A Study on Real-Time Scheduling for Holonic Manufacturing Systems --- Determination of Utility Values Based on Multi-agent Reinforcement Learning

Authors:
Koji Iwamura;Norihisa Mayumi;Yoshitaka Tanimizu;Nobuhiro Sugimura
Affiliations:
Gradutae School of Engineering, Osaka Prfecture University, Osaka, Japan 599-8531;Gradutae School of Engineering, Osaka Prfecture University, Osaka, Japan 599-8531;Gradutae School of Engineering, Osaka Prfecture University, Osaka, Japan 599-8531;Gradutae School of Engineering, Osaka Prfecture University, Osaka, Japan 599-8531
Venue:
HoloMAS '09 Proceedings of the 4th International Conference on Industrial Applications of Holonic and Multi-Agent Systems: Holonic and Multi-Agent Systems for Manufacturing
Year:
2009

Citing 3
Cited 0

Reference architecture for holonic manufacturing systems: PROSA

Computers in Industry - Special issue on manufacturing systems
Introduction to Reinforcement Learning

Introduction to Reinforcement Learning
A Study on Real-Time Scheduling for Holonic Manufacturing Systems --- Simulation for Estimation of Future Status by Individual Holons

HoloMAS '07 Proceedings of the 3rd international conference on Industrial Applications of Holonic and Multi-Agent Systems: Holonic and Multi-Agent Systems for Manufacturing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper deals with a real-time scheduling method for holonic manufacturing systems (HMS). In the previous paper, a real-time scheduling method based on utility values has been proposed and applied to the HMS. In the proposed method, all the job holons and the resource holons firstly evaluate the utility values for the cases where the holon selects the individual candidate holons for the next machining operations. The coordination holon secondly determine a suitable combination of the resource holons and the job holons which carry out the next machining operations, based on the utility values. Multi-agent reinforcement learning is newly proposed and implemented to the job holons and the resource holons, in order to improve their capabilities for evaluating the utility values of the candidate holons. The individual job holons and resource holons evaluate the suitable utility values according to the status of the HMS, by applying the proposed learning method.