Integrating flexible preventive maintenance activities into two-stage assembly flow shop scheduling with multiple assembly machines

• A MILP model with maintenance level and assembly machine assignment is developed. • Two heuristics with the latest PM strategy are designed. • A new algorithm with problem-specific evaluation and two local search is proposed. • The proposed algorithm outperforms 9 state-of-the-art methods in 720 instances. In this paper, preventive maintenance (PM) activities are incorporated into two-stage assembly flow shop scheduling where m 1 dedicated machines in fabrication stage and m 2 machines in assembly stage. Each machine is given a new feature maintenance level, whose initial value is determined based on the Weibull probability distribution. To ensure the machines’ reliability and production continuity, we need to find a fit product sequence along with PM execution time points. Hence this paper tries to tackle this new integration problem by a mixed integer linear programming model, two heuristics MCMT PM and NEH PM , and a PM-based iterated greedy algorithm (IG PM ). IG PM is embedded with a problem-specific solution evaluation and two types of local search methods. The final experimental results show that compared with the other 9 state-of-the-art methods, the proposed IG PM embedded with NEH PM and reference local search generates the best results in all benchmark instances.