Modified particle swarm optimization for solving machine-loading problems in flexible manufacturing systems

In flexible manufacturing systems (FMSs), the loading problem is considered as a vital pre-release decision because its operational effectiveness largely depends on a good quality solution to the loading problem. Difficulties arise in obtaining optimal solutions to such problems because of its combinatorial and NP-hard nature. In the past, numerous techniques have been suggested and found to be efficient, but they take long computational times when the problem size increases. In order to address the above issues, a meta-heuristic approach based on particle swarm optimization (PSO) has been proposed in this paper to improve the solution quality and reduce the computational effort. However, PSO has the tendency to suffer from premature convergence. Therefore, the PSO algorithm has been modified through the introduction of a mutation operator to improve efficiency of the algorithm. The proposed algorithm attempts to minimize the system unbalance while satisfying the technological constraints, such as the availability of machining time and tool slots. The proposed algorithm produces promising results in comparison to existing methods for ten benchmark instances available in the FMS literature.