Proposing a Model to Predict Efficiency and Related Risk by Using Stochastic Data Envelopment Analysis Technique and the Imperialist Competitive Algorithm

This paper proposes a new model which provides some insights on using Stochastic Data Envelopment Analysis (SDEA) technique in the future performance appraisal for similar units of an organization. This research contributes to advance current knowledge by proposing new model in which, regardless of having benefits of DEA, resolve some head problems of the context. For instance: 1) the efficiency estimate impossibility, 2) liability to measuring of acceptable risk taking for managers in regards to achieving each units predetermined efficiency, and 3) the unreal distribution of weights to inputs and outputs of the DEA model. We used from Imperialist Competitive Algorithm (ICA) refereeing to have a nonlinear and complexity model. ICA is one of the newest evolutionary optimization algorithms. Finally, in order to reach a better understanding of the proposed model by considering ICA, it was applied to predict efficiencies for a number of Iranian Bank branches. The high correlation between real efficiencies (were obtained with real outputs and DEA model) and predicted efficiencies (were obtained with proposed model) for all of branches in finish of the predicted financial period, which represented the validity of the proposed model.