Super-Efficiency DEA Models and Their Applications

Data Envelopment Analysis (DEA) offers several models for comparison the relative efficiency of decision making units (DMU) described by multiple inputs and outputs. In most DEA models the best DMUs reach the efficient score 100%. The necessity to classify the efficient units leads to super-efficiency definitions. The paper discusses several concepts of super-efficiency as Anderson and Petersen model and super SBM model. These models are based on removing the evaluated DMU from the set of units and measuring its distance from the modified efficient frontier. This point can be found by several ways, e.g. by minimisation of the sum of deviations, minimisation of the maximal deviation or by the combination of both the approaches. The paper compares super-efficiency scores given by several metrics and discusses the results. Numerical experiments are realized on the real data set of 185 bank branches of a Czech bank.