Enhanced methods for benchmarking and ranking in data envelopment analysis

Abstract

Data Envelopment Analysis (DEA) is a non-parametric method in operations research for estimating production frontier, and benchmarking and ranking Decision Making Units (DMUs). The current DEA techniques are not suitable for these assessments, thus, this study proposes novel robust methods to improve the capabilities of DEA for approximating the production frontier while simultaneously benchmarking and ranking DMUs. Firstly, the shortcomings in the DEA techniques are illustrated with several counter examples followed by new proposed methods to remove the shortcomings. Then, the techniques are combined and used in a linear programming model called Kourosh and Arash Model (KAM). KAM estimates the production frontier and allows decisions within the target regions instead of points in the benchmark of DMUs. In this study, KAM produces three efficiency indexes, namely: the lowest, technical and highest efficiency scores for each DMU. These efficiency indexes provide a sensitivity index for each DMU and rank DMUs completely. KAM is also able to measure the efficiency scores of DMUs inclusive of integer and real data. To sum up, the proposed techniques in this study have improved the capabilities of DEA to assess the production frontier, as well as benchmark and rank DMUs.