Title:
Cross-sector Variability in AI-Based Hazard Detection: A Comparative Analysis of YOLO Implementation Across Construction, Manufacturing, and Healthcare Environments

Authors:
Dr. Robb Shawe, USA

Abstract:

This study examines cross-sectoral variability in the performance and implementation of YOLO-based hazard-detection systems across construction, manufacturing, and healthcare environments in New York State. Using a comparative analytical approach, the study evaluates differences in detection accuracy, time-to-detection, and PPE compliance, as well as contextual factors influencing system effectiveness. Findings indicate that, while YOLO-based systems consistently outperform manual inspection across all sectors, variability arises from environmental complexity, operational dynamics, and task-specific conditions. Construction environments exhibited the greatest variability due to dynamic conditions, while manufacturing environments showed more stable performance. Healthcare settings posed unique challenges for contextual interpretation and compliance monitoring. The results highlight the importance of context-sensitive implementation strategies and reinforce the need for socio-technical alignment in AI-enabled safety systems. The study contributes to the advancement of Safety 4.0 by emphasizing the role of environmental and organizational factors in shaping AI performance.

DOI: http://dx.doi.org/10.51505/ijaemr.2026.11317