Open AccessComparative Study of ELECTRE Methods with Intuitionistic Fuzzy Sets Applied on Consumer Decision Making Case
Author(s) -
Mei-Yao Wu
Publication year - 2019
Publication title -
european journal of engineering and technology research
Language(s) - English
Resource type - Journals
ISSN - 2736-576X
DOI - 10.24018/ejeng.2019.4.10.1571
Subject(s) - electre , ranking (information retrieval) , rank (graph theory) , data mining , score , computer science , preference , product (mathematics) , weighting , ideal solution , function (biology) , mathematics , membership function , fuzzy logic , multiple criteria decision analysis , artificial intelligence , statistics , machine learning , fuzzy set , operations research , medicine , geometry , combinatorics , evolutionary biology , biology , radiology , physics , thermodynamics
This article proposes an intuitionistic fuzzy (IF) Elimination and Choice Translating Reality (ELECTRE) method to rank consumers’ alternatives ranking order with subjects’ questionnaires by using IF data and the ranking order applied the proposed method are closer to consumers their own ranking order. Moreover, the mean value of Spearman correlation coefficients are higher than 80% in each product category, and also higher than 90% at bank service product category especially. This study uses IF sets characteristics to handle uncertain decision environment and to classify the concordance and discordance sets according to their score function for measuring the degree of suitability of each alternative and also using the concept of the positive and negative ideal solutions to rank all candidate alternatives in the proposed method. Furthermore, analyzer can use this method to gain valuable information from questionnaires, and consumers rarely provide preference data directly. Additionally, an empirical study is given to illustrate the proposed method and also compared with Wu and Chen 2011’s paper which considered not only score function but also accuracy function. The results show that using the proposed method, decision makers can easily predict candidate alternatives ranking order and make decisions accurately.