Mathematical and Computer Modelling 46 (2007) 573–581

www.elsevier.com/locate/mcm

A grey-based decision-making approach to the supplier selection

problem
Guo-Dong Li a,∗ , Daisuke Yamaguchi a , Masatake Nagai b
a Graduate School of Engineering, Kanagawa University, 221-8686, Yokohama City, Japan
b Department of Engineering, Kanagawa University, 221-8686, Yokohama City, Japan

Received 17 July 2006; received in revised form 19 November 2006; accepted 21 November 2006

Abstract

Supplier selection is a multiple-attribute decision-making (MADM) problem. Since the decision makers (DMs) such as
preferences on alternatives or on the attributes of suppliers are often uncertain, supplier selection becomes more difficult. Grey
theory is one of the methods used to study uncertainty, being superior in the mathematical analysis of systems with uncertain
information. In this paper, we propose a new grey-based approach to deal with the supplier selection problem. The work procedure
is as follows: firstly, the weights and ratings of attributes for all alternatives are described by linguistic variables that can be
expressed in grey numbers. Secondly, using a grey possibility degree, the ranking order of all alternatives is determined. Finally,
an example of a selection problem of supplier was used to illustrate the proposed approach.
c 2006 Elsevier Ltd. All rights reserved.

Keywords: Supplier selection; Multiple attribute decision making (MADM); Grey number; Grey possibility degree

1. Introduction

With the globalization of the economic market and the development of information technology, many companies
consider that a well-designed and implemented supply chain management (SCM) system is an important tool for
increasing competitive advantage [1]. The supplier selection problem become one of the most important components
in SCM [2–4]. In the past, several methods have been proposed to solve the supplier selection problem, the main
ones being the linear weighting methods (LW) [5,6], the analytic hierarchy process (AHP) [7,8], the analytic network
process [9], total cost approaches [10,11] and mathematical programming (MP) techniques [12,13]. Although linear
weighting is a very simple method, it depends heavily on human judgement and also weights the attributes equally,
which rarely happens in practice. On the other hand, MP techniques cause a significant problem in considering
qualitative factors. However, AHP cannot effectively take into account risk and uncertainty in estimating the supplier’s
performance because it presumes that the relative importance of attributes affecting supplier performance is known
with certainty [14]. The drawback of MP is that it requires arbitrary aspiration levels and cannot accommodate
subjective attributes [15]. Supplier selection is a multiple- attribute decision-making (MADM) problem. The decision

∗ Corresponding author. Tel.: +81 467 46 4263; fax: +81 45 413 6565.
E-mail address: guodong li2006@yahoo.co.jp (G.-D. Li).

c 2006 Elsevier Ltd. All rights reserved.

0895-7177/$ - see front matter
doi:10.1016/j.mcm.2006.11.021
574 G.-D. Li et al. / Mathematical and Computer Modelling 46 (2007) 573–581

Fig. 1. The concept of a grey system.

makers (DMs) always express their preferences on alternatives or on the attributes of suppliers, which can be used to
help rank the suppliers or select the most desirable one. The preference information on alternatives of supplier and on
attributes belongs to the DMs’ subjective judgements. In conventional MADM methods, the ratings and weights of the
attributes are known precisely [16–18]. Generally, DMs’ judgements are often uncertain and cannot be estimated by
an exact numerical value. Thus, the problem of selecting suppliers has many uncertainties and becomes more difficult.
Grey theory [19] is one of the methods used to study uncertainty, being superior in the mathematical analysis of
systems with uncertain information. In grey theory, according to the degree of information, if the system information
is fully known, the system is called a white system; if the information is unknown, it is called a black system. A system
with information known partialy is called a grey system. In recent years, a fuzzy-based approach has been proposed to
deal with the supplier selection problem under uncertainty [20]. The advantage of grey theory over fuzzy theory [21,
22] is that grey theory considers the condition of the fuzziness; that is, grey theory can deal flexibly with the fuzziness
situation [19]. In this paper, we propose a new grey-based approach to deal with the problem of selecting suppliers
under an uncertain environment. The work procedure is briefly as follows: firstly, the weight and rating of attributes for
all supplier alternatives are described by linguistic variables that can be expressed in grey number. Secondly, a degree
of grey possibility is proposed to determine the ranking order of all alternatives of supplier. In the supplier selection
process, the degree of uncertainty of the attributes has to be taken into account [23]. In many situations, the preference
information on attributes is uncertain and inconsistent. In order to atone for the insufficiency of decision making, we
present a grey possibility degree to select the ideal supplier based on grey numbers. It will be more suitable for the
MADM system’s more uncertain environment than other approaches. Finally, an example of supplier selection is used
to illustrate the proposed approach.
This paper is organized as follows: Section 2 describes preliminaries which include grey theory and grey number
comparison. Section 3 introduces the proposed grey-based approach. In Section 4 the proposed approach is applied to
the supplier selection problem. Finally, conclusions are drawn in Section 5.

2. Preliminaries

2.1. Grey theory

Grey theory [19], which was proposed by Deng in 1982, is one of the new mathematical theories born out of
the concept of the grey set. It is an effective method used to solve uncertainty problems with discrete data and
incomplete information. The theory includes five major parts: grey prediction, grey relational analysis (GRA) [24,
25], grey decision, grey programming and grey control. Here, we give some basic definitions of the grey system, grey
set and grey number in grey theory.

Definition 1. A grey system is defined as a system containing uncertain information presented by a grey number and
grey variables. The concept of a grey system is shown in Fig. 1.

Definition 2. Let X be the universal set. Then a grey set G of X is defined by its two mappings µG (x) and µG (x).

µG (x) : x −→ [0, 1]

(1)
µG (x) : x −→ [0, 1]
µG (x) ≥ µG (x), x ∈ X, X = R, µG (x) and µG (x) are the upper and lower membership functions in G respectively.
When µG (x) = µG (x), the grey set G becomes a fuzzy set. It shows that grey theory considers the condition of the
fuzziness and can deal flexibly with the fuzziness situation.
 G.-D. Li et al. / Mathematical and Computer Modelling 46 (2007) 573–581 575

Definition 3. The grey number can be defined as a number with uncertain information. For example, the ratings of
attributes are described by the linguistic variables; there will be a numerical interval expressing it. This numerical
µ
interval will contain uncertain information. Generally, grey number is written as ⊗G, (⊗G = G |µ ).

Definition 4. Only the lower limit of G can be possibly estimated and G is defined as a lower-limit grey number.
⊗G = [G, ∞). (2)

Definition 5. Only the upper limit of G can be possibly estimated and G is defined as a lower-limit grey number.

⊗G = (−∞, G]. (3)

Definition 6. The lower and upper limits of G can be estimated and G is defined as an interval grey number.

⊗G = [G, G]. (4)

Definition 7. Grey number operation is an operation defined on sets of intervals, rather than real numbers. The modern
development of interval operation began with R.E. Moore’s dissertation [26]. We cite literatures [27,28] to define the
basic operation laws of grey numbers ⊗G 1 = [G 1 , G 1 ] and ⊗G 2 = [G 2 , G 2 ], on intervals where the four basic grey
number operations on the interval are the exact range of the corresponding real operation.

⊗G 1 + ⊗G 2 = [G 1 + G 2 , G 1 + G 2 ] (5)
⊗G 1 − ⊗G 2 = [G 1 − G 2 , G 1 − G 2 ] (6)
⊗G 1 × ⊗G 2 = [min(G 1 G 2 , G 1 G 2 , G 1 G 2 , G 1 G 2 ), max(G 1 G 2 , G 1 G 2 , G 1 G 2 , G 1 G 2 )] (7)
 
1 1
⊗G 1 ÷ ⊗G 2 = [G 1 , G 1 ] × , . (8)
G2 G2
Here we only give the proofs of addition and subtraction operations. Similarly, we can obtain the proofs of
multiplication and division operations.

Proof. Grey number ⊗G = [G, G] in real lineal set R can be expressed as

f (t) = G + (G − G)t, 0≤t ≤1 (9)

t = 0, f (t) is the lower limit of ⊗G; t = 1, f (t) is the upper limit of ⊗G. Eq. (9) may be also expressed as
f (t) = p + r t, 0 ≤ t ≤ 1. (10)
Setting a function set m 0 , and for ∀ f, g ∈ m 0 , assume f (t) = p1 + r1 t, g(t) = p2 + r2 t, 0 ≤ t ≤ 1, the addition
operation of f and g are expressed as
f + g = ( p1 + p2 ) + (r1 + r2 )t, 0 ≤ t ≤ 1. (11)
According to the above equation, we can obtain Eq. (5).
For a subtraction operation, −g may be expressed as −g = −( p2 + r2 ) + r2 t, 0 ≤ t ≤ 1. Then we can obtain the
subtraction operation as
f − g = p1 − ( p2 + r2 ) + (r1 + r2 )t, 0 ≤ t ≤ 1. (12)
According to the above equation, we can obtain Eq. (6). 

Definition 8. The lenth of grey number ⊗G is defined as

L(⊗G) = [G − G]. (13)

576 G.-D. Li et al. / Mathematical and Computer Modelling 46 (2007) 573–581

Table 1
The scale of attribute weights ⊗w

Scale ⊗w
Very low (VL) [0.0, 0.1]
Low (VL) [0.1, 0.3]
Medium low (ML) [0.3, 0.4]
Medium (M) [0.4, 0.5]
Medium high (MH) [0.5, 0.6]
High (H) [0.6, 0.9]
Very high (VH) [0.9, 1.0]

Table 2
The scale of attribute ratings ⊗G

Scale ⊗G
Very poor (VP) [0, 1]
Poor (P) [1, 3]
Medium poor (MP) [3, 4]
Fair (F) [4, 5]
Medium good (MG) [5, 6]
Good (G) [6, 9]
Very good (VG) [9, 10]

2.2. Comparison of grey numbers

We proposed a degree of grey possibility to compare the ranking of grey numbers.

Definition 9. For two grey numbers ⊗G 1 [G 1 , G 1 ] and ⊗G 2 = [G 2 , G 2 ], the possibility degree of ⊗G 1 ≤ ⊗G 2 can
be expressed as follows [29]:
max(0, L ∗ − max(0, G 1 − G 2 ))
P{⊗G 1 ≤ ⊗G 2 } = (14)
L∗
where L = L(⊗G 1 ) + L(⊗G 2 ).
∗

For the position relationship between ⊗G 1 and ⊗G 2 , there exist four possible cases on the real number axis. The
relationship between ⊗G 1 and ⊗G 2 is determined as follows:
(1) If G 1 = G 2 and G 1 = G 2 , we say that ⊗G 1 is equal to ⊗G 2 , denoted as ⊗G 1 = ⊗G 2 . Then P{⊗G 1 ≤ ⊗G 2 } =
0.5.
(2) If G 2 > G 1 , we say that ⊗G 2 is larger than ⊗G 1 , denoted as ⊗G 2 > ⊗G 1 . Then P{⊗G 1 ≤ ⊗G 2 } = 1.
(3) If G 2 < G 1 , we say that ⊗G 2 is smaller than ⊗G 1 , denoted as ⊗G 2 < ⊗G 1 . Then P{⊗G 1 ≤ ⊗G 2 } = 0.
(4) If there is an intercrossing part in them, when P{⊗G 1 ≤ ⊗G 2 } > 0.5, we say that ⊗G 2 is larger than ⊗G 1 ,
denoted as ⊗G 2 > ⊗G 1 . When P{⊗G 1 ≤ ⊗G 2 } < 0.5, we say that ⊗G 2 is samller than ⊗G 1 , denoted as
⊗G 2 < ⊗G 1 .

3. Proposed approach

A new approach based on a grey possibility degree is proposed for ordering the preference of supplier. This
method is very suitable for solving the group decision-making problem in an uncertain environment. Assume that
S = {S1 , S2 , . . . , Sm } is a discrete set of m possible supplier alternatives. Q = {Q 1 , Q 2 , . . . , Q n } is a set of n
attributes of suppliers. The attributes are additively independent. ⊗w = {⊗w1 , ⊗w2 , . . . , ⊗wn } is the vector of
attribute weights. In this paper, the attribute weights and ratings of suppliers are considered as linguistic variables [30].
Here, these linguistic variables can be expressed in grey numbers by the 1–7 scale shown in Table 1. The attribute
ratings ⊗G can also be expressed in grey numbers by the 1–7 scale shown in Table 2. The procedures are summarized
as follows:
 G.-D. Li et al. / Mathematical and Computer Modelling 46 (2007) 573–581 577

Step 1
Form a committee of decision makers and identify the attribute weights of suppliers. Assume that a decision group
has K persons, then the attribute weight of attribute Q j can be calculated as

1
⊗w j = [⊗w 1j + ⊗w 2j + · · · + ⊗w Kj ] (15)
K
where ⊗w Kj ( j = 1, 2, . . . , n) is the attribute weight of K th DMs and can be described by grey number ⊗w Kj =
[w Kj , w Kj ].
Step 2
Use linguistic variables for the ratings to make an attribute rating value. Then, the rating value can be calculated as

1
⊗G i j = [⊗G i1j + ⊗G i2j + · · · + ⊗G iKj ] (16)
K
where ⊗G iKj (i = 1, 2, . . . , m; j = 1, 2, . . . , n) is the attribute rating value of K th DMs and can be described by the
K
grey number ⊗G iKj = [G iKj , G i j ].
Step 3
Establish the grey decision matrix
 
⊗G 11 ⊗G 12 · · · ⊗G 1n
 ⊗G 21 ⊗G 22 · · · ⊗G 2n 
D= . .. .. ..  (17)
 
 .. . . . 
⊗G m1 ⊗G m2 ··· ⊗G mn

where ⊗G i j are linguistic variables based on the grey number.

Step 4
Normalize the grey decision matrix

⊗G ∗11 ⊗G ∗12 · · · ⊗G ∗1n

 
 ⊗G ∗21 ⊗G ∗22 · · · ⊗G ∗2n 
∗
D = . .. .. ..  (18)
 
 .. . . . 
⊗G ∗m1 ⊗G ∗m2 ··· ⊗G ∗mn

where for a benefit attribute, ⊗G i∗j is expressed as

" #
Gi j Gi j
⊗G i∗j = , max . (19)
G max
j Gj

G max
j = max1≤i≤m {G i j };
for a cost attribute, ⊗G i∗j is expressed as

G min G min
" #
j j
⊗G i∗j = , . (20)
Gi j Gi j

G min
j = min1≤i≤m {G i j }.
The normalization method mentioned above is to preserve the property that the ranges of the normalized grey
number belong to [0, 1].
578 G.-D. Li et al. / Mathematical and Computer Modelling 46 (2007) 573–581

Fig. 2. The selection structure of suppliers.

Step 5
Establish the weighted normalized grey decision matrix. Considering the different importance of each attribute, the
weighted normalized grey decision matrix can be established as
 
⊗V11 ⊗V12 · · · ⊗V1n
 ⊗V21 ⊗V22 · · · ⊗V2n 
D? =  . .. .. ..  (21)
 
 .. . . . 
⊗Vm1 ⊗Vm2 ··· ⊗Vmn
where ⊗Vi j = ⊗G i∗j × ⊗w j .
Step 6
Make the ideal alternative as a referential alternative. For m possible supplier alternatives set S = {S1 , S2 , . . . , Sm },
1 , ⊗G 2 , . . . , ⊗G n } can be obtained by
the ideal referential supplier alternative S max = {⊗G max max max

    
S max
= [ max V i1 , max V i1 ], max V i2 , max V i2 , . . . , max V in , max V in . (22)
1≤i≤m 1≤i≤m 1≤i≤m 1≤i≤m 1≤i≤m 1≤i≤m

Step 7
Calculate the grey possibility degree between compared suppliers alternatives set S = {S1 , S2 , . . . , Sm } and ideal
referential supplier alternative S max .
n
1X
P{Si ≤ S max } = P{⊗Vi j ≤ ⊗G max
j }. (23)
n j=1

Step 8
Rank the order of supplier alternatives. When P{Si ≤ S max } is smaller, the ranking order of Si is better. Otherwise,
the ranking order is worse.
According to the above procedures, we can determine the ranking order of all suppliers alternatives and select the
best from among a set of feasible suppliers.

4. Application and analysis

There are six suppliers Si (i = 1, 2, . . . , 6) selected as alternatives against four attributes Q j ( j = 1, 2, . . . , 4).
The four attributes are product quality, service quality, delivery time and price respectively [31]. Q 1 , Q 2 and Q 3 are
benefit attributes, the greater values being better. Q 4 is cost attributes, the smaller values are better. The selection
structure is shown in Fig. 2. The calculation procedures are as follows:
Step 1
Make the weights of attributes Q 1 , Q 2 , Q 3 and Q 4 . A committee of four DMs, D1 , D2 , D3 and D4 has been
formed to express their preferences and to select the best suppliers. According to Eq. (15), the evaluation values of
attribute weights from four MDs can be obtained and the results are shown in Table 3.
 G.-D. Li et al. / Mathematical and Computer Modelling 46 (2007) 573–581 579

Table 3
Attribute weights for six suppliers

Qj D1 D2 D3 D4 ⊗w j
Q1 VH H H H [0.675, 0.925]
Q2 H VH VH H [0.750, 0.950]
Q3 MH H H MH [0.550, 0.750]
Q4 M M MH MH [0.450, 0.550]

Table 4
Attribute rating values for suppliers

Qj Si D1 D2 D3 D4 ⊗G i j
Q1
S1 G MG G G [5.75, 8.25]
S2 MG G F MG [5.00, 6.50]
S3 F F MG G [4.75, 6.25]
S4 F MG MG F [4.50, 5.50]
S5 MG F F MG [4.50, 5.50]
S6 G MG MG MG [5.25, 6.75]
Q2
S1 G G MG MG [5.50, 7.50]
S2 G MG MG G [5.50, 7.50]
S3 F F P F [3.25, 4.50]
S4 P MP MP P [2.00, 3.50]
S5 MP MP M MP [2.50, 3.75]
S6 MP P P MP [2.00, 3.50]
Q3
S1 G MG MG G [5.50, 7.50]
S2 MG G G G [5.75, 8.25]
S3 G G F MG [5.25, 7.25]
S4 G MG MG G [5.50, 7.50]
S5 MG F F MG [4.50, 5.50]
S6 F F MG F [4.25, 5.25]
Q4
S1 F G G G [5.50, 8.00]
S2 G G F MG [5.75, 8.25]
S3 VG VG G G [7.50, 9.50]
S4 G MG G G [5.75, 8.25]
S5 MG MG G MG [5.25, 6.75]
S6 G VG VG G [7.50, 9.50]

Step 2
Make attribute rating values for six supplier alternatives. According to Eq. (16), the results of attribute rating values
are shown in Table 4.
Step 3
Establish the grey decision matrix. According to Eq. (17), we can obtain the grey decision matrix of suppliers.
Step 4
Establish the grey normalized decision table. According to grey normalized decision matrix shown in Eq. (18), the
grey normalized decision table is shown in Table 5.
Step 5
Establish the grey weighted normalized decision table. According to the grey weighted normalized decision matrix
shown in Eq. (21), the grey weighted normalized decision table is shown in Table 6.
580 G.-D. Li et al. / Mathematical and Computer Modelling 46 (2007) 573–581

Table 5
Grey normalized decision table

Si Q1 Q2 Q3 Q4
S1 [0.697, 1.000] [0.733, 1.000] [0.667, 0.909] [0.656, 0.955]
S2 [0.606, 0.788] [0.733, 1.000] [0.697, 1.000] [0.724, 1.000]
S3 [0.576, 0.758] [0.433, 0.600] [0.636, 0.879] [0.553, 0.700]
S4 [0.545, 0.667] [0.267, 0.467] [0.667, 0.909] [0.636, 0.913]
S5 [0.545, 0.667] [0.333, 0.500] [0.545, 0.545] [0.778, 1.000]
S6 [0.636, 0.818] [0.267, 0.467] [0.515, 0.636] [0.553, 0.700]

Table 6
Grey weighted normalized decision table

Si Q1 Q2 Q3 Q4
S1 [0.470, 0.925] [0.550, 0.950] [0.367, 0.682] [0.295, 0.525]
S2 [0.409, 0.729] [0.550, 0.950] [0.383, 0.750] [0.326, 0.550]
S3 [0.389, 0.701] [0.325, 0.570] [0.350, 0.659] [0.249, 0.385]
S4 [0.368, 0.617] [0.200, 0.443] [0.367, 0.682] [0.286, 0.502]
S5 [0.368, 0.617] [0.250, 0.475] [0.300, 0.500] [0.350, 0.550]
S6 [0.430, 0.757] [0.200, 0.443] [0.283, 0.477] [0.249, 0.385]

Step 6
Make the ideal supplier S max a referential alternative. According to Eq. (22), the the ideal supplier S max is shown as
follows:
S max = {[0.470, 0.925], [0.550, 0.950], [0.383, 0.750], [0.350, 0.550].}.

Step 7
Calculate the grey possibility degree between the compared alternatives of six suppliers Si (i = 1, 2, . . . , 6) and
the ideal referential supplier alternative S max . According to Eq. (23), the results of the grey possibility degree are
shown as follows:
P(S1 ≤ S max ) = 0.539, P(S2 ≤ S max ) = 0.549,
P(S3 ≤ S max
) = 0.789, P(S4 ≤ S max ) = 0.747,
P(S5 ≤ S max ) = 0.772, P(S6 ≤ S max ) = 0.841.

Step 8
Rank the order of six suppliers Si (i = 1, 2, . . . , 6). According to Step 7, the result of ranking order is shown as
follows:
S1 > S2 > S4 > S5 > S3 > S6 .
We can say that the supplier S1 is the best supplier out of the six. S1 should be an important alternative for the
company. The next important alternative is S2 . Because of the grey possibility, degrees of S1 and S2 against the ideal
S max are almost equal. S4 , S5 and S3 are good suppliers and S6 is the worst supplier.

5. Conclusions

In this paper, we proposed a new grey-based approach to deal with the supplier selection problem in an uncertain
environment. Supplier selection is a MADM problem. In conventional MADM methods, the ratings and the weights
of attributes must be known precisely [16–18]. However, in many situations DMs’ judgements are often uncertain and
cannot be estimated by an exact numerical value. Thus, the problem of selecting suppliers has many uncertainties
and becomes more difficult. We can change our perspective and look at the real world from a different angle.
System analysis can be treated from the point of view of the degree of information availabile. Grey theory is a new
 G.-D. Li et al. / Mathematical and Computer Modelling 46 (2007) 573–581 581

mathematical field born out of the concept of the grey set. It is one of the methods used to study the uncertainty of a
system.
The advantage of grey theory over fuzzy sets theory is that grey theory can deal flexibly with the fuzziness
situation. Supplier selection can be viewed as a grey system process. We may use grey theory to resolve it. The
ratings of attributes are described by linguistic variables that can be expressed in grey numbers. We also presented
a grey possibility degree to compare the ranking of grey numbers and select the most ideal supplier. An example of
a supplier selection problem was used to illustrate the proposed approach. The experimental result shows that the
proposed approach is reliable and reasonable.

References

[1] J. Choi, S.X. Bai, J. Geunes, H.E. Romeijn, Manufacturing delivery performance for supply chain management, Mathematical and Computer
Modelling 45 (2007) 11–20.
[2] W.D. Hong, B. Lyes, X.L. Xie, A simulation optimization methodology for supplier selection problem, International Journal of Computer
Integrated Manufacturing 18 (2–3) (2005) 210–224.
[3] N.O. Ndubisi, M. Jantan, L.C. King, M.S. Ayub, Supplier selection and management strategies and manufacturing flexibility, Journal of
Enterprise Information Management 18 (3) (2005) 330–349.
[4] R. Lasch, C.G. Janker, Supplier selection and controlling using multivariate analysis, International Journal of Physical Distribution and
Logistics Management 35 (6) (2005) 409–425.
[5] K.N. Thompson, Vendor profile analysis, Journal of Purchasing and Materials Management 26 (1) (1990) 11–18.
[6] E. Timmerman, An approach to vendor performance evaluation, Journal of Purchasing and Supply Management 1 (1986) 27–32.
[7] G. Barbarosoglu, T. Yazgac, An application of the analytic hierarchy process to the supplier selection problem, Production and Inventory
Management Journal 1st quarter (1997) 14–21.
[8] R. Narasimhan, An analytic approach to supplier selection, Journal of Purchasing and Supply Management 1 (1983) 27–32.
[9] J. Sarkis, S. Talluri, A model for strategic supplier selection, in: Ninth Int Conf on IPSERA, 2000, pp. 652–661.
[10] R.M. Monezka, S.J. Trecha, Cost-based supplier performance evaluation, Journal of Purchasing and Materials Management 24 (2) (1998)
2–7.
[11] D.L. Smytka, M.W. Clemens, Total cost supplier selection model: A case study, International Journal of Purchasing and Materials
Management 29 (1) (1993) 42–49.
[12] F.P. Buffa, W.M. Jackson, A goal programming model for purchase planning, Journal of Purchasing and Materials Management 19 (3) (1983)
27–34.
[13] S.S. Chaudhry, F.G. Forst, J.L. Zydiak, Vendor selection with price breaks, European Journal of Operational Research 70 (1993) 52–66.
[14] J.S. Dyer, P.C. Fishburn, R.E. Steuer, J. Wallenius, S. Zionts, Multiple criteria decision making, multi attribute utility theory: The next ten
years, Management Science 38 (5) (1992) 645–654.
[15] R. Khorramshahgol, H. Azani, Y. Gousty, An integrated approach to project evaluation and selection, IEEE Transactions on Engineering
Management 35 (1988) 265–270.
[16] M. Delgado, J.L. Verdegay, M.A. Vila, Linguistic decision-making models, International Journal of Intelligent Systems 7 (1992) 479–492.
[17] C.L. Hwang, K.P. Yoon, Multiple Attributes Decision Making: Methods and Applications, Springer, Berlin Heidelberg, 1981.
[18] A. Kaufmann, M.M. Gupta, Introduction to Fuzzy Arithmetic, Theory and Applications, Van Nostrand Reinhold, New York, 1991.
[19] J.L. Deng, The introduction of grey system, The Journal of Grey System 1 (1) (1989) 1–24.
[20] Y.X. Wang, Application of fuzzy decision optimum model in selecting supplier, The Journal of Science Technology and Engineering 5 (15)
(2005) 1100–1103.
[21] L.A. Zadeh, Fuzzy sets, Information and Control 8 (1965) 338–353.
[22] R.E. Bellman, L.A. Zadeh, Decision-making in a fuzzy environment, Management Science 17 (4) (1970) 141–164.
[23] C.T. Chen, C.T. Lin, S.F. Huang, A fuzzy approach for supplier evaluation and selection in supply chain management, International Journal
of Production Economics 102 (2006) 289–301.
[24] J.J. Zhang, D.S. Wu, D.L. Olson, The method of grey related analysis to multiple attribute decision making problems with interval numbers,
Mathematical and Computer Modelling 42 (9–10) (2005) 991–998.
[25] M.F. Chen, G.H. Tzeng, Combining grey relation and TPOSIS concepts for selecting an expatriate host country, Mathematical and Computer
Modelling 40 (13) (2004) 1473–1490.
[26] R.E. Moore, Interval Analysis, Prentice-Hall, Englewood Cliffs, NJ, 1966.
[27] Q. Wang, H. Wu, The concept of grey number and its property, in: Proc NAFIPS, pp. 45–49.
[28] Q. Wu, W. Zhou, S. Li, X. Wu, Application of grey numerical model to groundwater resource evaluation, Environmental Geology 47 (2005)
991–999.
[29] J.R. Shi, S.Y. Liu, W.T. Xiong, A new solution for interval number linear programming, Journal of Systems Engineering Theory and Practice
2 (2005) 101–106.
[30] G.D. Li, D. Yamaguchi, M. Nagai, A grey-based approach to suppliers selection problem, in: Proc Int Conf on Parallel, Distributed Processing
Techniques and Applications, 2006, pp. 180–186.
[31] S.J. Wang, H.A. Hu, Application of rough set on supplier’s determination, in: The Third Annual Conference on Uncertainty, 2005,
pp. 256–262.