A Learning Metaheuristic Algorithm for a Scheduling Application

Tabu Search is among one of the metaheuristic algorithms that are widely recognized as efficient approaches to solve many combinatorial problems. Studies to improve the performance of metaheuristics have increasingly relied on the use of various methods, either combining different metaheuristics or originating outside of the metaheuristic field. This paper presents a learning algorithm to improve the performance of tabu search by reducing its search space and the evaluation effort. We study its performance using classification methods in an attempt to select moves through the search space more intelligently. The experimental results demonstrate the benefit of using a learning mechanism under deterministic environment and with uncertainty conditions.