Best Paper Award
COMP Alumna Lai Sin Yi and Dr Mohammed, Receive Best Paper Award at ICWSNUCA-2026 for AI-Driven Construction Safety Research
We are proud to announce that our alumna (Class of 2025), LAI Sin-Yi Sage, and our Senior Lecturer, Dr MOHAMMED Aquil Mirza, have been honoured with the Best Paper Award at the International Conference on Wireless Sensor Networks, Ubiquitous Computing and Applications 2026 (ICWSNUCA-2026). The conference was held from 19-21 February 2026, in Hyderabad, India.
Their award-winning paper, titled "Smart Construction Safety Platform: Integrating AI and IoT for Proactive Hazard Management," addresses the critical safety challenges and rising fatality rates in the construction industry. By leveraging cutting-edge Artificial Intelligence and Internet of Things (IoT) technologies, the team developed a "one-stop" intelligent platform designed to automate site inspections and enhance worker protection.
Abstract:
This paper develops a one-stop intelligent construction site safety management platform using artificial intelligence and Internet of Things technologies to address crucial safety concerns in the construction industry. The construction sector is facing serious safety issues and rising numbers of fatalities and it is a highest-risk sector of occupational hazards always ranked but in terms of seriousness in safety challenges in managing workplaces. The difficulty for construction safety inspecting manually and the need for advanced technology were serious safety issues. The system combines real-time PPE (Personal Protective Equipment) monitoring through computer vision with environmental hazard detection using IoT sensors. The research compared four models of YOLO object detection algorithms (v8, v9, v10, v11) for PPE detection, with YOLOv11 showing better performance in accuracy and speed. The system can detect safety gear (helmets, gloves, vests, masks) and compute a safety compliance index while also monitoring environmental parameters such as temperature, humidity, air quality, and gas levels using attached sensors. Real-time alerting, historical trending of data, and anomaly detection capabilities provide safety managers with a powerful suite of monitoring capabilities through a web-based portal. The convergence of these technologies has tremendous potential to reduce workplace accidents by automating the proactive identification of hazards.