首页 | 本学科首页   官方微博 | 高级检索  
     

基于质量屋矩阵分解的设计问题简化方法的研究
引用本文:孙挪刚,张优云,梅雪松,陶涛.基于质量屋矩阵分解的设计问题简化方法的研究[J].计算机辅助设计与图形学学报,2004,16(3):319-323,330.
作者姓名:孙挪刚  张优云  梅雪松  陶涛
作者单位:西安交通大学机械工程学院,西安 710049
基金项目:国家自然科学基金 ( 5 9990 472 )资助
摘    要:针对质量屋(HOQ)的用户需求与产品工程性能关系矩阵以及产品工程性能的互相关矩阵的不同特点,提出一种对HOQ矩阵进行分解的方法。首先根据工程性能的互相关矩阵采用模糊聚类的方法,在给定某一阈值的情况下得到对工程性能的一种分类,并由此确定最后分解的模块数,在此基础上,基于惟一且满意度最大的原则,即每一个用户需求仅属于某一且惟一一个模块。并使分解得到的各模块中所包含关系系数之和为最大,将对用户需求与工程性能的关系矩阵的分解转化为一个带约束的整型规划问题。从而将基于HOQ的复杂的设计问题简化为一个个小模块的设计。

关 键 词:产品设计  质量屋  矩阵分解  模糊评判

Reduction of Complex Design Problem Based on Decomposition of HOQ Matrix
Sun Nuogang,Zhang Youyun,Mei Xuesong,Tao Tao.Reduction of Complex Design Problem Based on Decomposition of HOQ Matrix[J].Journal of Computer-Aided Design & Computer Graphics,2004,16(3):319-323,330.
Authors:Sun Nuogang  Zhang Youyun  Mei Xuesong  Tao Tao
Abstract:Considering the different characteristics of the relation matrix of customer requirements(CR) and engineering performances(EP) and the correlation matrix of engineering performances, a decomposition approach for the house of quality(HOQ) matrix is proposed First, on the basis of the correlation matrix of engineering performances, the fuzzy cluster method is applied to group the engineering performances under certain threshold and the number of the final modules is then determined Based on above and conforming to the unique criterion and maximization criterion of the satisfaction which mean that every customer requirement should be included in one and only one of the modules and the sum of the relation coefficients included in the modules should be maximum, the decomposition of the relation matrix of the customer requirements and the engineering performances is converted into an integer optimization question subject to certain constraints, thus complex design problem based on HOQ matrix is decomposed into the design of several simple modules
Keywords:house of quality(HOQ)  relation matrix  correlation matrix  decomposition  fuzzy cluster
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