Dynamic Sampling Measurement for Semiconductor Process Elements
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing sampling measurement methods in semiconductor manufacturing fail to effectively allocate measurement machine capacity, leading to inefficient use and potential long-term unmeasured lots due to not considering the process element information associated with arriving products, resulting in unstable yield and production issues.
Innovation Solution
A method and system that acquire preset and actual measurement ratios for each process element, controlling the measurement machine to measure lots when the actual ratio is less than the preset, ensuring balanced measurement across process elements and optimizing machine utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If sampling measurement is performed based on fixed rules (product number, time period, passed product count), then measurement machine load is reduced, but process element measurement coverage deteriorates
Solution Approach 1:
The system establishes a feedback mechanism that continuously monitors the measurement status of each process element and adjusts sampling decisions accordingly. The measurement status information of process elements is fed back to the sampling determination module, which uses this feedback to dynamically adjust sampling rates and ensure each process element meets its preset measurement ratio, resolving the contradiction between throughput and coverage
Solution Approach 2:
The patent transforms fixed sampling rules into dynamic sampling rates that adapt based on process element measurement status. Instead of static sampling based on product numbers or time periods, the system dynamically adjusts sampling probability for each lot based on real-time process element measurement status, allowing the measurement system to be both efficient and comprehensive
2Reliability
If measurement ratio for each process element is increased to ensure coverage, then measurement completeness is improved, but measurement machine load and production cycle time increase
Solution Approach 1:
The system applies different measurement priorities and sampling rates to different process elements based on their individual characteristics and preset measurement ratios. Rather than uniformly increasing measurement for all elements, the system locally optimizes measurement allocation for each process element, ensuring critical elements receive adequate measurement while less critical elements use lower sampling rates, thus maintaining coverage without proportionally increasing overall measurement time
Solution Approach 2:
The system performs measurement actions partially based on preset ratios rather than requiring complete measurement of all lots for each process element. By implementing measurement at controlled partial rates (preset measurement ratios) and using feedback to ensure minimum coverage requirements are met, the system achieves reliable process element coverage without the excessive time cost of measuring every single lot
Data Source
AI summary
Provided are a sampling measurement method and system, computer device and storage medium. The sampling measurement method includes: acquiring a preset measurement ratio of each process element in a process station; acquiring an actual measurement ratio of a process element associated with a lot of products to be measured that arrive at the measurement station in the process station; and, when the actual measurement ratio of the associated process element is less than the corresponding preset measurement ratio, controlling a measurement machine at the measurement station to measure the lot of products to be measured.