Sensor-Metrology Data Matching for ML-Based Corrective Action
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for associating sensor data and metrology data in manufacturing processes are inefficient, prone to errors, and costly, particularly due to manual association and the use of optical character recognition (OCR) systems, which consume significant resources and time.
Innovation Solution
A method that integrates sensor metrology data by identifying common identifiers between sensor and metrology data sets, generating aggregated data, and using this data to train a machine learning model for predictive analytics, thereby eliminating the need for manual association and OCR systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual association methods are used to link sensor data and metrology data, then data integration can be performed, but the process becomes inefficient and error-prone
Solution Approach 1:
The patent replaces manual mechanical association processes with automated electronic data processing systems. The system automatically links sensor data and metrology data using unique identifiers (wafer IDs, tool IDs, timestamps) without human intervention, eliminating manual errors while maintaining high integration efficiency through computational matching algorithms.
Solution Approach 2:
The patent introduces an intermediary data integration system that acts as a mediator between sensor data sources and metrology data sources. This intermediary system uses standardized data structures and identifier matching to automatically correlate data from different sources, resolving the contradiction by providing both automated efficiency and accurate association through its mediating function.
2Reliability
If OCR systems are used to associate sensor and metrology data, then data integration is achieved, but energy consumption and processor overhead increase significantly
Solution Approach 1:
The patent extracts and uses only the essential identifying features (unique IDs, timestamps, tool identifiers) from sensor and metrology data for association purposes. By taking out only the necessary matching elements rather than processing entire data sets or using complex OCR recognition, the system achieves accurate data association with minimal energy consumption and processor overhead.
3Reliability
If OCR systems are used for data association, then sensor and metrology data can be linked, but storage space requirements and costs increase
Solution Approach 1:
The patent extracts and stores only the essential identifying elements (unique wafer IDs, tool IDs, timestamps) needed for data association rather than storing complete sensor and metrology data sets. This extraction approach maintains accurate data linking capability while significantly reducing storage space requirements and associated costs.
4Reliability
If manual association methods are used, then data integration can be performed, but time consumption increases
Solution Approach 1:
The patent replaces manual mechanical association processes with automated electronic data processing systems that use identifier matching algorithms. This substitution eliminates the time-consuming manual review and verification steps while maintaining accurate data association through automated correlation of unique identifiers, timestamps, and tool IDs.
Data Source
AI summary
Methods, systems, and non-transitory computer readable medium are described for sensor metrology data integration. A method includes receiving sets of sensor data and sets of metrology data. Each set of sensor data includes corresponding sensor values associated with producing corresponding product by manufacturing equipment and a corresponding sensor data identifier. Each set of metrology data includes corresponding metrology values associated with the corresponding product manufactured by the manufacturing equipment and a corresponding metrology data identifier. The method further includes determining common portions between each corresponding sensor data identifier and each corresponding metrology data identifier. The method further includes, for each of the sensor-metrology matches, generating a corresponding set of aggregated sensor-metrology data and storing the sets of aggregated sensor-metrology data to train a machine learning model. The trained machine learning model is capable of generating one or more outputs for performing a corrective action associated with the manufacturing equipment.