首页 | 本学科首页   官方微博 | 高级检索  
     检索      

一种基于决策矩阵的DST-AHP多属性决策方法
引用本文:龚本刚,江文奇,张道武.一种基于决策矩阵的DST-AHP多属性决策方法[J].运筹与管理,2007,16(4):28-34.
作者姓名:龚本刚  江文奇  张道武
作者单位:1. 安徽工程科技学院,管理工程系,安徽,芜湖,241000
2. 南京理工大学,经济管理学院,江苏,南京,2100943
3. 清华大学,科技与社会研究所,北京,100084
基金项目:安徽省科技厅软科学基金;安徽省教育厅科研项目;安徽省高等学校优秀人才计划项目
摘    要:针对层次分析法决策时存在两两判断、一致性检验次数过多和判断矩阵残缺性等问题,本文提出了一种基于决策矩阵的DST-AHP多属性决策方法。该方法结合决策矩阵的特征值,构建DST-AHP方法层次结构模型和判断矩阵,并根据判断矩阵定义不同属性下各焦元的基本概率分配函数;然后利用Dempster合成法则对基本概率分配函数值进行合成,依据合成后值对方案进行排序。最后对AHP和DST-AHP两种方法进行比较分析,说明该方法的有效性。

关 键 词:决策矩阵  证据理论  层次分析法  多属性决策
文章编号:1007-3221(2007)04-0028-07
修稿时间:2007-01-10

A DST-AHP Method for the Multi-attribute Decision Making Based on Decision Matrix
GONG Ben-gang,JIANG Wen-qi,ZHANG Dao-wu.A DST-AHP Method for the Multi-attribute Decision Making Based on Decision Matrix[J].Operations Research and Management Science,2007,16(4):28-34.
Authors:GONG Ben-gang  JIANG Wen-qi  ZHANG Dao-wu
Institution:1. Department of Management Engineering, Anhui University of Technology and Science, Wuhu 241000, China;2. School of Economic and Management, Nanjing University of Science and Technology, Nanjing 210094, China, 3. Center of Science, Technology and Society, Tsinghua University, Beijing 100084, China
Abstract:To solve the problem of too many pairwise comparisons,too much consistency of comparisons,and incomplete matrices in AHP,a DST-AHP method for the multi-attribute decision making based on decision matrix is proposed.In this paper,by the trait values of decision matrix,we first construct the DST-AHP's hierarchical structure model and comparison matrixes so as to define the basic probability assignment function(BPA).Then the BPA values are integrated by Dempster's rules of combination so as to determine the alternative ranking.Moreover,we also compare the decision-making results of AHP and DST-AHP to illustrate its validity and wide applicability of this method.
Keywords:decision matrix  dempster-shafer theory(DST)  analytic hierarchy process(AHP)  multi-attribute decision making
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号