Nexen Tire America has rolled out an advanced, AI-driven automated system for tire product inspections, marking a significant leap in manufacturing quality control. The system uses machine vision technology to enable non-destructive inspections, replacing the reliance on human visual assessments. This technology leverages specialized equipment, including X-ray inspection tools designed to identify structural defects and laser interferometry inspection tools, such as Shearography, to detect air bubbles within the tire structure.
Nexen Tire explained that the AI system plays a key role in analyzing inspection images, enhancing the accuracy and efficiency of defect detection. The new system boasts an impressive reproducibility rate of defect detection, reaching up to 99.96%.
To further improve the system’s functionality, Nexen collaborated with Neurocell Inc., a company known for its automation solutions in machine learning (AutoML), and PDS Solution Inc., experts in tire design and data processing. This partnership began in the design phase and ensured the seamless integration of AI technology across the entire process.
Nexen also incorporated Machine Learning Operations (MLOps) technology, which optimizes and automates the entire lifecycle of AI models. This includes everything from selective data collection and AI model training to post-deployment monitoring. The approach significantly sped up the development process, reducing the time needed to create a deep learning model from 6-12 months to just two days.
Additionally, the platform-based system allows the AI to be quickly applied to new factories or equipment. According to Nexen, data from the factory where the system was first implemented helped stabilize the AI models when they were introduced into other production facilities.
This cutting-edge approach is set to revolutionize Nexen Tire’s inspection processes, improving both speed and precision in tire quality control.