Integrated data envelopment analysis: linear vs. nonlinear model

摘要: Abstract This paper develops a relationship between two linear and nonlinear data envelopment analysis (DEA) models which have previously been developed for the joint measurement of efficiency effectiveness decision making units (DMUs). It will be shown that DMU is overall efficient by model if only it model. We compare these demonstrate an alternative algorithm also show more computationally than model, does not potential estimation error heuristic search procedure used in determines global optimum solutions rather local optimum. Using 11 different sets from published papers 1000 simulated data, we explore models. set most frequently papers, with step size equal to 0.00001, requires running 1,955,573 programs (LPs) measure 24 DMUs compared LPs required Similarly, very small consists 5 DMUs, 7861 0.0001, whereas needs just LPs.