Zhang, Ming, Lu, Yang 
ORCID: https://orcid.org/0000-0002-0583-2688, Liu, Chao and Xu, Yuchun
(2022)
Dynamic Scheduling Method for Job-shop Manufacturing Systems by Deep Reinforcement Learning with Proximal Policy Optimization.
In: Connected Everything Conference 2022, 18/05/2022, Liverpool, UK.
Preview |
Text
MingZhang.pdf - Published Version Available under License Creative Commons Attribution. | Preview |
Abstract
With the rapid development of Industry 4.0, modern manufacturing systems have been experiencing profound digital transformation. Development of new technologies can help to improve the efficiency of production and the quality of products. With the increasingly complex production systems, operational decision-making has encountered challenges in the sustainable manufacturing process to satisfy customers and markets' ever-changing demands. Nowadays, the rule-based heuristics approaches are widely used for scheduling management in production systems, which however significantly depends on the expert domain knowledge. In this way, the efficiency of decision-making cannot be guaranteed nor meet the dynamic scheduling requirements in the job-shop manufacturing environment. In this study, we propose using deep reinforcement learning (DRL) methods to tackle the dynamic scheduling problem in the job-shop manufacturing system. The proximal policy optimization (PPO) algorithm has been used in the DRL framework to accelerate the learning process and improve performance. The proposed method has been testified within a real-world dynamic production environment, and it performs better compared with the state-of-the-art methods
University Staff: Request a correction | RaY Editors: Update this record
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)
