交叉效率评价方法中一种改进新单元导入保序性方法.docVIP

交叉效率评价方法中一种改进新单元导入保序性方法 　　摘要： 针对交叉效率方法在评估排序时所存在的局限性，即在导入新的决策单元时，可能会使原先的决策单元(DMU)排序发生变化，本文通过建立二次权重集结交叉效率保序性模型对各DMU的权重进行二次分配，进而对交叉效率矩阵进行集结以得到最终的均衡交叉效率值。利用该均衡交叉效率值进行排序时可以消除导入新DMU时所产生的逆序现象。算例说明了该方法有效，计算结果较纳什讨价还价交叉效率保序性模型有所改进，综合计算量更小，各DMU权重更加合理，与一般交叉效率模型的效率值差异更小。 Abstract: This paper firstly analyzes the the flaws that the fomer rank of the DMUs may be changed when new DMUs are added, when the cross efficiency method is used in ranking DMUs. Then, the paper allots the DMUs′ weight again through building the second concentration weight rank preservating(SCWRP) model, and may gain the final equilibrium efficiency by concentrating the cross efficiency matrix. The reversal of the rank when new DMUs added will be eliminated by using this method. An empirical example is illustrated to examine the validity of the proposed method. The result is improved by contrasting the Nash bargaining rank preservation(NBRP) model. The DMUs′ weight is more rational and the difference is littler by contrasting the general cross efficiency model in the SCWRP model than in the NBRP model. The total compute work is larger in the SCWRP model than in the NBRP. 关键词： DEA；交叉效率；保序；二次权重集结；纳什讨价还价 Key words: DEA；cross efficiency；rank preservation；second concentrating weight；Nash Bargaining 中图分类号：C934 文献标识码：A 文章编号：1006-4311（2012）03-0005-02 　　摘要： 针对交叉效率方法在评估排序时所存在的局限性，即在导入新的决策单元时，可能会使原先的决策单元(DMU)排序发生变化，本文通过建立二次权重集结交叉效率保序性模型对各DMU的权重进行二次分配，进而对交叉效率矩阵进行集结以得到最终的均衡交叉效率值。利用该均衡交叉效率值进行排序时可以消除导入新DMU时所产生的逆序现象。算例说明了该方法有效，计算结果较纳什讨价还价交叉效率保序性模型有所改进，综合计算量更小，各DMU权重更加合理，与一般交叉效率模型的效率值差异更小。 Abstract: This paper firstly analyzes the the flaws that the fomer rank of the DMUs may be changed when new DMUs are added, when the cross efficiency method is used in ranking DMUs. Then, the paper allots the DMUs′ weight again through building the second concentration weight rank preservating(SCWRP) model, and may gain the final equilibrium efficiency by concentrating the cross efficiency matrix. The reversal of the rank when new