., Zandieh, M., Soltani, R., & Vahdani, B. (2009). A hybrid multi-criteria
722 decision-making model for rms competence evaluation. Expert Systems with
723
Applications, 36, 1231412322.
724 Awasthi, A., Chauhan, S. S., & Goyal, S. K. (2010). A fuzzy multicriteria approach for
725
evaluating environmental performance of suppliers. International Journal of
726 Production Economics, 126(2), 370378.
727
Bai, C., & Sarkis, J. (2010). Green supplier development: Analytical evaluation using
728
rough set theory. Journal of Cleaner Production, 18(12), 12001210.
729 Cavallaro, F. (2010). Fuzzy TOPSIS approach for assessing thermal-energy
730
storage in concentrated solar power (CSP) systems. Applied Energy, 87(2),
731 496503.
732
Chan, F. T. S., & Kumar, N. (2007). Global supplier development considering risk
733
factors using fuzzy extended AHP-based approach. Omega, 35, 417431.
734 Chang, B., Chang, C.-W., & Wu, C.-H. (in press). Fuzzy DEMATEL method for
735
developing supplier selection criteria. Expert Systems with Applications
Q3
.
736 Chen, C.C., Tseng, M.L., & Lin, Y.H. (2008). Using fuzzy DEMATEL to develop a causal
737
and effect model of hot spring service quality expectation. In IEEE international
738 conference on industrial engineering and engineering management (pp. 1004
739
1008).
740
Chen, J.-K., & Chen, I-S. (2010). Using a novel conjunctive MCDM approach based on
741 DEMATEL, fuzzy ANP, and TOPSIS as an innovation support system for
742
Taiwanese higher education. Expert Systems with Applications, 37, 19811990.
743 Chen, S. J., & Hwang, C. L. (1992). Fuzzy multiple attribute decision-making methods
744
and application. In Lecture notes in economics and mathematical systems. New
745
York: Springer.
746 Chen-Yi, H., Ke-Ting, C., & Gwo-Hshiung, T. (2007). FMCDM with fuzzy DEMATEL
747
approach for customers choice behavior model. International Journal of Fuzzy
748 Systems, 9(4), 236246.
749
Dagdeviren, M., & Yksel,
_
I. (2010). A fuzzy analytic network process (ANP) model
750 for measurement of the sectoral competition level (SCL). Expert Systems with
751 Applications, 37(2), 10051014.
752
Faruk, A. C., Lamming, R. C., Cousins, P. D., & Bowen, F. E. (2002). Analyzing,
753 mapping, and managing environmental impacts along the supply chain. Journal
754
of Industrial Ecology, 5(2), 1336.
755 Fekri, R., Aliahmadi, A., & Fathian, M. (2009). Identifying the cause and effect factors
756
of agile NPD process with fuzzy DEMATEL method: The case of Iranian
757
companies. Journal of Intelligent Manufacturing, 20(6), 637648.
758 Gabus, A., & Fontela, E. (1972). World problems, an invitation to further thought within
759
the framework of DEMATEL. Switzerland, Geneva: Battelle Geneva Research
760 Centre.
761
Gabus, A., & Fontela, E. (1973). Perceptions of the world problematique:
762
Communication procedure, communicating with those bearing collective
763 responsibility. Switzerland Geneva: Battelle Geneva Research Centre. no. 1..
764
Goosey, M. (2004). End-of-life electronics legislation An industry perspective.
765 Circuit World, 30(2), 4145.
766
Gumus, A. T. (2009). Evaluation of hazardous waste transportation rms by using a
767 two step fuzzy-AHP and TOPSIS methodology. Expert Systems with Applications,
768
36, 40674074.
769
Hall, J. (2000). Environmental supply chain dynamics. Journal of Cleaner Production,
770 8(6), 206225.
771
Handeld, R. B., & Nichols, E. L. (2002). Supply chain redesign. Upper Saddle River, NJ:
772 Financial Times Prentice Hall..
773
Handeld, R., Walton, S., Sroufe, R., & Melnyk, S. (2002). Applying environmental
774
criteria to supplier assessment: A study of the application of the analytical
775 hierarchy process. European Journal of Operational Research, 141, 7087.
776
He, S., Chaudhry, S. S., Lei, Z., & Baohua, W. (2009). Stochastic vendor selection
777 problem: Chance-constrained model and genetic algorithms. Annals of
778
Operations Research, 168, 169179.
779 Humphreys, P. K., Wong, Y. K., & Chan, F. T. S. (2003). Integrating environmental
780
criteria into the supplier selection process. Journal of Materials Processing
781
Technology, 138, 349356.
782 Hwang, C. L., & Yoon, K. (1981). Multiple attribute decision-making: Methods and
783
application. New York: Springer.
784 I, Y. T., & Yurdakul, M. (2010). Development of a quick credibility scoring decision
785
support system using fuzzy TOPSIS. Expert Systems with Applications, 37,
786
567574.
787 Jabbour, A. B. L. S., & Jabbour, C. J. C. (2009). Are supplier selection criteria going
788
green? Case studies of companies in Brazil. Industrial Management & Data
789 Systems, 109, 477495.
790
Kelemenis, A., & Askounis, D. (2010). A new TOPSIS-based multi-criteria approach to
791 personnel selection. Expert Systems with Applications, 37(7), 49995008.
792
Ketchen, D. J., Jr., & Hult, G. T. M. (2007). Bridging organization theory and supply
793
chain management: The case of best value supply chains. Journal of Operations
794 Management, 25(2), 573580.
795
Kubokawa, S., & Saito, I. (2000). Manufacturing management strategies for
796 environmental protection: Toward the environmental upgrading of
797
management and manufacturing systems to cope with environmental laws.
798
Production Planning & Control, 11(2), 107112.
799 Kumar, M., Vrat, P., & Shankar, R. (2006). A fuzzy goal programming approach for
800
vendor selection problem in a supply chain. International Journal of Production
801 Economics, 101, 273285.
802 Kuo, R. J., Wang, Y. C., & Tien, F. C. (2010). Integration of articial neural network
803
and MADA methods for green supplier selection. Journal of Cleaner Production,
804 18(12), 11611170.
805
Kurk, F., & Eagan, P. (2008). The value of adding design-for the-environment to
806 pollution prevention assistance options. Journal of Cleaner Production, 16(6),
807
722726.
808
Lee, A. H. I., Kang, Y., Hsu, H. C-F., & Hung, H.-C. (2009). A green supplier selection
809 model for high-tech industry. Expert Systems with Applications, 36, 79177927.
810
Lin, C.-L., Hsieh, M.-S., & Tzeng, G.-H. (2010). Evaluating vehicle telematics system
811 by using a novel MCDM techniques with dependence and feedback. Expert
812
Systems with Applications, 37(10), 67236736.
813
Lin, C.-J., & Wu, W.-W. (2008). A causal analytical method for group decision-
814 making under fuzzy environment. Expert Systems with Applications, 34(1),
815
205213.
816 Lin, R.-H. (2009). An integrated FANPMOLP for supplier evaluation and order
817
allocation. Applied Mathematical Modelling, 33, 27302736.
818 Liou, J. J. H., Yen, L., & Tzeng, G.-H. (2008). Building an effective safety management
819
system for airlines. Journal of Air Transport Management, 14(1), 2026.
820
Liu, K. F. R., & Lai, J.-H. (2009). Decision-support for environmental impact
821 assessment: A hybrid approach using fuzzy logic and fuzzy analytic network
822
process. Expert Systems with Applications, 36, 51195136.
823 Liu, H.-T., & Wang, C.-H. (2010). An advanced quality function deployment model
824
using fuzzy analytic network process. Applied Mathematical Modelling, 34(11),
825
33333351.
826 Lu, L. Y. Y., Wu, C. H., & Kuo, T.-C. (2007). Environmental principles applicable to
827
green supplier evaluation by using multiobjective decision analysis.
828 International Journal of Production Research, 45, 43174331.
829
Luo, Z.-M., Zhou, J.-Z., Zheng, L.-P., Mo, L., & He, Y.-Y. (2010). A TFNANP based
830
approach to evaluate Virtual Research Center comprehensive performance.
831 Expert Systems with Applications, 37(12), 83798386.
832
Meade, L., & Sarkis, J. (2002). A conceptual model for selecting and evaluating third-
833 party reverse logistics providers. Supply Chain Management: An International
834
Journal, 7(5), 283295.
835 Mohanty, R. P., Agarwal, R., Choudhury, A. K., & Tiwari, M. K. (2005). A fuzzy ANP-
836
based approach to R&D project selection: A case study. International Journal of
837
Production Research, 43, 51995216.
838 Ozgen, D., nt, S., Glsn, B., Tuzkaya, U. R., & Tuzkaya, G. (2008). A two-
839
phase possibilistic linear programming methodology for multi-objective
840 supplier evaluation and order allocation problems. Information Sciences,
841
178, 485500.
842
Ont, S., Kara, S. S., & Is ik, E. (2009). Long term supplier selection using a combined
843 fuzzy MCDM approach: A case study for a telecommunication company. Expert
844
Systems with Applications, 36, 38873895.
845 Punniyamoorthy, M., Mathiyalagan, P., & Parthiban, P. (in press). A strategic model
846
using structural equation modeling and fuzzy logic in supplier selection. Expert
847 Systems with Applications.
848
Rao, P., & Holt, D. (2005). Do green supply chains lead to competitiveness and
849
economic performance? International Journal of Operations & Production
850 Management, 25(9), 898916.
851
Ramik, J. (2007). A decision system using ANP and fuzzy inputs. International Journal
852 of Innovative Computing, Information and Control, 3(4), 825837.
853
Roghanian, E., Rahimi, J., & Ansari, A. (2010). Comparison of rst aggregation and
854
last aggregation in fuzzy group TOPSIS. Applied Mathematical Modelling, 34(12),
855 37543766.
856
Saaty, T. L. (1996). The analytic network process. Pittsburgh: RWS Publications.
857 Sadi-Nezhad, S., & Damghani, K. K. (2010). Application of a fuzzy TOPSIS method
858
base on modied preference ratio and fuzzy distance measurement in
859 assessment of trafc police centers performance. Applied Soft Computing,
860
10(4), 10281039.
861
Salehi, M., & Tavakkoli-Moghaddam, R. (2008). Project selection by using a fuzzy
862 TOPSIS technique. World Academy of Science, Engineering and Technology, 40,
863
8590.
864 Sarkis, J. (2003). A strategic decision framework for green supply chain
865
management. Journal of Cleaner Production, 11(4), 397409.
866
Sarkis, J., Meade, L. M., & Talluri, S. (2004). E-logistics and the natural environment.
867 Supply Chain Management: An International Journal, 9(4), 303312.
868
Sen, S., Bas ligil, H., Sen, C. G., & Barali, H. (2007). A framework for dening both
869 qualitative and quantitative supplier selection criteria considering the buyer-
870
supplier integration strategies. International Journal of Production Research,
871 46(7), 18251845.
872
Sevkli, M., Koh, S. C. L., Zaim, S., Demirbag, M., & Tatoglu, E. (2007). An application of
873
data envelopment analytic hierarchy process for supplier selection: A case
874 study of BEKO in Turkey. International Journal of Production Research, 45,
875
19732003.
876 Simpson, D. F., & Power, D. J. (2005). Use the supply relationship to develop lean
877
and green suppliers. Supply Chain Management: An International Journal, 10,
878
6068.
879 Srivastava, S. K. (2007). Green supply chain management: A state-of-the-art
880
literature review. International Journal of Management Reviews, 9(1),
881 5380.
882
Sun, C.-C., & Lin, G. T. R. (2009). Using fuzzy TOPSIS method for evaluating the
883
competitive advantages of shopping websites. Expert Systems with Applications,
884 36(9), 1176411771.
885
Theyel, G. (2001). Customer and supplier relations for environmental performance.
886 Greener Management International, 35, 6169.
G. Bykzkan, G. ifi / Expert Systems with Applications xxx (2011) xxxxxx 11
ESWA 6898 No. of Pages 12, Model 5G
13 September 2011
Please cite this article in press as: Bykzkan, G., & ifi, G. A novel hybrid MCDM approach based on fuzzy DEMATEL, fuzzy ANP and fuzzy TOPSIS to
evaluate green suppliers. Expert Systems with Applications (2011), doi:10.1016/j.eswa.2011.08.162
887 Tsai, W.-H., & Hung, S.-J. (2009). A fuzzy goal programming approach for green
888
supply chain optimization under activity-based costing and performance
889 evaluation with a value-chain structure. International Journal of Production
890
Research, 47(18), 49915017.
891 Tseng, M.-L. (2009a). A causal and effect decision making model of service quality
892
expectation using grey-fuzzy DEMATEL approach. Expert Systems with
893
Applications, 36, 77387748.
894 Tseng, M.-L. (2009b). Using the extension of DEMATEL to integrate hotel service
895
quality perceptions into a cause-effect model in uncertainty. Expert Systems
896 with Applications, 36, 90159023.
897
Tseng, M.-L. (2010). An assessment of cause and effect decision-making model for
898
rm environmental knowledge management capacities in uncertainty.
899 Environmental Monitoring and Assessment, 161, 549564.
900
Tseng, M.-L., & Lin, Y. H. (2009). Application of fuzzy DEMATEL to develop a cause
901 and effect model of municipal solid waste management in Metro Manila.
902
Environmental Monitoring and Assessment, 158, 519533.
903 Tuzkaya, U. R., & nt, S. (2008). A fuzzy analytic network process based approach
904
to transportation-mode selection between Turkey and Germany: A case study.
905
Information Sciences, 178, 31333146.
906 Tuzkaya, G., Ozgen, A., Ozgen, D., & Tuzkaya, U. R. (2009). Environmental
907
performance evaluation of suppliers: A hybrid fuzzy multi-criteria decision
908 approach. International Journal of Environmental Science and Technology, 6,
909
477490.
910
Vinodh, S., Ramiya, R. A., & Gautham, S. G. (in press). Application of fuzzy analytic
911 network process for supplier selection in a manufacturing organisation. Expert
912
Systems with Applications.
913 Vachon, S., &Klassen, R. D. (2006). Green project partnership in the supply chain: The
914
case of the package printing industry. Journal of Cleaner Production, 14, 661671.
915 van Hock, R. I., & Erasmus (2000). From reversed logistics to green supply chains.
916
Logistics Solutions, 2, 2833.
917 Walker, H., Sisto, L. D., & McBain, D. (2008). Drivers and barriers to environmental
918
supply chain management practices: Lessons from the public and private
919
sectors. Journal of Purchasing & Supply Management, 14, 6985.
920 Wu, D. (2009). Supplier selection: A hybrid model using DEA, decision tree and
921
neural network. Expert Systems with Applications, 36, 91059112.
922 Wu, W.-W., & Lee, Y.-T. (2007). Developing global managers competencies using
923
the fuzzy DEMATEL method. Expert Systems with Applications, 32, 499507.
924
Ye, F. (2010). An extended TOPSIS method with interval-valued intuitionistic fuzzy
925 numbers for virtual enterprise partner selection. Expert Systems with
926
Applications, 37(10), 70507055.
927 Yong, D. (2006). Plant location selection based on fuzzy TOPSIS. International Journal
928
of Advanced Manufacturing Technology, 28, 839844.
929 Yksel,
_
I., & Dagdeviren, M. (2010). Using the fuzzy analytic network process (ANP)
930
for balanced scorecard (BSC): A case study for a manufacturing rm. Expert
931
Systems with Applications, 37(2), 12701278.
932 Zadeh, L. A. (1965). Fuzzy sets. Information & Control, 8, 338353.
933
Zhu, Q., Sarkis, J., & Lai, K.-H. (2007). Green supply chain management: pressures,
934 practices and performance within the Chinese automobile industry. Journal of
935
Cleaner Production, 15(1112), 10411052.
936
Zhu, Q., Sarkis, J., & Lai, K.-H. (2008). Green supply chain management implications
937 for closing the loop. Transportation Research Part E: Logistics and Transportation
938
Review, 44(1), 118.
939
12 G. Bykzkan, G. ifi / Expert Systems with Applications xxx (2011) xxxxxx
ESWA 6898 No. of Pages 12, Model 5G
13 September 2011
Please cite this article in press as: Bykzkan, G., & ifi, G. A novel hybrid MCDM approach based on fuzzy DEMATEL, fuzzy ANP and fuzzy TOPSIS to
evaluate green suppliers. Expert Systems with Applications (2011), doi:10.1016/j.eswa.2011.08.162