An adaptable problem-space-based search method for flexible flow line scheduling

Problem-space-based neighbourhoods have been recently suggested in the literature for the approximate solution of scheduling problems. This paper explores how effectively these neighbourhoods can be adapted to different regular measures of performance in the context of flexible flow line scheduling. Specifically, makespan and mean tardiness are used in the experiments. Near-optimal solutions and significant improvements in the performance of single-pass heuristics are found when searching these spaces with simple local search techniques for industrial and randomly generated problems.