Knowledge-based multi-objective evolutionary algorithm for energy-efficient flexible job shop scheduling with mobile robot transportation

Energy-efficient production is a core requirement for manufacturing companies to transform and upgrade. The popularity of mobile robots (MRs) has led to a closer relationship between production and logistics in the workshop. However, the existing energy-efficient scheduling approaches ignore the energy consumption induced by logistics equipment. For this purpose, this paper investigates the energy-efficient flexible job shop scheduling problem with mobile robots (EFJSPMR). A knowledge-based multi-objective evolutionary algorithm (KBMOEA) is proposed for simultaneously optimizing the Makespan and total energy consumption (TEC), including the energy consumption of MRs. The proposed KBMOEA includes domain knowledge in two main aspects: i, a new active decoding method is introduced to improve the search efficiency while expanding the search space of the algorithm. ii, four problem properties are proposed passively incorporated into a local search strategy to enhance the performance of the algorithm. Additionally, an adaptive mechanism is constructed to balance the exploration and the exploitation by adjusting selection and mutation probabilities. Finally, three groups of the benchmarks (a total of 44 instances) are used for experiments to verify the effectiveness of the proposed algorithm. Comparison results with the other four state-of-the-art algorithms show that the proposed KBMOEA can obtain a Pareto front with better convergence and diversity on all instances.