A Customized IMOEA/D for Bi-Objective Single-Machine Scheduling with Adaptive Preventive Maintenance

The prolonged operation of machines in the production process can lead to continuous deterioration or even failure, and the necessary maintenance measures can alleviate the above negative effects. For this reason, this study investigates a joint optimization problem of single-machine production and preventive maintenance (PM) considering linear deterioration effects. The objective is to obtain an integrated sequence of degrading jobs and PM activities in order to simultaneously minimize the makespan and the total cost. Based on the problem characteristics, an adaptive PM strategy is first designed. To efficiently solve the problem, an improved multi-objective evolutionary algorithm based on decomposition (IMOEA/D) is tailored, where the biased-distribution weight vector is proposed to enhance the search capability at both ends of the Pareto front. Five instances are used to evaluate the performance of the customized IMOEA/D and two classical multi-objective evolutionary algorithms. Numerical studies show that the IMOEA/D can substantially improve the hypervolume metric, the maximum spread metric, and the distributivity of the Pareto front at a slight sacrifice of the spacing metric.