Neuro-fuzzy comprehensive assemblability and assembly sequence evaluation

• Published in: AI EDAM Vol. 15; no. 5; pp. 367 - 384
• Main Author: ZHA, X.F.
• Format: Journal Article
• Language: English
• Published: Cambridge Cambridge University Press 01.11.2001
• Subjects: Assemblability; Design for assembly; Integration; Intelligent CAD; Neuro-Fuzzy Approach; Integration; Neuro-Fuzzy Approach; Assemblability; Intelligent CAD; Design for assembly
• ISSN: 0890-0604; 1469-1760
• DOI: 10.1017/S0890060401155010

Analysis of assembly properties of a product is needed during the initial design stage in order to identify potential assembly problems, which affect product performance in the later stages of life cycle. Assemblability analysis and evaluation play a key role in assembly design, assembly operation analysis and assembly planning. This paper develops a novel approach to assemblability and assembly sequence analysis and evaluation using the concept of the fuzzy set theory and neuro-fuzzy integration. Assemblability is described by assembly-operation difficulty, which can be represented by a fuzzy number between 0 and 1. Assemblability evaluation is therefore fuzzy evaluation of assembly difficulty. The evaluation structure covers not only the assembly parts' geometric and physical characteristics, but also takes into account the assembly operation data necessary to assemble the parts. The weight of each assemblability factor is subject to change to match the real assembly environments based on expert advice. The approach has the flexibility to be used in various assembly methods and different environments. It can be used in a knowledge-based design for assembly expert system with learning ability. Through integration with the CAD system, the developed system can effectively incorporate the concurrent engineering knowledge into the preliminary design process so as to provide users with suggestions for improving a design and also helping to obtain better design ideas. The applications in assembly design and planning show that the proposed approach and system are feasible.