诱导型语言算子的多属性群决策方法研究
发布时间：2018-08-04 14:43
【摘要】：语言型多属性群决策在社会、市场分析、经济等方面应用广泛,它是决策科学理论的重要组成部分。由于在决策过程中决策环境的不确定性,在面对复杂问题进行决策时,往往不能利用定量的方法进行决策分析,因此,专家利用语言信息进行方案属性的评价往往更加直观和方便,减少了信息的丢失,同时,利用群决策可以避免因其中某个专家决策的失误而导致决策结果评判错误,避免不良后果的情况发生,保证了决策结果的准确性,特别是对于电力建设项目而言,决策的结果直接影响到项目实施能否达到预期。诱导型语言算子常用于多属性决策问题,其主要特征是利用诱导变量进行指标值的重新排列,根据排列后的位置进行加权集结。本文在诱导型语言算子的基础上进行多属性群决策问题研究,主要研究内容包括:(1)通过研究诱导有序加权平均(IOWA)算子和概率有序加权平均(POWA)算子的性质,了解诱导变量和概率变量形成的集成算子在多属性群决策中的应用,在此基础上研究诱导语言概率有序加权平均(ILPOWA)算子,分析不同重要程度下集成算子的集结结果。(2)将诱导语言概率有序加权平均(ILPOWA)算子推广到诱导不确定语言广义概率有序加权平均(IULGPOWA)算子,其中,对于指标值以不确定语言变量所表示的情形,利用诱导不确定语言广义有序加权平均(IULGOWA)算子和不确定语言广义概率加权平均(ULGPWA)算子形成的IULGPOWA算子进行集结,这个新的集成算子不仅考虑了参数所在位置重要性程度,还考虑了参数的概率,与其他集成算子更具一般性,同时,根据电力建设项目特点,对新的集成算子进行实证分析。(3)基于诱导二元语义广义有序加权平均(2TLGOWA)算子和二元语义广义概率加权平均(2TLGPA)算子,形成诱导二元语义广义概率有序加权平均(2TLIGPOWA)算子,利用二元语义描述指标值,有效地减少信息的丢失,同时,在广义的环境下考虑概率信息和决策者的态度特征,在最大值和最小值中提供了一个集结算子族,阐述了该决策模型的广义性,与其他集成算子更具一般性,并在电力建设项目背景下进行广义集成算子的多属性群决策实证分析,验证了其有效性。通过构建了两类集成算子,能够对描述诱导语言环境下大部分的多属性决策问题,同时所提供的算例有效的证明了其在实际中的可行性和有效性。
[Abstract]:Linguistic multi-attribute group decision making is widely used in society, market analysis, economy and so on. It is an important part of decision science theory. Because of the uncertainty of the decision-making environment in the decision-making process, the quantitative method can not be used to make decision analysis in the face of complex problems. It is more intuitive and convenient for experts to evaluate scheme attributes by using language information, which reduces the loss of information. At the same time, the use of group decision making can avoid the error of decision result because of the error of one expert decision. In order to avoid the adverse consequences and ensure the accuracy of the decision results, especially for the electric power construction projects, the decision results directly affect the implementation of the project to achieve the expected. Inductive language operators are often used in multi-attribute decision making problems. The main feature of the operators is that the index values are rearranged with induced variables and weighted aggregation is carried out according to the arranged positions. On the basis of inductive language operators, this paper studies the problem of multi-attribute group decision making. The main contents are as follows: (1) the properties of induced ordered weighted average (IOWA) operator and probabilistic ordered weighted average (POWA) operator are studied. The application of the integration operator formed by induced variables and probabilistic variables in multi-attribute group decision making is studied. On this basis, the probabilistic ordered weighted average (ILPOWA) operator of inductive language is studied. The aggregation results of integration operators with different degrees of importance are analyzed. (2) the generalized probabilistic ordered weighted average (ILPOWA) operator of induced language is extended to the generalized probabilistic ordered weighted average (IULGPOWA) operator of induced uncertain language. For the case where the index value is expressed as an uncertain language variable, the IULGPOWA operators formed by the generalized ordered weighted average (IULGOWA) operator of induced uncertain language and the generalized probabilistic weighted average (ULGPWA) operator of uncertain language are used to aggregate. This new integration operator takes into account not only the importance of the location of the parameters, but also the probability of the parameters, which is more general than other integration operators. At the same time, according to the characteristics of the electric power construction project, (3) based on the generalized ordered weighted average (2TLGOWA) operator and the generalized probability weighted average (2TLGPA) operator, the generalized ordered weighted average (2TLIGPOWA) operator is formed. The binary semantics is used to describe the index value, which effectively reduces the loss of information. At the same time, considering the probability information and the attitude characteristics of the decision maker in the generalized environment, a family of aggregation operators is provided in the maximum and minimum values. The generality of the decision model is expounded, which is more general than other integration operators, and an empirical analysis of multi-attribute group decision making of generalized integration operator under the background of electric power construction project is carried out, which verifies its validity. By constructing two kinds of ensemble operators, most of the multi-attribute decision making problems in the environment of description induction language can be solved. At the same time, the examples are provided to prove its feasibility and validity in practice.
【学位授予单位】：华北电力大学(北京)
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：O225
本文编号：2164200
[Abstract]:Linguistic multi-attribute group decision making is widely used in society, market analysis, economy and so on. It is an important part of decision science theory. Because of the uncertainty of the decision-making environment in the decision-making process, the quantitative method can not be used to make decision analysis in the face of complex problems. It is more intuitive and convenient for experts to evaluate scheme attributes by using language information, which reduces the loss of information. At the same time, the use of group decision making can avoid the error of decision result because of the error of one expert decision. In order to avoid the adverse consequences and ensure the accuracy of the decision results, especially for the electric power construction projects, the decision results directly affect the implementation of the project to achieve the expected. Inductive language operators are often used in multi-attribute decision making problems. The main feature of the operators is that the index values are rearranged with induced variables and weighted aggregation is carried out according to the arranged positions. On the basis of inductive language operators, this paper studies the problem of multi-attribute group decision making. The main contents are as follows: (1) the properties of induced ordered weighted average (IOWA) operator and probabilistic ordered weighted average (POWA) operator are studied. The application of the integration operator formed by induced variables and probabilistic variables in multi-attribute group decision making is studied. On this basis, the probabilistic ordered weighted average (ILPOWA) operator of inductive language is studied. The aggregation results of integration operators with different degrees of importance are analyzed. (2) the generalized probabilistic ordered weighted average (ILPOWA) operator of induced language is extended to the generalized probabilistic ordered weighted average (IULGPOWA) operator of induced uncertain language. For the case where the index value is expressed as an uncertain language variable, the IULGPOWA operators formed by the generalized ordered weighted average (IULGOWA) operator of induced uncertain language and the generalized probabilistic weighted average (ULGPWA) operator of uncertain language are used to aggregate. This new integration operator takes into account not only the importance of the location of the parameters, but also the probability of the parameters, which is more general than other integration operators. At the same time, according to the characteristics of the electric power construction project, (3) based on the generalized ordered weighted average (2TLGOWA) operator and the generalized probability weighted average (2TLGPA) operator, the generalized ordered weighted average (2TLIGPOWA) operator is formed. The binary semantics is used to describe the index value, which effectively reduces the loss of information. At the same time, considering the probability information and the attitude characteristics of the decision maker in the generalized environment, a family of aggregation operators is provided in the maximum and minimum values. The generality of the decision model is expounded, which is more general than other integration operators, and an empirical analysis of multi-attribute group decision making of generalized integration operator under the background of electric power construction project is carried out, which verifies its validity. By constructing two kinds of ensemble operators, most of the multi-attribute decision making problems in the environment of description induction language can be solved. At the same time, the examples are provided to prove its feasibility and validity in practice.
【学位授予单位】：华北电力大学(北京)
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：O225
【参考文献】
相关期刊论文 前10条
1 陈启明;陈华友;;基于IOWA算子的两类准则下的最优组合预测模型及其应用[J];数理统计与管理;2013年05期
2 刘培德;张新;;直觉不确定语言集成算子及在群决策中的应用[J];系统工程理论与实践;2012年12期
3 万树平;;直觉模糊多属性决策方法综述[J];控制与决策;2010年11期
4 相辉;;语言型时序多属性群决策方法及在服务创新中的应用[J];运筹与管理;2009年04期
5 梁昌勇;张恩桥;戚筱雯;陆青;;一种评价信息不完全的混合型多属性群决策方法[J];中国管理科学;2009年04期
6 胡彦;李秀美;陈华友;;基于IOWA算子的税收组合预测模型[J];统计与决策;2009年10期
7 卫贵武;林锐;;基于二元语义多属性群决策的灰色关联分析法[J];系统工程与电子技术;2008年09期
8 毛开翼;;诱导有序加权平均算子在城市公共交通预测中的应用[J];河南科技大学学报(自然科学版);2007年06期
9 徐泽水;陈剑;;一种基于区间直觉判断矩阵的群决策方法[J];系统工程理论与实践;2007年04期
10 王坚强;;一种信息不完全确定的多准则语言群决策方法[J];控制与决策;2007年04期
相关硕士学位论文 前2条
1 洪俊;电网建设项目风险管理研究[D];华北电力大学（北京）;2011年
2 李溪;火电建设项目投资风险评价研究[D];华北电力大学（河北）;2009年
，本文编号：2164200
本文链接：https://www.wllwen.com/kejilunwen/yysx/2164200.html
