This work presents the development of an Advanced Runway Health Monitoring System (ARHMS) that combines acoustic sensors with machine learning-based computer vision to enable real-time runway condition assessment. The proposed system addresses critical safety issues such as surface defects, foreign object debris, and structural integrity that are often overlooked during conventional manual inspections. By employing YOLOv8 for accurate object detection and OpenCV for image processing, the prototype was trained on over 600 images from Kaggle datasets and evaluated using a scaled physical runway model. The results demonstrate strong accuracy in hazard detection and highlight the system’s potential for mobile deployment and future predictive capabilities. Key contributions include a hybrid acoustic–vision monitoring framework, analysis of real-world runway incidents, and practical recommendations for preventive maintenance. Despite challenges such as daylight dependency and occasional false positives, this study establishes a promising proof-of-concept for non-intrusive, AI-assisted runway monitoring aimed at significantly enhancing aviation safety.

The results showcase impressive accuracy in hazard detection, with exciting potential for

mobile deployment and advanced predictive capabilities in the future. Key contributions of

this research include the development of a hybrid acoustic–vision workflow, in-depth

analysis of real-life incidents, and actionable recommendations for preventive maintenance.

While we acknowledge challenges such as daylight dependency and the occurrence of false

positives, we also outline a clear roadmap for scaling this system to operational use. This

work not only serves as a compelling proof-of-concept for AI-assisted runway safety but also

underscores the critical need for non-intrusive, real-time monitoring to significantly reduce runway-related accidents.