Panagiotis D. Paraschos , Georgios K. Koulinas , Dimitrios E. Koulouriotis
{"title":"Parametric and reinforcement learning control for degrading multi-stage systems","authors":"Panagiotis D. Paraschos , Georgios K. Koulinas , Dimitrios E. Koulouriotis","doi":"10.1016/j.promfg.2021.10.055","DOIUrl":null,"url":null,"abstract":"<div><p>This paper addresses the joint control problem in the context of a two-stage stochastic manufacturing/remanufacturing system, which involve both manufacturing and remanufacturing processes. Its operability is affected by frequent deterioration failures. Along with the condition of the system, the manufactured items are affected as well. Thus, the system obtains lesser revenues due to the low-quality products and the downtimes of the deteriorated system. For this purpose, the state and the condition of the systems and the manufactured products must be monitored dynamically so as to devise an optimal strategy for manufacturing, maintenance, and quality control. The present paper proposes a novel two-agent reinforcement learning framework that incorporates parametric production and maintenance activities. The aim is to improve the productivity of the system and keep the system operational with minimal maintenance activities so as to maximize the overall profitability. The performance of the presented approach is evaluated through experimental scenarios.</p></div>","PeriodicalId":91947,"journal":{"name":"Procedia manufacturing","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2351978921002523/pdf?md5=972ef873dd66b3e1057b0153fcb026be&pid=1-s2.0-S2351978921002523-main.pdf","citationCount":"4","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Procedia manufacturing","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2351978921002523","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 4
Abstract
This paper addresses the joint control problem in the context of a two-stage stochastic manufacturing/remanufacturing system, which involve both manufacturing and remanufacturing processes. Its operability is affected by frequent deterioration failures. Along with the condition of the system, the manufactured items are affected as well. Thus, the system obtains lesser revenues due to the low-quality products and the downtimes of the deteriorated system. For this purpose, the state and the condition of the systems and the manufactured products must be monitored dynamically so as to devise an optimal strategy for manufacturing, maintenance, and quality control. The present paper proposes a novel two-agent reinforcement learning framework that incorporates parametric production and maintenance activities. The aim is to improve the productivity of the system and keep the system operational with minimal maintenance activities so as to maximize the overall profitability. The performance of the presented approach is evaluated through experimental scenarios.