A New Model for Determining Anchor Points in Data Envelopment Analysis

Data envelopment analysis (DEA) is a method for evaluating performance of organizations and decision making units (DMUs). This method divide decision making units (DMUs) in four different categories: inefficient، weak efficient، extreme efficient and non-extreme efficient. In this paper، we investigate a new DMU category which called «anchor point». An anchor point places on common region between the efficiency frontier and free-disposability. Indeed، an anchor point is extreme efficient which makes weak efficiency frontier. Omission of the anchor points will change efficiency frontier and eliminate a region of generating possibility set. Anchor point also has another characteristic، changing its input or output will change efficiency frontier and by increasing input or decreasing output، the new point will be still on frontier and is extreme efficient. Therefore، the characteristics of the anchor point demonstrate its importance. As for its importance، we propose faster methods to identify these points. At first، we express identifying algorithms for these points and then by using their characteristics، we propose some methods like supper efficiency method which identifies anchor points by using less calculation than others. We will give a numerical example to explain proposed method and compare it with other methods.