Ranking Stochastic Efficient Dmus Based on Reliability

Data envelopment analysis (dea) is a nonparametric mathematical programming approach for evaluating efficiency of a set of decision making units (dmus). since the number of efficient dmus is more than one, there is a necessity of having methods to discriminate between efficient dmus. in classic dea it is assumed that data have deterministic values. through real world applications this assumption may not be satisfied. on the other hand, data may have stochastic essence. in this paper a method for ranking stochastic dmus is suggested which is based on the reliability of efficiency of dmus. using numerical example, we demonstrate how to use the results.