Effects of Demographic Factors for Fatigue Detection in Manufacturing

Abstract

Over the years, advancement in automation technology is allowed the increased integration of humans and machines in a manufacturing environment, these days fewer humans. The use of Knowledge-based Systems in improving and converting human overall performance has been restrained in truth because of a lack of expertise of the way an individual’s overall performance deteriorates with fatigue buildup, which may range from employees to the work environment. As a result, the performance benefits of increased automation in a manufacturing environment, as well as the impact of human factors, must be taken into account. To predict fatigue in physically demanding tasks, this study takes a data-driven strategy. The influence of demographic characteristics, their physical fatigue states, detected workloads, and reactivity to physiological changes are investigated through sensors (Inertial Measurement Unit; IMU and Heart Rate Variability; HRV) in this paper. A framework is established for the selection of key features, machine learning algorithms, and evaluating subjective measures. To attain that, specific application scenarios of the framework are shown, each for different sorts of manufacturing-related tasks.