A Compromise Solution Approach for Fuzzy Data Envelopment Analysis: A Case of the Efficiency Prediction

Abstract

The data envelopment analysis (DEA) a data-oriented approach for evaluating the relative performance of decision-making units (DMUs). The traditional DEA applies only to crisp data, whereas the data collected in the real world may be ambiguous and imprecise. The fuzzy DEA is an extension of the DEA using the fuzzy variable to deal with uncertain or imprecise data. This paper proposes two new fuzzy arithmetic-based DEA models with dynamic weights and common weights, formulated as multiple objective decision-making (MODM), and proposed models are represented as the linear programs providing the compromise solutions. The numerical experiment is illustrated to examine the validity of the proposed models, and the experiment shows that the proposed models give better results than other models. The proposed fuzzy DEA models are applied to predict the energy efficiency of 40 iron and steel enterprises in China.