PRODUCTION FAILURE ANALYSIS SYSTEM

• US 20140277673A1
• Filed: 03/13/2013
• Published: 09/18/2014
• Est. Priority Date: 03/13/2013
• Status: Active Grant

• Current Assignee: Western Digital Technologies Incorporated (Western Digital Corporation)
• Original Assignee: Western Digital Technologies Incorporated (Western Digital Corporation)
• Inventors: SIM, KIAM HWEE, KOJARUNCHITT, NATTAPHAT, BOONSENG, APISIT, NG, MENG SENG
• Granted Patent: US 9,417,628 B2
• Time in Patent Office: Days
• US Class Current: 700/109
• CPC Class Codes: G05B 23/0275 Fault isolation and identification, e.g. classify fault; estimate cause or root of failure
• CPC Class Codes: G05B 23/0281 Quantitative, e.g. mathematical distance; Clustering; Neural networks; Statistical analysis

Abstract

A production failure analysis system including a factory quality control unit configured to transmit a trigger signal when a production failure is detected, and an analytics unit. The analytics unit is configured to determine a root cause of the production failure by at least receiving the trigger signal from a factory quality control unit, extracting production data from a database when the trigger signal is received, wherein the production data includes process input variables, and identifying one or more potential root causes of the production failure based in part on an analysis of the process input variables.

Claims

1. A storage device production failure analysis system comprising:

a factory quality control unit configured to transmit a trigger signal when a storage device production failure is detected; and

an analytics unit configured to determine a root cause of the storage device production failure by at least;

receiving the trigger signal from a factory quality control unit;

extracting storage device production data from a database when the trigger signal is received, wherein the storage device production data comprises process input variables; and

identifying one or more potential root causes of the storage device production failure based in part on an analysis of the process input variables.

2. The system of claim 1 wherein the factory quality control unit is configured to generate the storage device production data.

3. The system of claim 2 wherein the storage device production data comprises a cumulative yield, and the factory quality control unit detects the storage device production failure based on the cumulative yield.

4. The system of claim 2 wherein the storage device production data comprises a station yield, and the factory quality control unit detects the storage device production failure based on the station yield.

5. The system of claim 1 wherein the analytics unit is further configured to determine a root cause of the storage device production failure by at least:

cleaning the storage device production data.

6. The system of claim 5 wherein the process input variables comprise one or more configuration data.

7. The system of claim 6 wherein the cleaning the storage device production data further comprises:

separating the process input variables for analysis based on an amount of the configuration data.

8. The system of claim 1 wherein the analytics unit is further configured to determine a root cause of the storage device production failure by at least:

generating a report indicating the process input variables which correspond to the root cause of the storage device production failure.

9. The system of claim 8 wherein the report comprises correlation data indicating a correlation of the process input variables which correspond to the root cause of the storage device production failure, and the root cause of the storage device production failure.

10. The system of claim 9 wherein the correlation data comprises logworth data.

11. The system of claim 9 wherein the correlation data comprises G² data.

12. The system of claim 8 wherein the report comprises a decision tree graph.

13. The system of claim 8 wherein the report comprises a leaf report.

14. The system of claim 8 wherein the report comprises a correlation data chart.

15. The system of claim 8 wherein the report comprises storage device fail data and storage device pass data for the process input variables which correspond to the root cause of the storage device production failure.

16. The system of claim 1 wherein the identifying one or more potential root causes of the storage device production failure based in part on an analysis of the process input variables further comprises:

determining probabilities that the process input variables are the root causes of the storage device production failure; and

identifying the process input variables which are potential root causes of the storage device production failure based in part on the determined probabilities.

17. The system of claim 1 wherein the identifying one or more potential root causes of the storage device production failure based in part on an analysis of the process input variables further comprises:

determining sum squared errors that the process input variables are the root causes of the storage device production failure; and

identifying the process input variables which are potential root causes of the storage device production failure based in part on the determined sum squared errors.

18. A method for determining a root cause of a storage device production failure for a storage device production failure analysis system, the method comprising at least:

receiving a trigger signal from a factory quality control unit, the trigger signal generated when a storage device production failure is detected;

extracting storage device production data from a database when the trigger signal is received, wherein the storage device production data comprises process input variables; and

identifying one or more potential root causes of the storage device production failure based in part on an analysis of the process input variables.

19. The method of claim 18 wherein the factory quality control unit is configured to generate a storage device production data.

20. The method of claim 19 wherein the storage device production data comprises a cumulative yield, and the factory quality control unit detects the storage device production failure based on the cumulative yield.

21. The method of claim 19 wherein the storage device production data comprises a station yield, and the factory quality control unit detects the storage device production failure based on the station yield.

22. The method of claim 18 further comprising:

cleaning the storage device production data.

23. The method of claim 22

wherein the process input variables comprise one or more configuration data, and

wherein the cleaning the storage device production data further comprises:

separating the process input variables for analysis based on an amount of the configuration data.

24. The method of claim 18 further comprising generating a report indicating the process input variables which correspond to the root cause of the storage device production failure.

25. The method of claim 24 wherein the report comprises correlation data indicating a correlation of the process input variables which correspond to the root cause of the storage device production failure, and the root cause of the storage device production failure.

26. The method of claim 24 wherein the report comprises storage device fail data and storage device pass data for the process input variables which correspond to the root cause of the storage device production failure.

27. The method of claim 18 wherein the identifying one or more potential root causes of the storage device production failure based in part on an analysis of the process input variables further comprises:

determining probabilities that the process input variables are the root causes of the storage device production failure; and

identifying the process input variables which are potential root causes of the storage device production failure based in part on the determined probabilities.

28. The method of claim 18 wherein the identifying one or more potential root causes of the storage device production failure based in part on an analysis of the process input variables further comprises:

determining sum squared errors that the process input variables are the root causes of the storage device production failure; and

identifying the process input variables which are potential root causes of the storage device production failure based in part on the determined sum squared errors.

29. A non-transitory machine readable medium comprising instructions which, when executed, direct a computing system to perform operations comprising:

receiving a trigger signal from a factory quality control unit, the trigger signal generated when a storage device production failure is detected;

extracting storage device production data from a database when the trigger signal is received, wherein the storage device production data comprises process input variables; and

identifying one or more potential root causes of the storage device production failure based in part on an analysis of the process input variables.

30. The non-transitory machine readable medium of claim 29 wherein the computer system is further directed to perform operations comprising:

generating a report indicating the process input variables which correspond to a root cause of the storage device production failure.

31. The non-transitory machine readable medium of claim 29 wherein the identifying one or more potential root causes of the storage device production failure based in part on an analysis of the process input variables further comprises:

determining probabilities that the process input variables are a root cause of the storage device production failure; and

identifying the process input variables which are potential root causes of the storage device production failure based in part on the determined probabilities.

32. The non-transitory machine readable medium of claim 29 wherein the identifying one or more potential root causes of the storage device production failure based in part on an analysis of the process input variables further comprises:

determining sum squared errors that the process input variables are a root cause of the storage device production failure; and

identifying the process input variables which are potential root causes of the storage device production failure based in part on the determined sum squared errors.

33. A production failure analysis system comprising:

a factory quality control unit configured to transmit a trigger signal when a production failure is detected; and

an analytics unit configured to determine a root cause of the production failure by at least;

receiving the trigger signal from a factory quality control unit;

extracting production data from a database when the trigger signal is received, wherein the production data comprises process input variables; and

identifying one or more potential root causes of the production failure based in part on an analysis of the process input variables.

34. The system of claim 33 wherein the analytics unit is further configured to determine a root cause of the production failure by at least:

generating a report indicating the process input variables which correspond to the root cause of the production failure.

35. The system of claim 33 wherein the identifying one or more potential root causes of the production failure based in part on an analysis of the process input variables further comprises:

determining probabilities that the process input variables are the root causes of the production failure; and

identifying the process input variables which are potential root causes of the production failure based in part on the determined probabilities.

36. The system of claim 33 wherein the identifying one or more potential root causes of the production failure based in part on an analysis of the process input variables further comprises:

determining sum squared errors that the process input variables are the root causes of the production failure; and

identifying the process input variables which are potential root causes of the production failure based in part on the determined sum squared errors.