Substrate Defect Source Analysis With Automated Corrective Action
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional manual inspection of substrates for defects is time-consuming, inaccurate, and costly, leading to lower throughput, equipment damage, and production interruptions, with manual defect source tracing and root cause identification being subjective and inefficient.
Innovation Solution
A system and method for substrate defect analysis using property data from metrology equipment, including SEM and EDX images, to identify defect categories and sub-categories, and automate corrective actions such as cleaning, repair, or parameter updates based on machine learning models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual inspection is used to identify and trace defect sources, then subjectivity and inaccuracy are introduced, but automation and precision are reduced
Solution Approach 1:
The patent replaces manual mechanical inspection with an automated image processing system that uses computer algorithms to analyze substrate images, identify defects, and trace defect sources. The system automatically processes images from inspection equipment and generates defect reports without human intervention in the analysis phase, thereby eliminating subjectivity while maintaining high precision through consistent algorithmic application.
Solution Approach 2:
The system enables self-service by allowing the image processing algorithm to automatically identify defects, categorize them by type and severity, and trace their sources without requiring manual inspection. The automated system serves itself by using predefined criteria and machine learning models to make decisions about defect classification and root cause identification, reducing dependency on human operators.
2Loss of time
If manual defect tracing is performed, then time consumption increases, but productivity is reduced
Solution Approach 1:
The system performs preliminary action by pre-processing images and pre-identifying potential defects before final analysis. The image processing system continuously monitors substrates and prepares defect data in advance, so when a defect is detected, the tracing and analysis are already partially completed, significantly reducing the time required for full defect characterization and source identification.
Solution Approach 2:
The patent replaces time-consuming manual defect tracing with automated image processing that rapidly analyzes substrate images, compares them against known defect patterns, and identifies defect sources through algorithmic processing. This mechanical-to-automated substitution reduces analysis time from minutes or hours to seconds, thereby increasing substrate throughput and overall productivity.
3Measurement precision
If comprehensive defect analysis is performed, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the comprehensive defect analysis into distinct modular components: image acquisition, pre-processing, defect detection, defect classification, source tracing, and reporting. Each module performs a specific function with well-defined inputs and outputs, allowing the complex analysis system to be managed through independent, testable components that can be developed and maintained separately while achieving high overall precision.
Data Source
AI summary
A method includes determining, by a processing device, a defect source associated with one or more regions of a substrate corresponding to a plurality of defect sub-categories of a first defect category, the substrate being processed by a substrate processing system. The method further includes, responsive to the determining of the defect source, causing, by the processing device, performance of a corrective action associated with the substrate processing system to reduce substrate defects.