Cleaning operations based on deposition thickness

1 . A method comprising:

identifying deposition thickness property data associated with an amount of material deposited via one or more semiconductor processing operations of a process recipe performed in a processing chamber;

providing the deposition thickness property data as input to a variable cleaning relationship of the process recipe to determine cleaning operation parameters comprising two or more of cleaning intensity, cleaning temperature, or cleaning cycle quantity in the processing chamber, wherein the variable cleaning relationship comprises a function that outputs the cleaning operation parameters based on a current value of the deposition thickness property data; and

causing, based on the cleaning operation parameters, a cleaning operation in the processing chamber.

2 . The method of claim 1 , wherein the deposition thickness property data comprises a deposition thickness counter that accumulates deposition thickness values from multiple semiconductor processing operations.

3 . The method of claim 2 , further comprising resetting the deposition thickness counter after the cleaning operation is performed.

4 . The method of claim 1 , wherein the cleaning operation parameters further comprise a quantity of clean time.

5 . The method of claim 1 , wherein:

the processing chamber comprises a plurality of substrate processing areas; and

determining the cleaning operation parameters is based on the deposition thickness property data of at least one of the plurality of substrate processing areas and the variable cleaning relationship of the process recipe.

6 . The method of claim 1 , wherein the providing of of the deposition thickness property data as the input to the variable cleaning relationship to determine the cleaning operation parameters comprises:

providing the deposition thickness property data as the input to a trained machine learning model associated with the variable cleaning relationship; and

receiving, from the trained machine learning model, output associated with predictive data, wherein the cleaning operation parameters are based on the predictive data.

7 . The method of claim 6 , wherein the trained machine learning model is trained using data input comprising historical deposition thickness property data and target output comprising historical performance data.

8 . A non-transitory computer-readable storage medium storing instructions which, when executed, cause a processing device to perform operations comprising:

identifying deposition thickness property data associated with an amount of material deposited via one or more substrate processing operations of a process recipe performed in a processing chamber;

providing the deposition thickness property data as input to a variable cleaning relationship of the process recipe to determine cleaning operation parameters comprising two or more of cleaning intensity, cleaning temperature, or cleaning cycle quantity in the processing chamber, wherein the variable cleaning relationship comprises a function that outputs the cleaning operation parameters based on a current value of the deposition thickness property data; and

causing, based on the cleaning operation parameters, a cleaning operation in the processing chamber.

9 . The non-transitory computer-readable storage medium of claim 8 , wherein the deposition thickness property data comprises a deposition thickness counter that accumulates deposition thickness values from multiple semiconductor processing operations.

10 . The non-transitory computer-readable storage medium of claim 9 , the operations further comprising resetting the deposition thickness counter after the cleaning operation is performed.

11 . The non-transitory computer-readable storage medium of claim 8 , wherein the cleaning operation parameters further comprise a quantity of clean time.

12 . The non-transitory computer-readable storage medium of claim 8 , wherein the processing chamber comprises a plurality of substrate processing areas, and wherein determining the cleaning operation parameters is based on the deposition thickness property data of at least one of the plurality of substrate processing areas and the variable cleaning relationship of the process recipe.

13 . The non-transitory computer-readable storage medium of claim 8 , wherein the providing of of the deposition thickness property data as the input to the variable cleaning relationship to determine the cleaning operation parameters comprises:

providing the deposition thickness property data as the input to a trained machine learning model associated with the variable cleaning relationship; and

receiving, from the trained machine learning model, output associated with predictive data, wherein the cleaning operation parameters are based on the predictive data.

14 . The non-transitory computer-readable storage medium of claim 13 , wherein the trained machine learning model is trained using data input comprising historical deposition thickness property data and target output comprising historical performance data.

15 . A system comprising:

a memory; and

a processing device coupled to the memory, the processing device to:

identify deposition thickness property data associated with an amount of material deposited via one or more substrate processing operations of a process recipe performed in a processing chamber;

provide the deposition thickness property data as input to a variable cleaning relationship of the process recipe to determine cleaning operation parameters comprising two or more of cleaning intensity, cleaning temperature, or cleaning cycle quantity in the processing chamber, wherein the variable cleaning relationship comprises a function that outputs the cleaning operation parameters based on a current value of the deposition thickness property data; and

cause, based on the cleaning operation parameters, a cleaning operation in the processing chamber.

16 . The system of claim 15 , wherein the deposition thickness property data comprises a deposition thickness counter that accumulates deposition thickness values from multiple semiconductor processing operations, and wherein the processing device is to further reset the deposition thickness counter after the cleaning operation is performed.

17 . The system of claim 15 , wherein the cleaning operation parameters further comprise a quantity of clean time.

18 . The system of claim 15 , wherein:

the processing chamber comprises a plurality of substrate processing areas; and

the processing device is to further determine the cleaning operation parameters based on the deposition thickness property data of at least one of the plurality of substrate processing areas and the variable cleaning relationship of the process recipe.

19 . The system of claim 15 , wherein to provide the deposition thickness property data as the input to the variable cleaning relationship to determine cleaning operation parameters, the processing device is to:

provide the deposition thickness property data as the input to a trained machine learning model associated with the variable cleaning relationship; and

receive, from the trained machine learning model, output associated with predictive data, wherein the cleaning operation parameters are based on the predictive data.

20 . The system of claim 19 , wherein the trained machine learning model is trained using data input comprising historical deposition thickness property data and target output comprising historical performance data.