考虑设备可用性的航天制造车间动态调度方法

Dynamic scheduling method for aerospace manufacturing workshops considering equipment availability

摘要

航天产品研制具有多品种、变批量及研制与批产共线生产的特点,多任务、高负载的设备成为制约生产效率的关键因素。加工设备潜在的故障扰动严重影响车间生产效率。因此,本文提出了考虑设备可用性的航天制造车间动态调度方法。该方法借助统计预测开展设备可用性分析,获取相邻两次故障之间的平均工作时间等设备可用性信息,实现设备可用性约束的量化分析;同时,建立工序任务指派到设备、设备上工序任务排序的调度规则库,设计6种规则组合的调度动作,并结合深度强化学习的D3QN(deep double dueling Q-networks)算法,输出以最小化工件总加权完工时间为优化目标的车间动态调度方案。为验证所提动态调度方法的性能,将其与常见的组合式规则方法、DQN(deep Q-networks)、遗传算法进行对比试验,试验结果表明所提算法在五个算例上均能取得最好效果。

Abstract

Aerospace product development is characterized by multiple varieties, variable batch sizes, and concurrent development and mass production on the same production line. This makes multi-task, high-load equipment a key constraint on production efficiency, while potential equipment failures significantly disrupt workshop productivity. To address these challenges, this paper proposes a dynamic scheduling method for aerospace manufacturing workshops considering equipment availability. The method uses statistical prediction-based analysis to quantify availability constraints, deriving metrics such as the mean operating time between consecutive failures. Meanwhile, it establishes a scheduling rule library for assigning process tasks to equipment and sequencing operations on each machine. By designing scheduling actions through rule combinations and integrating the deep double dueling Q-networks (D3QN) algorithm from deep reinforcement learning, the method generates dynamic workshop schedules aimed at minimizing the total weighted completion time of workpieces. To evaluate the performance of the proposed dynamic scheduling method, comparative experiments were conducted with common combinatorial rule-based methods, deep Q-networks (DQN), and genetic algorithms. Experimental results demonstrate that the proposed algorithm achieves the best performance across all five test cases.

关键词

设备可用性 / 动态调度 / 设备故障 / 深度强化学习 / 组合规则

Key words

equipment availability / dynamic scheduling / equipment failure / deep reinforcement learning / combination rule

参考文献

[1] DING C, QIAO F, LIU J, et al. Semantic-driven real-time scheduling approach for aerospace product manufacturing[J]. Computers & Industrial Engineering, 2025, 210: 111525.

[2] LI Y, LIU Q, ZHANG C, et al. Graph-based dual-agent deep reinforcement learning for dynamic human-machine hybrid reconfiguration manufacturing scheduling[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2025, 55(11): 8729-8741.

[3] ROJAS-HERNÁNDEZ J J, SOLÍS-QUESADA A, GONZALEZ-LLORENTE J D. Development of an integrated system for screening and testing of electrochemical cells for cubesats[C]// 2025 IEEE 12th International Workshop on Metrology for AeroSpace (Metro-AeroSpace). IEEE, 2025: 12-17.

[4] 李想, 任晓羽, 周永兵, 等. 基于改进D3QN算法的随机工时下柔性综合调度问题研究[J]. 系统仿真学报, 2025, 37(2): 474-486.

[5] LI X, REN X Y, ZHOU Y B, et al. Research on flexible integrated scheduling under stochastic processing times based on improved D3QN algorithm[J]. Journal of System Simulation, 2025, 37(2): 474-486.

[6] ZHANG W, ZHAO F, FENG B, et al. A reinforcement learning control framework based on scalable graph transformer for large-scale fuzzy job shop scheduling problems[J]. IEEE Transactions on Neural Networks and Learning Systems, 2025, 36(9): 16521-16533.

[7] ROSHANAEI V, AZAB A, ELMARAGHY H. Mathematical modelling and a meta-heuristic for flexible job shop scheduling[J]. International Journal of Production Research, 2013, 51(20): 6247-6274.

[8] WANG S J, LIU M. A branch and bound algorithm for single-machine production scheduling integrated with preventive maintenance planning[J]. International Journal of Production Research, 2013, 51(3): 847-868.

[9] CHEN B C, MATIS T I. A flexible dispatching rule for minimizing tardiness in job shop scheduling[J]. International Journal of Production Economics, 2013, 141(1): 360-365.

[10] ZANDIEH M, KHATAMI A R, RAHMATI S H A. Flexible job shop scheduling under condition-based maintenance: Improved version of imperialist competitive algorithm[J]. Applied Soft Computing, 2017, 58: 449-464.

[11] 杜阳宇, 张枫. 考虑设备退化与预防维护的柔性作业车间调度[J]. 数字制造科学, 2018, 16(2): 133-138.

[12] DU Y Y, ZHANG F. Flexible job shop scheduling with degenerate and preventive maintenance[J]. Digital Manufacture Science, 2018, 16(2): 133-138.

[13] BOUAZZA W, SALLEZ Y, BELDJILALI B. A distributed approach solving partially flexible job-shop scheduling problem with a Q-learning effect[J]. IFAC-PapersOnLine, 2017, 50(1): 15890-15895.

[14] CHEN R H, YANG B, LI S, et al. A self-learning genetic algorithm based on reinforcement learning for flexible job-shop scheduling problem[J]. Computers & Industrial Engineering, 2020, 149: 106778.

[15] PARK J, CHUN J, KIM S H, et al. Learning to schedule job-shop problems: representation and policy learning using graph neural network and reinforcement learning[J]. International Journal of Production Research, 2021, 59(11): 3360-3377.

[16] 胡一凡, 张利平, 白雪, 等. 深度强化学习求解柔性装配作业车间调度问题[J]. 华中科技大学学报(自然科学版), 2023, 51(2): 153-160.

[17] HU Y F, ZHANG L P, BAI X, et al. Deep reinforcement learning for flexible assembly job shop scheduling problem[J]. Journal of Huazhong University of Science and Technology (Natural Science Edition), 2023, 51(2): 153-160.

[18] SHAIKAT F B, ISLAM R, HAPPY A T, et al. Optimization of production scheduling in smart manufacturing environments using machine learning algorithms[J]. Lett High Energy Phys, 2025, 12(1): 1-15.

[19] GUI L, LI X Y, ZHANG Q F, et al. Domain knowledge used in meta-heuristic algorithms for the job-shop scheduling problem: review and analysis[J]. Tsinghua Science and Technology, 2024, 29(5): 1368-1389.

[20] 王强, 李仁旺. 基于改进DQN算法的柔性作业车间调度问题[J]. 轻工机械, 2025, 43(3): 97-103.

[21] WANG Q, LI R W. Flexible job shop scheduling problem based on improved DQN algorithm[J]. Light Industry Machinery, 2025, 43(3): 97-103.

[22] 汪亚顺, 张书锋, 陈循. 机电设备可靠性验证试验统计方案设计方法研究[J]. 机械工程学报, 2021, 57(12): 247-257.

[23] WANG Y S, ZHANG S F, CHEN X. Study on the design of statistical schemes for reliability demonstration test of electromechanical equipment[J]. Journal of Mechanical Engineering, 2021, 57(12): 247-257.

[24] LIU P, XIA B Y, YANG Z W, et al. A deep reinforcement learning method for multi-stage equipment development planning in uncertain environments[J]. Journal of Systems Engineering and Electronics, 2022, 33(6): 1159-1175.

[25] 郭具涛, 吕佑龙, 戴铮, 等. 基于复合规则和强化学习的混流装配线调度方法[J]. 中国机械工程, 2023, 34(21): 2600-2606.

[26] GUO J T, LÜ Y L, DAI Z, et al. Compound rules and reinforcement learning based scheduling method for mixed model assembly lines[J]. China Mechanical Engineering, 2023, 34(21): 2600-2606.

[27] QIAO Y, WU N Q, HE Y F, et al. Adaptive genetic algorithm for two-stage hybrid flow-shop scheduling with sequence-independent setup time and no-interruption requirement[J]. Expert Systems with Applications, 2022, 208: 118068.

[28] 王霄汉, 张霖, 任磊, 等. 基于强化学习的车间调度问题研究简述[J]. 系统仿真学报, 2021, 33(12): 2782-2791.

[29] WANG X H, ZHANG L, REN L, et al. Brief review on applying reinforcement learning to job shop scheduling problems[J]. Journal of System Simulation, 2021, 33(12): 2782-2791.

[30] 刘建伟, 高峰, 罗雄麟. 基于值函数和策略梯度的深度强化学习综述[J]. 计算机学报, 2019, 42(6): 1406-1438.

[31] LIU J W, GAO F, LUO X L. Survey of deep reinforcement learning based on value function and policy gradient[J]. Chinese Journal of Computers, 2019, 42(6): 1406-1438.

[32] 王凌, 潘子肖. 基于深度强化学习与迭代贪婪的流水车间调度优化[J]. 控制与决策, 2021, 36(11): 2609-2617.

[33] WANG L, PAN Z X. Scheduling optimization for flow-shop based on deep reinforcement learning and iterative greedy method[J]. Control and Decision, 2021, 36(11): 2609-2617.

[34] 肖鹏飞, 张超勇, 孟磊磊, 等. 基于深度强化学习的非置换流水车间调度问题[J]. 计算机集成制造系统, 2021, 27(1): 192-205.

[35] XIAO P F, ZHANG C Y, MENG L L, et al. Non-permutation flow shop scheduling problem based on deep reinforcement learning[J]. Computer Integrated Manufacturing Systems, 2021, 27(1): 192-205.

基金资助

上海市启明星项目(22QB1404200)

国家自然科学基金项目(52375486)