Production control system and integrated production control system

A production control system of a factory, the system comprising:

• a plurality of machines;
• an air conditioner;
• a power computation unit for monitoring power consumption of an entire factory, including power consumption of the plurality of machines and the air conditioner;
• a temperature information generation unit for generating information on temperature inside the factory,
• wherein running status and processing condition of the plurality of machines and running status of the air conditioner are controlled to produce a product dictated by a production plan by a dictated delivery deadline;
• and
• a machine learning unit that learns a relationship of operational status, which includes the running status and processing condition of the plurality of machines and running status of the air conditioner, and environmental status, which includes production completion time of the product according to the operational status, the temperature inside the factory and the power consumption of the entire factory, and outputs the operational status that brings the environmental status to a desired condition in accordance with a production plan, wherein
• the machine learning unit includes a state observation unit, a learning device and a learning control unit;
• the state observation unit obtains data on the production plan, control data provided for the plurality of machines and the air conditioner, the total power consumption received from the power computation unit, and temperature information received from the temperature information generation unit, as state variables;
• the learning device performs learning based on the state variables obtained by the state observation unit, and outputs control data provided for the plurality of machines and the air conditioner in accordance with the data on the production plan;
• and
• the learning control unit includes a reward computation unit, which computes rewards, which indicate degrees of desirability of production operations executed in the factory in accordance with the control data and a storage unit, which store state variables and sets of rewards, and the learning control unit controls learning performed by the learning device on the basis of the state variables and the sets of rewards.