【作者】 李宁；

【导师】 杨印生；

【作者基本信息】 吉林大学 ， 系统工程， 2005， 硕士

【摘要】 通过对系统综合评价方法的归类和分析,综合评价方法现存有三方面的主要矛盾:均一评价与非均一评价的矛盾;输入、输出指标数据的确定性与不确定性的矛盾;决策的主观性与客观性的矛盾。鉴于每种评价模型都有自身的优点和缺点,本文研究的目的就是利用综合集成的数学规划方法,将不同的系统评价模型扬长避短的整合成为新的综合模型,从而使新模型具有更大的优点。本文的研究内容主要包括: (1)在介绍具有经济含义的DEA 模型的基础上,利用含最大利润的DEA模型应用于农产品生产企业的经济效益评价中。(2)利用AHP 构造偏好锥,首次提出AHP/C2WH 模型与方法,并应用于全国各地区综合经济状况评价中。(3)利用日本运筹学家Sueyoushi,T 提出的DEA-DA 模型,应用于区域规划管理中,并将客观全面的定量判别分析结果,作为制定区域规划方针的科学依据。(4)利用锥比率DEA 模型—C2WH 与DA 结合,首次提出C2WH-DA模型与方法,并应用于全国各省市电信通信服务水平的评价中。(5)在日本运筹学家Sueyoushi,T 提出的DEA-DA 模型基础上,首次提出能够处理观测样本数据为L-R 模糊数的FDEA-DA 模型与方法,并对其判别效果的灵敏性和处理多维数据的能力进行了分析。综合集成模型可以为区域管理提供量化的模型依据,并为国民经济的宏观调控提出建设性的规划方案;另外,综合集成模型还可以为部门之间的绩效评价提供量化工具,为企业或非盈利部门内部管理提供准确可靠的管理信息,因此,本论文的研究对国民经济的不断完善和发展具有重要意义。更多还原

【Abstract】 When China joined in WTO (World Trade Organization), every enterprise and department of our country will challenge to more intensity world contention. Only improving the capability of contention in the enterprises and departments can our country keeps in the advantage status of world. Both enterprise and department need adopt synthetic and overall systematic evaluation methods to obtain some theoretical conclusion as the next stage direction of work and production. In generally, synthetic evaluation methods are divided into quantitative and qualitative methods. Since the system of economy and society become more and more complication, the reliability on pure quantitative analysis conclusion become lower. We can obtain objective and overall conclusion when we integrate evaluation methods which is composed by quantitative and qualitative methods. So it will have more reality meaning to using integrated evaluation methods to assessment economy and society system in the background of global-content. The author of this paper participates in the project of Studies on non-uniform evaluation and Discriminate Analysis approaches based on DEA, which is authorized by the country natural science commission and consults a lot of academic documents about system evaluation method in china and abroad. The paper main introduces Data Envelopment Analysis method and draws some conclusion on the development of DEA Model study. Based on the opinion of System, we established three synthetic evaluation models in the aspect of uniform evaluation and non-uniform evaluation. At last, the paper added all of the reference documents. The paper is divided into seven chapters as follow: Chapter one, Introduction, discusses the background and meaning of the research. In this part, we also introduce the main content, technological route and methods. Chapter two, uniform system evaluation method —AHP, firstly introduces the theoretical system structure and methods of system integrated methods. The integrated models in this paper are structured in the opinion of system. Then, the uniform evaluation methods —AHP is discussed in this charpter. Integrated methods can be divided into uniform methods and non-uniform methods in the aspect of allocation weight. Chapter three, non-uniform system evaluation method, discusses the non-uniform evaluation methods of DA and DEA. The part of this paper summaries the development of DEA method research and detailed introduces three basic models of DEA method which are C2R model, C2GS2 model and C2WH model. Following, this part also introduces the DEA model with economic meaning which in the view of cost, revenue and profit efficiency. At last, this chapter summaries the types of efficiency and discusses the relationship of them.Being every evaluating model has itself characters, we can analysis different models and integrate them into a new model. Chapter four named as AHP/C2WH Model and Methodology Based on Definite Region. Using AHP method form a predilection cone of decision maker and incorporating above cone in C2WH model can reflect decision maker’s predilection more objective and overall. It’s very essential to build such AHP/C2WH model based on definite region. This paper also discusses this new model’s DEA (AHP/C2WH) efficiency and its relative superior character comparing to other models. Chapter five, The Study of DEA-DA Model and Method, introduces threer models of integrating DEA and DA. (1) Region Planning Studies in view of DEA-DA Model and Method DEA-DA is a discriminant analysis method in view of Data Envelopment Analysis. This method not only gives a discriminant function, but also can successfully dispose the overlap of discriminiant analysis by incorporating added DEA model into DA formulation, which makes the DA method more flexibility. If the value of discriminant function equal to discriminate value, we can use discriminant analysis method to calculate, then we can obtain overall and objective conclusions. This model can successfully used in Region Planning. This paper also gives an example to show its application. (2) Study on DEA-DA Model Based on Preference Cone C2WH model, a cone ratio DEA model, can embody the evaluator’s predilection in indexes and decision making unite (DMU) by using preference cone. Therefore, we can form a DEA-DA model based on preference cone by更多还原