Automated Methods to Estimate Duty Cycle for Evaluating and Reducing the Risk of Upper Limb Disorders and Fatigue

Research Trainee: Oguz Akkas, PhD Student, Dept. of Industrial and Systems Engineering at the University of Wisconsin at Madison

Faculty Sponsor: Robert G. Radwin, PhD, Professor of Industrial and Systems Engineering and Biomedical Engineering at the University of Wisconsin at Madison

This project will investigate automated methods for estimating duty cycle, which is the ratio of movement or exertion time to cycle time. Duty cycle is important for evaluating localized muscle fatigue and is a component of the American Conference of Government Industrial Hygienists (ACGIH) Threshold Limit Value (TLV) for Hand Activity Level (HAL). Chen et al. (2013) recently demonstrated a semi-automatic marker-less hand tracking algorithm for measuring hand kinematics from conventional video to calculate HAL. Akkas et al. (2014) developed an equation to estimate HAL by using hand motion speed and duty cycle. This pilot research project will explore how duty cycle can be estimated for the HAL equation based solely on the kinematics data obtained using marker-less tracking, utilizing videos of laboratory tasks and actual workers performing numerous industrial tasks. This research can ultimately lead to fully automatic evaluation of the TLV for HAL.

The research will first identify exertions for each specific tasks using multimedia video task analysis single frame-by-frame video analysis. The frame-by-frame analysis will be used for obtaining ground truth data for each task. Upper extremity kinematics including displacement, speed, and acceleration will be measured from the video and used to create an algorithm to predict the duty cycle of each task. Prediction results of duty cycle will be compared against ground truth.

Oguz Akkas is currently a PhD student at the University of Wisconsin-Madison, and is a project assistant in the Occupational Ergonomics and Biomechanics Laboratory. He has extensive education and training in assessing risk of musculoskeletal disorders in the workplace. He intends to pursue a course of graduate study that will apply these skills for solving problems in occupational safety and health.

Publications resulting from this project:
Akkas O, Lee CH, Hu YH, Yen TY, Radwin RG. Measuring Elemental Time and Duty Cycle Using Automated Video Processing. Ergonomics. 2016:1-12. doi: 10.1080/00140139.2016.1146347.