Video Analytics for Workplace Safety: Red Zone Monitoring Through AI
Marco Memmoli, Lorenzo Corallo, Pierluigi Nunzi · 2025
This paper presents the design and deployment of an AI-based system, Video Analytics for Workplace Safety, intended to enhance operational safety aboard offshore drilling vessels. The system addresses monitoring of Red Zone, which is designated high-risk area on the drill floor where the potential for injury due to interaction with moving equipment is especially critical. Continuous and reliable supervision of red zones is essential for the prevention of accidents and the maintenance of safe operating conditions. Ensuring safety in offshore environments presents a distinctive set of challenges, these include the presence of harsh and dynamic environmental conditions, the complexity of drilling operations, and the inherently hazardous nature of the tasks performed. While traditional HSE procedures, such as scheduled site inspections, direct human supervision, toolbox talks, etc. are essential components of safety management systems, their effectiveness is sometimes constrained by limitations, including operator fatigue, delayed detection of unsafe behaviors, and an inability to ensure continuous coverage of all critical areas. Till now, vessels have relied on passive video surveillance systems that merely record footage without enabling real-time analysis or intervention. As such, these systems offer limited value for proactive risk mitigation. The VAWS solution introduces a paradigm shift by embedding computer vision algorithms, enabling automated detection of safety violations, including unauthorized Red Zone entry.