The shifting bottleneck procedure for job shop scheduling
Management Science
An Efficient Optimal Solution to the Two-Hoist No-Wait Cyclic Scheduling Problem
Operations Research
| Hi-index | 0.00 |
In this paper, we aim at solving flexible job shop scheduling problems with transportation constraints and bounded processing times. We propose a hybrid method of genetic algorithm, tabu local search and a modified shifting bottleneck procedure. The genetic algorithm is used to generate and evolve assignment for each task (processing tasks and transportation tasks). The modified shifting bottleneck procedure is used to generate initial solutions and regenerate solutions when no improvement occurs during some generations. The tabu local search is then used to improve initial solutions during a limited number of iterations. To evaluate solutions, we elaborate a modified disjunctive graph which contains not only processing nodes, but also transportation and storage nodes. There are positive and negative arcs for bounded processing times, transportation times and minimum and maximum allowed storage time before and after each processing task. Our objective is to minimize makespan. Various types of instances with fixed or bounded processing times are tested. Computational results show that this hybrid method is able to solve efficiently these kinds of problems.