Increasing the discrimination power the decision making units based on reducing dispersion of weights in the data envelopment analysis

Abstract

<span style="font-size: medium;"><strong>Data envelopment analysis</strong><strong> which is a nonparametric technique for evaluating relative efficiency of the decision making units with multiple inputs and outputs, has been a very popular method among researchers. While this nonparametric technique is popular, it has some drawbacks such as lack of discrimination in efficient units and weights dispersion .The present study, which is a model based on a multi-criteria data envelopment analysis has been proposed to moderate the homogeneity of weights dispersion by using goal programming. The proposed model minimized variances of input and output weights. The result shows that the dispersion of input and output weights has been balanced. Furthermore, the power of discrimination has been improved in DEA.</strong></span>