• Tidak ada hasil yang ditemukan

Method to Determine the Criteria Weights

N/A
N/A
Protected

Academic year: 2024

Membagikan "Method to Determine the Criteria Weights"

Copied!
13
0
0

Teks penuh

(1)

Method to Determine the Criteria Weights

Reza Tavakkoli-Moghaddam1(&), Hossein Gitinavard2, Seyed Meysam Mousavi3, and Ali Siadat4

1 School of Industrial Engineering, College of Engineering, University of Tehran, Tehran, Iran

[email protected]

2 School of Industrial Engineering,

Iran University of Science and Technology, Tehran, Iran [email protected]

3 Industrial Engineering Department, Faculty of Engineering, Shahed University, Tehran, Iran

[email protected]

4 LCFC, Arts et Métier Paris Tech, Centre de Metz, Metz, France [email protected]

Abstract. In a multi-criteria group decision analysis, numerous methods have been developed and proposed to determine the weight of each criterion; how- ever, the group decision methods, except AHP, have rarely considered for obtaining the criteria weights. This study presents a new TOPSIS method based on interval-valued hesitant fuzzy information to compute the criteria weights. In this respect, the weight of each expert and the experts’ judgments about the criteria weights are considered in the proposed procedure. In addition, an application example about the location problem is provided to show the capa- bility of the proposed weighting method. Finally, results of the proposed method are compared with some methods from the related literature in the presented illustrative example to show the validation of the proposed interval-valued hesitant fuzzy TOPSIS method.

Keywords: Criteria weights

Group decision making

Interval-valued hesitant fuzzy set

Utility degree

Individual regret

1 Introduction

In modern group decision analysis, multi-criteria group decision making (MCGDM) problems are important part of operations research, which can rank the candidate potential alternative regarding to experts’judgments. One of the main factors that can affect ranking results is the criteria weight. In some decision problems, the authors focused on determining the criteria weights based on subjective, objective and inte- grated methods.

The subjective methods compute the criteria weights based on preferences and judgments of experts, such as ranking ordering method of criteria [1–3], direct rating method [4, 5], Delphi method [6], eigenvector method [7], point allocation method

©Springer International Publishing Switzerland 2015

B. Kamiński et al. (Eds.): GDN 2015, LNBIP 218, pp. 157169, 2015.

DOI: 10.1007/978-3-319-19515-5_13

(2)

[8,9], linear programming of preference comparisons [10], and linear programming techniques for multidimensional analysis of preferences [11]. In addition, the objective methods calculate the criteria weights by utilizing the information of objective decision matrix, such as criteria’ importance through inter-criteria correlation method [12], entropy method [6,12], maximizing deviation method [13,14], and standard deviation method [12,15]. Thus, in integrated methods, the weight of each criterion is deter- mined based on considering the objective decision matrix and experts’ subjective judgments [16,17].

In real-world, the natures of the objects have been uncertain and imprecise, because the preferences and judgments of experts are hesitant or vague. Therefore, the criteria of group decision-making problems in an uncertain condition should be expressed by fuzzy values [18,19], such as interval values [20, 21], linguistic variables [21, 22], intuitionistic fuzzy values [23, 24], hesitant fuzzy sets [25,26], and interval-valued hesitant fuzzy sets [27,28].

Fuzzy sets theory and its extensions have widely utilized in imprecise situations for evaluating the problems in manyfields, such as artificial intelligence [29], management [30], pattern recognition [31] and group decision making [32, 33]. Hence, in fuzzy group decision-making problems the criteria weights is an important issue to provide the best solution. Therefore, some researchers have studied on determining the criteria weights regarding to the uncertain environment. In this respect, Fan et al. [34] proposed an optimization model to determine the criteria weights by according to the experts’ fuzzy judgments and objective fuzzy decision matrices. Wang and Parkan [35] pre- sented a general multi-attribute decision making framework by considering the objective information and subjective preferences to determine the criteria weights under fuzzy environment. Chen and Lee [36] proposed a fuzzy AHP method regarding to triangular fuzzy numbers for determining the criteria weights of professional con- ference organizer.

In some complex situations, the experts have defined their preferences and judg- ments by assigning some interval-values membership degrees for an object under a set to decrease the uncertainty risk and margin of errors. Therefore, the interval-valued hesitant fuzzy set (IVHFS), whichfirst introduced by Chen et al. [27] is a powerful tool to deal with these situations. Thus, each criterion can be defined based on IVHFS and expressed in terms of experts’preferences. In this case, Zhang et al. [37] proposed an objective weighting approach by utilizing the Shannon information entropy under a hesitant fuzzy environment. Xu and Zhang [38] developed an optimization model regarding to the maximizing deviation method to determine the criteria weights under hesitant fuzzy and interval-valued hesitant fuzzy-environments. They proposed a hybridized group decision making method under some steps to specify the criteria weighs which led to be more easy to use versus the optimization model. Beg and Rashid [39] proposed a method to aggregate the preferences expert’s judgments among the different criteria, in which the experts’opinions are expressed based on the hesitant fuzzy linguistic variables sets. Zhang et al. [40] constructed a hesitant fuzzy multiple attribute group decision making approach based on the distance measure to avoid the aggregation complexity of the hesitant fuzzy information. Feng et al. [41] utilized the TOPSIS (technique for order performance by similarity to ideal solution) method to solve the hesitant fuzzy multiple attribute decision making problems, in which the

(3)

weight information are completely known. The literature review shows that deter- mining the criteria weights based on ranking methods and especially under the IVHF- environments is still an open problem. In this paper, a hybridized group decision making approach is proposed based on the TOPSIS method and preferences experts’ judgments about the criteria weights to determine the weight of each criterion under hesitant fuzzy environment. In addition, a group of experts is established to assess the problem based on the linguistic variables that indicate their subjective preferences. In sums, some merits and advantages of this study, which provide the proposed method to be more precise are expressed as follows: (1) Proposing a new TOPSIS method in an interval-valued hesitant fuzzy setting; (2) a group of experts is established to evaluate the problem by assigning their opinions by linguistic terms based on the interval-valued hesitant fuzzy information, which converted to interval-valued hesitant fuzzy elements;

and (3) proposing a new relative closeness index to obtain the criteria weights. In addition, the weight of experts is applied in procedure of the proposed method. The validation of the proposed approach is obtained by comparing with other weighting methods for determining criteria weights.

The rest of this paper organized as follows. In Sect.2, some methods to determine the criteria weights are explained and the interval-valued hesitant fuzzy TOPSIS (IVHF-TOPSIS) method for estimating the criteria weights are elaborated. In Sect.3, an illustrative example in the selection of the best site for building a new factory provided to show the implementation process of the proposed approach. Finally, in Sect.4, the paper is concluded.

2 Proposed Method

In this section, three techniques for computing the weight of criteria under the interval- valued hesitant fuzzy environment are extended to compare the computational results of criteria weights with the proposed approach.

2.1 Methods of Determining the Criteria Weights

In this subsection, these three approaches are considered to compute the criteria weights. Firstly, the criteria weights can be computed based on maximizing a deviation method introduced by Xu and Zhang [38] when the information completely unknown.

In this paper, the method is extended based on the interval-valued hesitant fuzzy Hamming distance measure to determine the optimal weight vector as follows; let h¼hhLij;hUiji

mn is an interval-valued hesitant fuzzy element:

wj¼

P

m i¼1

P

m r¼1

1 2l

P

l k¼1

hr kijð ÞL hr krjð ÞL

þhr kijð ÞU hr krjð ÞU

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Pn

j¼1

Pm

i¼1

Pm

r¼1 1 2l

Pl

k¼1

hr kð Þ

L

ij hr kð Þ

L

rj

þ hr kð Þ

U

ij hr kð Þ

U

rj

2

s ð1Þ

(4)

wj ¼ wj

Pn

j¼1

wj

; j¼1;2;. . .;n ð2Þ

where thehr kð Þ

L

ij ,hr kð Þ

U

ij ,hr kð Þ

L

rj and hr kð Þ

U

rj are the kth largest value inhLij,hUij,hLrj and hUrj, respectively and also thewj is the normalized criteria weight vector.

Secondly, the decision makers (DMs) specify the relative importance of criteria weights by linguistic variables that can be converted to the IVHFS and denoted bytj¼hlLj;lUj i

. In this respect, the final aggregated criteria weights regarding to DMs’judgments are obtained by the HIVFG operator are as follows:

tj¼HIVFG ~h1;~h2;. . .;~hK

¼ K

k¼1 kfk~hk

1K

¼ [~c

12~h1;~c22~h2;...;~ck2~hk

YK

j¼1

kLkcLk

K1

; YK

j¼1

kUkcUk

K1

" #

( )

ð3Þ

where the weight of each DM is represented askfk¼kLk;kUk

and is considered in the computational process of the criteria weights to decrease the errors.

Thirdly, the above-mentioned methods can be hybridized, and thus the following relation for computing the criteria weights can be proposed by:

xj¼tj:wj ð4Þ

tj¼lLj þlUj

2 8j ð5Þ

xj ¼ xj

Pn

j¼1

xj

; j¼1;2; :. . .n ð6Þ

wherexj is the normalized criteria weight vector.

2.2 TOPSIS Method with IVHFS

In this subsection, the proposed novel TOPSIS method is introduced based on the IVHFSs. In this respect, the DMs’ opinions about the relative importance of each criterion are considered in process of the proposed method. Therefore, the procedure of the proposed method is defined based on the following steps:

Step 1. Construct an interval-valued hesitant fuzzy decision matrix (IVHF-decision matrix) for each criterion (Cj; 1,2,…,n) regarding to the possible alternatives (Ai; 1,2,

…,m) and opinions of each DM (k; 1,2,…,K).

(5)

k1 k2 kK

Gj¼ A1

... Am

lL11j;lU11j

h i

lL212;lU212

hlLk1j;lUk1ji

...

... .. . ... lL1mj;lU1mj

h i

lL2mj;lU2mj

h i

hlLkmj;lUkmji 0

B B B B B

@

1 C C C C C A

mk

8j ð7Þ

wherehlLkmj;lUkmji

is the interval-value membership degree for them-th alternative that expressed by thek-th expert to construct thej-th IVHF-decision matrix.

Step 2. Normalize each IVHF-decision matrix GNj ¼hgLkij ;gLkiji

mk

based on the following relations [42]:

gLkij ¼ lLkij

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Pm

i¼1

PK

k¼1

½ðlLkijÞ2þ ðlUkij Þ2Š

s 8i;j;k ð8Þ

gUkij ¼ lUkij

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi P

m i¼1

P

K k¼1

½ðlLkij Þ2þ ðlUkij Þ2Š

s 8i;j;k ð9Þ

Step 3. Determine the interval-valued hesitant fuzzy positive ideal solution matrix (IVHF-PIS) and the interval-valued hesitant fuzzy negative ideal solution matrix (IVHF-NIS) by regarding to normalize IVHF-decision matrix as follows:

k1 k2 kK

A¼lLki ;lUki

mk¼

A1

... Am

lL11 ;lU11

lL21 ;lU21

lLK1 ;lUK1

...

... .. .

... lL1m ;lU1m

lL2m ;lU2m

lLKm ;lUKm 0

B B B

@

1 C C C A

mk

ð10Þ

k1 k2 kK

A ¼liLk;liUk

mk¼ A1

... Am

l1L1;l1U1

l1L2;l1U2

l1LK;l1UK ...

... .. .

... lmL1;lmU1

lmL2;lmU2

lmLK;lmUK 0

B B B

@

1 C C C A

mk

ð11Þ where the average of the group decision matrix is calculated by the following relations:

(6)

lLki ¼1 n

Xn

j¼1

lLkij 8i;k ð12Þ

lUki ¼1 n

Xn

j¼1

lUkij 8i;k ð13Þ

li Lk¼min

j nlLkijo

8i;k ð14Þ

li Uk¼max

j nlUkij o

8i;k ð15Þ

Step 4. Compute the separation measure for each normalized IVHF-decision matrix from the IVHF-PIS matrix and the IVHF-NIS matrix by using the IVHF-Euclidean distance measure, which indicates bySj and Sj , respectively.

Sj ¼

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1

2lxi

X

m i¼1

X

K k¼1

X

lxi

k¼1

lLkr ki ð Þð Þxi ALkr ki ð Þð Þxi

2

þlUkr ki ð Þð Þxi AkUr ki ð Þð Þxi

2

v u u

t 8j

ð16Þ

Sj ¼

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1

2lxi

Xm

i¼1

XK

k¼1

X

lxi

k¼1

lLkr ki ð Þð Þxi AiLkr kð Þð Þxi

2

þlUkr ki ð Þð Þxi AiUkr kð Þð Þxi

2

v u u

t 8j:

ð17Þ Step 5. Specify the relative closenessCj

for determining the most important criterion.

Cj¼

d S j ;S d S j ;S

þd S j;S 8j ð18Þ

S¼1 n

Xn

j¼1

Sj ð19Þ

S ¼1 n

Xn

j¼1

Sj ð20Þ

where S the average of Sjðj¼1;2; ::;nÞ, and also the average ofSj ðj¼1;2; ::;nÞ represented asS . The DM often specifies their opinion by linguistic variables, which applied in our proposed method and established a new hybrid approach. In this respect, the hybrid relative closenessCjh

is defined as follows:

(7)

Cjh¼ -j:d S j ;S d S j ;S

þd S j;S 8j ð21Þ

-j¼HIVFG ~h1;~h2;. . .;~hK

¼ K

k¼1 kfk~hk

K1

¼ [~c

12~h1;~c22~h2;...;~ck2~hk

Y

K j¼1

kLkcLk K1

; Y

K j¼1

kUkcUk

K1

" #

( )

ð22Þ

-j¼lLj þlUj

2 8j ð23Þ

where thefinal weight of each DM is indicated bykk¼kLk;kUk

that is computed by:

kLk¼ Pm

i

Pn

j

lLkij

P

K k

P

m i

P

n j

lLkij

ð24Þ

kUk ¼ Pm

i

Pn

j

lUkij

P

K k

P

m i

P

n j

lUkij

ð25Þ

kfk¼ Pm

i

Pn

j

lUkij þlLkij

P

K k

P

m i

P

n j

lUkij þlLkij

ð26Þ

where, thefinal DMs’weightskfk is P

K k¼1

kfk¼1.

Step 6. Estimate the weight of each criterionwjaccording to the relative closeness.

wj¼ Cj Pn

j¼1

Cj

8j ð27Þ

3 Illustrative Example

In this section, an illustrative example about the location problem from the literature [28]

is considered to show the capability of the proposed method for determining the weight of each criterion. In this respect, for showing the verification of the proposed method,

(8)

the application example is solved under three decision approaches, and then we compare them with our decision method. The application example is about the best site selection for building a new factory that is established by four DMs (k = 1, 2, …, 4), three potential alternative (A1,A2,A3) and the following six criteria:C1: Climate, Condition;

C2: Regional demand; C3: Expansion possibility; C4: Transportation availability;

C5: Labor force; andC6: Investment cost.

As represented in Tables 1 and 2, the relative importance of each hesitant fuzzy linguistic term for rating the importance of each criteria and rating of potential alter- natives are defined, respectively. In addition, the evaluation of alternatives expressed by DMs’opinions with hesitant fuzzy linguistic terms that shown in Table3. Similarly, the DMs’ judgments about relative importance of each criterion are expressed as hesitant fuzzy linguistic terms in Table4. Then, these linguistic variables are converted to interval-valued hesitant fuzzy elements (IVHFEs).

The IVHF-decision matrix is normalized by regarding Eqs. (8) and (9); then, the IVHF-PIS matrix and the IVHF-NIS matrix are constructed by considering the relations Eqs. (10)−(15). In addition, the separation measure for each criterion is calculated by applying Eqs. (16) and (17). Then, the relative closeness is specified by Eqs. (18)−(20).

Also, in hybrid decision approach, the DMs’judgments about the relative importance of each criterion are considered and computed by Eqs. (21)−(26). The Eq. (27) is utilized to compute the weight of each criterion in the proposed approach and in the

Table 1. Hesitant fuzzy variables for rating the importance of criteria and the DMs Hesitant fuzzy linguistic variables IVHFE

Very important (VI) [0.90,0.90]

Important (I) [0.75, 0.80]

Medium (M) [0.50, 0.55]

Unimportant (UI) [0.35, 0.40]

Very unimportant (VUI) [0.10,0.10]

Table 2. Hesitant variables for the rating of possible alternatives Hesitant fuzzy linguistic variables IVHFE

Extremely good (EG) [1.00,1.00]

Very very good (VVG) [0.90,0.90]

Very good (VG) [0.80, 0.90]

Good (G) [0.70, 0.80]

Medium good (MG) [0.60, 0.70]

Fair (F) [0.50, 0.60]

Medium bad (MB) [0.40, 0.50]

Bad (B) [0.25, 0.40]

Very bad (VB) [0.10, 0.25]

Very very bad (VVB) [0.10,0.10]

(9)

proposed hybrid approach. The above-mentioned results have shown in Tables5and6.

Other weighting techniques for determining the relative importance of criteria are extended and illustrated in Subsect. 2.1. In this respect, each extended technique in our application example is applied and for showing the low difference between them and proposed approaches, the mean value and the variance of criteria weights are provided.

Utilizing different techniques commonly leads to different results. It is unsuitable to say which method is powerful and capable because every method has various results underlying assertion or theory. However, the extended TOPSIS for the criteria weights is more capable for compromise of nearby to the ideal and farther from the negative ideal. Since the criteria weights of the proposed approaches are generated from the DMs’ judgments,“biased”or “false”judgments lead to a low weight [43].

Table 3. Linguistic evaluations by the decision makers Main

criteria

Alternatives k1 k2 k3 k4

C1 A1 MG MG G VG

A2 VG G G G

A3 F MG F MG

C2 A1 G VG VG MG

A2 F MG F F

A3 VG VG VG G

C3 A1 F MG MG MG

A2 VG G VG VG

A3 MG MG G MG

C4 A1 VG VG G G

A2 G G G MG

A3 G G VG F

C5 A1 MG G G VG

A2 MG MG VG G

A3 VG G G G

C6 A1 VVG VG VVG EG

A2 F B B MG

A3 VVG VVG VG VVG

Table 4. Linguistic evaluations for weights of criteria assigned by the decision makers k1 k2 k3 k4

C1 I VI M I C2 M I M I C3 I M I VI C4 I UI M M C5 M I I M C6 M M I I

(10)

In this regard, the proposed approach versus the classical/modern methods is more precise, because of two main features as considering the IVHFSs and the DMs’ opinions about the criteria weights. The IVHFSs aid to DMs by assigning some interval-values membership degrees for an element under a set to margin of errors. In addition, the preferences DMs’ judgments about the relative significance of the cri- teria are provided in procedure of the proposed method to decrease the errors.

4 Conclusions and Future Direction

In group decision making problems, determining the relative importance of each cri- terion has been very important issue. In this regard, a novel TOPSIS method has been proposed by utilizing the IVHFS regarding to the experts weights and their opinions about the criteria weights. In the proposed approach, the preferences experts’ judg- ments have been expressed by linguistic variables which transformed to interval-valued hesitant fuzzy element. Hence, an illustrative example about the location problem has been considered to illustrate the steps of the proposed decision method. Finally, the

Table 5. Computational results ofSj*

, SjandCj

Sj*

Sj Cj

S1*

0.111190 S1 0.200000 C1 0.034653 S2*

0.098327 S2 0.191599 C2 0.514718 S3* 0.119171 S3 0.208143 C3 0.38869 S4*

0.073633 S4 0.186870 C4 0.285241 S5* 0.060673 S5 0.182734 C5 0.273389 S6*

0.173451 S6 0.229551 C6 0.306194 S 0.1060746 S 0.199816

Table 6. Final criteria weight by the proposed method and hybridized method and comparative analysis

wj Proposed approach

Proposed hybridized approach

Maximizing deviation method

Linguistic variables

Hybridized

maximizing deviation method and linguistic variables

w1 0.019221 0.021825 0.146788 0.182418 0.163515

w2 0.285496 0.283351 0.183486 0.159441 0.178648

w3 0.215593 0.244808 0.146788 0.182418 0.163515

w4 0.158213 0.130957 0.091743 0.132973 0.074495

w5 0.151639 0.150499 0.091743 0.159441 0.089324

w6 0.169835 0.168558 0.339449 0.159441 0.330500

X 0.166667 0.166666667 0.166666667 0.1626893 0.16666667 r2 0.0077124 0.008448048 0.008430828 0.00033864 0.00829643

(11)

results have been compared with several approaches implemented in a practical example to show the validation of the proposed approach. For future direction, the proposed method can be enhanced by considering the hierarchical structure of the criteria.

Acknowledgments. This work has been supported financially by the Center for International Scientific Studies & Collaboration (CISSC) and the French Embassy in Tehran as well as the Partenariats Hubert Curien (PHC) program in France. Additionally, the authors would like thank anonymous reviewers for their valuable comments.

References

1. Ahn, B.S., Park, K.S.: Comparing methods for multiattribute decision making with ordinal weights. Comput. Oper. Res.35(5), 1660–1670 (2008)

2. Barron, F.H., Barrett, B.E.: Decision quality using ranked attribute weights. Manage. Sci.42 (11), 1515–1523 (1996)

3. Solymosi, T., Dombi, J.: A method for determining the weights of criteria: the centralized weights. Eur. J. Oper. Res.26(1), 35–41 (1986)

4. Bottomley, P.A., Doyle, J.R.: A comparison of three weight elicitation methods: good, better, and best. Omega29(6), 553–560 (2001)

5. Goodwin, P., Wright, G., Phillips, L.D.: Decision Analysis for Management Judgment.

Wiley, London (2004)

6. Tzeng, G.-H., Huang, J.-J.: Multiple Attribute Decision Making: Methods and Applications.

CRC Press, New York (2011)

7. Takeda, E., Cogger, K., Yu, P.: Estimating criterion weights using eigenvectors: a comparative study. Eur. J. Oper. Res.29(3), 360–369 (1987)

8. Roberts, R., Goodwin, P.: Weight approximations in multi-attribute decision models.

J. Multi-Criteria Decis. Anal.11(6), 291–303 (2002)

9. Doyle, J.R., Green, R.H., Bottomley, P.A.: Judging relative importance: direct rating and point allocation are not equivalent. Organ. Behav. Hum. Decis. Process.70(1), 65–72 (1997) 10. Horsky, D., Rao, M.: Estimation of attribute weights from preference comparisons. Manage.

Sci.30(7), 801–822 (1984)

11. Srinivasan, V., Shocker, A.D.: Linear programming techniques for multidimensional analysis of preferences. Psychometrika38(3), 337–369 (1973)

12. Xu, X.: A note on the subjective and objective integrated approach to determine attribute weights. Eur. J. Oper. Res.156(2), 530–532 (2004)

13. Wu, Z., Chen, Y.: The maximizing deviation method for group multiple attribute decision making under linguistic environment. Fuzzy Sets Syst.158(14), 1608–1617 (2007) 14. Wei, G.-W.: Maximizing deviation method for multiple attribute decision making in

intuitionistic fuzzy setting. Knowl.-Based Syst.21(8), 833–836 (2008)

15. Deng, H., Yeh, C.-H., Willis, R.J.: Inter-company comparison using modified TOPSIS with objective weights. Comput. Oper. Res.27(10), 963–973 (2000)

16. Wang, Y.-M., Luo, Y.: Integration of correlations with standard deviations for determining attribute weights in multiple attribute decision making. Math. Comput. Modell.51(1), 1–12 (2010)

17. Ma, J., Fan, Z.-P., Huang, L.-H.: A subjective and objective integrated approach to determine attribute weights. Eur. J. Oper. Res.112(2), 397–404 (1999)

(12)

18. Mousavi, S.M., Torabi, S.A., Tavakkoli-Moghaddam, R.: A hierarchical group decision- making approach for new product selection in a fuzzy environment. Arab. J. Sci. Eng.38 (11), 3233–3248 (2013)

19. Mousavi, S.M., Jolai, F., Tavakkoli-Moghaddam, R.: A fuzzy stochastic multi-attribute group decision-making approach for selection problems. Group Decis. Negot.22(2), 207– 233 (2013)

20. Vahdani, B., Tavakkoli-Moghaddam, R., Mousavi, S.M., Ghodratnama, A.: Soft computing based on new interval-valued fuzzy modified multi-criteria decision-making method. Appl.

Soft Comput.13(1), 165–172 (2013)

21. Vahdani, B., Zandieh, M.: Selecting suppliers using a new fuzzy multiple criteria decision model: the fuzzy balancing and ranking method. Int. J. Prod. Res.48(18), 5307–5326 (2010) 22. Parreiras, R., et al.: Aflexible consensus scheme for multicriteria group decision making

under linguistic assessments. Inf. Sci.180(7), 1075–1089 (2010)

23. Xu, Z.: Intuitionistic preference relations and their application in group decision making. Inf.

Sci.177(11), 2363–2379 (2007)

24. Hashemi, H., Bazargan, J., Mousavi, S.M.: A compromise ratio method with an application to water resources management: an intuitionistic fuzzy set. Water Resour. Manage27(7), 2029–2051 (2013)

25. Torra, V., Narukawa, Y.: On hesitant fuzzy sets and decision. In: IEEE International Conference on Fuzzy Systems, 2009, FUZZ-IEEE 2009. IEEE (2009)

26. Torra, V.: Hesitant fuzzy sets. Int. J. Int. Syst.25(6), 529–539 (2010)

27. Chen, N., Xu, Z., Xia, M.: Interval-valued hesitant preference relations and their applications to group decision making. Knowl.-Based Syst.37, 528–540 (2013)

28. Wang, J.-q: Interval-valued hesitant fuzzy linguistic sets and their applications in multi- criteria decision-making problems. Knowl.-Based Syst.288, 55–72 (2014)

29. Greco, S., Matarazzo, B., Giove, S.: The Choquet integral with respect to a level dependent capacity. Fuzzy Sets Syst.175(1), 1–35 (2011)

30. Doria, S.: Characterization of a coherent upper conditional prevision as the Choquet integral with respect to its associated Hausdorff outer measure. Ann. Oper. Res. 195(1), 33–48 (2012)

31. Demirel, T., Demirel, N.Ç., Kahraman, C.: Multi-criteria warehouse location selection using Choquet integral. Expert Syst. Appl.37(5), 3943–3952 (2010)

32. Wang, J.Q.: Multi-criteria outranking approach with hesitant fuzzy sets. OR Spectr.36(4), 1–19 (2013)

33. Qin, J., Liu, X.: Study on interval intuitionistic fuzzy multi-attribute group decision making method based on Choquet integral. Procedia Comput. Sci.17, 465–472 (2013)

34. Fan, Z.-P., Ma, J., Zhang, Q.: An approach to multiple attribute decision making based on fuzzy preference information on alternatives. Fuzzy Sets Syst.131(1), 101–106 (2002) 35. Wang, Y.-M., Parkan, C.: A general multiple attribute decision-making approach for

integrating subjective preferences and objective information. Fuzzy Sets Syst. 157(10), 1333–1345 (2006)

36. Chen, C.-F., Lee, C.-L.: Determining the attribute weights of professional conference organizer selection: an application of the fuzzy AHP approach. Tourism Econ.17(5), 1129– 1139 (2011)

37. Zhang, Y., Wang, Y., Wang, J.: Objective attributes weights determining based on shannon information entropy in hesitant fuzzy multiple attribute decision making. Math. Probl. Eng.

2014, 7 (2014)

38. Xu, Z., Zhang, X.: Hesitant fuzzy multi-attribute decision making based on TOPSIS with incomplete weight information. Knowl.-Based Syst.52, 53–64 (2013)

(13)

39. Beg, I., Rashid, T.: TOPSIS for hesitant fuzzy linguistic term sets. Int. J. Intell. Syst.28(12), 1162–1171 (2013)

40. Zhang, J.L., Qi, X.W., Huang, H.B.: A hesitant fuzzy multiple attribute group decision making approach based on TOPSIS for parts supplier selection. Appl. Mech. Mater.357, 2730–2737 (2013)

41. Feng, X.: TOPSIS method for hesitant fuzzy multiple attribute decision making. J. Intell.

Fuzzy Syst.26(5), 2263–2269 (2014)

42. Jahanshahloo, G.R., Lotfi, F.H., Davoodi, A.: Extension of TOPSIS for decision-making problems with interval data: Interval efficiency. Math. Comput. Modell.49(5), 1137–1142 (2009)

43. Yue, Z.: An extended TOPSIS for determining weights of decision makers with interval numbers. Knowl.-Based Syst.24(1), 146–153 (2011)

Referensi

Dokumen terkait

An integrated Analytic Hierarchy Process (AHP) and Technique of Order Preference by Similarity to Ideal Solution (TOPSIS) are the multiple-criteria decision making problem

• SMART (The Simple Multi Attribute Rating Technique ) • ANP (Analytic network process).. Multiple Criteria

The contents of this course including: Introduction to Decision Making Techniques Decision Making in Conditions of Uncertainty and Risk Decision Tree Theory of Utility Analytic

Past efforts in determining suitable normalization methods for multi-criteria decision- making: A short survey ABSTRACT The use of a multi-criteria decision-making MCDM technique

ORIGINAL ARTICLE Determining the Criteria and Their Weights for Medical Schools' Ranking: A National Consensus Rita Mojtahedzadeh1, Aeen Mohammadi1, Noushin Kohan2, Mitra Gharib1,

Knowing that the both type of weights are essential to determine the importance of sub-criteria, then the decision makers may use the integrated approach defined by the following

In the current study using the AHP method to determine the weight of the criteria and sub-criteria, then these weights will be used in data processing using the rating scale method to

The optimal method that can be suggested is the Multi-criteria Decision Making Analytic Hierarchy Process MCDM-AHP which is used to determine the weighting of a number of criteria used