Tire Tread Compound Prediction Using Constraint-Aware Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the static properties of rubber compounds for tire treads require extensive laboratory testing, leading to increased lead times, costs, and variability in data, hindering efficient product development.
Innovation Solution
A predictive method using machine learning, involving data normalization, pre-processing, and a stacked machine learning algorithm to simulate laboratory tests, ensuring accurate prediction of static properties while respecting physical constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If extensive laboratory testing is performed to determine static properties of rubber compounds, then measurement precision is improved, but loss of time and productivity deteriorate
Solution Approach 1:
The patent applies preliminary action by pre-processing and normalizing data before building predictive models. Historical laboratory test data is collected, normalized to reduce variability, and used to train machine learning models in advance. This allows the system to make rapid predictions without performing new physical tests, thus reducing time to market while maintaining prediction precision through the use of pre-processed high-quality data
Solution Approach 2:
The patent uses copying by creating a virtual model that replicates physical laboratory testing. Machine learning models are trained to copy the behavior and outcomes of extensive laboratory tests. Once trained, these models can predict static properties without requiring actual physical testing, thereby maintaining measurement precision while dramatically reducing the time and resources needed
2Manufacturing precision
If iterative experimental campaigns are conducted to optimize formulation, then manufacturing precision is improved, but loss of time and productivity deteriorate
Solution Approach 1:
The system performs preliminary action by pre-processing historical formulation data and normalizing it to reduce variability. This pre-processed data is then used to train machine learning models that can predict the outcomes of formulation changes without requiring iterative physical experiments. The models are trained in advance to understand formulation-property relationships, enabling rapid optimization
Solution Approach 2:
The patent replaces the mechanical system of physical experimentation with an information-based system. Instead of conducting iterative laboratory tests to optimize formulation, the system uses machine learning models that have been trained on historical data. These models can predict formulation outcomes computationally, substituting physical experimentation with algorithmic prediction, thereby maintaining formulation precision while dramatically increasing development speed
3Measurement precision
If data normalization and pre-processing are applied, then measurement precision is improved, but device complexity increases
Solution Approach 1:
Data normalization and pre-processing are applied as preliminary actions before model training. Historical laboratory data undergoes normalization to reduce variability and pre-processing to clean and structure the data. This upfront preparation improves measurement precision by ensuring high-quality input data for the predictive models, while the complexity is concentrated in the initial setup phase rather than in ongoing operations
Data Source
AI summary
The present invention refers to a predictive method based upon machine learning for the development of composites for tyre tread compounds.