Virtual Memory Device Test Interpolation for Faster Process Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The high cost and time-consuming nature of testing in the manufacturing of memory devices and integrated circuits limits the number of tests that can be performed, leading to incomplete data and reduced yields, as some tests render samples unusable.
Innovation Solution
The application of machine learning to interpolate virtual test results for devices that were not tested, based on correlations with other sets of tests, allowing for more efficient and accurate feedback on processing parameters and reducing the need for physical testing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If more testing is performed on memory devices, then quality control data availability improves, but testing cost and time consumption increase
Solution Approach 1:
The patent creates virtual copies of test results through machine learning interpolation. Instead of physically testing every device, the system generates virtual test data that replicates what would have been obtained from actual testing, allowing comprehensive quality control without the time and resource costs of exhaustive physical testing
Solution Approach 2:
The patent replaces the mechanical testing system with a computational system. Machine learning models substitute for physical test equipment, using algorithms to predict test outcomes based on correlated parameters from actually tested devices, thereby eliminating the need for time-consuming physical testing of every device
2Loss of information
If more testing is performed on memory devices, then quality control data availability improves, but testing cost increases
Solution Approach 1:
The patent creates virtual copies of test results through machine learning interpolation. Instead of physically testing every device, the system generates virtual test data that replicates what would have been obtained from actual testing, allowing comprehensive quality control without the time and resource costs of exhaustive physical testing
Solution Approach 2:
The patent uses inexpensive computational models to generate test data instead of expensive physical testing resources. The virtual test data serves as a disposable substitute for costly actual testing, providing sufficient quality control information at a fraction of the cost
3Measurement precision
If tests are performed that prepare test samples in specific ways, then test accuracy improves, but the samples become subsequently unusable
Solution Approach 1:
The patent creates virtual copies of the destructive testing process. Instead of physically destroying devices through exhaustive testing, the system generates virtual test data that captures all the quality information would have been obtained, preserving the actual devices for sale and use
Solution Approach 2:
The patent introduces machine learning models as intermediaries between the need for quality data and the physical devices. The models mediate by predicting test outcomes from non-destructive measurements, eliminating the need for destructive sample preparation and preserving device usability
Data Source
AI summary
To provide more test data during the manufacture of non-volatile memories and other integrated circuits, machine learning is used to generate virtual test values. Virtual test results are interpolated for one set of tests for devices on which the test is not performed based on correlations with other sets of tests. In one example, machine learning determines a correlation study between bad block values determined at die sort and photo-limited yield (PLY) values determined inline during processing. The correlation can be applied to interpolate virtual inline PLY data for all of the memory dies, allowing for more rapid feedback on the processing parameters for manufacturing the memory dies and making the manufacturing process more efficient and accurate. In another set of embodiments, the machine learning is used to extrapolate limited metrology (e.g., critical dimension) test data to all of the memory die through interpolated virtual metrology data values.