A new multi-criterion decision making (MCDM) method based on proximity indexed value for minimizing rank reversals. (May 2018)

• Record Type: Journal Article
• Authors: Mufazzal, Sameera; Muzakkir, S.M.
• Is Part Of: Computers & industrial engineering. Volume 119(2018)
• Page Start: 427
• Page End: 438
• Publication Date: 2018-05
• Subjects: Decision making -- Normalization -- Proximity index -- Rank reversal -- Ideal solution; Engineering -- Data processing -- Periodicals; Industrial engineering -- Periodicals; 620.00285
• DOI: 10.1016/j.cie.2018.03.045
• Languages: English
• ISSNs: 0360-8352

Abstract

Highlights: The addition/deletion of alternatives causes rank reversals in existing MCDM techniques. Several methods tried to minimize rank reversals but to no avail. A new method is proposed that minimizes rank reversals under these conditions. The accuracy of the method has been established by comparison with existing MCDM methods. The feasibility of the proposed method has been established under addition/deletion of alternative/s. Abstract: The identification of best alternative from amongst the available choices is a complex task dependent upon the user priorities that needs to be graded on a rating scale requiring careful consideration of all influencing characteristic features of individual alternative. Several multi criterion decision making techniques are available to facilitate the decision maker arrive at a best alternative by ranking the alternatives in an order of preference. However, it has been observed that with the addition of new alternatives or deletion of existing alternatives, the ranks of the available alternatives, indicating their suitability to a particular set of requirements, is not maintained. This is often described as rank reversal phenomenon by many researchers. The commonly used MCDM methods are particularly incapable of preventing this rank reversal phenomenon. Thus, addition of new alternative/s or deletion of existing alternative/s creates a modified order of preference which may, sometimes, lead to erroneous decisions/results. In the present research work, an effort has been made to critically examine the rank reversal phenomenon with an aim to propose a new method to obviate this problem. In order to establish the feasibility and effectiveness of the proposed method in preventing the rank reversal phenomenon, several case studies covering different technological specializations from the reported literature work have been considered. The results indicate that the rank reversal issue was found to be minimal with the use of proposed method and a good correlation was found to exist between the rankings obtained by the proposed method and the other commonly used MCDM methods. The proposed method is thus capable of preventing the rank reversal phenomenon, arising out of change in available alternatives.