Details
| Translated title of the contribution | Intelligent order sequencing in manufacturing |
|---|---|
| Original language | German |
| Pages (from-to) | 212-216 |
| Number of pages | 5 |
| Journal | WT Werkstattstechnik |
| Volume | 111 |
| Issue number | 4 |
| Publication status | Published - 2021 |
Abstract
Conventional approaches for order sequencing are usually put into practice by rule-based heuristics, requiring manual adjustments if changes to the production system occur. This article presents an approach for decentralized sequencing using deep q-learning. By considering different production key figures for evaluation, the sequencing can be adapted automatically to changes of the production system, thus achieving a reduction of the cycle time.
ASJC Scopus subject areas
- Engineering(all)
- Control and Systems Engineering
- Engineering(all)
- Automotive Engineering
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: WT Werkstattstechnik, Vol. 111, No. 4, 2021, p. 212-216.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Selbstoptimierende Reihenfolgebildung in der Fertigung
AU - Denkena, Berend
AU - Dittrich, Marc André
AU - Fohlmeister, Silas
N1 - Funding information: Die Autoren danken der Deutschen Forschungsgesellschaft (DFG) für die Förderung des Projekts DE 447/181–1 „Selbstoptimierende dezentrale Fertigungssteuerung“.
PY - 2021
Y1 - 2021
N2 - Conventional approaches for order sequencing are usually put into practice by rule-based heuristics, requiring manual adjustments if changes to the production system occur. This article presents an approach for decentralized sequencing using deep q-learning. By considering different production key figures for evaluation, the sequencing can be adapted automatically to changes of the production system, thus achieving a reduction of the cycle time.
AB - Conventional approaches for order sequencing are usually put into practice by rule-based heuristics, requiring manual adjustments if changes to the production system occur. This article presents an approach for decentralized sequencing using deep q-learning. By considering different production key figures for evaluation, the sequencing can be adapted automatically to changes of the production system, thus achieving a reduction of the cycle time.
UR - http://www.scopus.com/inward/record.url?scp=85108259972&partnerID=8YFLogxK
M3 - Artikel
AN - SCOPUS:85108259972
VL - 111
SP - 212
EP - 216
JO - WT Werkstattstechnik
JF - WT Werkstattstechnik
IS - 4
ER -