基于案例的推理
计算机科学
产品设计
特里兹
重新使用
概念设计
新产品开发
高级设计
产品(数学)
数据挖掘
情报检索
软件工程
系统工程
人工智能
人机交互
工程类
数学
几何学
业务
营销
废物管理
作者
Song Li,Yan Li,Wenqiang Li,Chen Chen
标识
DOI:10.1177/0954406219854901
摘要
During product design process, conventional case-based reasoning (CBR) has shown significant applications in coping with new problems by recalling and reusing solutions in old context. However, conventional CBR still lacks effective methods to assist designers in generating creative design solutions of identified problems during modification stage. To make up for this deficiency, an extended CBR (ECBR) method characterized by two retrieval stages is proposed in this paper. The former retrieval stage is aimed to retrieve the most similar source case for design reusing, and the latter one is designed to assist designers in searching for relevant function units to generate creative design ideas for final design solution. A five-tuple knowledge model of product case is presented. It includes both specific design information for design reusing and abstract function units for generating creative design ideas. To effectively organize these case information, a hierarchy knowledge model of case base is established. It consists of case type layer, product case layer, and function unit layer. A corresponding retrieval strategy is presented to deal with different retrieval goals. To improve the performance of ECBR method in modification stage, the substance-attribute-function-cause analysis model and TRIZ tools are integrated together to aid problem analysis and problem-solving of retrieved product case. Finally, a new product conceptual design process is proposed based on ECBR method and the conceptual design of a smokeless barbecue equipment is employed as an example to verify the feasibility and effectiveness of ECBR method.
科研通智能强力驱动
Strongly Powered by AbleSci AI