指标结构同质是数据包络分析（DEA）方法的基本假设之一；然而，现实问题的复杂性使得该假设常常难以完全被满足.针对具有包容关系的产出结构异质问题，通过解析决策单元（DMU）之间生产结构的内在关系来构建一种分阶段的DEA效率评价方法.该方法充分考虑了不同结构DMU的主观偏好，较好地规避了传统DEA方法在结构异质DMU效率评价过程中的不公平性.随后，该方法分别被拓展至投入结构异质和多重结构异质的情境中.最后，通过两个算例来说明本文方法的有效性与实用性.
The homogeneity of the index structure is one of the basic assumptions for the data envelopment analysis (DEA) method. However, the complexity of the practical problems always makes this assumption difficult to be fully satisfied. Aiming at the heterogeneous problem of outputs structure with the containment relationship, this paper constructs a phased DEA efficiency evaluation method by analyzing the internal relationship of production structure between decision making units (DMUs). This method takes the subjective preferences of DMUs with different structures into consideration, and avoids the unfairness of traditional DEA method in the process of efficiency evaluation, which has the DMUs with structural heterogeneity. Subsequently, this method is extended to the context of the inputs structure heterogeneity and the multiple structural heterogeneities, respectively. Finally, two examples are used to illustrate the effectiveness and practicability of the method proposed by this paper.
[1] Charnes A, Cooper W W, Rhodes E. Measuring the efficiency of DMUs[J]. European Journal of Operational Research, 1978, 2(6):429-444.
[2] Liu J S, Lu L Y Y, Lu W M. Research fronts in data envelopment analysis[J]. Omega, 2016, 58:33-45.
[3] Zhou H, Yang Y, Chen Y, et al. Data envelopment analysis application in sustainability:The origins, development and future directions[J]. European Journal of Operational Research, 2018, 264:1-16.
[4] Cook W D, Harrison J, Rouse P, et al. Relative efficiency measurement:The problem of a missing output in a subset of decision making units[J]. European Journal of Operational Research, 2012, 220(1):79-84.
[5] Zhu W W, Yu Y, Sun P P. Data envelopment analysis cross-like efficiency model for non-homogeneous decision-making units:The case of United States companies' low-carbon investment to attain corporate sustainability[J]. European Journal of Operational Research, 2018, 269(1):99-110.
[6] 吴永强, 贺昌政. 非同质决策单元的资源分配模型[J]. 系统工程, 2013, 31(10):87-92.
[7] Du J, Chen Y, Huo J. DEA for non-homogenous parallel networks[J]. Omega, 2015, 56:122-132.
[8] Cook W D, Harrison J, Imanirad R, et al. Data Envelopment Analysis with Non-Homogeneous DMUs[J]. Operations Research, 2013, 61(3):666-676.
[9] Li W H, Liang L, Cook W D, et al. DEA models for non-homogeneous DMUs with different input configurations[J]. European Journal of Operational Research, 2016, 254(3):946-956.
[10] 王秀贞, 丁慧平, 胡毅. 基于DEA方法的我国中小企业融资效率评价[J]. 系统工程理论与实践, 2017, 37(4):865-874.
[11] Ding J, Dong W, Liang L, et al. Goal congruence analysis in multi-division organizations with shared resources based on data envelopment analysis[J]. European Journal of Operational Research, 2017, 263(3):961-973.
