Abstract
Safety hazard identification is an essential method for mitigating non-fatal fall injuries and improving construction workers' safety performance. Current safety hazard identification methods mostly rely on experts' judgment to identify hazards, and thereby they are unable to continuously identify hazards in the diverse and dynamic nature of the construction environment. To identify safety hazards and improve workers' safety performance, a better understanding of the relationship between workers' gait disruption patterns and the presence of a safety hazard is vital. Toward achieving this goal, the objective of this study was to propose a non-invasive approach to examine the feasibility of using workers' gait disruption patterns to identify safety hazards among construction workers. To test the hypothesis of this study, ten asymptomatic participants conducted four simulated experiments in a laboratory setting to examine the feasibility of the proposed approach. The participants' gait disruption patterns were collected using a wearable insole pressure system to compute five gait variability parameters and a gait abnormality based on ground reaction force (GRF) deviation. The results showed that workers' gait disruption patterns measured by the gait abnormality based on GRF deviation values are highly correlated with the location of hazards, which indicated that workers' gait disruption patterns in hazardous areas are more distinct than non-hazardous areas. The findings of this study can serve as the basis for developing a non-intrusive and automated wearable insole pressure system that uses workers' gait disruption patterns as a useful data source to enable safety manager to identify safety hazards in construction.