Simulation of Stichastic Data Envelopment Analysis in Selection of Research and Development Projects

Uncertainty is the obvious attribute of any activity in the real world, and performance analysis of units in terms of uncertainty is one of the most important concerns of managers and planners of companies. Various approaches to data envelopment analysis have been presented so far to provide a good tool for evaluating the efficiency of units in uncertain situations, including random, fuzzy and robust approaches. In a situation where the probability distribution function of random variables is known, stochastic approach is a precise solution for problem solving. Nevertheless, previous research in this field has focused on the transformation of a random problem into a definite problem. In this research, a method for modeling data envelopment analysis is presented using computer programming. In this method, random processes are simulated, and finally, the performance distribution function is presented instead of a definite number as the unit's efficiency. In this model, a new concept is presented as the probability of efficiency, which represents the probability that the unit will be located on the efficient boundary. The proposed method is empirically applied in a real case of selecting research and development projects.