Scheduling efficiency on correlated parallel machine scheduling problems

Abstract

This research examines correlated parallel machine scheduling problems since they better reflect real world manufacturing environments. We consider the problem of scheduling correlated parallel machines to minimize makespan and scheduling correlated parallel machines with release times to minimize total weighted tardiness. We consider different levels and combination of machine correlations and job correlations in the processing times. Mathematical models are applied to evaluate the influence of machine correlation and job correlation on computation results and computation time. Computational results show that as the machine and job correlations increase, the problem instances become more difficult for mathematical models to solve. This implies that branch-and-bound based algorithms might have more difficulty solving parallel machine scheduling problems with correlations than without correlations. We can use this result to forecast branch-and-bound based algorithm performance on correlated parallel machine scheduling problems and thus improve process behavior by selecting the most suitable algorithms.