Ranking methods within data envelopment analysis

Abstract

Within the field of data envelopment analysis is a sub-group of papers in which many researchers have sought to improve the differential capabilities of DEA and to fully rank both efficient, as well as inefficient, decision-making units. We have divided the ranking concepts into seven general areas based on the following concepts: Super-efficiency, benchmarking, cross-efficiency, common set of weights, multivariate statistics, multi-criteria decision-making and inefficiency dominance. After describing the approaches, we compare and contrast them using an illustration drawn from a set of universities. It is apparent that the approaches succeed in strengthening the results of the non-parametric data envelopment analysis models and frequently enable an almost complete ranking of decision-making units. However, the results may diverge substantially between the different models, suggesting that the choice of framework must be context dependent and chosen with great care.