Wafer Hotspot Correction via Machine Learning Layout Hints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The traditional method of correcting wafer hotspots in integrated circuit fabrication is time-consuming and requires substantial experience, making it challenging for less experienced layout engineers to effectively address these defects.
Innovation Solution
A machine learning system is implemented to receive input layout files with identified wafer hotspots, categorize them, and output proposed layout modifications, utilizing artificial neural networks and image-to-image translation predictors to facilitate quick and accurate corrections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional manual layout modification is used to correct wafer hotspots, then correction accuracy can be achieved through engineer expertise, but the process becomes extremely time-consuming and requires substantial experience
Solution Approach 1:
The patent replaces the manual mechanical process of layout modification with an automated machine learning system. The ML model automatically generates layout modifications by learning from training data of hotspot patterns and corresponding corrections, substituting the engineer's manual intervention with an automated computational system that maintains high accuracy while dramatically reducing time requirements
Solution Approach 2:
The system performs preliminary action by pre-training the machine learning model on extensive datasets of hotspot patterns and their corrections before actual use. This preliminary training enables the model to quickly generate accurate corrections during production without requiring time-consuming manual analysis, effectively preparing the solution in advance
2Reliability
If traditional manual layout modification is used to correct wafer hotspots, then experienced engineers can achieve satisfactory corrections, but the method is not accessible to less experienced engineers
Solution Approach 1:
The machine learning system performs self-service by autonomously analyzing hotspot patterns and generating corrections without requiring human expertise in layout modification. The system serves itself by using its learned knowledge base to independently produce reliable corrections, making the capability accessible to any engineer who can operate the tool regardless of their experience level
Solution Approach 2:
The patent introduces the machine learning model as an intermediary between the hotspot detection and layout modification processes. This intermediary translates the complex task of expert-level correction into an automated process that any engineer can initiate, bridging the gap between detection and reliable correction without requiring deep expertise
3Manufacturing precision
If multiple trial-and-error iterations are performed to verify layout modifications, then satisfactory corrections can be achieved, but the overall productivity is significantly reduced
Solution Approach 1:
The system implements feedback by using simulation verification to check the accuracy of generated layout modifications. The ML model learns from both successful and unsuccessful corrections in its training data, and the feedback loop validates predictions before final implementation, ensuring high accuracy while maintaining efficient throughput through targeted rather than exhaustive verification
Data Source
AI summary
A system that includes a machine learning model that is configured to receive an input layout file that includes a portion of an integrated circuit layout that has a previously identified wafer hotspot, match the previously identified wafer hotspot to one of a plurality of categories of wafer hotspot types, and output a proposed layout modification associated with the matching category of wafer hotspot types.