A performance-enhancement algorithm for the part-machine grouping problem

Abstract

This paper introduces a novel meta-heuristic algorithm called a “performance-enhancement algorithm”, to solve the part-machine grouping problem. The process of applying the algorithm begins with allocating the parts and machines randomly into different clusters and assigning each of them a uniform initial membership index to each cluster. Each of the parts and machines are then placed in different clusters in turn, and the fitness function is recorded in order to adjust the membership index. Finally, each part and machine is reallocated to the cluster with the highest membership index to get the algorithm ready for next generation. To evaluate the solution quality of this type of grouping scheme, four indexes (total bond energy, exceptional elements, machine utilization, and grouping efficacy) are used as the performance measure of the proposed algorithm. The algorithm has been applied to solve several well-cited problems, and the computational results show that the novel algorithm is effective in finding the best solution and revolution ability.