Spatiotemporal deep learning with Digital Twin support for sustainable occupancy management

Introduction The convergence of Deep Learning (DL) and Digital Twin (DT) technologies is transforming smart building management by enabling intelligent, adaptive, and energy-efficient operations. However, most existing occupancy prediction models primarily focus on predictive performance while overlooking broader sustainability considerations such as energy consumption, computational efficiency, and environmental impact. Methods To address this gap, this study proposes a hybrid Convolutional Neural Network–Long Short-Term Memory (CNN-LSTM) framework for occupancy prediction in smart buildings. The model is designed to capture complex spatial and temporal patterns from multivariate building sensor data. In addition to prediction accuracy, a sustainability-aware optimisation strategy is incorporated by evaluating energy consumption, inference latency, and carbon footprint as integral performance metrics. Furthermore, the trained model is integrated within a Digital Twin environment to assess its real-time deployment feasibility and interaction with Building Management System (BMS) operations. Results Experimental evaluation using benchmark occupancy datasets demonstrates that the proposed framework achieves high predictive reliability while maintaining computational efficiency and reduced environmental impact. The Digital Twin-enabled cyber-physical validation confirms the model’s capability to operate effectively in real-time scenarios, supporting adaptive and energy-aware decision-making within smart building environments. Discussion The findings highlight the potential of integrating sustainability-aware deep learning with Digital Twin technology to advance intelligent building management beyond conventional data-centric approaches. By combining predictive intelligence, operational validation, and environmental responsibility, the proposed framework provides a scalable pathway toward resource-efficient and sustainable smart buildings, contributing directly to the United Nations Sustainable Development Goals (SDGs) on Affordable and Clean Energy, Sustainable Cities and Communities, and Climate Action.