Reinforcement learning approach to support setup decisions in distributed manufacturing systems

Patrick McDonnell, Sanjay B. Joshi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

A reinforcement learning approach to specifying payoffs for setup games is presented. Setup games are normal form, non-cooperative games used by heterarchical machine controllers to evaluate reconfiguration decisions. While past work utilizing heuristic measures to approximate the effect of setup decisions has demonstrated promising performance, the lack of an accurate long-term model of system dynamics in these heuristic approaches limits their usefulness. The reinforcement learning approach iteratively learns the long term costs of setup decisions, accounting for both immediate decision effects and the effects of likely downstream decisions.

Original languageEnglish (US)
Title of host publicationIEEE Symposium on Emerging Technologies & Factory Automation, ETFA
PublisherIEEE
Pages221-225
Number of pages5
StatePublished - 1997
EventProceedings of the 1997 IEEE 6th International Conference on Emerging Technologies and Factory Automation, ETFA'97 - Los Angeles, CA, USA
Duration: Sep 9 1997Sep 12 1997

Other

OtherProceedings of the 1997 IEEE 6th International Conference on Emerging Technologies and Factory Automation, ETFA'97
CityLos Angeles, CA, USA
Period9/9/979/12/97

Fingerprint

Reinforcement learning
Dynamical systems
Controllers
Costs

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Cite this

McDonnell, P., & Joshi, S. B. (1997). Reinforcement learning approach to support setup decisions in distributed manufacturing systems. In IEEE Symposium on Emerging Technologies & Factory Automation, ETFA (pp. 221-225). IEEE.
McDonnell, Patrick ; Joshi, Sanjay B. / Reinforcement learning approach to support setup decisions in distributed manufacturing systems. IEEE Symposium on Emerging Technologies & Factory Automation, ETFA. IEEE, 1997. pp. 221-225
@inproceedings{c7f7ba3eef8e42faaaa5df6aa884168a,
title = "Reinforcement learning approach to support setup decisions in distributed manufacturing systems",
abstract = "A reinforcement learning approach to specifying payoffs for setup games is presented. Setup games are normal form, non-cooperative games used by heterarchical machine controllers to evaluate reconfiguration decisions. While past work utilizing heuristic measures to approximate the effect of setup decisions has demonstrated promising performance, the lack of an accurate long-term model of system dynamics in these heuristic approaches limits their usefulness. The reinforcement learning approach iteratively learns the long term costs of setup decisions, accounting for both immediate decision effects and the effects of likely downstream decisions.",
author = "Patrick McDonnell and Joshi, {Sanjay B.}",
year = "1997",
language = "English (US)",
pages = "221--225",
booktitle = "IEEE Symposium on Emerging Technologies & Factory Automation, ETFA",
publisher = "IEEE",

}

McDonnell, P & Joshi, SB 1997, Reinforcement learning approach to support setup decisions in distributed manufacturing systems. in IEEE Symposium on Emerging Technologies & Factory Automation, ETFA. IEEE, pp. 221-225, Proceedings of the 1997 IEEE 6th International Conference on Emerging Technologies and Factory Automation, ETFA'97, Los Angeles, CA, USA, 9/9/97.

Reinforcement learning approach to support setup decisions in distributed manufacturing systems. / McDonnell, Patrick; Joshi, Sanjay B.

IEEE Symposium on Emerging Technologies & Factory Automation, ETFA. IEEE, 1997. p. 221-225.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Reinforcement learning approach to support setup decisions in distributed manufacturing systems

AU - McDonnell, Patrick

AU - Joshi, Sanjay B.

PY - 1997

Y1 - 1997

N2 - A reinforcement learning approach to specifying payoffs for setup games is presented. Setup games are normal form, non-cooperative games used by heterarchical machine controllers to evaluate reconfiguration decisions. While past work utilizing heuristic measures to approximate the effect of setup decisions has demonstrated promising performance, the lack of an accurate long-term model of system dynamics in these heuristic approaches limits their usefulness. The reinforcement learning approach iteratively learns the long term costs of setup decisions, accounting for both immediate decision effects and the effects of likely downstream decisions.

AB - A reinforcement learning approach to specifying payoffs for setup games is presented. Setup games are normal form, non-cooperative games used by heterarchical machine controllers to evaluate reconfiguration decisions. While past work utilizing heuristic measures to approximate the effect of setup decisions has demonstrated promising performance, the lack of an accurate long-term model of system dynamics in these heuristic approaches limits their usefulness. The reinforcement learning approach iteratively learns the long term costs of setup decisions, accounting for both immediate decision effects and the effects of likely downstream decisions.

UR - http://www.scopus.com/inward/record.url?scp=0030674586&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030674586&partnerID=8YFLogxK

M3 - Conference contribution

SP - 221

EP - 225

BT - IEEE Symposium on Emerging Technologies & Factory Automation, ETFA

PB - IEEE

ER -

McDonnell P, Joshi SB. Reinforcement learning approach to support setup decisions in distributed manufacturing systems. In IEEE Symposium on Emerging Technologies & Factory Automation, ETFA. IEEE. 1997. p. 221-225