A Cross-Efficiency Profiling For Increasing Discrimination in Data Envelopment Analysis

Abstract

Data Envelopment Analysis (DEA) cannot provide adequate discrimination among effieient decision making units (DMUs). To discriminate these effieient DMUs is an interesting research subject. The purpose of this paper is to present a Cross-Efficiency Prollling (CEP) model which can be used to improve discriminalitin power of DEA and condiicl a methodological comparison of CEP anti the other developed methods without a priori information. CEP retains ihe original spirilot DEA in trying to extract as much information as ptissible from the data without requiring pre-selecied weights on inputs and outputs. We propose ihat inputs which are not substitutes for each other be assessed separately and only with respect to outputs which consume them or to which they are otherwise related. In this way input-specific ratings based on the concept of cross-efficiency measure arc derived giving a profile for each DMU. We will demonstrate that CEP is more discriminating through an example taken from Baker and Talluri [Computer and Industrial Engineering, 32(1), 101–108 (1997)].