Supplementary information files for "Predictive automation of window ventilation in Green Buildings: a data‐driven framework for integrating IoT sensors, machine learning, and occupant behavior modeling"

Supplementary files for article "Predictive automation of window ventilation in Green Buildings: a data‐driven framework for integrating IoT sensors, machine learning, and occupant behavior modeling"<br><br>This paper presents an Internet of Things (IoT)–enabled automated control framework for optimizing residential window ventilation systems. The system combines sensors, AI prediction, and automated motors to control ventilation based on environmental conditions and human behaviors. Unlike traditional fixed rule–based systems, our framework uses machine learning trained on 1.2 million data points to predict and automatically adjust ventilation beforehand. The modular automation system features message queuing telemetry transport (MQTT) and building automation and control network (BACnet) communication protocols for seamless integration with existing building management systems (BMSs), fail‐safe mechanisms for operational reliability, and mobile override capabilities preserving user autonomy. A 1‐month pilot study across 50 suburban Chinese households achieved an 18% reduction in heating, ventilation, and air‐conditioning (HVAC) energy use. The framework is modular and scalable, working with current building systems. It includes technical specs, integration methods, and security protections. While behavioral insights inform control logic, the core innovation lies in the automation architecture. This study advances intelligent building systems by fusing human‐centric design with robust automation engineering.<br><br>© The Author(s), CC-BY 4.0