Thickness estimation method and processing control method

A processing control method comprising:

• obtaining a test spectrum image from reflected light obtained by irradiating a test substrate with light;
• generating a regression analysis model using a correlation between test spectrum data included in the test spectrum image and thickness data of a test layer formed on the test substrate, and storing the regression analysis model in a memory;
• performing a semiconductor processing on a first semiconductor substrate to which a first processing control parameter is applied;
• obtaining a spectrum image of the first semiconductor substrate on a transfer path of the first semiconductor substrate subjected to the semiconductor processing;
• obtaining spectrum data included in a predetermined wavelength band from the spectrum image;
• updating the regression analysis model using a correlation between the spectrum data and the first processing control parameter;
• determining a second processing control parameter using the regression analysis model;
• and
• performing the semiconductor processing on a second semiconductor substrate, which is the same as the first semiconductor substrate, and to which the second processing control parameter is applied.

A thickness estimation method comprising:

• obtaining a test spectrum image corresponding to a plurality of positions on a test substrate;
• generating a regression analysis model using a correlation between test spectrum data included in the test spectrum image and thickness data of a test layer formed on the test substrate, and storing the regression analysis model in a memory;
• irradiating a semiconductor substrate that is being transferred with light and obtaining a spectrum image corresponding to an entire area of the semiconductor substrate;
• and
• estimating a thickness of a target layer over the entire area of the semiconductor substrate by applying spectrum data based on the spectrum image of the semiconductor substrate to the regression analysis model.