Automatic cave-in safety risk identification in construction excavation Conference

Wang, J, Pradhananga, N, Teizer, J. (2014). Automatic cave-in safety risk identification in construction excavation . 130-139. 10.1061/9780784413517.0014

cited authors

  • Wang, J; Pradhananga, N; Teizer, J


  • Safety continues to be among the top issues in the construction industry after experiencing 738 fatalities in the United States in 2011. Among all construction operations, excavation is one of the most hazardous because of its possible cave-ins, contact with objects and equipment, bad air, and so on. Until today, most safety inspectors still are inspecting the site manually, making the inspection time consuming and error prone. This paper presents a method that automatically identifies cave-in safety risks in construction excavation. It first extracts relevant safety rules from OHSA standards and industrial best practices. Then it collects a set of point cloud data of a construction site under excavation using laser scanning, registering, and cleaning the point cloud data afterward. Finally, it develops an automated identification algorithm on the basis of those rules and applies the algorithm to the data to identify potential cave-in risks by analyzing geometrical properties. An experimental trial also is conducted in this paper, and results show that the method identifies those cave-in risks successfully. The presented method actively monitors the fast changing situations of construction sites under excavation and helps inspectors and project managers make good safety decisions, preventing accidents and fatalities. © 2014 American Society of Civil Engineers.

publication date

  • January 1, 2014

Digital Object Identifier (DOI)

start page

  • 130

end page

  • 139