Adjusting weighting of a parameter relating to fault detection based on a detected fault

• US 8,676,538 B2
• Filed: 11/02/2004
• Issued: 03/18/2014
• Est. Priority Date: 11/02/2004
• Status: Active Grant

• Current Assignee: Ocean Semiconductor LLC (Fullbrite Capital Partners, LLC)
• Original Assignee: Advanced Micro Devices, Inc.
• Inventors: Purdy, Matthew A.
• Primary Examiner(s): COSIMANO, EDWARD R
• Application Number:
  US10/979,309
• Publication Number:
  US 20060095232A1
• Time in Patent Office: 3,423 Days
• Field of Search: 702/185, 702/1, 702/33, 702/34, 702/35, 702/36, 702/40, 702/57, 702/58, 702/59, 702/81, 702/82, 702/83, 702/84, 702/108, 702/113, 702/114, 702/115, 702/117, 702/118, 702/127, 702/179, 702/181, 702/182, 702/183, 702/187, 702/189, 738/658, 738/659, 438/5, 438/6, 438/7, 438/8, 438/9, 438/10, 438/11, 438/12, 438/13, 438/14, 438/15, 438/16, 438/17, 438/18, 700/1, 700/11, 700/21, 700/79, 700/90, 700/95, 700/96, 700/108, 700/109, 700/110, 700/117, 700/118, 700/119, 700/120, 700/121, 700/159, 700/174, 700/175, 708/100, 708/105, 708/200, 714/1, 714/25, 714/37, 714/48, 714/100
• US Class Current: 702/183
• CPC Class Codes: G05B 23/0221 Preprocessing measurements, e.g. data collection rate adjustment; Standardization of measurements; Time series or signal analysis, e.g. frequency analysis or wavelets; Trustworthiness of measurements; Indexes therefor; Measurements using easily measured parameters to estimate parameters difficult to measure; Virtual sensor creation; De-noising; Sensor fusion; Unconventional preprocessing inherently present in specific fault detection methods like PCA-based methods

Abstract

A method, apparatus and a system, for provided for performing a dynamic weighting technique for performing fault detection. The method comprises processing a workpiece and performing a fault detection analysis relating to the processing of the workpiece. The method further comprises determining a relationship of a parameter relating to the fault detection analysis to a detected fault and adjusting a weighting associated with the parameter based upon the relationship of the parameter to the detected fault.

Claims

1. A method, comprising:

performing in a computer a fault detection analysis relating to processing of a workpiece;

determining in a said computer a relationship of a parameter relating to said fault detection analysis to a detected fault;

adjusting in said computer a weighting of said parameter based upon said relationship of said parameter to said detected fault; and

performing in said computer the fault detection analysis relating to processing of a subsequent workpiece using said adjusted weighting.

2. The method of claim 1, wherein said workpiece comprises a semiconductor wafer.

3. The method of claim 1, wherein determining in said computer a relationship of a parameter relating to said fault detection analysis to a detected fault further comprises determining in said computer a causation of a parameter relating to said fault detection analysis to a detected fault.

4. The method of claim 1, wherein determining a relationship of a parameter relating to said fault detection analysis to a detected fault further comprises determining in said computer an importance of a parameter relating to said fault detection analysis to a detected fault.

5. The method of claim 1, further comprising:

designated in said computer whether said detected fault is a significant fault; and

adjusting said weighting associated with said parameter based responsive to designating said detected fault as a significant fault.

6. The method of claim 1, wherein performing in said computer said fault detection analysis further comprises utilizing a fault detection model to perform said fault detection, wherein said parameter is an input parameter to said fault detection model.

7. The method of claim 1, wherein determining in said computer said relationship of a parameter relating to said fault detection analysis to a detected fault further comprises determining in said computer whether said parameter is a significant factor associated with said fault.

8. The method of claim 1, wherein adjusting in said computer said weighting associated with said parameter based upon said relationship of said parameter to said detected fault further comprises decreasing in said computer said weighting associated with said parameter based upon said relationship.

9. The method of claim 1, wherein determining in said computer said relationship of a parameter relating to said fault detection analysis to said detected fault further comprises determining in said computer a relationship between at least one of pressure data, temperature data, humidity data, or gas flow rate data associated with said processing of said workpiece, to said detected fault.

10. The method of claim 1, further comprising:

receiving in said computer metrology data relating to processing said workpiece;

receiving in said computer tool state data relating to processing said workpiece; and

correlating in said computer said metrology data and said tool state data with said fault data to characterize a fault.

11. The method of claim 10, wherein said tool state data further comprises at least one of pressure data, temperature data, humidity data, or gas flow rate data associated with said processing of said workpiece.

12. The method of claim 1, wherein performing in said computer said fault detection analysis further comprises performing a principal component analysis (PCA) relating to said processing of said workpiece.

13. The method of claim 12, wherein performing in said computer said principal component analysis further comprises utilizing a PCA model in said computer to perform said PCA, wherein said parameter is an input parameter to said PCA model.

14. The method of claim 1, wherein adjusting in said computer said weighting associated with said parameter based upon said relationship of said parameter to said detected fault further comprises increasing in said computer said weighting associated with said parameter based upon said relationship.

15. The method of claim 14, wherein increasing in said computer said weighting associated with said parameter based upon said relationship further comprises requiring in said computer a smaller fluctuation of said parameter during said fault detection analysis to determine that a fault associated with said processing of said workpiece has occurred.

16. The method of claim 15, wherein increasing in said computer said weighting associated with said parameter based upon said relationship further comprises requiring in said computer a larger fluctuation of said parameter during said fault detection analysis to determine that a fault associated with said processing of said workpiece has occurred.

17. An apparatus, comprising:

means for performing a fault detection analysis in a computer relating to processing of a workpiece;

means for determining in said computer a relationship of a parameter relating to said fault detection analysis to a detected fault;

means for adjusting in said computer a weighting of said parameter based upon said relationship of said parameter to said detected fault; and

means for performing in said computer the fault detection analysis relating to processing of a subsequent workpeice using adjusted weighting.

18. A non-transitory computer readable program storage device encoded with instructions that, when executed by a computer, performs a method, comprising:

performing a fault detection analysis relating to processing of a workpeice;

determining a relationship of a parameter relating to said fault detection analysis to a detected fault;

adjusting a weighting of said parameter based upon said relationship of said parameter to said detected fault; and

performing the fault detection analysis relating to processing of a subsequent workpeice using said adjusting weighting.

19. The non-transitory a computer readable program storage device encoded with instructions that, when executed by a computer, performs the method of claim 18, wherein performing said fault detection analysis further comprises utilizing a fault detection model to perform said fault detection, wherein said parameter is an input parameter to said fault detection model.

20. The non-transitory computer readable program storage device encoded with instructions that, when executed by a computer, performs the method of claim 18, wherein determining said relationship of a parameter relating to said fault detection analysis to a detected fault further comprises determining whether said parameter is a significant factor associated with said fault.

21. The non-transitory computer readable program storage device encoded with instructions that, when executed by a computer, performs the method of claim 18, wherein determining said relationship of a parameter relating to said fault detection analysis to said detected fault further comprises determining a relationship between at least one of pressure data, temperature data, humidity data, or gas flow rate data associated with said processing of said workpiece, to said detected fault.

22. The non-transitory computer readable program storage device encoded with instructions that, when executed by a computer, performs the method of claim 18, wherein performing said fault detection analysis further comprises performing a principal component analysis (PCA) relating to said processing of said workpiece.

23. The non-transitory computer readable program storage device encoded with instructions that, when executed by a computer, performs the method of claim 22, wherein performing said principal component analysis further comprises utilizing a PCA model to perform said PCA, wherein said parameter is an input parameter to said PCA model.

24. The non-transitory computer readable program storage device encoded with instructions that, when executed by a computer, performs the method of claim 18, wherein adjusting said weighting associated with said parameter based upon said relationship of said parameter to said detected fault further comprises increasing said weighting associated with said parameter based upon said relationship.

25. The non-transitory computer readable program storage device encoded with instructions that, when executed by a computer, performs the method of claim 24, wherein increasing said weighting associated with said parameter based upon said relationship further comprises requiring a smaller fluctuation of said parameter during said fault detection analysis to determine that a fault associated with said processing of said workpiece has occurred.

26. The non-transitory computer readable program storage device encoded with instructions that, when executed by a computer, performs the method of claim 18, wherein adjusting said weighting associated with said parameter based upon said relationship of said parameter to said detected fault further comprises decreasing said weighting associated with said parameter based upon said relationship.

27. The non-transitory computer readable program storage device encoded with instructions that, when executed by a computer, performs the method of claim 26, wherein increasing said weighting associated with said parameter based upon said relationship further comprises requiring a larger fluctuation of said parameter during said fault detection analysis to determine that a fault associated with said processing of said workpiece has occurred.

28. The non-transitory computer readable program storage device encoded with instructions that, when executed by a computer, performs the method of claim 18, further comprising:

acquiring metrology data relating to processing said workpiece;

acquiring tool state data relating to processing said workpiece; and

correlating said metrology data and said tool state data with said fault data to characterize a fault.

29. The non-transitocv computer readable program storage device encoded with instructions that, when executed by a computer, performs the method of claim 28, wherein acquiring said tool state data relating to processing said workpiece further comprises acquiring at least one of pressure data, temperature data, humidity data, or gas flow rate data associated with said processing of said workpiece.

30. The non-transitory computer readable program storage device encoded with instructions that, when executed by a computer, performs the method of claim 28, further comprising determining a relationship of a parameter relating to said fault detection analysis to a detected fault further comprises determining a causation of a parameter relating to said fault detection analysis to a detected fault.

31. The non-transitory computer readable program storage device encoded with instructions that, when executed by a computer, performs the method of claim 28, further comprising determining a relationship of a parameter relating to said fault detection analysis to a detected fault further comprises determining an importance of a parameter relating to said fault detection analysis to a detected fault.