An energy-saving real-time scheduling method based on bi-level multi-agent architecture with bargaining game for flexible job shops

Under the interference of random disturbances, frequent machine switching and excessive idle times lead to significant energy wastage in manufacturing processes. The job shop scheduling for reasonable machine selection and machine processing sequence adjustment is an effective way to reduce energy consumption. Therefore, this paper proposes a real-time energy-saving scheduling method for effective managing a shopfloor with multi-service units through a multi agents bargaining game between the shopfloor level and service unit level (bi-levels). First, we establish a real-time energy-saving scheduling model for flexible job shops with optimization objectives of makespan, idle rate and total production energy consumption. Then, a hierarchical collaborative energy-saving scheduling strategy based on this model is designed. Uniquely, this model and strategy facilitate reduced energy usage by allocating operations of all jobs at the shop level and adjusting machine processing sequences at the service unit level. Upon disturbance, the method prioritizes adjustments within the service unit to enhance the optimization efficacy and robustness of the energy-saving schedule. Moreover, a bi-level multi-agents bargaining game method is introduced to enable better optimization outcomes in this distributed system. This method allows for competition among agents at the same level and cooperation between agents at different levels. Finally, the efficiency and advancedness of the method is illustrated by examples and comparisons with existing studies.