A Genetic Cluster Algorithm For The Machine-component Grouping Problem

Abstract

This research presents the usage of a genetic algorithm for the clustering of parts and machines. A detailed analysis is shown comparing GCA results with single link cluster analysis, rank order clustering, and the direct clustering algorithm. GCA was also compared with several additional cell formation heuristics described in the recent literature, including GRAPHICS, MODROC, and a cost-based heuristic. Results showed that the GCA was far superior over single link cluster analysis and provided equivalent results to those of the direct clustering algorithm and rank order clustering. GCA was also found to provide superior results to the other heuristics. The discussion explains these findings by illustrating the inflexibility of traditional cell formation heuristics in the selection of final machine-component groupings. © 1996 Chapman & Hall.

Department(s)

Mechanical and Aerospace Engineering

Keywords and Phrases

Cell formation; Cellular manufacturing; Genetic algorithms; Group technology

International Standard Serial Number (ISSN)

0956-5515

Document Type

Article - Journal

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2023 Springer, All rights reserved.

Publication Date

01 Jan 1996

Link to Full Text

Share

 
COinS