关联网络DEA在多层次指标体系中的应用研究  

The Application of Relational Network DEA in Multi-level Index Systems

作　　者：张静娴[1,2] 周文[2] 陈伟[2] 向宇 

机构地区：[1]北京大学汇丰商学院,广东深圳518000 [2]哈尔滨工程大学经济管理学院,黑龙江哈尔滨150001

出　　处：《运筹与管理》2016年第4期187-194,共8页Operations Research and Management Science

基　　金：国家软科学项目(2010GXS5B173);黑龙江省自然科学基金项目(G201209)

摘　　要：对复杂系统进行效率评价,是改善其运作性能的必要之举。DEA作为一种典型的效率评价方法,现有研究已提出了若干串联、并联系统的DEA效率评价问题。然而,为评价复杂系统的效率,建立的评价指标体系往往具有多层次性,常有串并联混合结构出现。现有研究囿于将复杂系统中的层次关系作为黑箱处理,或者只研究串联或并联系统,没有有效考察DEA在串并联混合结构中效率评价的应用,对复杂系统的"多层次指标体系"不能进行有效评价。与传统的DEA评价方法相比,本文首次将网络DEA引入多层次指标体系的评价当中,以串并联混合结构为基础,构建了用于多层次指标体系评价的关联网络DEA模型;从模型创新的角度,本文的关联网络DEA模型融合了串联和并联系统DEA模型,为处理子系统间复杂的关联关系提供了新思路。最后,通过对军民结合企业的实证研究验证了模型的合理性。To improve a complex system operation performance, it is necessary to evaluate its efficiency. DEA is a kind of typical efficiency evaluation methods. The existing researches on DEA have put forward efficiency evaluation problems of series or parallel systems. However, the index system established to evaluate the efficiency of complex systems often appears with multi-level, series-parallel mixed structures. The existing researches limit the complex hierarchical relationships in the system in a black box, or only simplify the complex system into series or parallel system. Few studies have investigated the application of DEA efficiency evaluation in the series-parallel mixed structure effectively, thus they cannot make effective evaluation of ＂multi-level index system＂ of the complex system. Compared with traditional DEA evaluation methods, this paper introduced the application of network DEA in multi-level index systems based on the series - parallel mixed structures for the first time. From the perspective of model innovation, the relational network DEA model is a blend of series and parallel system DEA model in this paper, and the model provides a new way of dealing with complex relationships between subsystems in a complex system. Finally, this paper verifies the rationality of the model by evaluating the civil-military integration enterprise innovation efficiency.

关 键 词：数据包络分析 关联网络DEA 串并联混合结构 多层次指标体系 

分 类 号：F224.3[经济管理—国民经济]