Prediction of slaughterhouse workers' RULA scores and knife edge using low-cost inertial measurement sensor units and machine learning algorithms

Abstract
The high prevalence of work-related musculoskeletal disorders (WRMSDs) has been a concern in the meatprocessing industry, owing to the manual nature of the work and the high upper-limb and neck exposure to movements that can lead to WRMSD. The ability to perform an accurate and fast assessment of WRMSDs remains a challenge in industrial environments. Most assessment methodologies rely on standard survey-based methods, which are time- and labor-intensive. In this paper, we present an application of inertial measurement units (IMUs) to measure human activity, and the use of artificial intelligence and machine learning techniques to perform task classification and ergonomic assessments in workplace settings. We present the results obtained by using simple low-cost IMUs worn on slaughterhouse worker wrists to capture information on their movements. We describe the use of this information to detect the risk factors of the wrists/hands that can lead to WRMSDs. The results indicate that by using low-cost IMU-based sensors on the wrists of slaughterhouse workers, we can accurately classify the sharpness of the knife and predict the worker RULA score.
Description
Keywords
Work-related musculoskeletal disorders, Machine learning, Sensors, RULA scores, Slaughterhouse workers, MUSCULOSKELETAL DISORDERS, SHARPNESS
Citation