Assessing the Impact of AR-Assisted Warnings on Roadway Workers' Stress Under Different Workload Conditions
Fatemeh Banani Ardecani, Amit Kumar, Omidreza Shoghli
TL;DR
Rising fatalities in roadway work zones motivate enhanced safety systems. The study uses a high-fidelity VR environment to evaluate how AR-assisted multisensory warnings affect workers' physiological stress under light and moderate workloads. Using wearable sensors, the researchers monitor $PPG$, $EDA$, and $ST$, along with HRV metrics such as $RMSSD$, $NN50$, $pNN50$, and $HF-HRV$, to quantify stress during multisensory alerts. The findings indicate that moderate workload elevates physiological stress (higher HR and lower HRV) while AR warnings remain feasible, supporting a noninvasive, real-time stress monitoring framework to improve safety and worker well-being in construction sites.
Abstract
Recent data from the Federal Highway Administration highlights an alarming increase in fatalities and injuries in roadway work zones, emphasizing the need for enhanced worker safety measures. This study addresses this concern by evaluating stress levels among roadway workers equipped with AR-assisted multi-sensory warning technology during varying work intensities. The research leverages a high-fidelity Virtual Reality environment to simulate realistic work scenarios, enabling safe evaluation of high-risk situations. Unlike previous studies focusing on external factors, this research investigates the internal physiological impact on workers. Utilizing wearable sensors, the study collected physiological data, including photoplethysmography (PPG), electrodermal activity (EDA), and skin temperature (ST), to assess stress levels continuously and non-invasively. Our findings from 18 participants reveal significant differences between light- and medium-intensity activities in heart rate variability metrics. These metrics commonly used to assess autonomic nervous system function and stress levels, included mean heart rate, NN50, pNN50, and HF-HRV. By examining the relationship between AR-enabled warnings, work intensity, and stress levels, the study contributes to enhancing worker safety and well-being. The proposed methodology offers potential for active stress monitoring in the field, contributing to enhanced safety practices and worker productivity in construction sites. By providing real-time physiological data, this approach enables informed stress management and more effective hazard warning systems in roadway work zones. This research bridges a gap in understanding the physiological impacts of AR-assisted warnings on roadway workers. The insights gained from this study can inform future safety interventions and guide the development of more effective warning systems.