Existing deterministic approaches to real-time tracking of construction workers for safety purposes provide alerts for specific incidents that are isolated in space and time, once they have occurred. In contrast, this paper presents a statistical model that can support a more dynamic form of safety control, by utilizing real-time tracking data to control the exposure of construction workers to safety risks that accumulate and change over time. The model addresses risks that are the result of concurrent activities on the construction site, and provides proactive alerts in case of an increasing risk exposure for a worker or crew. Statistical zones that are related to medium risk areas on the site are defined in the model. A number of statistical rules are then used to identify deviations from a predefined maximum allowable risk exposure for workers located in the statistical zones. The model can thus prevent potential accidents from occurring, without unnecessarily affecting the efficiency of the activities carried out on site. Laboratory tests of the model were carried out, using a Wi-Fi-based RTLS. The results of the tests demonstrate that the model can identify an excessive exposure to risk for workers, support an initial analysis of the causes for the excessive risk exposure, and compensate for errors in the RTLS measurements.
- Construction safety
- Dynamic risk control
- Statistical Process
ASJC Scopus subject areas
- Control and Systems Engineering
- Civil and Structural Engineering
- Building and Construction