Details
| Original language | English |
|---|---|
| Pages (from-to) | 389-392 |
| Number of pages | 4 |
| Journal | CIRP Annals - Manufacturing Technology |
| Volume | 69 |
| Issue number | 1 |
| Early online date | 18 May 2020 |
| Publication status | Published - 2020 |
Abstract
Multi-agent systems can limit the control problem in complex production systems and solve them more efficiently. However, they often show local optimization tendencies. This paper presents a novel approach for a cooperative multi-agent system, which uses reinforcement learning and considers global key performance indicators. For this purpose, a central deep q-learning module transfers its knowledge to the cooperative order agents. The order agent's experience is stored in a replay memory for subsequent reinforcement learning. Interdependencies between the characteristics of nonlinear production systems and learning parameters are investigated and the performance is evaluated in comparison to conventional methods of production control.
Keywords
- Machine learning, Multi-agent system, Production planning
ASJC Scopus subject areas
- Engineering(all)
- Mechanical Engineering
- Engineering(all)
- Industrial and Manufacturing Engineering
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: CIRP Annals - Manufacturing Technology, Vol. 69, No. 1, 2020, p. 389-392.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Cooperative multi-agent system for production control using reinforcement learning
AU - Dittrich, Marc-André
AU - Fohlmeister, Silas
N1 - Funding information: The presented investigations were conducted within the research project DE 447/181-1. We would like to thank the German Research Foundation (DFG) for the support of this project. In addition, we would like to thank Berend Denkena for his valuable comments and his support.
PY - 2020
Y1 - 2020
N2 - Multi-agent systems can limit the control problem in complex production systems and solve them more efficiently. However, they often show local optimization tendencies. This paper presents a novel approach for a cooperative multi-agent system, which uses reinforcement learning and considers global key performance indicators. For this purpose, a central deep q-learning module transfers its knowledge to the cooperative order agents. The order agent's experience is stored in a replay memory for subsequent reinforcement learning. Interdependencies between the characteristics of nonlinear production systems and learning parameters are investigated and the performance is evaluated in comparison to conventional methods of production control.
AB - Multi-agent systems can limit the control problem in complex production systems and solve them more efficiently. However, they often show local optimization tendencies. This paper presents a novel approach for a cooperative multi-agent system, which uses reinforcement learning and considers global key performance indicators. For this purpose, a central deep q-learning module transfers its knowledge to the cooperative order agents. The order agent's experience is stored in a replay memory for subsequent reinforcement learning. Interdependencies between the characteristics of nonlinear production systems and learning parameters are investigated and the performance is evaluated in comparison to conventional methods of production control.
KW - Machine learning
KW - Multi-agent system
KW - Production planning
UR - http://www.scopus.com/inward/record.url?scp=85084728013&partnerID=8YFLogxK
U2 - 10.1016/j.cirp.2020.04.005
DO - 10.1016/j.cirp.2020.04.005
M3 - Article
VL - 69
SP - 389
EP - 392
JO - CIRP Annals - Manufacturing Technology
JF - CIRP Annals - Manufacturing Technology
SN - 0007-8506
IS - 1
ER -