Substrate processing apparatus, substrate processing method, training data generation method, training method, training device, trained model creation method, and trained model

A substrate processing apparatus comprising:

• a substrate holding section configured to hold a processing target substrate in a rotatable manner, the processing target substrate including a resist layer with a hardened layer formed therein;
• a chemical liquid supply section configured to supply a chemical liquid to remove the resist layer to the processing target substrate;
• a substrate information acquiring section configured to acquire substrate information of the processing target substrate including hardened layer thickness information of the processing target substrate or ion implantation condition information of the processing target substrate, the hardened layer thickness information of the processing target substrate indicating a thickness of the hardened layer in the processing target substrate, the ion implantation condition information of the processing target substrate indicating a condition for ion implantation by which the hardened layer in the processing target substrate has been formed in the resist layer of the processing target substrate;
• a chemical liquid processing condition information acquiring section configured to acquire chemical liquid processing condition information of the processing target substrate from a trained model based on the substrate information of the processing target substrate, the chemical liquid processing condition information of the processing target substrate indicating a chemical liquid processing condition to remove the resist layer from the processing target substrate;
• and
• a controller configured to control the substrate holding section and the chemical liquid supply section to perform processing to remove the resist layer with a chemical liquid on the processing target substrate based on the chemical liquid processing condition information of the processing target substrate acquired in the chemical liquid processing condition information acquiring section, wherein
• the trained model is built through machine training of training data in which substrate information of a training target substrate including a resist layer with a hardened layer formed therein, chemical liquid processing condition information of the training target substrate, and processing result information of the training target substrate are associated with each other, the substrate information of the training target substrate including hardened layer thickness information indicating a thickness of the hardened layer in the training target substrate or ion implantation condition information indicating a condition for ion implantation by which the hardened layer in the training target substrate has been formed in the resist layer of the training target substrate, the chemical liquid condition information of the training target substrate indicating a condition for the processing to remove the resist layer with the chemical liquid having been performed on the training target substrate, the processing result information indicating a result of the processing with the chemical liquid having been on the training target substrate,
• the chemical liquid processing condition information of the processing target substrate contains information indicating any of a concentration of the chemical liquid, a temperature of the chemical liquid, a supply amount of the chemical liquid, a discharge pattern of the chemical liquid, and a rotational speed of the processing target substrate in supply of the chemical liquid, and
• the chemical liquid processing condition information of the training target substrate contains information indicating any of the concentration of the chemical liquid, the temperature of the chemical liquid, the supply amount of the chemical liquid, the discharge pattern of the chemical liquid, and a rotational speed of the training target substrate when the chemical liquid has been supplied.