Predictive data selection for model calibration to reduce model prediction uncertainty

摘要

Systems and methods are described for reducing prediction uncertainty in a predictive model associated with a patterning process. They may be used, for example, to calibrate a process model associated with a patterning process. Reducing uncertainty in the predictive model may include determining a predictive uncertainty parameter based on predictive data. Predictive data may be determined using a predictive model. The predictive model may have been calibrated with calibration data. The prediction uncertainty parameter may be associated with variations in the prediction data. Reducing uncertainty in a predictive model may include selecting a subset of process data based on a predictive uncertainty parameter; and recalibrating the predictive model using the selected subset of calibration data and process data.