• CN:11-2187/TH
  • ISSN:0577-6686

›› 2010, Vol. 46 ›› Issue (19): 108-116.

• 论文 • 上一篇    下一篇

基于模糊多属性决策的复杂产品配置方法

任彬;张树有;伊国栋   

  1. 浙江大学流体传动及控制国家重点实验室
  • 发布日期:2010-10-05

Configuration Design of Complex Products Based on Fuzzy Multi-attribute Decision-making

REN Bin;ZHANG Shuyou;YI Guodong   

  1. State Key Laboratory of Fluid Power Transmission and Control, Zhejiang University
  • Published:2010-10-05

摘要: 目前产品配置设计主要是标准模块基础上的功能实现,针对模糊型复杂产品的配置问题,将模糊多属性决策理论引入到配置设计方法中。考虑客户主观不确定偏好,通过模糊需求的知识获取以及模糊需求与产品属性间的描述,实现客户模糊需求到配置需求的非线性映射。将配置需求作为复杂产品模糊层次配置的输入,通过非标准模块的相似性标识和模糊层次配置的求解,实现复杂产品模糊层次配置设计,模糊层次配置的输出为配置方案集。利用神经网络技术建立模糊多属性决策的配置方案评价模型,对配置方案集进行比较和排序,优选出最接近客户满意的配置结果。结合注塑机产品的配置设计,分析客户需求的模糊、不足和冗余问题,在第一层精确配置和第二层模糊相似配置的基础上,获得四个配置方案集,通过模糊神经网络的多属性决策模型,求解最优方案,并开发了注塑机配置设计系统。

关键词: 复杂产品, 模糊多属性决策, 配置需求, 神经网络技术, 7075铝合金, X射线衍射, 电导率, 喷射成形, 强度, 时效

Abstract: Currently, the product configuration design is mainly functional realization on the basis of standard modules. For the configuration problem of complex products with fuzzy demands, the fuzzy multi-attribute decision-making method is introduced into the configuration design. Considering customers subjective uncertain preferences, the non-linear mapping from customer fuzzy demands to configuration demands is achieved by the knowledge acquisition of fuzzy requirements and the description between fuzzy requirements and product attributes. The configuration demands are input to the fuzzy-level configuration design, and the program collections of configuration design are output. The fuzzy-level configuration design is supported by the similarity identification of non-standard modules and the fuzzy-level solution. Finally, the evaluation model of configuration design of fuzzy multi-attribute decision-making is established with neural network method. The program collections of configuration design are compared and sorted out. Therefore, the optimum results with closest customer satisfaction are obtained. In practice, the proposed method is applied to the configuration design of injection molding machine. The configuration design system of injection molding machine is developed based on the acquisition of customer requirements and the realization of the fuzzy-level configuration design.

Key words: Configuration design, Fussy multiple-attribute decision, Neural network technology, Complex product, 7075 alloy, Conductivity, Retrogression and re-aging, Spray forming, Strength, XRD

中图分类号: