Study on interval data envelopment analysis model with preference restraint cone based on unascertained rational number

Abstract

Data envelopment analysis (DEA) is a non-parametric method for evaluating the relative efficiency of decision-making units (DMUs) with multiple inputs and multiple outputs. Unascertained math is a useful tool for dealing with unascertained information. The unascertained theory was introduced to data envelopment analysis in this paper. A new model, which is called interval data envelopment analysis with preference restraint cone model, was constructed to make approval of the decision making units. In this model, the input data and output data could be disposed by unascertained rational number method in advance and the weight of index should be dominated by evaluator through preference restraint cone. Therefore, interval data envelopment analysis with preference restraint cone model is convenient for integrating decision-maker's preferences and precision of imprecise data. At the end of the text, a case was applied to qualify scientific and feasibility of the method.