Machine Learning-Driven Predictive Maintenance Data

This paper presents a comprehensive analysis of machine learning applications in predictive maintenance for IoT-enabled smart buildings. Machine learning algorithms enable real-time fault detection and predictive analytics in building operations. IoT sensor networks provide continuous monitoring of HVAC, lighting, and mechanical systems. The integration achieves 8-19% energy reduction and 93-98% fault detection accuracy [1]. Digital twin technologies enhance system modelling and performance optimisation. The study analyses 34 peer-reviewed publications from 2019 to 2025. Implementation challenges include data quality, system integration, and computational requirements. Future research directions focus on explainable AI, federated learning, and edge computing applications.