Adaptive order allocation and cross-factory collaborative scheduling for distributed group manufacturing with heterogeneous resources

Against the backdrop of globalised production, production scheduling in distributed group manufacturing has garnered significant attention. The uneven distribution of regional resources and the vicious competition among factories with heterogeneous resource usually affect the overall production efficiency of the manufacturing group. To address this issue, this paper investigates a novel collaborative scheduling problem for distributed group manufacturing, focusing on the optimisation of cross-factory collaborative scheduling and adaptive order allocation. A mathematical model is developed to simultaneously optimise three objectives: maximising the profit of the least profitable factory and minimizing total order tardiness as well as transportation costs. And a reinforcement learning enriched particle swarm optimisation (RPSO) algorithm is proposed to solve it. In RPSO, a high-quality initial population generation method based on three heuristics is designed to improve the quality of initial solutions. Then, a reinforcement learning-based order allocation strategy considering geographical constraints is constructed to assign orders to the most appropriate production factories. Furthermore, a hybrid update approach based on crossover operators is integrated to enhance solution search efficiency. Comparative experiments with state-of-the-art algorithms are conducted to validate the efficacy and advantages of the proposed RPSO approach. Finally, a case study is used to provide solutions for managers.