A group genetic algorithm for the machine cell formation problem

The facility layout is a fundamental element to achieve the company's manufacturing strategy. To obtain a good layout design, appropriate computational algorithms as an aid to the layout designer in the decision making process during the factory layout design should be applied. This work presents a new genetic algorithm (GA) for the cell layout design. It is based on group encoding instead of simple machine encoding, generally used by most of the GA implementations encountered in the literature. Moreover, new crossover and mutation operators are proposed. Tests using published data show that the proposed algorithm can find the group structure present in the data sets.