A genetic algorithm with proper parameters for manufacturing cell formation problems


JOURNAL OF INTELLIGENT MANUFACTURING, vol.23, no.4, pp.1047-1061, 2012 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 23 Issue: 4
  • Publication Date: 2012
  • Doi Number: 10.1007/s10845-010-0446-8
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.1047-1061
  • Keywords: Cell formation problem, Genetic algorithms, Grouping efficacy, Design of experiments, GROUP-TECHNOLOGY, CLUSTERING-ALGORITHM, PART-FAMILIES, TABU SEARCH, NEURAL-NETWORK, MACHINE-CELLS, MODEL, FORMULATION, EXTENSION, DESIGN
  • Eskisehir Osmangazi University Affiliated: Yes


One fundamental problem in cellular manufacturing is the formation of product families and machine cells. Many solution methods have been developed for the cell formation problem. Since efficient grouping is the prerequisite of a successful Cellular Manufacturing installation the research in this area will likely be continued. In this paper, we consider the problem of cell formation in cellular manufacturing systems with the objective of maximizing the grouping efficacy. We propose a Genetic Algorithm (GA) to obtain machine-cells and part-families. Developed GA has three different selection and crossover operators. The proper operators and parameters of the GA were determined by design of experiments. A set of 15 test problems with various sizes drawn from the literature is used to test the performance of the proposed algorithm. The corresponding results are compared to several well-known algorithms published. The comparative study shows that the proposed GA improves the grouping efficacy for 40% of the test problems.