Machine Learning Metrology Estimation Model for Semiconductor Overlay Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current semiconductor metrology methods require multiple measurements and complex hardware settings to achieve accurate overlay analysis, leading to inefficiencies in measurement time and throughput.
Innovation Solution
Implementing machine learning algorithms to derive estimation models from metrology metrics, using single images per site for measurement data, and incorporating deep learning to enhance accuracy and speed, while deriving confidence scores for process control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple measurements and complex hardware settings are used to achieve accurate overlay analysis, then measurement precision is improved, but measurement time and throughput deteriorate
Solution Approach 1:
The patent replaces traditional mechanical measurement systems with machine learning algorithms that process single images to derive overlay information. The machine learning model substitutes for complex hardware settings and multiple mechanical measurements, achieving accurate overlay analysis through computational methods rather than repeated physical measurements.
Solution Approach 2:
The patent performs preliminary training of machine learning models using simulation data and initial measurements to create estimation models before actual production measurements. This preliminary action enables the system to make accurate overlay estimations from single images without requiring complex real-time measurements during production.
2Measurement precision
If multiple measurements and complex hardware settings are used to achieve accurate overlay analysis, then measurement precision is improved, but productivity deteriorates
Solution Approach 1:
The patent replaces time-consuming mechanical measurement processes with computational machine learning algorithms. The estimation models process single images rapidly through neural networks, eliminating the need for multiple sequential measurements and complex hardware adjustments, thereby significantly improving throughput while maintaining accuracy.
Solution Approach 2:
The patent changes the fundamental parameters of the measurement process by transitioning from multiple physical measurements to single image analysis. The machine learning model learns to extract overlay information from single images by analyzing patterns and features, changing the measurement approach from quantitative repeated measurements to qualitative single-image analysis.
3Device complexity
If traditional metrology methods are used, then measurement procedures are simple, but measurement precision and accuracy deteriorate
Solution Approach 1:
The patent replaces simple but inaccurate traditional measurement procedures with machine learning-based estimation models. The neural network algorithms process single images to derive accurate overlay information, substituting procedural simplicity with computational intelligence that achieves higher precision without requiring complex hardware or multiple measurements.
Data Source
AI summary
Metrology methods, modules and systems are provided, for using machine learning algorithms to improve the metrology accuracy and the overall process throughput. Methods comprise calculating training data concerning metrology metric(s) from initial metrology measurements, applying machine learning algorithm(s) to the calculated training data to derive an estimation model of the metrology metric(s), deriving measurement data from images of sites on received wafers, and using the estimation model to provide estimations of the metrology metric(s) with respect to the measurement data. While the training data may use two images per site, in operation a single image per site may suffice—reducing the measurement time to less than half the current measurement time. Moreover, confidence score(s) may be derived as an additional metrology and process control, and deep learning may be used to enhance the accuracy and/or speed of the metrology module.