Applying K-harmonic means clustering to the part-machine classification problem

Cellular manufacturing system (CMS) is an application of group technology (GT) to the production environment. There are many advantages of CMS over traditional manufacturing systems like reduction in setup-time, throughput time, etc. The grouping of machine cells and their associated part families so as to minimize the cost of material handling is a major step in CMS and it is called as cell formation (CF) problem. Cell formation is important to the effective performance of manufacturing. In this paper, an attempt has been made to effectively apply the K-harmonic means clustering technique to form machine cells and part families simultaneously, which we call K-harmonic means cell formation (KHM-CF). A set of 20 test problems with various sizes drawn from the literature are used to test the performance of the proposed algorithm. Then, the results are compared with the optimal solution, and the efficacy of the proposed algorithms is discussed. The comparative study shows that the proposed KHM-CF algorithm improves the grouping efficacy for 70% of the test problems, and gives the same results for 30% of the test problems.