Do Construction Workers Failing Image-Based Safety Tests Engage in Unsafe Behavior? A VR-Based Human–Robot Collaboration Study

Résumé fourni par la source

Unsafe behaviors among construction workers are a leading cause of site accidents, making their prevention a key strategy for reducing risks. Early identification of unsafe tendencies can help mitigate potential hazards before workers enter human–robot collaboration sites. This study aims to identify two tendencies of unsafe behavior using image-based safety tests and assess on-site unsafe behavior of workers with different tendencies. To ensure feasibility and safety, virtual reality (VR) technology was employed to simulate construction environments. The results of 35 participants revealed a strong correlation between image-based safety test performance and behavioral responses in VR scenarios with physiological indicators, validating the effectiveness of image testing as a low-cost evaluation tool. These findings suggest that image-based safety tests can serve as an initial evaluation method for site admission, enabling the early identification of high-risk workers and the targeted allocation of training resources. This study provides empirical support for integrating image-based safety tests into proactive accident prevention strategies, balancing feasibility with effectiveness.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.