网络结构决策单元效率评价方法及其应用研究
发布时间：2018-07-17 03:26
【摘要】：效率评价是管理学和经济学领域中的经典问题之一。由于网络数据包络分析(DEA)与传统DEA方法相比,具有结果更准确、可识别系统无效性来源等优点,因此网络DEA是近年来的研究热点之一。本文在现有研究的基础上,进一步研究不同网络结构下的决策单元效率评价问题,研究结果有望为决策者提供有效的绩效管理决策支持。本文主要分为六章,内容组织如下:第一章介绍DEA与网络DEA的基本理论,对网络DEA的相关研究开展文献回顾与总结,并深入阐述了本文的研究意义。第二章针对子系统产出的非同质性问题开展研究。本章结合香港医院中存在的子系统产出非同质性问题,拓展了现有的非同质性决策单元绩效评价模型,并对香港医院的绩效进行了实证分析。研究首次关注到医院中子系统产出非同质性的问题,有助于拓展DEA方法的应用研究范围。第三章针对两阶段DEA关系模型中的规模收益可变问题开展理论研究,并将拓展的两阶段DEA关系模型应用到奥运会参赛国(地区)效率评价问题中。本章将两阶段DEA关系模型被扩展为一个非线性规划模型,并采用启发式搜寻算法求解非线性模型。同时本章还首次将参赛国(地区)在奥运会上的表现看成是一个两阶段过程(运动员准备阶段和运动员竞技阶段)开展效率评价。研究发现参赛国(地区)的总系统效率受运动员竞技效率影响较大。第四章针对平行系统子系统效率非唯一性问题开展研究,并将所提出的方法应用到奥运会参赛国(地区)效率评价中。本章考虑子系统所有可能的效率,计算得到每个子平行系统的效率值区间。同时本章还首次将参赛国(地区)的夏季奥运会与冬季奥运会看作平行结构的两个子系统进而开展效率评价,可以有效避免采用夏季奥运会进行评价对高纬度寒冷地区的国家（地区）不公平,采用冬季奥运会又对低纬度热带地区的国家（地区）不公平的问题。研究发现参赛国(地区)在夏季奥运会与冬季奥运会上的表现与其地理位置高度相关。第五章基于全局权重思想评价平行结构特征下DMU的占优关系和排名区间问题。通过构建平行结构效率占优关系模型,可以得到DMU的占优关系和效率排名区间,可以有效避免权重不确定性带来的不利影响。第六章总结了本文的主要研究工作和创新之处,在分析本文研究存在的不足的基础上,提出了本文未来可能的研究方向。本文的创新点主要有:(1)基于全局权重思想研究了平行结构特征下的决策单元占优关系和效率排名区间;(2)首次在医院系统绩效评价中考虑了子系统产出非同质性,扩展了已有的非同质性决策单元绩效评价方法;(3)首次考虑奥运会参赛国(地区)内部的两阶段与平行系统结构,扩展了已有的两阶段与平行系统DEA评价方法。
[Abstract]:Efficiency evaluation is one of the classic problems in the field of management and economics. Network data Envelopment Analysis (DEA) is one of the research hotspots in recent years because it has more accurate results and can identify the source of system inefficiency than the traditional DEA method. Based on the existing research, this paper further studies the efficiency evaluation of Decision-making units under different network structures, and the results are expected to provide effective decision support for decision makers. This paper is divided into six chapters, the content is organized as follows: the first chapter introduces the basic theory of DEA and network DEA, carries out literature review and summary of the related research of network DEA, and expounds the significance of this paper. The second chapter studies the heterogeneity of subsystem output. In this chapter, the existing performance evaluation model of non-homogeneous decision-making unit is extended, and the performance of Hong Kong hospitals is analyzed empirically in combination with the non-homogeneity problem of subsystem output in Hong Kong hospitals. The research focuses on the heterogeneity of subsystem output in hospital for the first time, which is helpful to expand the application of DEA method. The third chapter studies the variable scale income problem in the two-stage DEA model, and applies the extended two-stage DEA relationship model to the efficiency evaluation of the countries (regions) competing in the Olympic Games. In this chapter, the two-stage DEA model is extended to a nonlinear programming model, and a heuristic search algorithm is used to solve the nonlinear model. At the same time, the performance of the participating countries (regions) in the Olympic Games is considered as a two-stage process (athlete preparation stage and athlete competitive stage) for the first time. It is found that the total system efficiency of the participating country (region) is greatly affected by the athletes' competitive efficiency. In chapter 4, the non-uniqueness problem of the efficiency of parallel system subsystem is studied, and the proposed method is applied to the evaluation of the efficiency of the countries (regions) competing in the Olympic Games. In this chapter, all possible efficiency of subsystems is considered, and the efficiency interval of each sub-parallel system is calculated. At the same time, for the first time, the Summer Olympic Games and the Winter Olympic Games in the participating countries (regions) are regarded as two subsystems with parallel structure, and then the efficiency evaluation is carried out. It can effectively avoid the unfair evaluation of countries (regions) in the cold regions with high latitude and the countries (regions) in the tropical regions of low latitudes with the use of the Summer Olympic Games, and the use of the Winter Olympics is not fair to the countries (regions) in the tropical regions of the lower latitudes. The study found that the performance of the participating countries in the Summer and Winter Olympic Games is highly correlated with their geographical location. The fifth chapter evaluates the dominance relation and ranking interval problem of DMU under parallel structure based on global weight idea. By constructing a parallel structure efficiency dominant relation model, the dominant relationship and efficiency ranking interval of DMU can be obtained, and the adverse effects of uncertainty of weights can be avoided effectively. The sixth chapter summarizes the main research work and innovation of this paper, based on the analysis of the shortcomings of this study, the possible future research direction of this paper is put forward. The innovations of this paper are as follows: (1) based on the idea of global weight, the dominant relationship of decision units and the efficiency ranking interval under parallel structure characteristics are studied; (2) the heterogeneity of subsystem output is considered in the performance evaluation of hospital system for the first time. This paper extends the existing non-homogeneous decision unit performance evaluation methods. (3) considering the two-stage and parallel system structure within the countries (regions) of the Olympic Games for the first time, the existing two-stage and parallel system DEA evaluation methods are extended.
【学位授予单位】：中国科学技术大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：F224
[Abstract]:Efficiency evaluation is one of the classic problems in the field of management and economics. Network data Envelopment Analysis (DEA) is one of the research hotspots in recent years because it has more accurate results and can identify the source of system inefficiency than the traditional DEA method. Based on the existing research, this paper further studies the efficiency evaluation of Decision-making units under different network structures, and the results are expected to provide effective decision support for decision makers. This paper is divided into six chapters, the content is organized as follows: the first chapter introduces the basic theory of DEA and network DEA, carries out literature review and summary of the related research of network DEA, and expounds the significance of this paper. The second chapter studies the heterogeneity of subsystem output. In this chapter, the existing performance evaluation model of non-homogeneous decision-making unit is extended, and the performance of Hong Kong hospitals is analyzed empirically in combination with the non-homogeneity problem of subsystem output in Hong Kong hospitals. The research focuses on the heterogeneity of subsystem output in hospital for the first time, which is helpful to expand the application of DEA method. The third chapter studies the variable scale income problem in the two-stage DEA model, and applies the extended two-stage DEA relationship model to the efficiency evaluation of the countries (regions) competing in the Olympic Games. In this chapter, the two-stage DEA model is extended to a nonlinear programming model, and a heuristic search algorithm is used to solve the nonlinear model. At the same time, the performance of the participating countries (regions) in the Olympic Games is considered as a two-stage process (athlete preparation stage and athlete competitive stage) for the first time. It is found that the total system efficiency of the participating country (region) is greatly affected by the athletes' competitive efficiency. In chapter 4, the non-uniqueness problem of the efficiency of parallel system subsystem is studied, and the proposed method is applied to the evaluation of the efficiency of the countries (regions) competing in the Olympic Games. In this chapter, all possible efficiency of subsystems is considered, and the efficiency interval of each sub-parallel system is calculated. At the same time, for the first time, the Summer Olympic Games and the Winter Olympic Games in the participating countries (regions) are regarded as two subsystems with parallel structure, and then the efficiency evaluation is carried out. It can effectively avoid the unfair evaluation of countries (regions) in the cold regions with high latitude and the countries (regions) in the tropical regions of low latitudes with the use of the Summer Olympic Games, and the use of the Winter Olympics is not fair to the countries (regions) in the tropical regions of the lower latitudes. The study found that the performance of the participating countries in the Summer and Winter Olympic Games is highly correlated with their geographical location. The fifth chapter evaluates the dominance relation and ranking interval problem of DMU under parallel structure based on global weight idea. By constructing a parallel structure efficiency dominant relation model, the dominant relationship and efficiency ranking interval of DMU can be obtained, and the adverse effects of uncertainty of weights can be avoided effectively. The sixth chapter summarizes the main research work and innovation of this paper, based on the analysis of the shortcomings of this study, the possible future research direction of this paper is put forward. The innovations of this paper are as follows: (1) based on the idea of global weight, the dominant relationship of decision units and the efficiency ranking interval under parallel structure characteristics are studied; (2) the heterogeneity of subsystem output is considered in the performance evaluation of hospital system for the first time. This paper extends the existing non-homogeneous decision unit performance evaluation methods. (3) considering the two-stage and parallel system structure within the countries (regions) of the Olympic Games for the first time, the existing two-stage and parallel system DEA evaluation methods are extended.
【学位授予单位】：中国科学技术大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：F224
【相似文献】
相关期刊论文 前10条
1 黄凌灵;;3种套保效率评价模型的比较研究[J];北方工业大学学报;2009年02期
2 朱慧新;骆Z，
本文编号：2128797
本文链接：https://www.wllwen.com/shoufeilunwen/jjglbs/2128797.html
