Flexible factors in categorized data for data envelopment analysis

Evaluating the performance of a Decision Making Unit (DMU) is a significant topic for many researchers. One of the most appropriate methods for measuring the relative efficiency of peer DMUs is Data Envelopment Analysis (DEA), which has been widely used in recent decades. In conventional DEA models, the efficiency of DMUs is determined without considering their conditions and capabilities, which is not fair at all. In this study, DMUs are categorized based on their uncontrollable input variables as well as flexible measures which play either input or output roles. In this vein, for the first time, a novel algorithm is proposed to assess the efficiency of DMUs and introduce the role of flexible measures simultaneously via solving a Mixed Binary Linear Programming (MBLP). Since the role of flexible measures may be different in each classification, an Ordered Weighted Averaging (OWA) operator is applied to select the status of those measures. Finally, the applicability of the proposed approach is evaluated at the hospitals in thirteen districts of the city of Ahvaz, Iran, which are categorized based on their population and facilities in presence of flexible measures.