Solving the job shop scheduling problem with operators by depth-first heuristic search enhanced with global pruning rules

Abstract

The job shop scheduling problem with an additional resource type has been recently proposed to model the situation where each operation in a job shop has to be assisted by one of a limited set of human operators. We confront this problem with the objective of minimizing the total flow time, which makes the problem more interesting from a practical point of view and harder to solve than the version with makespan minimization. To solve this problem we propose an enhanced dept-first search algorithm. This algorithm exploits a schedule generation schema termed OG&T, two admissible heuristics and some powerful pruning rules. In order to diversify the search, we also consider a variant of this algorithm with restarts. We have conducted an experimental study across several benchmarks. The results of this study show that the global pruning rules are really effective and that the proposed algorithms are quite competent for solving this problem.