An improved fuzzy MULTIMOORA approach for multi-criteria decision making based on objective weighting method (CCSD) and its application to technological forecasting method selection. (March 2019)

• Record Type: Journal Article
• Authors: Dahooie, J. Heidary; Zavadskas, E.K.; Firoozfar, H.R.; Vanaki, A.S.; Mohammadi, N.; Brauers, W.K.M.
• Is Part Of: Engineering applications of artificial intelligence. Volume 79(2019)
• Page Start: 114
• Page End: 128
• Publication Date: 2019-03
• Subjects: Technological forecasting method -- Fuzzy MULTIMOORA -- Correlation coefficient and standard deviation (CCSD) -- Multi-attribute decision making; Engineering -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Ingénierie -- Informatique -- Périodiques; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; Artificial intelligence; Engineering -- Data processing; Expert systems (Computer science); Periodicals; 620.00285
• DOI: 10.1016/j.engappai.2018.12.008
• Languages: English
• ISSNs: 0952-1976

Abstract

Abstract: MULTIMOORA (Multi-Objective Optimization on the basis of Ratio Analysis plus full multiplicative form) is a somewhat new multi criteria decision-making (MCDM) method which provides high efficiency and effectiveness in problem solving. To evaluate different alternatives and calculate their scores, it uses three major approaches; namely, ratio system (RS), reference point (RP), and full multiplicative form (FM). Based on the scores, alternatives are individually ranked in each approach. The obtained ranks are the basis for final ranking, which is determined under the rules of dominance theory. However, this method impose some limitations on the evaluation model such as the need for aggregation and final ranking based on each alternative's ranks already obtained by three different approaches (i.e., RS, RP, and FM) with no consideration to score differences, taking the same importance level for these approaches, the lack of attention to deviation of scores, the complexity of aggregation in the dominance theory and the need for multiple comparisons to obtain final rankings, and the probability of a circular reasoning with no distinction between alternatives included. In order to overcome shortcomings, we applied an objective weight determination method called CCSD (Correlation Coefficient and Standard Deviation) method to enhance the MULTIMOORA performance. In this regard, the scoring distance of every alternative is completely included in the aggregation. Using regression statistics, standard deviations and correlation coefficients; the dispersion in the set of scores is also computed in relation to three approaches of RS, RP, and FM. The application of unique weight for each approach will yield more realistic results. Further, it is no longer needed to use the dominance theory and the problems such as multiple comparisons and a circular reasoning are also eliminated. Finally, a real decision making problem is applied to select the proper technological forecasting method which illustrates the validity and practicality of the proposed approach. Graphical abstract: Highlights: Selecting the technological forecasting method using new hybrid MADM method. Overcome MULTIMOORA shortcomings and enhance its performance. Decrease complexity of aggregation process using CCSD. Reduce probability of a circular reasoning phenomenon using CCSD. Enhance distinctiveness of aggregated ranks by using alternatives scores instead of ranks.