Method for dynamically targeting a batch process

• US 20030220709A1
• Filed: 06/17/2003
• Published: 11/27/2003
• Est. Priority Date: 12/27/2000
• Status: Active Grant

• Current Assignee: ADA ANALYTICS ISRAEL LTD (Advanced Emissions Solutions, Inc.)
• Original Assignee: Insyst, Ltd. (Advanced Emissions Solutions, Inc.)
• Inventors: Hartman, Jehuda, Brill, Eyal, Yulevitch, Oren
• Granted Patent: US 7,123,978 B2
• Time in Patent Office: Days
• US Class Current: 700/121
• CPC Class Codes: G05B 19/41875 characterised by quality surveillance of production
• CPC Class Codes: G05B 2219/32201 Build statistical model of past normal proces, compare with actual process
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: Y02P 90/02 Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Abstract

A method for controlling at least one characteristic of a product of an industrial batch process. The method comprising the steps of creating a hierarchical knowledge tree describing the process. Following the creation of a knowledge tree a learning process occurs. This leads to the creation of a global model. During the execution of a batch process, the global model is applied to dynamically target subsequent phase parameters based on already executed phases.

Claims

1. A method for controlling a quality of a product of an industrial batch process comprising the steps of:

a) assigning the product a predetermined criteria with respect to a performance of the product;

b) linking said performance to a feature of at least one structural element of the product; and

c) associating said feature of said at least one structural element with at least one stage in a process which forms said at least one structural element wherein at least two said process stages are controlled by a modular process control.

2. A method for controlling a quality of a product of an industrial batch process as in claim 1 wherein said modular process control includes a process control selected from the group consisting of a semi-automatic process control and an automatic process control.

3. A method for controlling a quality of a product of an industrial batch process as in claim 2 wherein said automatic process control includes the use of a quantitative modeler.

4. A method according to claim 3, wherein said quantitative modeler is an empirical modeler.

5. A method according to claim 4, wherein said quantitative modeler comprises a theoretical modeler.

6. A method according to claim 3, wherein said automatic process control comprises use of expert-provided input.

7. A method for controlling a quality of a product of an industrial batch process as in claim 3 wherein said quantitative empirical modeler includes at least one statistical technique selected from the group consisting of linear regression, nearest neighbor, clustering, process output empirical modeling (POEM), classification and regression tree (CART), chi-square automatic interaction detector (CHAID), decision trees and neural network empirical modeling.

8. A method for controlling a quality of a product of an industrial batch process as in claim 1 wherein said product quality is assessed with regard to products selected from the group consisting of semiconductor devices and semiconductor components.

9. A method for controlling a quality of a product of an industrial batch process as in claim 1 wherein said steps are effected at a microelectronic manufacturing facility.

10. A method for controlling a quality of a product of an industrial batch process as in claim 8 wherein said performance includes electronic specifications of said product.

11. A method for controlling a quality of a product of an industrial batch process as in claim 1 wherein said at least one stage in a process is carried out with at least one tool, said process control of said at least one stage affecting an operation of said tool.

12. A method for controlling a quality of a product of an industrial batch process as in claim 11 wherein said at least one stage is carried out by at least one first tool and at least one second tool.

13. A method for controlling a quality of a product of an industrial batch process as in claim 12 wherein an output of said at least one first tool is a measurable input to said at least one second tool.

14. A method for controlling a quality of a product of an industrial batch process as in claim 11 wherein said structural element is formed by at least one first process and at least one second process, further comprising relating said first process to said second process using a Knowledge Tree map.

15. A method for controlling a quality of a product of an industrial batch process as in claim 14 wherein an output of said at least one tool of said at least one first process is a measurable input to said at least one tool in said at least one second process.

16. A method for controlling a quality of a product of an industrial batch process as in claim 12 further comprising controlling a controllable input of said at least one first tool according to an output of said at least one second tool.

17. A method for controlling a quality of a product of an industrial batch process as in claim 14 further comprising controlling a controllable input of said at least one tool in said at least one process according to an output of said at least one tool of said at least one said second process.

18. A method for controlling a quality of a product of an industrial batch process as in claim 13 wherein a throughput of said at least one first tool differs substantially than a throughput of said at least one second tool.

19. A method for controlling a quality of a product of an industrial batch process as in claim 14 wherein a throughput of said at least one tool of said at least one first process differs substantially than a throughput of said at least one tool in said at least one second process.

20. A method for controlling a quality of a product of an industrial batch process as in claim 11 wherein said at least one stage is carried out by at plurality of tools, each tool among said plurality of tools has a ranking score according to an hierarchy of said each tool in said manufacturing, the method further comprising the step of controlling a controllable input of a first said tool according to an output of a second said tool that has a higher said rank than said first tool.

21. A method for controlling an overall batch process target specification, wherein the overall process includes a plurality of sub-processes, each sub-process having an assigned value of an intermediate target, the method comprising reassigning the value of an output of a first one of said sub-process in accordance with an output of at least one second sub-process.

22. A method for controlling an overall batch process target specification as in claim 21, wherein said first one of said sub-process first occurs at a time after said at least one second sub-process.

23. A method for controlling at least one characteristic of a product of an industrial batch process, the method comprising the steps of:

a) creating a hierarchical knowledge tree describing the process;

b) learning the process;

c) creating a model; and

d) executing the model using dynamic targeting.

24. A method of improving the characteristic of a product of a batch process having a plurality of phases each phase having inputs and outputs where some outputs of a preceding phase are inputs of a succeeding phase comprising, upon receipt of an output of an intermediate phase not conforming to a predetermined value:

modifying conditions of a current phase so as to influence outputs of said current phase wherein said influenced outputs of said current phase are inputs of the succeeding phase and whereby outputs of the succeeding phase are influenced;

until the quality of the product is improved.