Finding a Common Set of Weights for Ranking Decision-Making Units in Data Envelopment Analysis

Data envelopment analysis (DEA) has been a popular method of measuring production performance of a group of decision-making units (DMUs) utilizing same types of inputs to produce outputs. DEA has been applied to many studies including studies on the performance of hospitals, banks, airlines, and schools. However, despite of its popularity, DEA still suffers from some inadequacies. For instance, it sometimes may overestimate efficiencies of some DMUs by applying weight sets with extreme values. Furthermore, it lacks discriminating power among efficient DMUs, since all efficient DMUs have the same efficiency scores. In this paper, we propose a methodology which attempts to identify DMUs which efficiencies are possibly overstated by DEA and provides a common set of input and output weights for ranking DMUs. A common set of weights sometimes is useful when studying tradeoffs between different inputs and outputs in DEA especially when different inputs can be traded and obtaining a full rank of DMUs.