US 12,443,153 B2
Deep learning model in high-mix semiconductor manufacturing
Yulei Sun, Flower Mound, TX (US); Shelby Crain, Sachse, TX (US); and Stephen McWilliams, Carrollton, TX (US)
Assigned to Onto Innovation Inc., Wilmington, MA (US)
Filed by Onto Innovation Inc., Wilmington, MA (US)
Filed on Sep. 23, 2022, as Appl. No. 17/934,912.
Claims priority of provisional application 63/247,904, filed on Sep. 24, 2021.
Prior Publication US 2023/0102925 A1, Mar. 30, 2023
Int. Cl. G05B 13/02 (2006.01)
CPC G05B 13/027 (2013.01) 20 Claims
OG exemplary drawing
 
1. A method to set at least one processing parameter for manufacturing a semiconductor device, the method comprising:
receiving context information regarding the at least one processing parameter for performing a processing step by a manufacturing machine to manufacture the semiconductor device;
inputting the context information into a machine learning network;
receiving a predicted value for the at least one processing parameter from the machine learning network;
setting the at least one processing parameter for the manufacturing machine based on the predicted value to perform the processing step to manufacture the semiconductor device;
receiving, from a metrology instrument, a measured result of the processing step on the semiconductor device associated with the at least one processing parameter; and
feeding back the measured result into the machine learning network,
wherein the machine learning network is trained based on a plurality of features identified by performing a regression technique to estimate a variance of the at least one processing parameter to identify the plurality of features as having a direct relationship to the variance,
wherein the plurality of features includes at least one manufacturing context feature, at least one equipment hardware parameter feature, and at least one upstream parametric data feature.