利用粒子群算法计算AHP判断矩阵排序权重  

A Computing Method of AHP Judgement Matrix Rank Weights by Particle Swarm Optimization

作　　者：张建华[1] 李富萍[2] 

机构地区：[1]中北大学电子与计算机科学技术学院,太原030051 [2]太原科技大学计算机学院,太原030024

出　　处：《太原科技大学学报》2012年第2期149-154,共6页Journal of Taiyuan University of Science and Technology

基　　金：太原科技大学青年基金(20093014)

摘　　要：根据判断矩阵的基本性质,可以将判断矩阵的排序权重计算归结为一个最小化一致性指标的最小优化问题。针对这个最优化问题,提出一种利用改进的粒子群算法计算排序权重的算法。首先对判断矩阵排序权重计算及一致性检验、改进的粒子群算法进行了介绍,然后对排序权重计算进行了描述,最后给出仿真实验数据及其分析。实验结果验证了此算法的有效性,并显示该算法具有很高的精度和稳定性。The rank weights computation of judgment matrix can be attributed to a optimization problem with the minimum consistency index.An algorithm for computing rank weights by the Particle Swarm Optimization is presented in this paper.Firstly,the computing method of rank weights and the consistency check of judgment matrix were introduced.Then,the algorithm of rank weights computation based on the Particle Swarm Optimization was described.Finally,the experimental data and the analysis were given.Experimental results verified the effectiveness of the algorithm,and this algorithm was shown to have high accuracy and stability.

关 键 词：层次分析法 判断矩阵 一致性 权重排序 粒子群优化算法 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]