Human fatigue due to repetitive and physically challenging jobs may result in poor performance and a Work-related Musculoskeletal Disorder (WMSD). Thus, the importance of being able to monitor fatigue to implement preventative interventions cannot be overstated. This study was designed to monitor fatigue through the development of a methodology that objectively classifies an individual’s level of fatigue in the workplace by utilizing the motion sensors embedded in smartphones. An experiment consisting of squatting tasks, primarily involving the lower extremity musculature, was conducted with 24 participants using a smartphone attached to their right shank. Using Borg’s Ratings of Perceived Exertion (RPE) to label gait data, we developed machine learning algorithms to classify each individual’s gait into two (no- vs. strong-fatigue), three (no-, medium-, and strong-fatigue) and four (no-, low-, medium-, and strong-fatigue) levels of fatigue, for which accuracy of 91%, 76%, and 61% were obtained, respectively. The outcomes of this study may facilitate the implementation of a proactive approach supporting the continuous monitoring of a worker’s fatigue level, which may subsequently enhance workers’ performance and reduce the risk of WMSDs.
Industrial and Systems Engineering (MS)
Department, Program, or Center
Industrial and Systems Engineering (KGCOE)
Karvekar, Swapnali Babasaheb, "Smartphone-based Human Fatigue Detection in an Industrial Environment Using Gait Analysis" (2019). Thesis. Rochester Institute of Technology. Accessed from
RIT – Main Campus