基于聚类分析的数量差异较大决策单元DEA有效性评价  

The DEA Efficiency Evaluation of Decision Making Units with Large Quantity Difference Based on Cluster Analysis

作　　者：张雪怡 马生昀[1] 马占新[2] 黄沙日娜 ZHANG Xue-yi;MA Sheng-yun;MA Zhan-xin;HUANG Sharina(School of Science,Inner Mongolia Agricultural University,Hohhot 010018,China;School of Economics and Management,Inner Mongolia University,Hohhot 010021,China)

机构地区：[1]内蒙古农业大学理学院,内蒙古呼和浩特010018 [2]内蒙古大学经济管理学院,内蒙古呼和浩特010021

出　　处：《数学的实践与认识》2021年第5期169-176,共8页Mathematics in Practice and Theory

基　　金：国家自然科学基金(71661027);内蒙古自然科学基金(2019MS01005)。

摘　　要：传统DEA方法对每个决策单元基于决策单元集进行相对效率评价,如果它们之间在数量上存在较大差异,得到的评价结果可能无法体现数量规模的影响.为了解决这类问题,首先利用聚类分析方法按一定标准将决策单元进行分类,然后利用传统和广义DEA方法对决策单元按类进行自评和他评,最后根据各类决策单元的数量规模情况进行赋权进而给出决策单元的综合相对效率值.Traditional data envelopment analysis(DEA) method evaluates the relative efficiency of each decision making unit(DMU) based on the DMU set,the obtained evaluation results may not reflect the influence of the quantity scale if there exist large quantity differences among all DMUs.To solve this kind of problem,first,all DMUs are divided into several clusters according to a certain standard by cluster analysis,then they are conducted self-evaluation and other-evaluation by each cluster with traditional DEA method and generalized DEA method,and finally weights of them are assigned according to quantity scale of each cluster and the comprehensive efficiency is given.

关 键 词：传统DEA方法 广义DEA方法 决策单元 样本单元 聚类分析 DEA有效 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]