Guardbands in substrate processing systems
A method includes identifying trace data including a plurality of data points, the trace data being associated with production, via a substrate processing system, of substrates having property values that meet threshold values. The method further includes determining, based on a guardband, guardband violation data points of the plurality of data points of the trace data. The method further includes determining, based on the guardband violation data points, guardband violation shape characterization. Classification of additional guardband violation data points of additional trace data is to be based on the guardband violation shape characterization. Performance of a corrective action associated with the substrate processing system is based on the classification.
1 . A method comprising:
identifying trace data comprising a plurality of data points, the trace data being associated with production, via a substrate processing system, of substrates having property values that meet acceptable threshold substrate property values;
determining, based on a guardband, guardband violation data points of the plurality of data points of the trace data associated with the substrates that have the property values that meet the acceptable threshold substrate property values;
determining, based on the guardband violation data points, guardband violation shape characterization that classifies the guardband violation data points as being acceptable by encircling the guardband violation data points on a multivariable graph, wherein classification of additional guardband violation data points of additional trace data is to be based on the guardband violation shape characterization via a probability graph; and
causing, based on the classification, performance of a corrective action to update substrate processing via the substrate processing system.
2 . The method of claim 1 , wherein the guardband violation shape characterization is a weighted combination of one or more of guardband violation duration, guardband violation magnitude, guardband violation area, guardband violation position, or guardband violation intermittency.
3 . The method of claim 1 , wherein the classification of the additional guardband violation data points comprises concatenation of successive or future violations into a single violation based on the guardband violation shape characterization.
4 . The method of claim 3 , wherein the concatenation of the successive or future violations into the single violation is further based on one or more non-violation shapes between two violations of the successive or future violations.
5 . The method of claim 1 , wherein the determining of the guardband violation data points is via multi-variable analysis.
6 . The method of claim 1 , wherein the determining of the guardband violation data points comprises characterizing the guardband violation data points via multi-variable analysis.
7 . The method of claim 1 , wherein the determining of the guardband violation data points is further based on: segmentation of portions of the trace data associated with change in values that exceed a threshold change; and extraction of features from the trace data.
8 . The method of claim 1 , wherein the determining of the guardband violation shape characterization comprises training a machine learning model using data input comprising the trace data to form a trained machine learning model associated with the guardband violation shape characterization.
9 . The method of claim 1 further comprising:
identifying the additional trace data;
determining, based on the guardband, the additional guardband violation data points of the additional trace data;
providing the additional guardband violation data points as data input to a trained machine learning model;
receiving, from the trained machine learning model, output data comprising predictive data; and
determining, based on the predictive data, the classification of the additional guardband violation data points.
10 . A method comprising:
identifying trace data comprising a plurality of data points, the trace data being associated with production, via a substrate processing system, of substrates;
determining, based on a guardband, guardband violation data points of the plurality of data points of the trace data;
determining, based on guardband violation shape characterization being associated with historical substrates that have historical property values that meet acceptable threshold substrate property values, classification of the guardband violation data points via a probability graph, wherein the guardband violation shape characterization classifies historical guardband violation data points of the historical property values as being acceptable by encircling the guardband violation data points on a multivariable graph; and
causing, based on the classification, performance of a corrective action to update substrate processing via the substrate processing system.
11 . The method of claim 10 , wherein the guardband violation shape characterization is a weighted combination of one or more of guardband violation duration, guardband violation magnitude, guardband violation area, guardband violation position, or guardband violation intermittency.
12 . The method of claim 10 , wherein the determining of the classification of the guardband violation data points comprises concatenation of successive or future violations into a single violation based on the guardband violation shape characterization.
13 . The method of claim 10 , wherein the determining of the guardband violation data points is via multi-variable analysis.
14 . The method of claim 10 , wherein the determining of the guardband violation data points is further based on: segmentation of portions of the trace data associated with change in values that exceed a threshold change; and extraction of features from the trace data.
15 . The method of claim 10 further comprising:
receiving historical trace data associated with historical production, via the substrate processing system, of the historical substrates having the historical property values that meet the acceptable threshold substrate property values; and
training a machine learning model using data input comprising the historical trace data to form a trained machine learning model associated with the guardband violation shape characterization.
16 . The method of claim 10 further comprising:
providing the guardband violation data points as data input to a trained machine learning model;
receiving, from the trained machine learning model, output data comprising predictive data; and
determining, based on the predictive data, the classification of the guardband violation data points.
17 . A non-transitory computer-readable storage medium storing instructions which, when executed, cause a processing device to perform operations comprising:
identifying trace data comprising a plurality of data points, the trace data being associated with production, via a substrate processing system, of substrates having property values that meet acceptable threshold substrate property values;
determining, based on a guardband, guardband violation data points of the plurality of data points of the trace data associated with the substrates that have the property values that meet the acceptable threshold substrate property values;
determining, based on the guardband violation data points, guardband violation shape characterization that classifies the guardband violation data points as being acceptable by encircling the guardband violation data points on a multivariable graph, wherein classification of additional guardband violation data points of additional trace data is to be based on the guardband violation shape characterization via a probability graph; and
causing, based on the classification, performance of a corrective action to update substrate processing via the substrate processing system.
18 . The non-transitory computer-readable storage medium of claim 17 , wherein the guardband violation shape characterization is a weighted combination of one or more of guardband violation duration, guardband violation magnitude, guardband violation area, guardband violation position, or guardband violation intermittency.
19 . The non-transitory computer-readable storage medium of claim 17 , wherein the classification of the additional guardband violation data points comprises concatenation of successive violations into a single violation based on the guardband violation shape characterization.
20 . The non-transitory computer-readable storage medium of claim 17 , wherein the determining of the guardband violation data points is via multi-variable analysis.