Labor fatigue is increasingly recognized worldwide as a problem of modern industry, being one of the main causes of occupational accidents (Kant et al, 2003; Sadeghniiat-Haghighi and Yazdi, 2015). The International Labour Organization (ILO) estimates that economic losses from work-related diseases and accidents are significant: about 4% of world GDP (Directorate of Pension Studies, 2015).
The systems currently available use, for the most part, subjective methods, with response bias, and they are ex-post, not in real time, so they do not generate alerts to sudden events. Those that use biosensors are invasive and therefore difficult to implement. In addition, they do not have good predictability, since they do not use advanced Data Analytics, such as Deep Learning or other, nor do they use signals such as Heart Rate Variability and Electrodermal Activity, or environmental variables in an integrated manner.
This project seeks to address the previous problem in the Chilean manufacturing industry by characterizing mental and physical fatigue induced by various work tasks.
Our technological development will allow the industry to:
The focus is the delivery of methodological tools for monitoring and forecasting fatigue that support the management of the manufacturing workforce in different areas, such as scheduling and reconfiguration of shifts or the prevention of psychosocial illnesses and accidents at work, among others.
The work team is formed by the group of prediction of labor fatigue and integration of sensors of the Department of Industrial Engineering of the University of Chile and the Institute of Complex Systems of Engineering, which has experience in dozens of R&D projects that have required the design of experiments, acquisition, processing and analysis of digital signals, as well as data analysis with statistical models, machine and deep learning in various industrial sectors (transport, manufacturing, health, banking, agriculture, education, advertising, health, smart cities, etc.). In addition, this group has developed projects applied in the public sector, such as the analysis with unsupervised methods of the behavior of the processing of court cases or the automation of medical image diagnoses with deep learning in public hospitals, among others.
Dr. Ángel Jiménez
Dr. Enrique López