An inverse DEA model for inputs/outputs estimation with respect to decision makers preferences: The case of Refah bank of IRAN

Inverse DEA problems introduced to answer the following question: among a group of DMUs, if we change certain inputs (outputs) of a particular unit, how much should the outputs (inputs) of that unit be changed in order that the DMU retains its efficiency level with respect to the other DMUs? To solve such problems, it has been transformed into an MOLP problem, where every efficient solution of the MOLP problem can be regarded as a forecasting tool for the desired outputs (inputs). Although different solution methods have been developed for the MOLP problems, the interactive approaches are generally believed to be the most promising methods in which the preference information of the decision maker is progressively articulated during the solution process and incorporated into it. In this paper we propose a methodology that uses interactive MOLP for solving the inverse DEA problems. Its robustness is illustrated via a practical problem, for which data set is taken from Refah bank of IRAN.