# US 2023/0058166 A1 · Method for determining an inspection strategy for a group of substrates in a semiconductor manufacturing process

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2023/0058166 A1
- Document ID: US20230058166A1
- Status: Published application
- Kind: A1
- Application number: 17/910,454
- Filed: 2021-03-01
- Published: 2023-02-23
- Earliest claimed: 2020-04-02
- Assignees: ASML Netherlands B.V.
- Applicants: Asml Netherlands B.V.
- Inventors: Eleftherios KOULIERAKIS; Carlo LANCIA; Juan Manuel GONZALEZ HUESCA; Alexander YPMA; Dimitra GKOROU; Reza SAHRAEIAN
- Web page: https://patentgrep.com/patent/US20230058166A1

## CPC

- G03F 7/705: {Modelling or simulating from physical phenomena up to complete wafer processes or whole workflow in wafer productions}
- G03F 7/70525: {Controlling normal operating mode, e.g. matching different apparatus, remote control or prediction of failure}
- G03F: PHOTOMECHANICAL PRODUCTION OF TEXTURED OR PATTERNED SURFACES, e.g. FOR PRINTING, FOR PROCESSING OF SEMICONDUCTOR DEVICES; MATERIALS THEREFOR; ORIGINALS THEREFOR; APPARATUS SPECIALLY ADAPTED THEREFOR; (phototypographic composing devices B41B; photosensitive materials or processes for photographic purposes G03C; electrophotography, sensitive layers or processes therefor G03G)

## Abstract

A method for determining an inspection strategy for at least one substrate, the method including: quantifying, using a prediction model, a compliance metric value for a compliance metric relating to a prediction of compliance with a quality requirement based on one or both of pre-processing data associated with the substrate and any available post-processing data associated with the at least one substrate; and deciding on an inspection strategy for the at least one substrate, based on the compliance metric value, an expected cost associated with the inspection strategy and at least one objective value describing an expected value of the inspection strategy in terms of at least one objective relating to the prediction model.

## Claims

### Claim 1

A method for determining an inspection strategy for at least one substrate, the method comprising:
quantifying, using a prediction model and based on one or both of pre-processing data associated with the at least one substrate and/or any available post-processing data associated with the at least one substrate, a compliance metric value for a compliance metric relating to a prediction of compliance with a quality requirement; and
deciding on an inspection strategy for the at least one substrate, based on the compliance metric value, an expected cost associated with the inspection strategy and at least one objective value describing an expected value of the inspection strategy in terms of at least one objective relating to the prediction model.

### Claim 2 (depends on 1)

The method as claimed in claim 1, wherein the at least one objective comprises at least one selected from:
monitoring and/or improving performance of the prediction model according to one or more performance metrics;
accelerating learning of the prediction model;
discovering new patterns to be recognized by the prediction model.

### Claim 3 (depends on 1)

The method as claimed in claim 1, wherein the at least one substrate comprises a group of substrates, the deciding on an inspection strategy comprises deciding whether to inspect each substrate of the group of substrates, and the compliance metric comprises a probability value describing a probability of complying with the quality requirement.

### Claim 4 (depends on 3)

The method as claimed in claim 3, further comprising assigning a respective probability value, using the prediction model, to each substrate of the group of substrates based on its corresponding pre-processing data.

### Claim 5 (depends on 3)

The method as claimed in claim 3, wherein the deciding is configurable such that the decision can be configured to impose a bias towards corresponding quantiles of the probability of a substrate complying with the quality requirement.

### Claim 6 (depends on 5)

The method as claimed in claim 5, wherein the configurability of the deciding is realized via one or more configurable parameters of an objective function used in the deciding.

### Claim 7 (depends on 6)

The method as claimed in claim 6, wherein the objective function comprises a binary vector indicating which substrates are selected for inspection weighted by a coefficient dependent on the probability values, expected cost and the one or more configurable parameters and the deciding comprises minimizing the objective function for different candidate binary vectors, subject to a constraint based on the expected cost.

### Claim 8 (depends on 7)

The method as claimed in claim 7, wherein the constraint imposes a maximum number of substrates which can be inspected over a time period.

### Claim 9 (depends on 7)

The method as claimed in claim 7, wherein the constraint is implemented by a constraint parameter in the weighting.

### Claim 10 (depends on 6)

The method as claimed in claim 6, further comprising using a reinforcement learning agent to tune the one or more configurable parameters to improve a performance metric and/or cost metric of the model.

### Claim 11 (depends on 1)

The method as claimed in claim 1, wherein the compliance metric comprises a confidence metric describing a level of confidence that the prediction model is valid, and the method further comprises:
iteratively performing one or more metrology actions to obtain other post-processing data; and
updating the prediction model based on the other post-processing data and assessing the confidence metric or related metric until one or more criteria for the confidence metric or related metric is satisfied.

### Claim 12 (depends on 11)

The method as claimed in claim 11, further comprising:
determining an acquisition score based on the confidence metric and a cost of additional metrology action; and
deciding whether to perform an additional metrology action based on the acquisition score.

### Claim 13 (depends on 3)

The method as claimed in claim 3, further comprising:
labeling selected substrates according to a result of an inspection performed on the selected substrates;
labeling remaining substrates according to the probability value assigned thereto; and
evaluating performance of the prediction model based on the labeling.

### Claim 14 (depends on 13)

The method as claimed in claim 13, further comprising triggering retraining of the prediction model based on the evaluating the performance.

### Claim 15 (depends on 1)

The method as claimed in claim 1, wherein the quantifying is based at least one pre-processing data and the pre-processing data comprises one or both selected from:
context data describing a processing history of the at least one substrate; and/or
pre-processing metrology data acquired prior to an exposure step on the at least one substrate.

### Claim 16

A computer program product comprising a non-transitory computer-readable medium comprising program instructions therein, the instructions, when run on a suitable apparatus, configured to cause the apparatus to at least:
quantify, using a prediction model, a compliance metric value for a compliance metric relating to a prediction of compliance with a quality requirement based on one or both of pre-processing data associated with at least one substrate and/or any available post-processing data associated with at least one substrate; and
decide on an inspection strategy for the at least one substrate, based on the compliance metric value, an expected cost associated with the inspection strategy and at least one objective value describing an expected value of the inspection strategy in terms of at least one objective relating to the prediction model.

### Claim 17 (depends on 16)

The computer program product of claim 16, wherein the at least one objective comprises at least one selected from:
monitoring and/or improving performance of the prediction model according to one or more performance metrics;
accelerating learning of the prediction model;
discovering new patterns to be recognized by the prediction model.

### Claim 18 (depends on 16)

The computer program product of claim 16, wherein the at least one substrate comprises a group of substrates, the instructions configured to cause the apparatus to decide on an inspection strategy are further configured to cause the apparatus to decide whether to inspect each substrate of the group of substrates, and the compliance metric comprises a probability value describing a probability of complying with the quality requirement.

### Claim 19 (depends on 18)

The computer program product of claim 18, wherein the instructions are further configured to cause the apparatus to assign a respective probability value, using the prediction model, to each substrate of the group of substrates based on its corresponding pre-processing data.

### Claim 20 (depends on 18)

The computer program product of claim 18, wherein the instructions configured to cause the apparatus to decide are configurable such that the decision can be configured to impose a bias towards corresponding quantiles of the probability of a substrate complying with the quality requirement.
