Critical Dimension Prediction Using Sparse Spectrum Training
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current semiconductor manufacturing processes face challenges in accurately predicting critical dimensions with limited sample data, leading to inefficiencies and increased costs due to the need for extensive data collection and processing.
Innovation Solution
A critical dimension prediction system that utilizes a measuring device to acquire sample data, a training data selection device to assign sparsity scores and select relevant spectrums, and a critical dimension predicting model generating device to train an AI model for accurate dimension prediction, allowing for high accuracy even with a small amount of data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If extensive data collection is performed to improve prediction accuracy, then manufacturing costs and time increase
Solution Approach 1:
The patent extracts only the most informative spectrums from the complete dataset by calculating sparsity scores and selecting spectrums with highest scores. This extraction approach maintains prediction accuracy while reducing the amount of data that needs to be collected and processed, thereby reducing manufacturing time and costs
Solution Approach 2:
The patent changes the parameter of data quantity by selecting a subset of spectrums based on sparsity scores rather than using all available spectrums. This parameter change allows the system to achieve high prediction accuracy with reduced data collection requirements
2Measurement precision
If extensive data collection is performed to improve prediction accuracy, then manufacturing costs increase
Solution Approach 1:
The system extracts only the essential spectrums needed for accurate prediction by using sparsity score calculations. This reduces the data processing burden and associated manufacturing costs while maintaining high prediction accuracy
Solution Approach 2:
The patent applies partial action by selecting only the necessary portion of spectrums (those with highest sparsity scores) rather than processing the complete dataset. This partial processing approach reduces computational resources and manufacturing costs while achieving sufficient prediction accuracy
3Measurement precision
If all spectrums are used for training, then model accuracy improves but processing complexity increases
Solution Approach 1:
The patent extracts the most valuable spectrums for training by calculating sparsity scores and selecting those with highest scores. This extraction reduces the dimensionality of the training data, simplifying processing while maintaining or improving model accuracy through focus on the most informative features
Solution Approach 2:
The patent segments the complete spectrum data into individual spectrums and evaluates each independently using sparsity scores. This segmentation allows selective training on only the most informative spectrums, reducing processing complexity while maintaining model accuracy
Data Source
AI summary
A critical dimension prediction system includes a measuring device configured to acquire sample data from a sample semiconductor chip, the sample data including a plurality of spectrums, a training data selection device configured to select a training data set based on the sample data, a critical dimension predicting model generating device configured to generate a critical dimension predicting model by training an artificial intelligence model based on the training data set, and a critical dimension predicting device configured to predict a critical dimension of a target layer by inputting input data into the critical dimension predicting model, the input data including information about the target layer, where the training data selection device is further configured to assign a sparsity score to each of the plurality of spectrums and select at least one of the plurality of spectrums as the training data set based on the sparsity score.