Abstract:

In recent years， distributed production scheduling has received significant attention from both researchers and practitioners. However， existing studies often neglect two crucial aspects： （1） job transfer processes among machines within distributed systems and （2） the distribution of finished jobs to customers. To address these gaps， an integrated scheduling framework is proposed that combines distributed flexible job shops with logistics distribution， explicitly incorporating automated guided vehicle （AGV） transportation operations. First， a mixed-integer programming （MIP） model is formulated to minimize two objectives： the makespan and total tardiness. Second， a learning-driven multi-objective artificial bee colony （ABC） algorithm is developed to efficiently solve the proposed model， leveraging problem-specific heuristics to enhance the search performance. Finally， the effectiveness of the proposed approach is validated through extensive experiments on benchmark instances and compared against two state-of-the-art metaheuristics. The results demonstrate that the proposed model and algorithm achieve superior performance in both solution quality and computational efficiency.

Key words: distributed flexible job shop scheduling, integrated scheduling of production and distribution, automated guided vehicle transportation, artificial bee colony algorithm, Q-learning method