Uncertain Data Envelopment Analysis for Cross Efficiency Evaluation with Imprecise Data

Abstract

Self evaluation and peer evaluation in data envelopment analysis (DEA) are effective means to comprehensively reflect the efficiencies of decision-making units (DMUs). However, when some of the inputs and outputs of DMUs cannot be accurately observed, the traditional evaluation methods will lose their applicability. This paper attempts to treat the imprecise inputs and outputs as uncertain variables based on uncertainty theory and hence to propose a new uncertain DEA model for cross-efficiency evaluation via the evaluation of both self efficiency and peer efficiency. Moreover, the equivalent form and the proof of the new model are also presented for accurate calculation. Finally, a numerical example is given to illustrate the evaluation results.