Researcher at the Business Studies and Development Office, Saipa Yadak, Islamic Republic of Iran
Abstract
A remarkable result of Webber et al. (1996) shows the scale used, format and order in which the comparisons are entered in the matrix, may affect the AHP results. For the purpose of the experiment, in this investigation, we will concentrate our attention on format effect. In other words, we argue that, what happens if we change the format of alternative or attribute presentation in decision making matrix and redo our assessment of alternatives, in some of MADM models (i.e. TOPSIS, PROMETHEE II, ELECTRE, and three versions of AHP (distributive, ideal, and multiplicative modes)). The result showed that the stability of the TOPSIS, PROMETHEE II, and ELECTRE results. Meanwhile, the distributive, ideal and multiplicative modes have bad and unreasonable performance. An interesting finding is that in distributive and ideal modes, for change the format of alternative presentation and in multiplicative mode, for change the format of attribute presentation,
rank reversal phenomenon can be happened. The
finding in this paper can help the decision maker select suitable prioritization methods.
Keywords: MADM, TOPSIS, PROMETHEE II, ELECTRE, AHP, Format Effect, Rank Reversal
View PDF