Dynamic Resource Matching and Adaptive Scheduling for Complex Test Systems under Multi-Task Concurrent Scenarios

Weihan Zhang; Huaqiang Chen; Yuanyuan Liang; Tongqing Wang; Hongle Zhang; Honghui Chen; Yi Xie; Zhentao Han

doi:10.54097/zfanpm24

Authors

Weihan Zhang
Huaqiang Chen
Yuanyuan Liang
Tongqing Wang
Hongle Zhang
Honghui Chen
Yi Xie
Zhentao Han

DOI:

https://doi.org/10.54097/zfanpm24

Keywords:

Dynamic Resource Matching, Multi-Task Concurrency, Reinforcement Learning, Sequence Pattern Mining

Abstract

The intelligent scheduling of large-scale Test Support Systems (TSS) is a critical cornerstone of modern aerospace and industrial engineering. However, under multi-task concurrent scenarios, traditional static planning and local resource mapping struggle to cope with dynamic demand variations, sudden equipment failures, and the high-dimensional spatial-temporal competition of global auxiliary resources. To address this strongly coupled NP-hard problem, this paper proposes a multi-scenario adaptive resource matching and collaborative scheduling framework. First, a dynamic evolution mechanism utilizing hybrid SPADE and Apriori sequence pattern mining is introduced to autonomously recalibrate multi-dimensional resource conflict matrices in real-time. Second, a dual-layer intelligent scheduling architecture is proposed: at the macroscopic parallel layer, a multi-objective Genetic Algorithm (GA) is designed for global space exploration to minimize the makespan and balance global auxiliary loads; at the microscopic execution layer, a Reinforcement Learning (RL) agent, formulated as a Markov Decision Process (MDP), is deployed for real-time adaptive rescheduling against dynamic disturbances. Finally, a closed-loop visual evaluation system is constructed for multi-scenario verification. Experimental results demonstrate that the proposed framework effectively eliminates resource deadlocks, prevents peak load overruns, and significantly minimizes task response delays under strict safety boundaries, providing a novel autonomous paradigm for the robust operation of complex test facility clusters.

Downloads

Download data is not yet available.

References

[1] X. Li, Y. Chen, and Z. Wang, "Capturing topological dependencies in complex process flows via graph neural networks," IEEE Transactions on Industrial Informatics, vol. 18, no. 12, pp. 8891-8902, 2022.

[2] A. Smith, B. Johnson, and C. Davis, "Temporal logic-based sequential pattern mining for identifying resource competition in large-scale scientific facilities," IEEE Access, vol. 10, pp. 45123-45135, 2022.

[3] Y. Li, et al., "Digital twin-enabled dynamic scheduling with preventive maintenance using a double-layer Q-learning algorithm," Robotics and Computer-Integrated Manufacturing, vol. 82, 2023.

[4] H. Chen, et al., "Hierarchical Model Predictive Control for Energy-Aware Scheduling of Digital Twin-Based Batch Manufacturing Systems," IEEE Transactions on Automation Science and Engineering, 2025.

[5] T. Tu, et al., "Fair Resource Allocation in Weakly Coupled Markov Decision Processes," in Proceedings of the Conference on Uncertainty in Artificial Intelligence (UAI), PMLR, 2024.

[6] D. Bertsimas, A. Gupta, and V. Kallus, "Robust Markov decision processes for resource allocation under parameter uncertainty," Operations Research, vol. 69, no. 1, pp. 1-18, 2021.

[7] C. Zhang, W. Song, Z. Cao, J. Zhang, P. S. Tan, and X. Chi, "Learning to dispatch for job shop scheduling via deep reinforcement learning," in Advances in Neural Information Processing Systems (NeurIPS), vol. 33, pp. 1621-1632, 2020.

[8] W. Liu, et al., "Dynamic Job-Shop Scheduling Problems Using Graph Neural Network and Deep Reinforcement Learning," IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 54, no. 1, pp. 1-13, 2024.

[9] H. Zhang, et al., "A Bidding-Based Deep Reinforcement Learning Approach for Multi-Agent Job Shop Scheduling Problem," IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 55, no. 8, pp. 1-14, 2025.

[10] J. A. Killian, A. Biswas, S. Sharma, and M. Tambe, "Robust policies for restless bandits via deep multi-agent reinforcement learning," in Proceedings of the 39th International Conference on Machine Learning (ICML), pp. 11046-11059, 2022.

[11] X. Du, et al., "Situation-Dependent Causal Influence-Based Cooperative Multi-agent Reinforcement Learning," in Proceedings of the 38th AAAI Conference on Artificial Intelligence (AAAI), 2024.

[12] S. Kim, et al., "Navigating the Social Welfare Frontier: Portfolios for Multi-objective Reinforcement Learning," in Proceedings of the 42nd International Conference on Machine Learning (ICML), 2025.

[13] M. Liu, J. Zhao, and T. Wu, "Enhancing global convergence in complex scheduling using Latin hypercube sampling initialized evolutionary algorithms," Computers & Industrial Engineering, vol. 175, 2023.

[14] M. Debner, et al., "Scheduling conditional task graphs with deep reinforcement learning," in Proceedings of the 41st International Conference on Machine Learning (ICML), 2024.

[15] C. Chang, et al., "Dynamic Measurement Scheduling for Event Forecasting using Deep RL," in Proceedings of the 36th International Conference on Machine Learning (ICML), pp. 950-959, 2019.

[16] J. Ding, Y. Ye, J. Xia, and M. Chen, "Effective Reinforcement Learning-based Dynamic Flexible Job Shop Scheduling Using Two-Stage Dispatching," Journal of Systems Architecture, vol. 158, 2025.

[17] L. Wang, S. Zheng, and X. Li, "Robust scheduling strategies in uncertain environments utilizing knowledge-transfer reinforcement learning," IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 52, no. 11, pp. 6904-6916, 2022.

[18] F. Song, et al., "Offloading dependent tasks in multi-access edge computing: A multi-objective reinforcement learning approach," Future Generation Computer Systems, vol. 128, pp. 333-348, 2022.

[19] H. Shen, Q. Xiao, and T. Chen, "On Penalty-based Bilevel Gradient Descent Method," in Proceedings of the 40th International Conference on Machine Learning (ICML), 2023.

[20] Z. Wang, et al., "A Reinforcement-Learning-Enhanced Knowledge-Guided Genetic Algorithm for Flexible Job-Shop Scheduling Problems With Lot Streaming," IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 55, no. 12, pp. 8863-8876, 2025.