Towards railways safety: A systematic review on predictive diagnostics for axle bearings
Jaromír Konecny, Štěpán Ožana, Jan Choutka, Michal Prauzek
Abstract
Axle bearings are critical for safe railway operation. Exposed to high mechanical loads and harsh environmental conditions, their failure can cause serious accidents. This systematic review analyses 75 studies on predictive diagnostics of axle bearings, published between 2014 and 2024. It focuses on methods such as vibration analysis, acoustic emissions, temperature monitoring, and AI-driven fault detection. The strengths and weaknesses of each approach are discussed, with attention to challenges like noise interference, real-world deployment, and the scalability of AI models. The review highlights the need for standardized validation procedures and better integration with predictive maintenance frameworks. Its goal is to support the development of reliable, real-time monitoring systems that enhance safety, reduce costs, and extend bearing life.