Wafer Defect Risk Scoring and Correlation Graph Display
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In semiconductor production, existing systems face challenges in efficiently selecting and managing surface defects on wafers that could fatally affect integrated circuit operation, as they rely heavily on human observation and are overwhelmed by large volumes of defect data, making it difficult to prioritize defects based on their impact.
Innovation Solution
A surface defect data display and management system calculates a risk score for each defect based on size and pattern concentration, using a correlation graph to enable operators to easily select defects with high risk scores for further analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automatic defect review and classification systems are used to process large volumes of defect data, then productivity increases, but the operator still faces difficulty in selecting which defects to visually inspect among the overwhelming amount of data
Solution Approach 1:
The system changes the parameter of defect selection from manual visual scanning to automated risk score-based filtering. By calculating risk scores based on defect size, pattern concentration, and other parameters, the system automatically prioritizes defects, making it easy for operators to select high-risk defects for visual inspection without being overwhelmed by data volume.
Solution Approach 2:
The patent introduces an intermediary mechanism (risk score calculation and correlation graph) between the automatic defect review system and the operator. This intermediary processes the large volume of defect data and presents it in a prioritized manner, bridging the gap between automated data collection and human decision-making.
2Measurement precision
If operators visually inspect all detected surface defects, then measurement precision improves, but the time required for defect analysis increases significantly
Solution Approach 1:
The system extracts only the most critical defects from the complete defect dataset by calculating risk scores and selecting defects above a certain threshold. This extraction principle allows operators to focus visual inspection on a small subset of high-risk defects rather than examining all defects, maintaining assessment accuracy while dramatically reducing time consumption.
Solution Approach 2:
The system performs preliminary action by automatically calculating risk scores and prioritizing defects before operator visual inspection. This preliminary filtering and scoring process prepares the defect data in advance, so operators receive pre-sorted defect lists requiring minimal time for analysis.
3Loss of information
If the system displays all defect information in detailed lists, then information completeness improves, but the complexity of the display interface increases
Solution Approach 1:
The patent segments defect information display into multiple hierarchical levels: an overview correlation graph showing defect distribution, a prioritized defect list with key parameters, and detailed defect images accessible upon selection. This segmentation allows complete information to be available while presenting it in organized, manageable sections rather than overwhelming a single complex interface.
Solution Approach 2:
The system transitions from traditional one-dimensional list displays to two-dimensional correlation graphs that visualize defect risk scores and characteristics spatially. This dimensional change allows operators to grasp defect priorities and patterns at a glance while maintaining access to complete detailed information through interactive exploration.
Data Source
AI summary
A surface defect data display and management system comprises a risk score calculation unit for calculating the magnitude of an influence a surface defect on a wafer detected by a wafer inspection system or review system has upon a reduction in the yield of a final product as a risk score of the surface defect based on a defect size of the surface defect on the wafer and a pattern concentration obtained from design data of a pattern figure nearby a location corresponding to the position of the surface defect, and a correlation graph and defect image display unit for preparing a correlation graph showing the correlation between the defect size and the risk score of each defect, displaying the prepared correlation graph on the display apparatus and displaying additionally a defect image list of one or more defects selected by using the correlation graph.