Research Topic 3: Predicting Fatigue and Fall Risks Among Roofers: A Hybrid Model Combining Computer Vision and Sensor Technologies

Research Description

Diagram of Closed Circuit cameras working with internet of things to provide real time monitoring

Roofing is one of the most hazardous occupations, with a high incidence of falls leading to severe injuries or fatalities. One of the primary causes of these accidents is fatigue and distraction, which can impair a worker’s judgment, reaction time, movement, and balance. This research aims to develop a hybrid model that combines computer vision and sensor technologies to predict fatigue and assess fall risks among roofers in real-time. By integrating these technologies, we can monitor physical movements, posture, and other fatigue indicators to provide timely alerts and interventions, potentially saving lives and reducing injuries.
Computer vision will analyze visual data (joints information) to detect signs of fatigue, such as slowed movements, frequent breaks, or unstable postures. Concurrently, wearable sensors will monitor physiological indicators like heart rate, body temperature, and sweat levels, which can provide insights into a worker’s fatigue levels. By merging data from both sources, the hybrid model will offer a comprehensive view of a roofer’s fatigue status and predict potential fall risks.

REU Research Plan

REU students will be at the forefront of this innovative project. The research will commence with a literature review of existing fatigue detection methods and their efficacy in the construction industry. Students will then collaborate with graduate student mentors and faculty advisors to design and test a prototype system. They will be involved in data collection, where they will monitor participants in controlled environments to gather data. With guidance from mentors, students will develop algorithms to analyze the data and predict fatigue and fall risks. By the end of the research period, students will have the opportunity to present their findings at conferences and contribute to journal publications.

More Details

Keywords: Fatigue prediction, fall risks, roofers, computer vision, sensor technology, wearable sensors, construction safety.

Required Skills: Basic knowledge of construction safety protocols, coding skills, and statistics.

Undergraduate Degrees: Civil Engineering, Construction Science/Management, Industrial Engineering, Computer Science, Occupational Safety.

Potential Faculty Advisors: Dr. Chukwuma Nnaji, Dr. Zhenyu Zhang, Dr. Ashrant Aryal and Dr. Xi Wang.