Algorithmic Module Selection for Semiconductor Defect Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current semiconductor fabrication processes face challenges in accurately and efficiently detecting defects in ultra-large-scale integration devices due to the need for high precision and uniformity, which requires automated examination methods but lacks effective automation for defect classification and review.

Innovation Solution

A system utilizing a processor and memory circuitry that selects and adapts algorithmic modules based on supervised feedback to identify the most effective module for defect detection in semiconductor specimens, improving defect classification accuracy and efficiency by dynamically updating scores based on feedback.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple algorithmic modules are used for defect detection, then defect classification accuracy is improved, but system complexity and processing time increase

Engineering Contradiction:

Improvedefect classification accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system implements feedback mechanisms where examination results from algorithmic modules are evaluated and used to update module scores. Supervised feedback from operators and automated performance metrics continuously refine the selection process, allowing the system to learn from past examinations and improve accuracy without linearly increasing complexity.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system dynamically changes parameters by adjusting module selection probabilities based on scores. Instead of using a fixed set of algorithms, the system modifies which modules are selected and their weighting based on performance parameters, enabling adaptive accuracy improvement without proportional complexity increase.

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If multiple algorithmic modules are used for defect detection, then defect classification accuracy is improved, but processing time increases

Engineering Contradiction:

Improvedefect classification accuracyVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system dynamically adjusts the number and type of algorithmic modules executed based on real-time requirements. The selection module adapts module usage dynamically rather than statically processing through all modules, allowing the system to maintain high accuracy while minimizing processing time for each specific examination case.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system applies partial action by selecting only the necessary subset of algorithmic modules for each examination based on scores and requirements. Instead of running all available modules, the system executes only those with highest relevance and probability, achieving sufficient accuracy with reduced processing time.

Inventive Principle:

Principle #16Partial or excessive action

3Extent of automation

If automated selection of algorithmic modules is implemented, then operator intervention is reduced, but system complexity increases

Engineering Contradiction:

Improveoperator intervention reductionVSAvoidsystem complexity

Core Design Contradiction:

Extent of automationVSDevice complexity

Solution Approach 1:

The system performs self-service through automated module selection and scoring mechanisms. The selection module autonomously evaluates examination results, updates module scores, and makes selection decisions without operator intervention. This self-managing capability reduces the need for human operators while the modular architecture keeps complexity manageable.

Inventive Principle:

Principle #25Self-service

4Measurement precision

If dynamic updating of module scores based on feedback is implemented, then defect detection accuracy is improved, but computational requirements increase

Engineering Contradiction:

Improvedefect detection accuracyVSAvoidcomputational requirements

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The system implements periodic action by updating module scores at structured intervals rather than continuously after every single examination. This periodic update mechanism maintains detection accuracy while reducing computational overhead, as score updates occur at optimal intervals rather than requiring constant recalculation.

Inventive Principle:

Principle #19Periodic action

Data Source

PatentUS20210343000A1Automatic selection of algorithmic modules for examination of a specimen

Publication Date: 2021.11.04 APPL MATERIALS ISRAEL LTD

AI summary

There is provided a system comprising a processor configured to obtain a set of images of a semiconductor specimen, (1) for an image of the set of images, select at least one algorithmic module MS out of a plurality of algorithmic modules, (2) feed the image to MS to obtain data DMS representative of one or more defects in the image, (3) obtain a supervised feedback regarding rightness of data DMS, (4) repeat (1) to (3) for a next image until a completion criterion is met, wherein an algorithmic module selected at (1) is different for at least two different images of the set of images, generate, based on the supervised feedback, a score for each of a plurality of the algorithmic modules, and use scores to identify one or more algorithmic modules Mbest as the most adapted for providing data representative of one or more defects in the set of images.