Evidential Reasoning approach for multiple-criteria decision making : A simulation-based formulation

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journalpeer-review

12 Scopus Citations
View graph of relations



Original languageEnglish
Pages (from-to)4381-4396
Journal / PublicationExpert Systems with Applications
Issue number9
Online published9 Jan 2015
Publication statusPublished - 1 Jun 2015


Multiple-criteria decision making (MCDM) permeates in almost every industrial and management setting. The Evidential Reasoning (ER) approach, pioneered and developed by Yang, Xu and their colleagues since the early 1990's and currently with applications in a wide ranging set of domains, is among the premier methods for MCDM. While it is hard to dispute the versatility of the ER approach, a key disadvantage in the existing ER framework is that its formulation involves complex formulas with logically non-trivial proofs. This complexity forces the non-specialists to use ER as a black-box technique, and presents definite impediment for the specialists to further develop ER. A contribution of the present article is that through a conceptually simple recasting of ER into a simulation-based framework (termed SB-ER), we show that the complexity seen in the existing ER framework can be radically reduced - it now becomes logically straightforward to comprehend the inner working of ER. Further, we show that the capability of the existing ER approach can be readily extended via this simulation-based framework. Thus, owing its intellectual debt to and building upon the firm foundation of ER, SB-ER paves a promising shortcut for fine-tuning and further developing ER. Finally, we demonstrate the utility of SB-ER using a small industrial dataset. To facilitate further development, a set of Matlab source codes, which complements currently available ER-based software, is available from the author upon request.

Research Area(s)

  • Dempster-Shafer theory, Evidential Reasoning, Fuzzy uncertainty, Multiple-criteria decision making, Simulation