AI and Machine Learning-Based Predictive Maintenance and Remaining Useful Life (RUL) Prediction for Railway HVAC Systems

—Railway Heating, Ventilation and Air Conditioning(HVAC) systems are essential for maintaining passenger comfortand suitable environmental conditions inside railway coaches.Failures in HVAC equipment can result in passenger discomfort,service interruptions, increased maintenance costs, and reducedequipment availability. Traditional corrective and time-basedpreventive maintenance approaches may not always identifyequipment degradation before failure. Predictive maintenanceprovides an alternative by continuously monitoring equipmentcondition and using data-driven techniques to identify potentialfailures in advance.Artificial Intelligence (AI) and Machine Learning (ML) techniques provide effective approaches for analysing HVAC telemetry, detecting abnormal operating conditions, assessing equipment health, and predicting future degradation. One importantapplication is Remaining Useful Life (RUL) prediction, whichestimates the amount of useful operating time remaining beforean equipment unit reaches a defined failure or maintenance condition. This review examines the application of AI and ML techniques for predictive maintenance and RUL prediction in railway HVAC systems. The review discusses commonly monitoredHVACparameters, machine learning approaches, RUL predictionmethods, predictive maintenance workflows, challenges, researchgaps, and future research directions. The study highlights theimportance of integrating condition monitoring, fault detection,health assessment, RUL prediction, and maintenance planninginto a unified railway HVAC maintenance framework.