An investigation on the sensitivity and stability radius of returns to scale and efficiency in data envelopment analysis

An issue which has received widespread attention in rapidly growing field of DEA is the sensitivity of the results of analysis to perturbations in the data. The sensitivity of the returns to scale (RTS) classifications in data envelopment analysis is studied by means of linear programming problems. This is surprising, since RTS classifications provide important information for improving an individual DMU's performance when scale inefficiencies are detected. But an important point in this paper is a new method for calculating the 'stability radius', in which, the data variation neither change the class of efficiency units, nor the class of return to scale. In fact, obtaining the 'stability radius', meets both objectives simultaneously. (C) 2012 Elsevier Inc. All rights reserved.