User configurable multivariate time series reduction tool control method

• US 20030055523A1
• Filed: 09/17/2002
• Published: 03/20/2003
• Est. Priority Date: 03/19/1999
• Status: Active Grant

• Current Assignee: GlobalFoundries, Inc.
• Original Assignee: Asif Ghias, James J. Mcgill, John Z. Colt Jr, Maheswaran Surendra, Marc A. Taubenblatt, Nancy T. Pascoe, Raymond J. Bunkofske
• Inventors: Taubenblatt, Marc A., McGill, James J., Surendra, Maheswaran, Ghias, Asif, Colt, John Z. Jr., Pascoe, Nancy T., Bunkofske, Raymond J.
• Granted Patent: US 6,584,368 B2
• Time in Patent Office: Days
• US Class Current: 700/108
• CPC Class Codes: G05B 2219/45031 Manufacturing semiconductor wafers
• 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
• CPC Class Codes: H01L 22/20 Sequence of activities consisting of a plurality of measurements, corrections, marking or sorting steps

Abstract

A method and structure for controlling a manufacturing tool includes measuring different manufacturing parameters of the tool, transforming a plurality of time series of the manufacturing parameters into intermediate variables based on restrictions and historical reference statistics, generating a surrogate variable based on the intermediate variables, if the surrogate variable exceeds a predetermined limit, identifying a first intermediate variable, of the intermediate variables, that caused the surrogate variable to exceed the predetermined limit and identifying a first manufacturing parameter associated with the first intermediate variable, and inhibiting further operation of the tool until the first manufacturing parameter has been modified to bring the surrogate value within the predetermined limit.

Claims

1. A method of controlling a manufacturing tool, comprising:

measuring different manufacturing parameters of said tool;

transforming a plurality of time series of said manufacturing parameters into intermediate variables based on restrictions and historical reference statistics;

generating at least one surrogate variable based on said intermediate variables;

if said surrogate variable exceeds a predetermined limit, identifying a first intermediate variable, of said intermediate variables, that caused said surrogate variable to exceed said predetermined limit and identifying a first manufacturing parameter associated with said first intermediate variable; and

inhibiting further operation of said tool until said first manufacturing parameter has been modified to bring said surrogate value within said predetermined limit.

2. The method according to claim 1, wherein said surrogate variable comprises T², where:

3. The method according to claim 2, wherein x and σ

are user adjustable.

4. The method according to claim 1, wherein said transforming reduces a dimensionality of said time series of said manufacturing parameters.

5. The method according to claim 1, wherein said transforming includes one or more of selecting, modifying and combining said time series of said manufacturing parameters based on said restrictions and said historical reference statistics.

6. The method according to claim 1, wherein said restrictions comprise user-defined, batch-specific restrictions.

7. A computer implemented method of controlling a tool using multivariate analysis, comprising:

measuring a set of tool variables with sensors;

storing values of said variables in a computer connected to receive data from said sensors, said computer generating a surrogate variable having a value representative of whether said tool is in control, said surrogate variable being a function of a plurality of intermediate variables;

determining if said value of said surrogate variable is outside of a predetermined limit, said computer identifying an out-of-control intermediate variable of said intermediate variables that primarily contributed to said value of said surrogate variable;

identifying an out-of-control tool variable of said tool variables that primarily contributed to said out-of-control intermediate variable; and

inhibiting further operation of said tool until said out-of-control tool variable has been modified to bring said surrogate value within said predetermined limit.

8. The method according to claim 7, wherein said surrogate variable comprises T², where:

9. The method according to claim 8, wherein x and σ

are user adjustable.

10. The method according to claim 7, further comprising transforming a plurality of time series of said tool variables into said intermediate variables based on restrictions and historical reference statistics.

11. The method according to claim 10, wherein said transforming reduces a dimensionality of said time series of said tool variables.

12. The method according to claim 10, wherein said transforming includes one or more of selecting, modifying and combining said time series of said tool variables based on said restrictions and said historical reference statistics.

13. The method according to claim 10, wherein said restrictions comprise user-defined, batch-specific restrictions.

14. The method according to claim 7, further comprising:

inputting reference statistics and a transformation program into said computer;

inputting raw data from process runs into said computer;

using said transformation program to compute said intermediate variables from said raw data; and

applying said reference statistics to a scaled version of a Hotelling function and computing said T² score for said surrogate variable based on said intermediate variables.

15. A computerized method of determining corrections for a manufacturing process, comprising:

measuring a set of manufacturing parameters;

performing a first T² score computation using said manufacturing parameters;

providing a first inverse matrix using intermediate variables obtained from said first T² score computation;

using said first inverse matrix to create a generalized formulation for such matrices;

using said generalized formulation to compute a second T² score for said inverse matrix with one variable missing; and

subtracting said second T² score from said first T² score.

16. The method according to claim 15, wherein:

17. The method according to claim 16, wherein x and σ

are user adjustable.

18. The method in claim 15, further comprising transforming a plurality of time series of said manufacturing parameters into said intermediate variables based on restrictions and historical reference statistics.

19. The method according to claim 15, further comprising:

repeating said computing and said subtracting for each of said intermediate variables; and

ranking results of said subtracting to identify an out-of-control intermediate variable.

20. A computer interface for creating a file for transformation of various process recipes, process tools and products in a statistical analysis of a manufacturing process, said computer interface comprising:

an input device for inputting data;

a correction device for correcting missing data in a time series of said data by determining a length of said times series and an interpolation interval of said time series;

a first selector for selecting a transformation function;

a generator for generating one or more control charts from a Hotelling function using said data corrected by said correction device and using said transformation function; and

a second selector for selecting between said control charts.

21. The computer interface according to claim 20, wherein said Hotelling function comprises T², where:

22. The computer interface according to claim 21, wherein x and σ

are user adjustable.

23. The computer interface according to claim 20, wherein said transformation function reduces a dimensionality of said time series of said manufacturing parameters.

24. The computer interface according to claim 20, wherein said transformation function selects, modifies and combines said time series of said manufacturing parameters based on restrictions and historical reference statistics.

25. The computer interface according to claim 24, wherein said restrictions comprise user-defined, batch-specific restrictions.