The most efficient unit without explicit inputs: An extended MILP-DEA model

Abstract Data envelopment analysis (DEA) has been a very popular method for measuring and benchmarking relative efficiency of each decision making units (DMUs) with multiple inputs and multiple outputs. DEA and Discriminant Analysis (DA) are similar in classifying units to exhibit either good or poor performance. On the other hand, selecting the most efficient unit between several efficient ones is one of the main issues in multi-criteria decision making (MCDM). Some proponents have suggested some approaches and claimed their methodologies involve discriminating power to determine the most efficient DMU without explicit input. This paper focuses on the weakness of a recent methodology of these approaches and to avoid this drawback presents a mixed integer programming (MIP) approach. To illustrate this drawback and compare discriminating power of the recent methodology to our new approach, a real data set containing 40 professional tennis players is utilized.