Past research in part family identification has focused mainly on the development of efficient procedures for manufacturing-oriented part family formation in which similarities among parts are established primarily on machine or operation requirements. While these part families are essential in cellular manufacturing, they are not welt suited for other areas of production, in particular, part design and process planning. A new part family identification technique using a simple genetic algorithm is proposed in this paper to first determine a set of part family differentiating attributes, and second to use these attributes to guide the formation of part families. The technique is implemented in C using a SUN SPARC workstation 1+. Empirical analyses of the technique on both artificially generated data and a real application are performed and discussed.
|Number of pages||18|
|Journal||International Journal of Production Research|
|State||Published - Mar 10 2000|
ASJC Scopus subject areas
- Strategy and Management
- Management Science and Operations Research
- Industrial and Manufacturing Engineering