View in EDS

Constrained continuous-action reinforcement learning for supply chain inventory management.

Saved in:
Bibliographic Details
Title: Constrained continuous-action reinforcement learning for supply chain inventory management.
Authors: Burtea, Radu1 (AUTHOR), Tsay, Calvin1 (AUTHOR) c.tsay@imperial.ac.uk
Source: Computers & Chemical Engineering. Feb2024, Vol. 181, pN.PAG-N.PAG. 1p.
Subjects: Reinforcement learning, Supply chain management, Inventory control, Machine learning, Constraint satisfaction, Supply chains
Abstract: Reinforcement learning (RL) is a promising solution for difficult decision-making problems, such as inventory management in chemical supply chains. However, enabling RL to explicitly consider known environment constraints is crucial for safe deployment in practical applications. This work incorporates recent tools for optimization over trained neural networks to introduce two algorithms for safe training and deployment of RL, with a focus on supply chains. Specifically, we use optimization over trained neural-network state–action value functions (i.e., a critic function) to directly incorporate constraints when computing actions in a continuous action space. Furthermore, we introduce a second algorithm that guarantees constraint satisfaction during deployment by directly implementing actions from constrained optimization of a trained value function. The algorithms are compared against state-of-the-art algorithms TRPO, CPO, and RCPO using a computational supply chain case study. • Reinforcement learning algorithms are proposed to directly embed constraints. • Greedy actions subject to constraints can be found using optimization tools. • Supply chain case study demonstrates "few-shot learner" behavior. • Optimal constrained actions can be directly implemented to guarantee feasibility. [ABSTRACT FROM AUTHOR]
Copyright of Computers & Chemical Engineering is the property of Pergamon Press - An Imprint of Elsevier Science and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Description
Abstract:Reinforcement learning (RL) is a promising solution for difficult decision-making problems, such as inventory management in chemical supply chains. However, enabling RL to explicitly consider known environment constraints is crucial for safe deployment in practical applications. This work incorporates recent tools for optimization over trained neural networks to introduce two algorithms for safe training and deployment of RL, with a focus on supply chains. Specifically, we use optimization over trained neural-network state–action value functions (i.e., a critic function) to directly incorporate constraints when computing actions in a continuous action space. Furthermore, we introduce a second algorithm that guarantees constraint satisfaction during deployment by directly implementing actions from constrained optimization of a trained value function. The algorithms are compared against state-of-the-art algorithms TRPO, CPO, and RCPO using a computational supply chain case study. • Reinforcement learning algorithms are proposed to directly embed constraints. • Greedy actions subject to constraints can be found using optimization tools. • Supply chain case study demonstrates "few-shot learner" behavior. • Optimal constrained actions can be directly implemented to guarantee feasibility. [ABSTRACT FROM AUTHOR]
ISSN:00981354
DOI:10.1016/j.compchemeng.2023.108518