Modeling and optimizing flexible manufacturing systems

In this research, the loading and scheduling problems for a random flexible manufacturing system (FMS) are studied. A hierarchical approach to the solution of these problems is first considered. Heuristic loading algorithms are developed to improve the workload balance and the number of late jobs obtained by existing loading procedures. The algorithms provide significantly better loading results than the existing procedures in workload balance as well as the number of late jobs.; The loading results obtained from the loading algorithms are then used as input data for the scheduling problem. Several dispatching rules are applied to solve the scheduling problem. The scheduling results obtained show that the dispatching rules applied have a determining effect on the scheduling solution and that the loading results have a lesser effect on the scheduling solution than previously conjectured in the literature.; A concurrent approach is then considered for solving the loading and scheduling problems and concurrent heuristic procedures are developed. The performance of the concurrent heuristic procedures is evaluated and compared to the performance of existing hierarchical procedures for different loading and scheduling criteria using randomly generated data as well as data obtained from an industrial site. The loading and scheduling results obtained by the concurrent procedures developed here show a significant improvement over those provided by existing hierarchical procedures.