Ergonomics practice involves the identification, and control, of risk factors resulting from physically demanding work, including repetitive motions, forceful exertions, and awkward postures. However, most traditional tools available for identifying these risk factors are limited to characterizing the mean or peak postures, forces, or repetitions, and do not account for interactions between these factors. In addition, the tools often do not account for the development of fatigue. Research has shown that the stressors associated with physically demanding work can result in physical fatigue, which can subsequently lead to decreased productivity, changes in work performance, and increased injury risk (Kumar 2001, Côté et al. 2002, Selen, Beek, and van Dieën 2007).
Reduction of injury risk is dependent on identifying fatigue so that interventions can be put in place. Since the traditional ergonomics assessment tools fail to capture risk factor interactions or fatigue, alternative methods are needed. In a recent survey of manufacturing workers, it was identified that workplace fatigue is typically accompanied by changes in work postures and pace (Lu et al. Under Review). Therefore, methods that capture any changes in worker behavior may provide reliable indicators for monitoring. How to capture this information and what should be done with the data, must first be considered. Wearable sensors provide a means for data collection of any necessary work parameter. Data can be collected continuously from a number of workers for relatively low cost across a range of conditions. However, to date there is a lack of consensus on how the collected data can be integrated for monitoring and individualized risk detection. For effective technological approaches to fatigue measurements, it is essential that the system can predict fatigue (before there is a detrimental productivity and safety impact), measure and monitor fatigue in the operational environment, and intervene when deficits are identified. In addition, considerations of individualized baseline conditions are necessary, but often ignored in a population-based approach to safety. This presentation will include an introduction to the development and examination of the use of sensors for data-driven occupational exposure assessments in work environments. In addition, it will include a description of predictive models based on analytics of the wearable sensor data.
