Neural-Network Processing Tool Control for Stable Semiconductor Yield
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Semiconductor processing tools experience performance degradation between preventive maintenance cycles, leading to reduced quality and yield in electronic device manufacturing, and varying performance across tools of similar functionality, resulting in decreased production efficiency.
Innovation Solution
Implementing an intelligent processing tool that uses an artificial neural network to monitor and adjust key parameters in real-time, learning from R&D and production data to maintain consistent performance without frequent maintenance, ensuring identical performance across matching tools.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If preventive maintenance is performed periodically on semiconductor processing tools, then reliability is improved, but productivity deteriorates due to downtime and recovery time
Solution Approach 1:
The system performs preliminary actions by continuously monitoring processing parameters and predicting tool performance degradation before it occurs. The neural network model forecasts future tool states and schedules maintenance only when necessary, preventing unexpected failures while avoiding unnecessary maintenance interruptions to production
Solution Approach 2:
The system implements feedback by continuously monitoring processing parameters, comparing actual performance against predicted performance using neural network models, and automatically adjusting operational parameters or scheduling maintenance based on the deviation. This closed-loop feedback ensures reliability while minimizing productivity impact by only intervening when actually needed
2Productivity
If processing tools are operated continuously between maintenance cycles, then productivity is improved, but manufacturing precision deteriorates as performance degrades
Solution Approach 1:
The system continuously monitors processing parameters and compares them against neural network predictions of optimal performance. When deviations indicate quality degradation, the system provides feedback to adjust operational parameters or trigger maintenance, ensuring manufacturing precision is maintained while maximizing continuous operation time between maintenance cycles
Solution Approach 2:
The system dynamically adjusts operational parameters in real-time based on neural network predictions of tool performance degradation. By continuously adapting process parameters to compensate for tool wear and drift, the system maintains manufacturing precision throughout extended operation periods between maintenance cycles
3Manufacturing precision
If artificial neural network parameter adjustment is implemented, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The system introduces an intermediary neural network software layer that sits between the processing tool and control system. This intermediary model translates complex tool performance data into simple maintenance recommendations and parameter adjustments, improving precision without requiring complex hardware modifications to the actual processing tool
Data Source
AI summary
One or more first parameters associated with an electronic device manufacturing process are monitored. An artificial neural network associated with the one or more first parameters is determined. One or more second parameters are determined using the artificial neural network. The one or more first parameters are adjusted using the one or more second parameters.