Development of opened-network data envelopment analysis models under uncertainty

Abstract

Efficiency evaluation is a very important and key issue in competitive conditions. Organizations and companies face various uncertainties, and this makes it extremely difficult and complex to evaluate their efficiency. In this research, opened-network data envelopment analysis models have been developed in uncertain space for three uncertain states, including: uncertain outputs, uncertain inputs, and simultaneous uncertain inputs and outputs. The proposed models have been used to evaluate the efficiency of 10 two-stage processes, seller and buyer in a supply chain, and the impact of data uncertainty is examined. The results obtained from the developed models have been compared with the results of traditional DEA network models. The validity and accuracy of the developed models have also been examined. The results have shown that the reliability of the proposed models is higher than the traditional DEA network model. Also, by examining the efficiency of decision-making units in different conditions of data uncertainty and deviations, it was determined that the greater the range of this deviation, the lower the efficiency score of different units will be, which is consistent with reality.