A New Model for Classifying Inputs and Outputs and Evaluating the DMUs Efficiency in DEA Based on Cobb-Douglas Production Function

In conventional data envelopment analysis it is assumed that the input versus output status of each chosen performance measures is known. In some conditions finding a statue of some variables from the point view of input or output is very difficult; these variables treat as both an input and output and are called flexible measures. This paper proposes a new model based on Cobb-Douglas production function for classifying inputs and outputs and evaluating the performance of decision making units by considering flexible measures. The model is capable to provide marginal output elasticities and marginal rates of output transformation. The presented model is evaluated comparing with the recent model in the literature. The result shows that the measure efficiencies of our model are statistically closer to true efficiencies and have higher rank correlation with true efficiencies. We illustrate the application of proposed model with the real data found in literature. Also results obtained from simulated and real data show that there are high correlation between our model and the recent model.