Machine Learning Equipment Constant Updates for Chamber Uniformity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for improving the performance of manufacturing equipment, such as semiconductor processing chambers, are inefficient and costly, as they often require adjusting processing recipes to match golden trace data, which can lead to increased energy usage, component stress, and reduced equipment lifespan, resulting in higher maintenance needs and downtime.
Innovation Solution
The method involves updating equipment constants using machine learning models that analyze trace data and equipment performance to recommend optimal adjustments, allowing for more precise and efficient improvements in manufacturing chamber performance without the need for frequent recipe changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If processing recipes are adjusted to match golden trace data, then manufacturing precision is improved, but energy consumption increases and equipment lifespan decreases
Solution Approach 1:
The patent changes the approach from adjusting processing recipes to updating equipment constants. By modifying equipment constants (such as chamber pressure, temperature, or flow rate baseline values) rather than process parameters, the system achieves performance uniformity while avoiding the excessive energy consumption associated with recipe adjustments. This parameter change enables the equipment to operate more efficiently while maintaining precision.
2Manufacturing precision
If processing recipes are adjusted to match golden trace data, then manufacturing precision is improved, but equipment lifespan decreases
Solution Approach 1:
The patent transitions from adjusting processing recipes to updating equipment constants. By changing equipment constants (fundamental equipment parameters) rather than process recipes, the system achieves performance uniformity without subjecting equipment to the stress and wear that results from frequent recipe adjustments, thereby extending equipment lifespan.
3Manufacturing precision
If processing recipes are adjusted to match golden trace data, then manufacturing precision is improved, but maintenance needs increase
Solution Approach 1:
The patent changes the adjustment target from processing recipes to equipment constants. By updating equipment constants, the system achieves performance uniformity while reducing maintenance needs, as equipment constant updates are less intrusive and require fewer subsequent maintenance interventions compared to recipe adjustments.
4Manufacturing precision
If processing recipes are adjusted to match golden trace data, then manufacturing precision is improved, but downtime increases
Solution Approach 1:
The patent transitions from recipe adjustment to equipment constant update. By modifying equipment constants, the system achieves performance uniformity with reduced downtime, as equipment constant updates can be implemented more quickly and require less validation time compared to comprehensive recipe adjustments.
Data Source
AI summary
A method includes providing, to a trained machine learning model configured to determine a recommended adjustment to an equipment constant of a substrate manufacturing system, first input data indicative of a state of the substrate manufacturing system. The method further includes providing, to the trained machine learning model as second input data, an indication of a performed adjustment to the equipment constant. The method further includes retraining the trained machine learning model based on a difference between the recommended adjustment to the equipment constant and the performed adjustment to the equipment constant to generate a retrained machine learning model.