Comparison of multi-criteria decision methods through a ranking stability index

Abstract

Various multi-criteria decision-making (MCDM) procedures have been developed over the last few decades to help decision-making in complex and seemingly intractable decision tasks. A major criticism of various MCDM methods is that they may yield different results when applied to the same problem. A comparative study is proposed here to identify similarities and divergences between the most used MCDM methods. Compared approaches include: ELECTRE III, PROMETHEE I and II, TOPSIS, AHP and PEG. Two multi-criteria case studies are presented. Studied methods are employed to establish an arrangement of a number of alternatives based on two and eight conflicting criteria, respectively. A Gini index is used to quantify rankings dispersion of Pareto optima obtained through studied MCDM methods. Results highlight the sensitivity of the Pareto-compromise design and ranking to the applied MCDM method.