An Immune Genetic algorithm for inter-cell layout problem in cellular manufacturing system

Abstract

The objective function of inter-cell layout problem minimizes the total inter-cellular material handling cost. It is mostly significant with moderate production quantity in cellular manufacturing systems (CMS). This problem is classified as quadratic assignment problem (QAP) which is NP-Hard in nature. Heuristic techniques are extremely effective for such problems. In this paper we proposed a novel Immune Genetic algorithm (Immune-GA-RS) to obtain competent inter-cell layout in the vicinity of CMS. It exploits an elitist replacement strategy in order to improve the rate of convergence. The proposed method is successfully tested upon 8 datasets which are being widely used for inter-cell layout design problems. Proposed Immune-GA-RS is compared with two variants of the Genetic Algorithms, GA-RS and Alt-GA-RS. It is further compared with other published layout design techniques. Immune-GA-RS is shown to acquire 11.11 % improved solutions with 7.72 % reduced CPU time on an average. Further Immune-GA-RS is tested on 36 structured QAP instances available through QAPLIB and shown to outperform other two GA variants while attaining optimal solutions for 33 instances. It is also shown to outpace other published algorithms while attaining smaller solution gap for 11 test instances and obtains at least equal or better quality solutions for 24 instances. We conclude our work with a statistical data test to signify the results.