Professional Documents
Culture Documents
Aset 2022 Fuzzy
Aset 2022 Fuzzy
Abstract—The selection of a facility location among alternative making process. Cagri et al. [11] proposed a fuzzy MCDM
locations is a multi-criteria decision-making problem that includes approach with fuzzy real option value theory for selecting
both subjective and objective criteria. The conventional approaches supermarket location problems using assessment criteria of
to facility location problems tend to be less effective in dealing with the retail location. Ta-chungchu [12] investigated the facility
the imprecise or vagueness nature of the linguistic assessment.
Thus, this situation can be regarded as a fuzzy multiple-criteria
location using a fuzzy TOPSIS model incorporating both
decision making problem considering the fuzziness and uncertainty subjective and objective criteria. Rahman et al. [13] used
of subjective perception. This paper proposes an alternative Analytic Hierarchy Process (AHP) to identify the right
approach, the additive fuzzy integral, to cope with the evaluation of location for the plastic manufacturing industries in
fuzzy MCDM problems. To illustrate the proposed procedure, the Bangladesh, India. Chatterjee and Mukherjee [14] considered
site selection of municipal waste is investigated. In this paper, the three major factors and eleven subfactors for evaluating the
classification of subjective and objective criteria is constructed. location of a hospital using the Analytical Hierarchy Process
Factor analysis is employed to extract four independent common in India and came up with the most significant factors such as
factors, subjective and objective, with seven sub-criteria for land cost, transport facility and population density. Similarly,
evaluating the four location alternatives. In this paper triangular
fuzzy number is employed to represent the decision-makers
Mandal and Mondal [15] investigated five alternative projects
subjective preferences on the considered criteria and alternatives. with ten criteria using the AHP method to identify and
The evaluation frame is constructed by using the Analytic Hierarchy evaluate the best project selection. Aydin and Kahraman [16]
Process which fuzzified to derive the relative weights with respect applied a modified fuzzy AHP method for measuring the
to each type of criteria. The fuzzy synthetic evaluation is aggregated performance excellence of companies applied for the
with fuzzy performance values and the best location is chosen. Malcolm Baldrige National Quality Award. Similarly, chou et
al. [17] and Aras et al. [18] selected the best location for tourist
Keywords—Municipal waste; Facility location; Fuzzy sets; hotel destinations and wind observation stations using
Group decision; Analytic Hierarchy Process. MCDM respectively. AHP is highly recommended over
I. INTRODUCTION TOPSIS for identifying the facility location, even though
TOPSIS is useful for evaluating alternatives’ specialties.
The selection of a facility location from a set of alternatives is Kengpol et al. [19] used fuzzy AHP and TOPSIS for the
a multicriteria decision-making problem that includes both selection of solar power plant locations free from flood.
quantitative and qualitative criteria. Due to the imprecision of Kaboli et al. [20] proposed a holistic approach of MCDM for
assessing the linguistic variables, the traditional methods for finding suitable locations preferable to investors and
solving facility location problems are less effective. In managers. Chan et al. [21] proved that the fuzzy integrated
general, many of the multi criteria decision making hierarchical MCDM is more precise than the analytical
approaches are stochastic and qualitative in nature [1]–[4]. hierarchy process for the identification of facility location.
The macro analysis involves the evaluation of communities, This study proposed a systematic approach to facility location
alternatives and sub-regions whereas the microanalysis selection by using the concepts of group decision-making,
involves the evaluation of specific sites in a particular fuzzy set theory and hierarchical structure analysis controlled
community. Linear programming, factor rating systems, by objective and subjective criteria.
analytic delphi model and center of gravity are some of the
techniques that support macroanalysis [5], [6]. Without taking II. METHODOLOGY
into consideration the global linkages between the choice A fuzzy system is devised using fuzzy set theory and
elements, the conventional approaches can only provide a set hierarchical structure for improving the accuracy of location
of systematic procedures for problem-solving. Moreover, the selection and justification. The entire process for the
outcome also depends on the decision by the analyst. Machine identification of the best facility location is as follows.
learning algorithms are suitable for solving multi criteria a) Form a decision-making committee with members from
decision making approaches [7]–[10], however fuzzy different divisions of owner, [E1, E2…., Ek]. Determine the
approach is used for this research. Identification of facility
location in a region has key importance in the decision-
numerous m alternatives (A1, A2…, Am) for each of the K scale. The transformation of linguistic scales to fuzzy
criteria (C1, C2…., Cn). numbers as defined in Table I, is constructed in these matrices
b) Choose an appropriate linguistic scale and allocate the corresponding to each owner as shown in Table II.
weights in triangle fuzzy numbers or indirectly through TABLE 1. MEMBERSHIP FUNCTION OF LINGUISTIC SCALE
pairwise comparisons. Fuzzy Linguistic scales Scale of
c) Sort the criteria into hierarchical order and categorize them number Fuzzy
using subjective and objective criteria. number
d) Subjective criteria and objective criteria are characterized ~ Equally important (Eq) (1,1,3)
1
by linguistic assessments and quantitative values. ~ Weakly important (Wk) (1,3,5)
e) Transform the linguistic variables into triangle fuzzy 3
~ Essentially important (Es) (3,5,7)
numbers using a defined rating scale. 5
~ Very strongly important (Vs) (5,7,9)
f) Create a fuzzy reciprocal matrix using criterion and sub- 7
~ Absolutely important (Ab) (7,9,9)
criteria. 9
~ −1 Less Absolutely important (LAb) (1/9,1/9, 1/7)
g) Aggregate the hierarchy over all criteria and sub-criteria. 9
h) Represent the final preference order of locations using Best ~ −1 Less Very strongly important (LVs) (1/9,1/7,1/5)
7
Non-fuzzy Performance (BNP) value. ~ −1
5 Less Essentially important (LEs) (1/7,1/5,1/3)
~ −1 Less Weakly important (LWk) (1/5,1/3,1)
III. HIERARCHICAL EVALUATION CRITERION 3
~ −1 Less Equally important (LEq) (1/3,1,1)
STRUCTURE 1
a~11 a~12 . . a~1n ( L11 , M11 , U11 ) ( L12 , M12 , U12 ) . . ( L1n , M1n , U1n )
~
a21 a~22 . . a~2n ( L21 , M 21 , U 21 ) ( L22 , M 22 , U 22 ) . . ( L2n , M 2n , U 2n )
~
D= . . . . . = . . . . . (2)
. . . . . . . . . .
a~ a~n 2 . . a~nn ( Ln1, M n1, U n1 ) ( Ln 2 , M n 2 , U n 2 ) . . ( Lmn , M mn , U nn
n1 (2)
Table V shows the constructed fuzzy Pairwise comparisons weights of sub-criteria wsj= ( Lwsj, Mwsj, Uwsj) are estimated in
of sub-criteria based on five owners. Therefore, Table VI.
normalization of fuzzy vector and also Fuzzy synthetic
TABLE V. FUZZY PAIRWISE COMPARISONS OF SUB-CRITERIA WITH FIVE OWNERS
Owner 1
Environmental Spatial Economy
C11 C12 C21 C22 C31 C32
Lij Mij Uij Lij Mij Uij Lij Mij Uij Lij Mij Uij Lij Mij Uij Lij Mij Uij
C11 1.00 1.00 1.00 0.33 1.00 1.00 C21 1.00 1.00 1.00 7.00 9.00 9.00 C31 1.00 1.00 1.00 3.00 5.00 7.00
C12 1.00 1.00 3.00 1.00 1.00 1.00 C22 0.11 0.11 0.14 1.00 1.00 1.00 C32 0.14 0.20 0.33 1.00 1.00 1.00
Owner 2
C11 1.00 1.00 1.00 0.20 0.33 1.00 C21 1.00 1.00 1.00 5.00 7.00 9.00 C31 1.00 1.00 1.00 1.00 3.00 5.00
C12 1.00 3.00 5.00 1.00 1.00 1.00 C22 0.11 0.14 0.20 1.00 1.00 1.00 C32 0.20 0.33 1.00 1.00 1.00 1.00
Owner 3
C11 1.00 1.00 1.00 0.14 0.20 0.33 C21 1.00 1.00 1.00 1.00 3.00 5.00 C31 1.00 1.00 1.00 7.00 9.0 9.0
C12 3.00 5.00 7.00 1.00 1.00 1.00 C22 0.20 0.33 1.00 1.00 1.00 1.00 C32 0.11 0.11 0.14 1.00 1.00 1.00
Owner 4
C11 1.00 1.00 1.00 0.111 0.143 0.200 C21 1.00 1.00 1.00 3.00 5.00 7.00 C31 1.00 1.00 1.00 7.00 9.00 9.00
C12 5.00 7.00 9.00 1.000 1.000 1.00 C22 0.14 0.20 0.33 1.00 1.00 1.00 C32 0.11 0.11 0.14 1.00 1.00 1.00
Owner 5
C11 1.00 1.00 1.00 0.111 0.111 0.143 C21 1.00 1.00 1.00 1.00 1.00 3.00 C31 1.00 1.00 1.00 7.00 9.00 9.00
C12 7.00 9.00 1.00 1.000 1.000 1.00 C22 0.33 1.00 1.00 1.00 1.00 1.00 C32 0.11 0.11 1.00 1.000 1.00 1.00
Fig. 4. Membership function of linguistic variables for the evaluator, E2 where 𝐸̃31 is the evaluation value of alternative number three
with respect to sub-criteria number one. Therefore, the fuzzy
performance matrix 𝐸̃𝑖𝑗 of each of the alternatives can also
be obtained from the fuzzy performance value of each
alternative under n criteria, that is, 𝐸̃= 𝐸̃𝑖𝑗 = (𝐿𝐸𝑖𝑗 , 𝑀𝐸𝑖𝑗 , 𝑈𝐸𝑖𝑗 ) as
shown in Table IX.
Fig. 5. Membership function of linguistic variables for the evaluator, E3 VII. FUZZY SYNTHETIC DECISION
The weights of each location evaluation criterion 𝐹𝑤 ̃𝑗 and
fuzzy performance values are integrated for the calculation of
fuzzy numbers. According to the weight of criteria 𝑤 ̃𝑗 =
(𝐿𝑤𝑗 , 𝑀𝑤𝑗 , 𝑈𝑤𝑗 ) and weight of sub-criteria 𝑤 ̃ 𝑠𝑗 =
(𝐿𝑤𝑠𝑗 , 𝑀𝑤𝑠𝑗 , 𝑈𝑤𝑠𝑗 ), the overall weight of criteria would be
Fig. 6. Membership function of linguistic variables for the evaluator, E4
𝐹𝑤̃𝑗 =(𝐿𝑤𝑗 , 𝑀𝑤𝑗 , 𝑈𝑤𝑗 ) x(𝐿𝑤𝑠𝑗 , 𝑀𝑤𝑠𝑗 , 𝑈𝑤𝑠𝑗 ) = (𝐿𝐹𝑤𝑗 , 𝑀𝐹𝑤𝑗 , 𝑈𝐹𝑤𝑗 ) [17].
The estimated overall criteria weight vectors are provided in
Table VIII.
TABLE IX. AVERAGE FUZZY PERFORMANCE VALUES OF ALTERNATIVES FOR EACH ALTERNATIVE
WITH OVERALL WEIGHTS
Criterion Overall weights A1 A2 A3 A4
Pollution 0.03 0.06 0.19 86.6 100 100 35.6 48.6 66.0 23.0 36.4 53.6 46.0 60.6 77.0
Nuisance 0.10 0.24 0.61 83.6 96.0 98.0 60.0 72.6 84.0 53.0 65.6 80.0 81.6 93.0 96.0
Growth 0.04 0.11 0.32 79.0 88.0 93.0 74.0 81.0 89.0 3.0 6.0 29.8 0.0 0.0 23.8
Neighborhood 0.01 0.03 0.08 54.6 66.0 78.0 52.6 64.6 75.0 52.6 64.6 74.0 53.0 65.4 76.6
Transportation 0.01 0.03 0.08 46.0 60.6 77.0 86.6 100 100 79.0 92.6 96.0 70.0 79.6 89.0
Investment 0.003 0.005 0.02 75.0 86.6 93.0 48.0 60.4 73.6 50.0 59.4 73.6 77.0 88.0 93.0
NIMBY 0.27 0.53 0.93 71.6 81.6 88.0 59.0 71.6 84.0 59.0 71.6 84.0 26.6 37.6 54.0
(𝐿𝐸11 , 𝑀𝐸11 , 𝑈𝐸11 ) (𝐿𝐸12 , 𝑀𝐸12 , 𝑈𝐸12 ) . . (𝐿𝐸1𝑛 , 𝑀𝐸1𝑛 , 𝑈𝐸1𝑛 ) (𝐿𝐹𝑤1 , 𝑀𝐹𝑤1 , 𝑈𝐹𝑤1 )
(𝐿𝐸21 , 𝑀𝐸21 , 𝑈𝐸21 ) (𝐿𝐸22 , 𝑀𝐸22 , 𝑈𝐸22 ) . . (𝐿𝐸2𝑛 , 𝑀𝐸2𝑛 , 𝑈𝐸2𝑛 ) (𝐿𝐹𝑤2 , 𝑀𝐹𝑤2 , 𝑈𝐹𝑤2 )
𝑅̃ = . . . . . ∗ . (3)
. . . . . .
[ (𝐿𝐸𝑚1 , 𝑀𝐸𝑚1 , 𝑈𝐸𝑚1 ) (𝐿𝐸𝑚2 , 𝑀𝐸𝑚2 , 𝑈𝐸𝑚2 ) . . (𝐿𝐸𝑚𝑛 , 𝑀𝐸𝑚𝑛 , 𝑈𝐸𝑛𝑛 ] [ (𝐿𝐹𝑤𝑛 , 𝑀𝐹𝑤𝑛 , 𝑈𝐹𝑤𝑛 )]
From the overall criteria weight vector 𝐹𝑤 ̃𝑗 and the of injection molding parameters for HDPE/TiO2 nanocomposites
fabrication with multiple performance characteristics using the
̃
alternatives fuzzy performance matrix 𝐸𝑖𝑗 , the final fuzzy Taguchi method and grey relational analysis,” Materials (Basel).,
synthetic decision matrix can be concluded in Table IX. vol. 9, no. 8, p. 710, 2016.
Therefore, the fuzzy appropriate index of locations are [3] I. Sabry, A.-H. I. Mourad, and D. T. Thekkuden, “Optimization of
metal inert gas-welded aluminium 6061 pipe parameters using
derived from the fuzzy synthetic decision matrix 𝑅̃ , that is, analysis of variance and grey relational analysis,” SN Appl. Sci.,
𝑅̃𝑖 = 𝐸̃ ∘ 𝐹𝑤̃ as constructed in equation (4), where ∘ vol. 2, no. 2, p. 175, 2020.
indicates the arithmetic operation. From Table IX, the [4] R. Arunachalam et al., “Optimization of stir–squeeze casting
parameters for production of metal matrix composites using a
calculation of fuzzy multiplication is complex, therefore it is hybrid analytical hierarchy process–Taguchi-Grey approach,”
denoted by the approximate multiplied result of the fuzzy Eng. Optim., vol. 52, no. 7, pp. 1166–1183, 2020.
multiplication and the approximate fuzzy number 𝑅̃𝑖 , where [5] A. Klose, “An LP-based heuristic for two-stage capacitated facility
𝑅̃𝑖 = (𝐿𝑅𝑖 , 𝑀𝑅𝑖 , 𝑈𝑅𝑖 ) a fuzzy synthetic decision of each location problems,” J. Oper. Res. Soc., vol. 50, no. 2, pp. 157–166,
1999.
alternative can be represented as shown in Table X, where [6] C. Canel, B. M. Khumawala, J. Law, and A. Loh, “An algorithm
𝐿𝑅𝑖 , 𝑀𝑅𝑖 and 𝑈𝑅𝑖 are the lower, middle and upper synthetic for the capacitated, multi-commodity multi-period facility location
performance values of the alternative as constructed in problem,” Comput. Oper. Res., vol. 28, no. 5, pp. 411–427, 2001.
equation 5. [7] A. H. Idrisi and A. H. I. Mourad, “Wear performance analysis of
Aluminum matrix composites using Artificial neural network,” in
𝐿𝑅𝑖 = ∑𝑛𝑖=1 𝐿𝐸𝑖𝑗 × 𝐿𝐹𝑤𝑗 ; 𝑀𝑅𝑖 = ∑𝑛𝑖=1 𝑀𝐸𝑖𝑗 × 𝑀𝐹𝑤𝑗 ; 2019 Advances in Science and Engineering Technology
𝑛
𝑈𝑅𝑖 = ∑𝑖=1 𝑈𝐸𝑖𝑗 × 𝑈𝐹𝑤𝑗 International Conferences (ASET), 2019, pp. 1–5.
[8] B. F. Yousef, A.-H. I. Mourad, and A. Hilal-Alnaqbi, “Prediction
of the mechanical properties of PE/PP blends using artificial neural
VIII. RANKING THE FUZZY NUMBER networks,” Procedia Eng., vol. 10, pp. 2713–2718, 2011.
[9] B. F. Yousef, A.-H. I. Mourad, and A. Hilal-Alnaqbi, “Modeling
The BNP value of the fuzzy number 𝑅̃𝑖 can be calculated of the mechanical behavior of polyethylene/polypropylene blends
using artificial neural networks,” Int. J. Adv. Manuf. Technol., vol.
according to the following equation: 𝐵𝑁𝑃𝑤𝑖 = [[(𝑈𝑅𝑖 − 64, no. 5–8, pp. 601–611, 2013.
𝐿𝑅𝑖 ) + (𝑀𝑅𝑖 − 𝐿𝑅𝑖 )]/3] + 𝐿𝑅𝑖 i as constructed in Table [10] D. T. Thekkuden and A.-H. I. Mourad, “Investigation of feed-
X. forward back propagation ANN using voltage signals for the early
prediction of the welding defect,” SN Appl. Sci., vol. 1, no. 12, p.
Table X.
10 FUZZY
Fuzzy appropriate indexINDEX
of locations 1615, 2019.
TABLE APPROPRIATE OF LOCATIONS
[11] A. Cagri Tolga, F. Tuysuz, and C. Kahraman, “A fuzzy multi-
LRi MRi URi BNP Rank
criteria decision analysis approach for retail location selection,”
A1 34.377 85.855 204.970 108.40 1 Int. J. Inf. Technol. Decis. Mak., vol. 12, no. 04, pp. 729–755,
A2 27.653 72.059 185.951 95.220 2 2013.
[12] T.-C. Chu, “Facility location selection using fuzzy TOPSIS under
A3 23.796 61.488 161.772 82.350 3 group decisions,” Int. J. uncertainty, fuzziness knowledge-based
A4 18.222 50.819 146.292 71.780 4 Syst., vol. 10, no. 06, pp. 687–701, 2002.
[13] M. S. Rahman, M. I. Ali, U. Hossain, and T. K. Mondal, “Facility
location selection for plastic manufacturing industry in
Bangladesh by using AHP method,” Int. J. Res. Ind. Eng., vol. 7,
IX. CONCLUSION no. 3, pp. 307–319, 2018.
[14] D. Chatterjee and B. Mukherjee, “Potential hospital location
Fuzzy group decision making approach was found to selection using AHP: a study in rural India,” Int. J. Comput. Appl.,
be an excellent tool to evaluate the qualitative and vol. 71, no. 17, 2013.
quantitative assessments for the selection of municipal waste [15] S. Mandal and S. S. Mondal, “Analytic hierarchy process (AHP)
approach for selection of open cast coal mine project,” Int. J. Ind.
facility location. A fuzzy group is constructed from five
Eng., vol. 7, no. 2, 2016.
owners based on their experiences. The owners’ decision [16] S. Aydin and C. Kahraman, “A New Fuzzy Analytic Hierarchy
depended on the proposed hierarchical analysis of the Process and Its Application to Vendor Selection Problem.,” J.
municipal waste location problem that included subjective Mult. Log. Soft Comput., vol. 20, 2013.
[17] T.-Y. Chou, C.-L. Hsu, and M.-C. Chen, “A fuzzy multi-criteria
and objective criteria. The concepts of fuzzy numbers and
decision model for international tourist hotels location selection,”
linguistic variables are used to evaluate four alternatives Int. J. Hosp. Manag., vol. 27, no. 2, pp. 293–301, 2008.
considering seven sub-criteria hierarchical from objective [18] H. Aras, Ş. Erdoğmuş, and E. Koç, “Multi-criteria selection for a
and subjective criteria. In the proposed technique, the owners wind observation station location using analytic hierarchy
process,” Renew. Energy, vol. 29, no. 8, pp. 1383–1392, 2004.
chose their scales in comparison to alternatives without any
[19] A. Kengpol, P. Rontlaong, and M. Tuominen, “A decision support
limitations. As a result, the suggested decision algorithm was system for selection of solar power plant locations by applying
implemented. Ranks were allotted based on the calculated fuzzy AHP and TOPSIS: An Empirical Study,” 2013.
best non-performance fuzzy number. The best location was [20] M. Tabari, A. Kaboli, M.-B. Aryanezhad, K. Shahanaghi, and A.
Siadat, “A new method for location selection: a hybrid analysis,”
evaluated using the proposed Fuzzy AHP algorithm.
Appl. Math. Comput., vol. 206, no. 2, pp. 598–606, 2008.
[21] F. T. S. Chan, N. Kumar, and K. L. Choy, “Decision-making
REFERENCES approach for the distribution centre location problem in a supply
chain network using the fuzzy-based hierarchical concept,” Proc.
[1] I. Sabry, A. H. Idrisi, and A.-H. I. Mourad, “Friction Stir Welding Inst. Mech. Eng. Part B J. Eng. Manuf., vol. 221, no. 4, pp. 725–
Process Parameters Optimization Through Hybrid Multi-Criteria 739, 2007.
Decision-Making Approach,” Int. Rev. Model. Simulations
(IREMOS); Vol 14, No 1, Feb. 2021.
[2] H. Pervez, M. S. Mozumder, and A.-H. I. Mourad, “Optimization