Sustainable building’s energy management with artificial intelligence and machine learning: A decadal scoping review (2014–2024)

Artificial Intelligence (AI) and Machine Learning (ML) are transforming energy management in buildings by enabling data-driven prediction, optimization, and real-time control. This scoping review synthesized the state of the art in this field over the past decade (2014-2024), based on a structured search of peer-reviewed studies. The review classified AI/ML applications, according to features like: forecasting horizons, algorithm families (classical ML, deep learning, hybrid/ensemble, reinforcement learning), building categories (residential, commercial, healthcare), and integration levels (standalone, IoT-enabled, microgrid-enabled). The analysis of 100+ comparative studies showed that ensemble and hybrid methods frequently outperform single algorithm methods, however, performance gains vary considerably depending on data quality, forecasting horizon, and building operational characteristics. Key challenges remain in data standardization, balance between model complexity and interpretability, privacy, and scalability from individual buildings to urban districts. Emerging opportunities were highlighted at the intersection of AI with digital twins, IoT, and transfer learning, which support small data contexts and real time adaptation. By bridging technical advances with practical challenges, this work aims at informing the scientific community, working towards intelligent, sustainable, and efficient energy management in buildings, by supporting with a clear classification of AI/ML techniques and their most promising applications across different building types.