Integrated optimization of assembly line balance and preventive maintenance based on improved multi-objective grey wolf algorithm

In the traditional assembly line operation management, the assembly line balance and the equipment maintenance on the station are completely separated, which resulting in huge production losses and great inconvenience in maintenance. Aiming at minimizing the cycle time and production adjustment in normal work and given preventive maintenance situations, an integrated optimization model of assembly line balancing and predictive maintenance was constructed to generate multiple allocation plans that could provide maintenance opportunities and promote the production continuity, and an improved grey wolf algorithm was proposed for this multi-objective optimization problem. Considering discrete features of the integration problem, a random key-based encoding and decoding method was employed. Facing with the multi-objective optimization characteristics, a wolf class classification method and update mechanism were designed, which integrated Pareto hierarchy and congestion distance, so as to preserve and better explore the near-optimal solution. A crossover operator was hired to improve the diversification of grey wolf optimization algorithm. Experimental results showed that the improved grey wolf algorithm had better convergence and diversity, and the obtained non-dominated solutions approach was closer to the Pareto optimal frontier. The effective integration of balancing and maintenance could significantly reduce the production costs. © 2020, Editorial Department of CIMS. All right reserved.