Determining equipment constant updates by machine learning
A method includes providing, as input to a first trained machine learning model, trace data associated with one or more substrate processing procedures. The input further includes equipment constants associated with the one or more substrate processing procedures. The input further includes trace data of a first processing chamber. The input further includes equipment constants of the first processing chamber. The method further includes obtaining, as output from the first trained machine learning model, a recommended update to a first equipment constant of the first processing chamber. The method further includes updated the first equipment constant of the first processing chamber responsive to obtaining the output from the first trained machine learning model.
1 . A method, comprising:
providing, as input to a first trained machine learning model:
first trace data that is associated with substrates exhibiting one or more target properties, wherein the first trace data was generated by a second trained machine learning model based on input to the second trained machine learning model comprising metrology data of first one or more substrates exhibiting the one or more target properties and associated second trace data,
equipment constants associated with one or more substrate processing procedures,
trace data of a first processing chamber, and
equipment constants of the first processing chamber;
obtaining, as output from the first trained machine learning model, a recommended update to a first equipment constant of the first processing chamber;
updating the first equipment constant of the first processing chamber responsive to obtaining the output from the first trained machine learning model; and
performing a processing operation to process a first substrate in accordance with the first equipment constant.
2 . The method of claim 1 , wherein the input to the first trained machine learning model further comprises metrology data of one or more substrates processed in the first processing chamber.
3 . The method of claim 1 , further comprising:
providing, as input to the second trained machine learning model, the metrology data of first one or more substrates processed in a processing chamber, and the associated second trace data; and
obtaining, as output from the second trained machine learning model, the second trace data associated with the one or more substrate processing procedures, wherein the second trace data is predicted to be associated with substrate processing procedures that are associated with substrates exhibiting the one or more target properties.
4 . The method of claim 3 , further comprising providing, as input to the second trained machine learning model, metrology data of second one or more substrates processed in a processing chamber, wherein the second one or more substrates were processed subsequent to the first one or more substrates, and wherein the second trained machine learning model is configured to apply a larger first weighting factor to the metrology data of the second one or more substrates are weighted more heavily in while generating output by the second trained machine learning model than a second weighting factor applied to the metrology data of the first one or more substrates.
5 . The method of claim 1 , further comprising obtaining, as output from the first trained machine learning model, a recommended update to a second equipment constant of the first processing chamber, and scheduling an update of the second equipment constant, wherein the update of the second equipment constant is scheduled to occur subsequent to updating the first equipment constant.
6 . The method of claim 1 , further comprising:
providing, as input to the first trained machine learning model, trace data of a second processing chamber, and equipment constants of the second processing chamber;
obtaining, as output from the first trained machine learning model, a recommended update to a second equipment constant of the second processing chamber; and
scheduling an update of the second equipment constant of the second processing chamber, wherein the update of the second equipment constant of the second processing chamber is scheduled to occur subsequent to updating the first equipment constant.
7 . The method of claim 1 , wherein the trace data associated with the one or more substrate processing procedures comprises an upper limit trace and a lower limit trace.
8 . The method of claim 7 , wherein the upper limit trace comprises one or more of:
a trace including a highest values of a plurality of traces;
a trace representing a statistical upper bound for the plurality of traces; or
a trace representing a predicted upper bound for traces associated with the plurality of traces.
9 . A system, comprising memory and a processing device coupled to the memory, wherein the processing device is to:
provide, as input to a first trained machine learning model:
first trace data that is associated with substrates exhibiting one or more target properties, wherein the first trace data was generated by a second trained machine learning model based on input to the second trained machine learning model comprising metrology data of first one or more substrates exhibiting the one or more target properties and associated second trace data,
equipment constants associated with one or more substrate processing procedures,
trace data of a first processing chamber, and
equipment constants of the first processing chamber;
obtain, as output from the first trained machine learning model, a recommended update to a first equipment constant of the first processing chamber;
update the first equipment constant of the first processing chamber responsive to obtaining the output from the first trained machine learning model; and
perform a processing operation to process a first substrate in accordance with the first equipment constant.
10 . The system of claim 9 , wherein the input to the first trained machine learning model further comprises metrology data of one or more substrates processed in the first processing chamber.
11 . The system of claim 9 , wherein the processing device is further to:
provide, as input to the second trained machine learning model, the metrology data of first one or more substrates processed in a processing chamber, and the associated second trace data associated with the first one or more substrates processed in the processing chamber; and
obtain, as output from the second trained machine learning model, the second trace data associated with the one or more substrate processing procedures, wherein the second trace data is predicted to be associated with substrate processing procedures that are associated with substrates exhibiting the one or more target properties.
12 . The system of claim 11 , wherein the processing device is further to provide, as input to the second trained machine learning model, metrology data of second one or more substrates processed in a processing chamber, wherein the second one or more substrates were processed subsequent to the first one or more substrates, and wherein the second trained machine learning model is configured to apply a larger first weighting factor to the metrology data of the second one or more substrates are weighted more heavily in while generating output by the second trained machine learning model than a second weighting factor applied to the metrology data of the first one or more substrates.
13 . The system of claim 9 , wherein the processing device is further to obtain, as output from the first trained machine learning model, a recommended update to a second equipment constant of the first processing chamber, and scheduling an update of the second equipment constant, wherein the update of the second equipment constant is scheduled to occur subsequent to updating the first equipment constant.
14 . The system of claim 9 , wherein the processing device is further to:
provide, as input to the first trained machine learning model, trace data of a second processing chamber, and equipment constants of the second processing chamber;
obtain, as output from the first trained machine learning model, a recommended update to a second equipment constant of the second processing chamber; and
schedule an update of the second equipment constant of the second processing chamber, wherein the update of the second equipment constant of the second processing chamber is scheduled to occur subsequent to updating the first equipment constant.
15 . The system of claim 9 , wherein the trace data associated with the one or more substrate processing procedures comprises an upper limit trace and a lower limit trace.
16 . They system of claim 15 , wherein the upper limit trace comprises one or more of:
a trace including a highest values of a plurality of traces;
a trace representing a statistical upper bound for the plurality of traces; or
a trace representing a predicted upper bound for traces associated with the plurality of traces.
17 . A non-transitory machine-readable storage medium storing instructions which, when executed, cause a processing device to perform operations comprising:
providing, as input to a first trained machine learning model:
first trace data that is associated with substrates exhibiting one or more target properties, wherein the first trace data was generated by a second trained machine learning model based on input to the second trained machine learning model comprising metrology data of first one or more substrates exhibiting the one or more target properties and associated second trace data,
equipment constants associated with one or more substrate processing procedures,
trace data of a first processing chamber, and
equipment constants of the first processing chamber;
obtaining, as output from the first trained machine learning model, a recommended update to a first equipment constant of the first processing chamber;
updating the first equipment constant of the first processing chamber responsive to obtaining the output from the first trained machine learning model; and
performing a processing operation to process a first substrate in accordance with the first equipment constant.
18 . The non-transitory machine-readable storage medium of claim 17 , wherein the operations further comprise:
providing, as input to the second trained machine learning model, the metrology data of first one or more substrates processed in a processing chamber, and the associated second trace data associated with the first one or more substrates processed in the processing chamber; and
obtaining, as output from the second trained machine learning model, the second trace data associated with the one or more substrate processing procedures, wherein the second trace data is predicted to be associated with substrate processing procedures that are associated with substrates exhibiting the one or more target properties.
19 . The non-transitory machine-readable storage medium of claim 18 , wherein the operations further comprise providing, as input to the second trained machine learning model, metrology data of second one or more substrates processed in a processing chamber, wherein the second one or more substrates were processed subsequent to the first one or more substrates, and wherein the second trained machine learning model is configured to apply a larger first weighting factor to the metrology data of the second one or more substrates are weighted more heavily in while generating output by the second trained machine learning model than a second weighting factor applied to the metrology data of the first one or more substrates.
20 . The non-transitory machine-readable storage medium of claim 17 , wherein the operations further comprise obtaining, as output from the first trained machine learning model, a recommended update to a second equipment constant of the first processing chamber, and scheduling an update of the second equipment constant, wherein the update of the second equipment constant is scheduled to occur subsequent to updating the first equipment constant.