Nexen advances use of AI in tire inspection, production processes

Seoul – Nexen Tire has claimed an “industry first” with the deployment of a new artificial intelligence (AI) system in its tire production processes.

The Korean tire maker has developed an AI-based automated tire product inspection platform which, it said, can be applied across factories and/or items of equipment.

With the new system, Nexen said it has extended the application of AI in the industry from the development process to manufacturing processes.

The system is applied to “non-destructive inspection equipment,” using machine vision technology to gather and analyse visual information.

This includes 'X-ray inspection equipment' for detecting structural defects and 'laser interferometry inspection equipment' for detecting air bubbles.

The AI tool assists in interpreting inspection images, which previously relied on human visual assessment, according to Nexen.

The technology, it continued, has achieved a defect-detection reproducibility rate of up to 99.96%, picking up “minute defects that human inspectors might overlook.”

This enhances the quality of finished products, which conventionally undergo “hundreds of tests during post-production inspection” to avoid defects.

Nexen also claims to have enhanced the system’s practicality by automating “the entire process of AI training and application.”

Here, Nexen as collaborated with machine-learning specialist Neurocle Inc. and PDS Solution, which focuses on tire design, analysis and data processing.

Furthermore, Nexen has applied ‘machine learning operations’ (MLOps) technology, which optimises and automates the entire lifecycle of AI models.

This, it explained, includes “selective data-collection for AI training, AI model training, model validation, actual application, and post-deployment monitoring.”

According to Nexen, adopting the approach has reduced the time needed to create a deep-learning model from six to 12 months to “as few as two days”.

Being “platform-based”, the system can be rolled out across many factories and across various parts of the production process.

Indeed, Nexen said it used the system, trained with data from another factory, for the “early stabilisation of systems introduced in other factories.”