New model for improving discrimination power in DEA based on dispersion of weights

One of the difficulties of data envelopment analysis (DEA) is the problem of deficiency discrimination among efficient decision making units (DMUs) and hence, yielding large number of DMUs as efficient ones. The main purpose of this paper is to overcome this inability. One of the methods for ranking efficient DMUs is minimising the coefficient of variation (CV) for inputs-outputs weights, which, was suggested by Bal et al. (2008). In this paper, we introduce a nonlinear model for ranking efficient DMUs based on modifying of the model suggested by Bal et al. and then we convert the nonlinear model proposed into a linear programming form. The motivation of this work is to linearise the existing nonlinear model which has the computational complexity.