Methods and apparatuses for utilizing adaptive predictive algorithms and determining when to use the adaptive predictive algorithms for virtual metrology

Patent metadata

Field	Value
Publication number	US-9002492-B2
Application number	US-201113027072-A
Country	US
Kind code	B2
Filing date	Feb 14, 2011
Priority date	Feb 16, 2010
Publication date	Apr 7, 2015
Grant date	Apr 7, 2015

Abstract

Described herein are methods, apparatuses, and systems for determining adaptive predictive algorithms for virtual metrology. In some embodiments, a computer implemented method identifies a plurality of predictive algorithms. The method determines when to use one or more of the plurality of predictive algorithms to predict one or more virtual metrology variables in a manufacturing facility.

First claim

What is claimed is:

1. A computer implemented method including a processor and memory comprising:
identifying a plurality of predictive algorithms;
identifying an error switching threshold that is based on a comparison of predictions of metrology data and actual metrology data;
determining when to invoke a first adaptive predictive algorithm, a second adaptive predictive algorithm, or both first and second adaptive predictive algorithms to predict one or more virtual metrology variables in a manufacturing facility based on the error switching threshold;
invoking the first adaptive predictive algorithm for updating a first component vector that is a zero'th order term of a virtual metrology prediction equation if an error is greater than the error switching threshold; and
subsequently invoking the second adaptive predictive algorithm for updating the first component vector and a second component vector that represents a matrix of the virtual metrology prediction equation, wherein the first adaptive predictive algorithm is different than the second adaptive predictive algorithm, wherein the virtual metrology prediction equation includes a predicted output that is based on an input factor, the first component vector, and the second component vector.

2. The computer implemented method of claim 1 , further comprising:
determining whether an error, which is the difference between predictions of metrology data and actual metrology measurement data, is greater than the error switching threshold.

3. The computer implemented method of claim 2 , further comprising:
invoking the first adaptive predictive algorithm if the error is less than or approximately equal to the error switching threshold.

4. The computer implemented method of claim 1 , further comprising:
identifying a quality threshold that is based on predictions of metrology data and actual metrology measurement data;
comparing a quality for prediction for a prediction algorithm with the quality threshold if the actual metrology measurement data is not available.

5. The computer implemented method of claim 4 , further comprising:
invoking the second adaptive predictive algorithm if the quality for prediction is greater than the quality threshold; and
invoking the first adaptive predictive algorithm if the quality for prediction is less than or approximately equal to the quality threshold.

6. The computer implemented method of claim 1 , further comprising:
identifying a quality for prediction for each predictive algorithm;
identifying a quality for prediction for each prediction; and
performing a normalized weighted calculation of the predictions where the weights are a function of the quality for prediction for each predictive algorithm if the actual metrology measurement data is not available.

7. A computer-readable non-transitory storage medium comprising executable instructions to cause a processor to perform operations, the instructions comprising:
identifying a plurality of predictive algorithms;
identifying an error switching threshold that is based on a comparison of predictions of metrology data and actual metrology data;
determining when to invoke a first adaptive predictive algorithm, a second adaptive predictive algorithm, or both first and second adaptive predictive algorithms to predict one or more virtual metrology variables in a manufacturing facility based on the error switching threshold;
invoking the first adaptive predictive algorithm for updating a first component vector that is a zero'th order term of a virtual metrology prediction equation if an error is greater than the error switching threshold; and
subsequently invoking the second adaptive predictive algorithm for updating the first component vector and a second component vector that represents a matrix of the virtual metrology prediction equation, wherein the first adaptive predictive algorithm is different than the second adaptive predictive algorithm, wherein the virtual metrology prediction equation includes a predicted output that is based on an input factor, the first component vector, and the second component vector.

8. The computer-readable non-transitory storage medium of claim 7 , further comprising:
determining whether an error, which is the difference between predictions of metrology data and actual metrology measurement data is greater than the error switching threshold.

9. The computer-readable non-transitory storage medium of claim 8 , further comprising:
invoking the first adaptive predictive algorithm if the error is less than or approximately equal to the error switching threshold.

10. The computer-readable non-transitory storage medium of claim 7 , further comprising:
identifying a quality threshold that is based on predictions of metrology data and actual metrology measurement data;
comparing a quality for prediction for a prediction algorithm with the quality threshold if the actual metrology measurement data is not available.

11. The computer-readable non-transitory storage medium of claim 10 , further comprising:
invoking the second adaptive predictive algorithm if the quality for prediction is greater than the quality threshold; and
invoking the first adaptive predictive algorithm if the quality for prediction is less than or approximately equal to the quality threshold.

12. The computer-readable non-transitory storage medium of claim 7 , further comprising:
identifying a quality for prediction for each predictive algorithm;
identifying a quality for prediction for each prediction;
perform a normalized weighed calculation of the predictions where the weights are a function of the quality for prediction for each predictive algorithm if the actual metrology measurement data is not available.

13. A computer system comprising:
a memory; and
a processor, coupled to the memory, to:
identify one or more of a plurality of predictive algorithms;
identify an error switching threshold that is based on a comparison of predictions of metrology data and actual metrology data;

determine when to invoke a first adaptive predictive algorithm, a second adaptive predictive algorithm, or both first and second adaptive predictive algorithms to predict one or more virtual metrology variables in a manufacturing facility based on the error switching threshold;
invoke the first adaptive predictive algorithm for updating a first component vector that is a zero'th order term of a virtual metrology prediction equation if an error is greater than the error switching threshold; and
subsequently invoke the second adaptive predictive algorithm for updating the first component vector and a second component vector that represents a matrix of the virtual metrology prediction equation, wherein the first adaptive predictive algorithm is different than the second adaptive predictive algorithm, wherein the virtual metrology prediction equation includes a predicted output that is based on an input factor, the first component vector, and the second component vector.

14. The computer system of claim 13 , wherein the processor is further to:
determine whether an error, which is the difference between predictions of metrology data and actual metrology measurement data, is greater than the error.

15. The computer system of claim 14 , wherein the processor is further to:
invoke the first adaptive predictive algorithm if the error is less than or approximately equal to the error switching threshold.

16. The computer system of claim 13 , wherein the processor is further to:
identify a quality threshold that is based on predictions of metrol

Assignees

• Moyne James
• Applied Materials Inc

Inventors

• Moyne James

Classifications

• G05B2219/31465

  Determine which variables of the system to be monitored · CPC title

• G05B13/026 Primary

  using a predictor · CPC title

• G05B2219/31483

  Verify monitored data if valid or not by comparing with reference value · CPC title

• G06N5/04 Primary

  Inference or reasoning models · CPC title

• G05B2219/31459

  Library with metrology plan for different type of workpieces · CPC title