Seeking alternative DEA benchmarks

Abstract

This paper identifies alternative, more easily attainable benchmarks for inefficient decision making units (DMUs). This is done by extending data envelopment analysis (DEA), a popular linear programming method that optimizes the ratio of multiple weighted outputs to multiple weighted inputs, using the resultant ratio as a proxy measure of efficiency of the DMU. DEA identifies peers for an inefficient DMU; DMUs that could serve as benchmarks for the inefficient DMU to improve its current efficiency level. In this paper, we propose a method for computing an alternative benchmark that is based on linear combination of the peer DMUs. The alternative benchmarks are in many instances and in some respects, more easily attainable, and hence provide a more meaningful target for the inefficient DMU. To derive the alternative benchmark, we combine results of the multiplier and the envelopment form of DEA in a logical and intuitive manner. The proposed model is simple and practical, and provides a more meaningful transition path for an inefficient DMU as is empirically demonstrated using the familiar “Hospital” data set.