A facility- simulator based job scheduling system using reinforcement deep learning

摘要

In the factory environment where a number of processes constitute a workflow that has an interrelationship with each other, and when the processes in the workflow proceed, the process is implemented by learning the neural network agent that determines the next operation action given the current state of the workflow in the factory environment. It relates to a factory simulator-based scheduling system using reinforcement learning for scheduling, comprising at least one neural network that outputs a next job to be processed in the corresponding state when receiving a factory workflow state (hereinafter referred to as a workflow state) as an input, the neural network is a neural network agent trained by reinforcement learning; Factory simulator simulating factory workflows; and a reinforcement learning module for simulating the factory workflow with the factory simulator, extracting reinforcement learning data from the simulation result, and learning the neural network of the neural network agent with the extracted reinforcement learning data. With the system as described above, by constructing learning data by extracting the next state and performance when a work action of a specific process is performed in the state of various processes through the simulator, it is possible to stably learn the neural network agent in a faster time. and, due to this, it is possible to instruct more optimized work in the field.