Uncertain RUSSEL data envelopment analysis model: A case study in iranian banks

Abstract

Data Envelopment Analysis (DEA) is recognized as a robust analytical tool extensively utilized in measuring the relative efficiency of a group of decision-making units (DMUs) with multiple inputs and outputs. The DEA models require inputs and outputs equipped with precise information. However, in real-world situations, inputs and outputs may be unstable and complicated, thus unable to be accurately measured. This problem resulted in the investigation of uncertain DEA models. The RUSSELL model was studied in this paper in an uncertain environment where uncertain inputs and outputs were belief degree-based uncertainty, useful for the cases for which no historical information of an uncertain event is available. As the solution method, the uncertain RUSSELL model was converted to a crisp form using two approaches of expected value model and expected value and dependent chance-constrained model separately. Finally, an applied example regarding the Iranian banking system was presented to document the proposed models.