A hybrid clonal algorithm for the cell formation problem with variant number of cells

Cellular manufacturing is an important application of the Group Technology that has been used in several real-world applications such as the electronics industry, offices, structural fabrication, service industries, and hospitals. The manufacturing cell formation problem is considered the first issue faced in designing cellular manufacturing systems in order to overcome difficulties related to multi-product and batch-production systems. The aim is to minimize the inter-cell movements of the parts and maximize the use of the machines. In this paper, a new approach based on the clonal selection algorithm is proposed for solving the problem where the number of cells is not fixed a priori. The approach integrates a local search mechanism to intensify the search of the solutions. To evaluate the effectiveness of the proposed algorithm, a set of 40 benchmark problems is used; the results are then compared to other methods recently developed. The results show that the proposed algorithm performs very well on all test problems since it can reach the best-known solution of 39 benchmark problems (97.5%).