Undesirable factors in stochastic DEA cross-efficiency evaluation: An application to thermal power plant energy efficiency

Abstract

In this study using an input-oriented data envelopment analysis (DEA) model with undesirable outputs a new stochastic model called Expected Ranking Criterion is proposed. The proposed model employs statistical techniques to evaluate the efficiency of decision making units (DMUs) with stochastic data. Based on the proposed model, a stochastic DEA (SDEA) cross-efficiency model is suggested for ranking and discrimination of DMUs. Then, given the non-uniqueness of resulting optimal solution, a stochastic model is introduced for rating priorities by which cross-efficiency evaluation is performed using aggressive approach. Finally, the proposed models are implemented for evaluating 32 thermal power plants. The results show the applicability of the proposed models.