Semi-robust layout design for cellular manufacturing in a dynamic environment

Abstract

In this article, the problems of cell formation (CF) and cellular layout (CL) encountered in the design of a cellular manufacturing system are studied. A semi-robust cellular approach is proposed, which is able to cope with the continuous change of product mix and demand. At the heart of the proposed robust approach, the facility layout is not being changed from one period to the other, but rather the positions of the pick-up/drop-off points of cells. The developed model and solution algorithms can concurrently make the decisions regarding the optimum number of cells, CF and CL (both inter-and intra-cellular layout of unequal-area facilities). The problem is formulated as a multi-objective mathematical programming model. A modified non-dominated sorting genetic algorithm (MNSGA-II) is then used to obtain Pareto-optimal solutions. In the proposed MNSGA-II, an improved non-dominated sorting strategy and a modified dynamic crowding distance procedure are implemented. The effectiveness of the MNSGA-II is evaluated against two well-known multi-objective optimization algorithms, namely multi-objective particle swarm optimization and non-dominated ranking genetic algorithm. To this aim, several numerical examples and computational experiments are carried out; five metrics are employed to evaluate the quality of the developed algorithms. The results demonstrate the efficiency of the proposed methodology.