Defect Analysis Apparatus for Efficient Recipe Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Recent defect inspection apparatuses produce thousands of defect coordinates, making it time-consuming and difficult to observe and analyze all defects using existing techniques, which struggle to efficiently detect desired defects.
Innovation Solution
An analyzing apparatus that totals defects with similar attribute data for each inspection recipe, calculates a reference value, and determines the number of defects within a predetermined threshold to efficiently specify the inspection recipe for detecting desired defects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all defect coordinates are observed by defect review apparatus, then detection accuracy is improved, but time consumption increases and efficiency decreases
Solution Approach 1:
The patent extracts and analyzes only the attribute data of defects (such as size, shape, intensity) rather than reviewing all defect coordinates. By extracting relevant characteristics and comparing them across inspection recipes, the system identifies suitable recipes without needing to manually review every defect, thus reducing time while maintaining detection accuracy.
Solution Approach 2:
The patent introduces attribute data as an intermediary between defect coordinates and inspection recipes. Instead of directly reviewing all defect coordinates, the system uses attribute data as a mediator to compare and evaluate inspection recipes, which simplifies the review process and reduces time consumption while maintaining effectiveness.
2Reliability
If multiple inspection recipes are analyzed manually, then inspection quality is improved, but operation complexity increases
Solution Approach 1:
The patent implements self-service by automatically comparing attribute data across multiple inspection recipes and determining the most suitable recipe without requiring manual intervention. The system autonomously analyzes the data, calculates comparisons, and identifies optimal inspection recipes, thereby maintaining high inspection quality while significantly reducing operational complexity.
Solution Approach 2:
The patent replaces manual mechanical review processes with automated data processing. Instead of manually reviewing and comparing defect information across multiple recipes, the system uses automated algorithms to process attribute data, perform comparisons, and determine suitable recipes, thus maintaining quality while simplifying operation.
3Quantity of substance
If all defect coordinates are reviewed, then detection completeness is improved, but productivity decreases
Solution Approach 1:
The patent extracts only the essential attribute data from defect coordinates for analysis. By focusing on extracting and analyzing key characteristics (size, shape, intensity) rather than reviewing all coordinates, the system maintains detection completeness for the most relevant defects while dramatically improving productivity by reducing the review scope.
Solution Approach 2:
The patent applies partial action by analyzing only the necessary attribute data rather than all defect coordinates. This selective approach ensures that sufficient information is gathered to determine suitable inspection recipes while avoiding the excessive time consumption associated with reviewing every defect coordinate, thus balancing completeness with productivity.
Data Source
AI summary
In order to allow to easily specify inspection recipe with which defects desired to be detected can be detected efficiently, a defect inspection apparatus performs defect inspection of a substrate in accordance with a plurality of inspection recipes and produces defect information associated with position of defect in the substrate and attribute data of the defect for each of the inspection recipes and a defect review apparatus produces review result information specifying a kind of defect selected from defects contained in the defect information. An analyzing apparatus obtains defect information and review result information and totalizes the number of defects having attribute data similar to attribute data possessed in defects corresponding to kind of defects to be analyzed for each inspection recipe.


