基于“专家打分法”的DEA交叉效率聚集方法

DEA cross-efficiency aggregation method based upon expert scoring

摘要:

现存的有关数据包络分析(DEA)交叉效率的文献多集中在,对DEA交叉效率矩阵的计算上,较少关注DEA交叉效率的集结问题。最常用的集结方法,就是没有考虑他们彼此重要性的等权处理,运用简单的数学平均方法得出每个决策单元最终的平均交叉效率值。考虑到彼此重要性的差异,基于“专家打分法”,提出了一个非等权的处理方法。该方法把每个进行权重选择的决策单元当作一个外部“专家”,交叉效率看作不同外部“专家”对被评价决策单元的效率“打分”。由于每个“专家”的学识背景、工作经验等方面的差异,他们的“打分”理应给予不同的权重,权重的确定主要基于他们效率“打分”上的差异,若某个“专家”的效率“打分”和其他“专家”存在较大差异,则此专家的权威性就会受到质疑,相应的他的效率“打分”在决策单元最终的效率“得分”上扮演的作用就会小。最后运用算例分析说明了此方法的有效性。

Abstract:

The existing literatures about the DEA cross-efficiency evaluation are mainly focused on the cross-efficiency matrix calculation. Little attention has been paid to the aggregation process of cross-efficiencies. The most extensively used approach is to aggregate cross-efficiencies with equal weights. Taking the relative importance into consideration, we propose a method to aggregate cross-efficiencies based on expert scoring. This method views the cross-efficiency matrix as the efficiency score by the different experts. Due to their difference on the education background and experience, their score should play different role in the final score. The weights allocated to them are based on their dissimilarity level to other ones. Finally, the numerical example shows its validity.