Worker Fatigue Monitoring

Our joint work with Profs. Qi Zhu, John Rogers, and Jian Cao on “Wearable network for multilevel physical fatigue prediction in manufacturing workers” has been online in PNAS Nexus. The work has also been featured in AAAS EurekAlert! and Popular Science.

Our research enhances occupational health by improving ergonomics and managing fatigue among manufacturing workers. Leveraging advanced multi-modal wearable sensors and lightweight machine learning algorithms, it enables continuous, real-time fatigue monitoring. The study addresses limitations in adaptive sensing technologies and explores complex biomarker-fatigue relationships. Data from 43 participants in diverse manufacturing tasks reveal insights such as the impact of non-dominant arm kinetics on fatigue, and the role of body mass, age, and gender. The research also highlights the significance of physiological signs in fatigue perception and confirms that fatigue characteristics are highly personalizable, with better prediction performance for users whose data was included in training