Enhancing Construction Hazard Recognition with High-Fidelity Augmented Virtuality

Abstract

Most construction safety management processes rely on the hazard recognition capability of workers. Hazards that remain unidentified can potentially result in catastrophic injuries and illnesses. As such, thorough hazard recognition is fundamentally essential to protect the health and well-being of the construction workforce. Despite its importance, recent research indicates that a large proportion of hazards remain unrecognized, exposing workers to unmitigated risks. Surprisingly, safety research has not adequately focused on developing specialized strategies to develop construction worker competency in hazard recognition. This paper reports a two-year research effort with the following objectives: (1) develop a high-fidelity augmented virtual environment [System for Augmented Virtuality Environment Safety (SAVES)] that helps develop workers’ hazard recognition skill through risk-free learning and immediate feedback; (2) embed cognitive retrieval mnemonics to improve long-term retention of cues for construction hazards; (3) evaluate the effectiveness of the strategy as an intervention on active construction crew by using the multiple baseline testing approach. The first two objectives were accomplished through a combined effort from a panel of 14 subject matter experts and five academic researchers. This was followed by field experiments to test the hypothesis that the experience with SAVES improves the proportion of hazards identified by participants during subsequent field operations. The findings revealed that crews, on average, were able to only identify 46% of hazards prior to the introduction of the intervention, but were able to recognize 77% of hazards in the postintervention phase. This study represents the first endeavor to measure the effectiveness of augmented virtuality and serious gaming in developing hazard signal detection skills in construction field settings.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 281 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.