Dynamic Sampling Plans for Wafer Metrology
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of semiconductor fabrication processes and increasing metrology requirements lead to optimization challenges in wafer transportation and metrology measurements, with current methods struggling to efficiently manage wafer sampling plans and measurement paths, resulting in increased throughput time and computational resource demands.
Innovation Solution
The implementation of dynamic sampling plans and optimized measurement paths derived using quantum computing resources, which are updated and managed in real-time to provide wafer-specific measurement sites and conditions, and integrated with FOUP transport optimization to enhance throughput and accuracy in semiconductor fabrication plants.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pre-defined metrology sampling plans are applied uniformly to all wafers, then computational resources and time are reduced, but measurement accuracy and process control quality deteriorate due to inability to adapt to specific wafer conditions
Solution Approach 1:
The patent implements dynamic sampling plans that adapt in real-time based on wafer-specific conditions, process variations, and measurement results. The sampling plan is no longer static but dynamically adjusted during fabrication, allowing the system to optimize measurement accuracy for each wafer while managing throughput time through efficient quantum computing resources.
Solution Approach 2:
The system changes measurement parameters (number of points, locations, frequency) based on actual wafer conditions and process variations. By using quantum computing to analyze real-time data, the system dynamically adjusts sampling parameters to achieve optimal measurement accuracy without uniformly increasing measurements for all wafers, thus balancing precision and throughput.
2Measurement precision
If the number of metrology measurement points per wafer is increased from several hundred to over a thousand, then measurement accuracy and process variation detection improve, but computational resource requirements and processing time increase significantly
Solution Approach 1:
The patent replaces classical computing systems with quantum computing resources to handle the computational complexity of analyzing over a thousand measurement points per wafer. Quantum algorithms can process the large datasets and complex correlations much more efficiently, enabling high-precision measurements without proportionally increasing computational burden.
Solution Approach 2:
The system performs preliminary quantum-based analysis of process conditions and wafer characteristics before final measurements to determine the optimal subset and locations of measurement points. This preliminary action allows the system to focus computational resources on the most critical measurements, reducing the overall computational load while maintaining detection sensitivity.
3Reliability
If sophisticated data analysis methods such as big data techniques and machine learning are applied to metrology data, then process control quality and insight improve, but computational infrastructure requirements and processing time increase
Solution Approach 1:
The patent replaces classical big data and machine learning infrastructure with quantum computing algorithms that can process metrology data more efficiently. Quantum machine learning algorithms can identify patterns and make predictions faster than classical systems, improving process control quality without proportionally increasing processing time or infrastructure requirements.
Solution Approach 2:
The system applies quantum-based data analysis selectively to the most critical data points and process parameters rather than analyzing all data uniformly. This partial action approach focuses computational resources on the measurements that most impact process control quality, reducing overall processing time while maintaining reliability.
4Productivity
If wafer transport between hundreds of production and metrology tools is optimized, then throughput improves, but the complexity of managing FOUP transportation and scheduling increases
Solution Approach 1:
The patent uses quantum computing algorithms to optimize wafer transport scheduling and FOUP management across hundreds of tools. Quantum optimization algorithms can solve the complex scheduling problem much more efficiently than classical methods, improving throughput while the quantum system handles the computational complexity of coordinating all transportation activities.
Data Source
AI summary
Systems and methods of optimizing wafer transport and metrology measurements in a fab are provided. Methods comprise deriving and updating dynamic sampling plans that provide wafer-specific measurement sites and conditions, deriving optimized wafer measurement paths for metrology measurements of the wafers that correspond to the dynamic sampling plan, managing FOUP (Front Opening Unified Pod) transport through the fab, transporting wafers to measurement tools while providing the dynamic sampling plans and the wafer measurement paths to the respective measurement tools before or as the FOUPs with the respective wafers are transported thereto, and carrying out metrology and/or inspection measurements of the respective wafers by the respective measurement tools according to the derived wafer measurement paths. Quantum computing resources may be used to solve the corresponding specific optimization problems, to reduce the required time, improve the calculated solutions and improve the fab yield and accuracy of the produced wafers.