Reinforcement learning and simulation based dispatching method in a factory, and an apparatus thereof
AssigneeVMS Solutions Co. Ltd.
InventorsWon-Jun Lee, Byung-Hee Kim, Goo-Hwan Chung
Life of this patent
Estimating the term…
Abstract
Provided is a dispatching method in a factory based on reinforcement learning. The dispatching method in a factory based on reinforcement learning may comprise: constructing a Markov decision process (MDP) for dispatching actions of a dispatcher in the factory and resulting rewards and states of the factory; performing learning by applying reinforcement learning (RL) to the constructed MDP; and as a result of said RL, selecting a job that maximizes a weighted sum of a plurality of scored dispatching rules.
Claims10
A dispatching method in a factory based on reinforcement learning, comprising:
- constructing a Markov decision process (MDP) for dispatching actions of a dispatcher in the factory and resulting rewards and states of the factory;
- performing learning by applying reinforcement learning (RL) to the constructed MDP;
- and
- as a result of said RL, selecting a job that maximizes a weighted sum of a plurality of scored dispatching rules.
A dispatcher configured to perform dispatching actions in a factory based on reinforcement learning, comprising:
- a Markov decision process (MDP) construction unit configured to construct an MDP for the dispatching actions of the dispatcher in the factory and resulting rewards and states of the factory;
- a reinforcement learning unit configured to perform learning by applying reinforcement learning (RL) to the constructed MDP;
- and
- a job selection unit configured to select a job that maximizes a weighted sum of a plurality of scored dispatching rules, as a result of said RL.
Dependencies are inferred from the claim text.