Wafer Defect Classification Accuracy Tracking for Recipe Update Timing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing defect classification systems in semiconductor manufacturing face challenges in accurately determining the need for recipe updates due to inappropriate inspection conditions, leading to decreased classification accuracy and increased operating time, with manual defect classification being cumbersome for large defect counts.
Innovation Solution
A defect classification device that compares automatic classification results with manual classification results across multiple classification spaces to determine the appropriate timing for updating inspection conditions, using a combination of automatic and manual defect classification methods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual defect classification is performed to calculate classification accuracy rate, then classification accuracy can be evaluated, but the process becomes very troublesome and difficult to apply when the number of defects is large
Solution Approach 1:
The patent uses automatic classification results as a copy or substitute for manual classification. Instead of requiring operators to manually classify all defects, the system performs automatic classification and uses these results to evaluate classification accuracy, thereby avoiding the cumbersome manual operation while maintaining evaluation capability
Solution Approach 2:
The system performs self-evaluation by using its own automatic classification results to assess its classification accuracy. The defect classification device automatically calculates accuracy metrics without requiring external manual classification, enabling the system to self-monitor and self-evaluate its performance
2Measurement precision
If recipe update is performed frequently to maintain classification accuracy, then classification accuracy is improved, but operating time of the device decreases
Solution Approach 1:
The patent implements a feedback mechanism where the system continuously monitors classification accuracy using automatic classification results. When accuracy drops below a threshold, the system triggers a recipe update. This feedback loop enables timely updates only when necessary, balancing accuracy maintenance with operational efficiency
Solution Approach 2:
The system performs preliminary evaluation of classification accuracy using automatic results before triggering a recipe update. By预先 assessing whether accuracy has degraded, the system avoids unnecessary updates and only performs recipe updates when actually needed, thereby preserving operating time
Data Source
AI summary
An object of the present disclosure is to provide a defect classification device capable of easily grasping an appropriate recipe update timing of an imaging device when classification accuracy for classifying defects existing on a semiconductor wafer is decreased. The defect classification device according to the present disclosure calculates classification accuracy by further acquiring a result of a manual classification for defects spanning a plurality of classification spaces as a result of an automatic classification, and comparing the result of the automatic classification with the result of the manual classification (see FIG. 5).