Hybrid tabu search for fuzzy job shop

Abstract

We consider the fuzzy job shop scheduling problem, which is a variant of the well-known job shop problem, with uncertainty in task durations that we model using fuzzy numbers. We propose a tabu search algorithm for minimising the expected makespan based on reversing arcs within critical blocks. We test the algorithm and then combine it with a genetic algorithm from the literature so we can observe the synergy effect, obtaining better results with the hybrid algorithm than with its components by separate. Finally we compare our hybrid algorithm with a memetic algorithm from the literature and show that even in similar times, our method is better in terms of expected makespan