.png){kind=icon}
返回
Optimal differentiated threshold characterization for multi-task stochastic deadline scheduling with queuing
.png){kind=icon}
PRE
AI
DOI：10.1016/j.automatica.2024.111545.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
We study the dynamic scheduling of multiple deadline constrained tasks in a serving system with N servers and a public queue with M positions. Under stochastic task arrivals and processing costs (with possibly unknown dynamics), we seek to minimize the long-term expected system cost (the sum of task processing cost and penalty cost resulting from missing tasks' deadlines). To mitigate the curse of dimensionality in the action space, we propose a new approach that utilizes optimal policy characterizations to reduce the action space dimensionality without loss of optimality. For a general setting with preemptive/non-preemptive queuing, we establish the optimality of deadline differentiated threshold policies under arbitrary system dynamics. Integrating the established optimal policy characterizations into the proximal policy optimization (PPO) method, the proposed approach is scalable in M+N. Numerical results demonstrate that the proposed approach significantly outperforms the classical PPO method and other priority rule based PPO approaches, in real-world applications including electric vehicle charging and live streaming scheduling. (c) 2024 Elsevier Ltd. All rights reserved.
Keyword:
Dynamic programming
Deep reinforcement learning
Queuing
Stochastic deadline scheduling
Electric vehicle charging
期刊
IF:
5.9
论文数:
1.2W
被引数:
5.2W
机构
引用论文
NOMA-Enabled Mobile Edge Computing for Internet of Things via Joint Communication and Computation Resource Allocations通过联合通信和计算资源分配实现物联网的NOMA移动边缘计算
.png){kind=icon}
PREAI
分享收藏Optimal Scheduling for Electric Vehicle Charging With Discrete Charging Levels in Distribution Grid
PREAI
分享收藏Study of bi-directional buck-boost converter topologies for application in electrical vehicle motor drives应用于电动汽车电机驱动的双向buck-boost变换器拓扑研究
PREAI
分享收藏
分享收藏Cross-linkable Polymer Matrix for Enhanced Thermal Stability of Succinonitrile-based Polymer Electrolyte in Lithium Rechargeable Batteries可交联聚合物基质，用于增强锂可充电电池中基于丁二腈的聚合物电解质的热稳定性
OAAI
OAAI分享收藏
查看更多内容