Abstract: Disclosed is a method of treating a substrate, the method including: seating a first container on a first load port; opening a door of the first container seated on the first load port; after the door of the first container is opened, sequentially taking out one or more sheets of substrates of a first group received in the first container and transferring the taken-out substrates to a substrate treatment unit, and treating the substrates of the first group in the substrate treatment unit; seating a second container on a second load port; opening a door of the second container seated on the second load port; and after the door of the second container is opened, sequentially taking out one or more sheets of the substrates of the second group received in the second container and transferring the taken-out substrates to the substrate treatment unit and treating the substrates of the second group in the substrate treatment unit, in which a timing of opening the door of the second container is determined based on a

Abstract: Disclosed is an apparatus for treating a substrate, the apparatus including: a housing having an interior space; a treatment container disposed in the interior space and having a treatment space in which the substrate is treated; a support unit for supporting the substrate in the interior space; a nozzle unit having a nozzle for supplying a treatment liquid to the substrate supported on the support unit; an exhaust unit for exhausting airflow in the treatment space; an airflow supply unit for providing downdraft airflow into the interior space; and an inspection module for inspecting the treatment space, in which the inspection module is installed in the airflow supply unit.

Abstract: The inventive concept provides a method to determine whether a substrate treatment process is normal using a deep learning model. The method comprising receiving input on a substrate treatment process video, preprocessing the inputted video, using the deep learning model to study a preprocessed video, and determining whether the substrate treatment process is normal by comparing the trained model and a real time substrate treatment process video.

Abstract: The inventive concept provides a method to determine whether a substrate treatment process is normal using a deep learning model. The method comprising receiving input on a substrate treatment process video, preprocessing the inputted video, using the deep learning model to study a preprocessed video, and determining whether the substrate treatment process is normal by comparing the trained model and a real time substrate treatment process video.