Machines are the major component of the cellular manufacturing systems (CMS). Usually, it is difficult to handle machine breakdowns as quickly as the production requirement dictates and therefore, the reliability consideration plays an important role in the overall performance of the CMS. We present a mathematical model of the cell formation problem with alternative process routings (APR) and machine reliability consideration. The proposed model tries to simultaneously minimize the intercellular movement costs and to maximize the reliability of the manufacturing system. In addition, we develop three sets of metaheuristics, namely simulated annealing, genetic algorithm and memetic algorithm to solve the proposed model. Using some numerical examples, we compare the performance of the proposed algorithms with an optimum algorithm, namely the branch and bound algorithm. The results show that in comparison with the branch and bound algorithm, the proposed metaheuristics can obtain better objective function values in less computational time.
