Smart Tyre Quality Assessment through AI for a Sustainable Automotive Future

Tyres play a crucial role in vehicle safety, efficiency, and performance, but the challenge for detecting defects in tyres remains a critical challenge in the automotive maintenance and manufacturing process. Regarding the shift in the global industry towards a smarter and more sustainable production system, there is a great opportunity to enhance safety efficiency using deep learning on the classification of defects in tyres. The deep learning models of ResNet50 and MobileNet architectures have been applied to classify tyres into defective and non- defective categories. High-precision identification of tyre defects is crucial to the elimination of accidents and the operation of vehicles at their maximum potential, especially in line with Sustainable Development Goal 9 regarding industry, innovation, and infrastructure: resiliency infrastructure and innovative infrastructures. Among those models, it is seen that MobileNet outperformed ResNet50 at an accuracy of 95.68% which was greater than the 76.06% achieved by ResNet50. MobileNet architecture is lightweight because of the depthwise separable convolutions, hence performing efficiently in binary classification tasks, especially when the resources are severely limited to either computational power or a smaller dataset. With this, MobileNet is a better contender for real-time industrial applications like automated tyre inspection systems that demand faster processing rates and higher accuracy levels. Further, this paper makes a very extensive analysis of transfer learning models in the classification of tyre quality, further opening new possibilities to support SDG 12, also called Responsible Consumption and Production, in more effective and automatic processes of tyre manufacturing. Improving accuracy and speed in the defectivity detection of tyres, the work lays out a model for intelligent manufacturing systems and driver safety in one go, which contributes to a more sustainable and secure automotive environment.