Tire Tread Compound Prediction Without Repeated Lab Testing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining the viscoelastic and processability properties of rubber compounds for tire treads require extensive laboratory testing, leading to increased lead times, costs, and variability in data, making it inefficient for product development and validation.
Innovation Solution
A predictive method using machine learning implemented on an electronic computer to simulate laboratory tests, generating a database of existing recipes and their properties, and applying data augmentation and transformation techniques to enhance the accuracy of processability property predictions for new recipes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If extensive laboratory testing is performed to determine viscoelastic and processability properties of rubber compounds, then measurement precision and reliability are improved, but loss of time and productivity deteriorate
Solution Approach 1:
The patent applies preliminary action by performing data augmentation and transformation on historical recipe data before model training, preparing the dataset in advance to enable rapid predictions without requiring time-consuming physical tests for each new compound formulation
Solution Approach 2:
The patent creates a digital copy of the physical testing process through machine learning models that predict viscoelastic and processability properties from recipe compositions, replacing the need for physical laboratory testing while maintaining prediction accuracy
2Manufacturing precision
If iterative experimental campaigns are conducted to optimize composite formulation, then manufacturing precision is improved, but loss of time and productivity deteriorate
Solution Approach 1:
The patent implements feedback by using the predicted properties from the machine learning model to guide formulation optimization, allowing rapid iteration and adjustment of recipe compositions based on predicted performance without requiring repeated physical testing cycles
Solution Approach 2:
The patent applies parameter changes by transforming the input recipe parameters and ingredient quantities through the machine learning model to predict output viscoelastic and processability properties, enabling virtual optimization of formulation parameters
3Measurement precision
If physical laboratory tests are performed for compound validation, then measurement precision is improved, but recurring costs increase
Solution Approach 1:
The patent replaces expensive physical laboratory tests with a digital copy through machine learning predictions, maintaining the ability to validate compound properties while eliminating recurring costs associated with raw materials, labor, and facility usage for physical testing
4Measurement precision
If data augmentation and transformation are applied to enhance prediction accuracy, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent replaces complex physical testing machinery and procedures with a computational system that uses data augmentation and transformation algorithms, substituting mechanical and manual processes with automated software-based solutions that enhance predictive precision without requiring additional physical equipment
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.