Towards autonomous cloud-based close call data management for construction equipment safety

Construction accounts for up to 25% of all occupational fatalities. About half are being struck by and caught-in/between objects or vehicles, resulting from insufficient or delayed detection of pedestrian workers. Research has proposed detecting visibility-related close calls in order to alarm the involved personnel and prevent negative consequences at the earliest possible time. This work has three objectives: First, a comprehensive synthesis on close call reporting processes and individual technologies. Second, a system focusing on (a) autonomous close call data generation from real-time proactive proximity detection and alerting technology and (b) cloud-based data processing and visualization in building information models at run time. Test results demonstrate that the developed system reaches purposes for decision making in safety management beyond the scope of existing detection and alarming devices. Third, guiding future research by outlining the role of proximity detection and alarming technology in developing autonomous safety systems for construction. [Display omitted] •Struck-by and caught-in/between accidents remain leading causes of fatal injuries.•Frequent worker-machine interactions demand research for autonomous safety systems.•Understanding and use of emerging technology for proactive assistance is limited.•A cloud-based close call data management based on automated reporting is proposed.•Model-based close call data visualization allows right-time personalized feedback.