Inspection Strategy for Semiconductor Substrates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current semiconductor manufacturing processes face inefficiencies in deciding which substrates to inspect, leading to high engineering operating expenses and yield loss due to random or expert-based selection strategies that are not systematically reliable, resulting in unnecessary inspections and undetected defects.
Innovation Solution
A method using a prediction model to quantify compliance metric values based on pre-processing and post-processing data, combined with expected costs and objectives, to determine an inspection strategy that optimizes the selection of substrates for inspection, leveraging machine learning techniques such as active learning and reinforcement learning to improve decision-making.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If random or expert-based selection strategies are used to decide which substrates to inspect, then the inspection process is simple to implement, but yield loss and operating expenses increase due to unreliable selection
Solution Approach 1:
The system uses pre-processing data and prediction models to enable substrates to self-identify their inspection needs. Each substrate is evaluated against quality requirements and risk factors, allowing the system to automatically determine which substrates require inspection without relying on random selection or expert judgment, thereby improving reliability while maintaining ease of implementation
Solution Approach 2:
The system performs preliminary evaluation of substrates using pre-processing data before inspection decisions are made. By assessing substrates against quality requirements and risk factors in advance, the system identifies high-risk substrates that need inspection, improving selection reliability without adding complexity to the actual inspection process
2Reliability
If all substrates are inspected to ensure quality, then quality assurance is maximized, but productivity and operating expenses decrease due to limited inspection capacity
Solution Approach 1:
The system applies different inspection strategies to different substrates based on their individual risk profiles and pre-processing data. High-risk substrates receive thorough inspection while low-risk substrates are exempted, allowing the system to maintain high quality assurance for critical items while preserving overall productivity by not inspecting all substrates uniformly
Solution Approach 2:
The system performs inspection on only the necessary portion of substrates - specifically those that exceed quality thresholds or exhibit risk factors. This partial inspection approach maintains adequate quality assurance by focusing resources on substrates that actually need inspection, thereby preserving productivity without compromising quality
3Measurement precision
If systematic inspection strategies are implemented to reduce yield loss, then inspection accuracy improves, but device complexity and implementation difficulty increase
Solution Approach 1:
The system uses a universal prediction model and quality requirement framework that can evaluate all substrates using the same criteria. This multi-functional approach allows the system to maintain high inspection accuracy through systematic evaluation while avoiding excessive complexity by applying a unified methodology rather than multiple specialized systems
Solution Approach 2:
The system incorporates feedback from pre-processing data, inspection results, and quality outcomes to continuously refine inspection decisions. By using feedback loops to improve accuracy over time based on actual performance data, the system achieves high measurement precision while the complexity remains manageable through automated learning rather than complex manual rules
Data Source
AI summary
A method for determining an inspection strategy for at least one substrate, the method including: quantifying, using a prediction model, a compliance metric value for a compliance metric relating to a prediction of compliance with a quality requirement based on one or both of pre-processing data associated with the substrate and any available post-processing data associated with the at least one substrate; and deciding on an inspection strategy for the at least one substrate, based on the compliance metric value, an expected cost associated with the inspection strategy and at least one objective value describing an expected value of the inspection strategy in terms of at least one objective relating to the prediction model.