Etch feedback for control of upstream process

A non-transitory computer readable medium comprising instructions that, when executed by a processing device, cause the processing device to perform operations comprising:

• receiving one or more optical measurements of a film on a substrate captured by an optical sensor at least one of during an etch process or after the etch process in a process chamber;
• determining, for each optical measurement of the one or more optical measurements, a film thickness of the film;
• determining an etch rate of the film based on the one or more optical measurements using the determined film thickness of each optical measurement of the one or more optical measurements;
• inputting the etch rate into a trained machine learning model, wherein the trained machine learning model outputs an estimated process parameter value of at least one process parameter for a previously performed process that was performed on the substrate prior to the etch process by a process chamber;
• determining that the estimated process parameter value of the at least one process parameter provided by the process chamber for the previously performed process had a deviation from a target process parameter value of the at least one process parameter;
• and
• causing an adjustment to an operation of the process chamber that corrects the deviation between the estimated process parameter value and the target process parameter value for future execution of the previously performed process on the process chamber.

A method comprising:

• receiving one or more optical measurements of a film on a substrate captured by an optical sensor at least one of during an etch process or after the etch process in a process chamber;
• determining, for each optical measurement of the one or more optical measurements, a film thickness of the film;
• determining an etch rate of the film based on the one or more optical measurements using the determined film thickness of each optical measurement of the one or more optical measurements;
• inputting the etch rate into a trained machine learning model, wherein the trained machine learning model outputs an estimated process parameter value of at least one process parameter for a previously performed process that was performed on the substrate prior to the etch process by a process chamber;
• determining that the estimated process parameter value of the at least one process parameter provided by the process chamber for the previously performed process had a deviation from a target process parameter value of the at least one process parameter;
• and
• causing an adjustment to an operation of the process chamber that corrects the deviation between the estimated process parameter value and the target process parameter value for future execution of the previously performed process on the process chamber.