A Discrete Bacterial Chemotaxis Approach to the Design of Cellular Manufacturing Layouts

The design of cellular manufacturing layouts is a very important process, because an adequate placement of machines can reduce costs and waiting times, and ultimately improve the yield of the system. The design process includes two main optimization sub-problems. The first one is a clustering problem, the so-called cell formation, consisting in the definition of groups (the cells) of machines that produce sets of related product parts. The second step is a location-allocation problem, which has to be solved to define the relative position of the cells and of the machines inside each cell. Both problems offer significant challenges from a computational point of view. This paper presents a novel approach for the design of cellular manufacturing layouts through an optimization algorithm based on bacterial chemotaxis. The proposed approach solves simultaneously the two optimization sub-problems mentioned above by minimizing transport cost and maximizing clustering of cells, taking into account the sequencing of production steps, the volume of production and the batch sizes. The performance of the proposed algorithm was tested through benchmark problems, and the results were compared with a genetic algorithm and analytical solutions modeled in GAMS. In all cases our proposal achieves better performance than Genetic Algorithm in quality and time, and comparable results with exact analytical solutions.