Using radial basis function networks and hyper-cubes for excursion classification in semi-conductor processing equipment

• US 9,852,371 B2
• Filed: 02/12/2016
• Issued: 12/26/2017
• Est. Priority Date: 01/17/2013
• Status: Active Grant

• Current Assignee: Applied Materials Incorporated
• Original Assignee: Applied Materials Incorporated
• Inventors: Cantwell, Dermot
• Primary Examiner(s): Hill, Stanley K
• Assistant Examiner(s): BUSS, BENJAMIN J
• Application Number:
  US15/043,229
• Publication Number:
  US 20160162778A1
• Time in Patent Office: 683 Days
• Field of Search: None
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 3/063 using electronic means
• CPC Class Codes: G06N 3/08 Learning methods

Abstract

A method and system for analysis of data, including creating a first node, determining a first hyper-cube for the first node, determining whether a sample resides within the first hyper-cube. If the sample does not reside within the first hyper-cube, the method includes determining whether the sample resides within a first hyper-sphere, wherein the first hyper-sphere has a radius equal to a diagonal of the first hyper-cube.

Claims

1. A method comprising:

identifying a node;

determining a hyper-cube associated with the node;

determining a hyper-sphere associated with the node and a size of the hyper-cube;

identifying a location of a sample associated with data from a semiconductor processing equipment;

determining whether the location of the sample is within the hyper-cube or the hyper-sphere;

determining, by a processing device, a class for the sample based on the determination of whether the location of the sample is within the hyper-cube or the hyper-sphere by assigning a lower confidence to the sample when the sample is located outside of the hyper-cube and within the hyper-sphere than when the sample is located inside of the hyper-cube; and

providing a corrective action to the semiconductor processing equipment based on the determination.

2. The method of claim 1, wherein the determining of the class for the sample comprises:

when the sample is located within the hyper-cube, identifying the sample as the class corresponding to the node.

3. The method of claim 1, wherein the lower confidence is based on a distance of the location of the sample from a side of the hyper-cube.

4. The method of claim 1, wherein the determining of the class for the sample comprises:

when the sample is not located within the hyper-cube or the hyper-sphere, identifying the sample as an unknown class.

5. The method of claim 1, wherein a radius of the hyper-sphere is based on a diagonal of the hyper-cube.

6. The method of claim 1, further comprising:

receiving a user-selection of an excursion;

creating an excursion node for the excursion;

determining an excursion hyper-cube for the excursion node;

determining whether the location of the sample is within the excursion hyper-cube about an excursion node origin;

when the location of the sample is not within the excursion hyper-cube, determining whether the location of the sample is within an excursion hyper-sphere, wherein the excursion hyper-sphere has a radius equal to a diagonal of the excursion hyper-cube; and

determining a likely sample class based on whether the location of the sample is within the excursion hyper-cube or the excursion hyper-sphere.

7. A system comprising:

a memory; and

a processing device, operatively coupled with the memory, to;

identify a node;

determine a hyper-cube associated with the node;

determine a hyper-sphere associated with the node and a size of the hyper-cube;

identify a location of a sample associated with data from a semiconductor processing equipment;

determine whether the location of the sample is within the hyper-cube or the hyper-sphere;

determine a class for the sample based on the determination of whether the location of the sample is within the hyper-cube or the hyper-sphere by assigning a lower confidence to the sample when the sample is located outside of the hyper-cube and within the hyper-sphere than when the sample is located inside of the hyper-cube; and

provide a corrective action to the semiconductor processing equipment based on the determination.

8. The system of claim 7, wherein to determine the class for the sample, the processing device is further to:

when the sample is located within the hyper-cube, identify the sample as the class corresponding to the node.

9. The system of claim 7, wherein the lower confidence is based on a distance of the location of the sample from a side of the hyper-cube.

10. The system of claim 7, wherein to determine the class for the sample, the processing device is further to:

when the sample is not located within the hyper-cube or the hyper-sphere, identify the sample as an unknown class.

11. The system of claim 7, wherein a radius of the hyper-sphere is based on a diagonal of the hyper-cube.

12. The system of claim 7, wherein the processing device is further to:

receive a user-selection of an excursion;

create an excursion node for the excursion;

determine an excursion hyper-cube for the excursion node;

determine whether the location of the sample is within the excursion hyper-cube about an excursion node origin;

when the location of the sample is not within the excursion hyper-cube, determine whether the location of the sample is within an excursion hyper-sphere, wherein the excursion hyper-sphere has a radius equal to a diagonal of the excursion hyper-cube; and

determine a likely sample class based on whether the location of the sample is within the excursion hyper-cube or the excursion hyper-sphere.

13. A non-transitory computer readable medium comprising instructions that, when executed by a processing device, cause the processing device to perform operations comprising:

identifying a node;

determining a hyper-cube associated with the node;

determining a hyper-sphere associated with the node and a size of the hyper-cube;

identifying a location of a sample associated with data from a semiconductor processing equipment;

determining whether the location of the sample is within the hyper-cube or the hyper-sphere;

determining a class for the sample based on the determination of whether the location of the sample is within the hyper-cube or the hyper-sphere by assigning a lower confidence to the sample when the sample is located outside of the hyper-cube and within the hyper-sphere than when the sample is located inside of the hyper-cube; and

providing a corrective action to the semiconductor processing equipment based on the determination.

14. The non-transitory computer readable medium of claim 13, wherein to determine the class for the sample, the operations further comprise:

when the sample is located within the hyper-cube, identifying the sample as the class corresponding to the node.

15. The non-transitory computer readable medium of claim 13, wherein the lower confidence is based on a distance of the location of the sample from a side of the hyper-cube.

16. The non-transitory computer readable medium of claim 13, wherein to determine the class for the sample, the operations further comprise:

when the sample is not located within the hyper-cube or the hyper-sphere, identifying the sample as an unknown class.

17. The non-transitory computer readable medium of claim 13, wherein a radius of the hyper-sphere is based on a diagonal of the hyper-cube.