Abstract: In practical production, due to the complex assembly process and diverse configuration requirements, the final assembly pulling production driven by orders is often adopted in the delivery of large and complex products, which are processed in distributed job shops first and assembled into final products in an assembly factory. Additionally, with the increasing global attention to sustainable development, green and energy-saving production has become the focus of enterprises and academia. Therefore, the distributed assembly job-shop scheduling problem with variable machine speed is investigated. To minimize the maximum completion time and the total energy consumption, a non-dominated sorting genetic algorithm II with variable neighborhood search is proposed. In the local search period of the algorithm, several machine speed adjustment strategies are designed to reduce maximum completion time and total energy consumption respectively, and the local search strategies based on critical path is used to optimize the operation sequencing problem. A large number of experiments are carried out on four groups of instances with different scales to verify the effectiveness of the proposed algorithm, which demonstrate that the adopted framework and the designed local search strategies effectively improve the scheduling scheme.

Key words: assembly job shop, distributed, multi-objective, energy consumption, variable neighborhood search