Improving the discriminating power of DEA: focus on globally efficient units
提出全局效率方法，通过共同权重结构筛选并排序DEA有效单元，提升区分能力，并与多准则DEA和交叉效率方法比较。
We introduce in this paper the global efficiency approach as a means to improve the discriminating power of Data Envelopment Analysis (DEA). To discriminate further among the DEA efficient units, we deal only with the units that can maintain their efficiency score under common weighting structures. Then we proceed further to ranking the whole set of DEA efficient units. We compare the global efficiency approach with the multi-criteria DEA and the cross-efficiency approaches on the basis of characteristic numerical examples drawn from the literature.