A Study on Efficiency Ranking in

Data Envelopment Analysis

Fan Yuqiang

Abstract

Recently, Data Envelopment Analysis (DEA) has gained widespread attention and research as an analysis method that can evaluate relative efficiency without the need to assume weights beforehand. In addition to traditional models such as CCR, BCC, and SBM, many extended models have been developed, such as cross-efficiency and super-efficiency models. However, among the numerous DEA models, no method can achieve the best performance in all evaluation scenarios, which often confuses decision-makers. Furthermore, many applications have only used the initial BCC and CCR models. Therefore, it is crucial to classify and analyze the pros and cons, and scope of different categories of DEA models that have been proposed. This paper presents a classification of different DEA models and explains their principles, advantages, disadvantages, and research progress to help decision-makers fully understand relevant knowledge. Moreover, we propose a three-step process that enables decision-makers to conveniently select the specific model they need according to actual conditions. The paper showcases a case study of Guilin University's academic department's efficiency using this process and conducts a result difference analysis.