A Selective and Gradual Method for Efficiency Improvement of DEA Models

Basic-DEA models have ability to calculate performance targets for inefficient DMUs which are used for efficiency improvement approaches. Unfortunately, these improvement approaches aren't selective and gradual. Therefore, this paper introduces an efficiency improvement algorithm under this circumstance and discusses on the convergence and computational aspects of the algorithm. Also, an illustrative example is presented to show the ability of suggested method, from computational point of view.