Research on Dynamic Energy-Efficient Scheduling of Flexible Job Shop Based on Deep Reinforcement Learning

DOI

10.2991/978-94-6239-758-3_17How to use a DOI?

Keywords

Flexible job shop scheduling; Deep reinforcement learning; Hierarchical decision-making; Energy-efficient scheduling; Dynamic disturbances

Abstract

The Flexible Job Shop Scheduling Problem (FJSP) is a core challenge in manufacturing optimization under dynamic disturbances and energy constraints. This paper proposes a hierarchical deep reinforcement learning method for dynamic energy-efficient scheduling. The framework decomposes decisions into two layers: the upper agent selects an efficiency-priority, energy-priority, or balanced mode based on global workshop state; the lower agent performs job sequencing and machine allocation guided by the selected mode. Deep Q-Networks (DQN) are employed with a phased training strategy. Experiments under the high disturbance scenario show that the proposed method reduces makespan by 12.3% and energy consumption by 9.6% compared to single-agent DQN, and by 24.3% and 19.9% respectively compared to the SPT rule, verifying the effectiveness of the hierarchical architecture.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Fanyao Gao
PY  - 2026
DA  - 2026/09/08
TI  - Research on Dynamic Energy-Efficient Scheduling of Flexible Job Shop Based on Deep Reinforcement Learning
BT  - Proceedings of the 2026 7th International Conference on Management Science and Engineering Management (ICMSEM 2026)
PB  - Atlantis Press
SP  - 172
EP  - 178
SN  - 2352-5428
UR  - https://doi.org/10.2991/978-94-6239-758-3_17
DO  - 10.2991/978-94-6239-758-3_17
ID  - Gao2026
ER  -