计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (27): 25-28.DOI: 10.3778/j.issn.1002-8331.2010.27.006

• 博士论坛 • 上一篇    下一篇

采用分布估计算法计算AHP判断矩阵排序权重

张建华1,2,3,曾建潮2   

  1. 1.兰州理工大学 电气工程与信息工程学院,兰州 730050
    2.太原科技大学 复杂系统与智能计算实验室,太原 030024
    3.中北大学 电子与计算机科学技术学院,太原 030051
  • 收稿日期:2010-05-20 修回日期:2010-08-14 出版日期:2010-09-21 发布日期:2010-09-21
  • 通讯作者: 张建华

Rank weights computing of AHP judgement matrix using estimation of distribution algorithm

ZHANG Jian-hua1,2,3,ZENG Jian-chao2   

  1. 1.College of Electrical and Information Engineering,Lanzhou University of Technology,Lanzhou 730050,China
    2.Complex System and Computational Intelligence Laboratory,Taiyuan University of Science and Technology,Taiyuan 030024,China
    3.School of Electronics and Computer Science and Technology,North University of China,Taiyuan 030051,China
  • Received:2010-05-20 Revised:2010-08-14 Online:2010-09-21 Published:2010-09-21
  • Contact: ZHANG Jian-hua

摘要: 判断矩阵排序权重计算及其一致性检验可以归结为一个使一致性指标最小化的非线性、变量耦合的优化问题。提出一个利用基于序贯重点采样粒子滤波和Cholesky分解的分布估计算法计算排序权重的算法,该算法采用的概率模型是多峰的并考虑了变量之间相关性。文中对判断矩阵排序权重计算及一致性检验、基于序贯重点采样粒子滤波和Cholesky分解的分布估计算法进行了介绍,描述了排序权重计算及一致性检验算法,最后给出实验数据及其分析。实验结果验证了算法的有效性,并具有很高的精度和稳定性。

关键词: 层次分析法, 判断矩阵, 序贯重点采样, 粒子滤波, Cholesky分解, 分布估计算法

Abstract: The rank weights computing and the consistency check of judgement matrix can be attributed to a nonlinear,coupled variable optimization problem to minimum the consistency index.An algorithm for computing rank weights using estimation of distribution algorithm based on sequential importance sampling particle filter and Cholesky decomposition(PFEDA2) is presented in this paper.The probability model of this algorithm is multi-peak and the correlation between variables is considered.The method of rank weights computing,the consistency check and PFEDA2 is introduced,and then the algorithms of rank weights computing and consistency check are described.Finally,the experimental data and the analysis is given.Experimental results verify the effectiveness of the algorithm,and show this algorithm is high accuracy and stability.

Key words: Analytic Hierarchy Process(AHP), judgement matrix, sequential importance sampling, particle filters, Cholesky decomposition, estimation of distribution algorithm

中图分类号: