An effective tabu search for job shop scheduling with parallel machines

In this work we consider a job shop scheduling problem with the objective of minimizing the makespan. As in the standard job shop problem, the operations of each job have to be processed on different stages in a given order, but in the problem variant studied here each stage disposes of parallel machines. The number of parallel machines may vary from stage to stage, yet if all stages have a single machine we have the standard job shop problem. We propose a tabu search to solve this problem, which searches for good job-stage assignments, and combine it with an effective heuristic rule to derive a complete schedule from them.