Assessing crash propensity in work zone queueing events

Work zones are essential for road maintenance and expansion, but lead to reduced roadway capacity, contributing to congestion and safety concerns. This study evaluates crash propensity during slow and stop queuing events in work zones compared with free-flow conditions to better incorporate safety into lane-closure planning. Data from Iowa's Traffic Critical Projects were analyzed. Traffic sensor data were used to monitor occupancy and speed, establishing thresholds to classify slow and stop queuing events, and crash data were used to identify incidents occurring within active work zones. Four scenarios were evaluated to examine different temporal and spatial crash associations with queuing events. A binary logistic regression model was developed to predict crash occurrence based on variables such as event duration, work zone length, speed limit, time of day, and event type. Results indicate that work zones on high-speed roads, longer work zones, and events occurring in the afternoon are associated with a higher crash risk. Stop events consistently showed nearly twice the crash risk of free-flow conditions, regardless of duration. For slow events, crashes are more likely when the event lasts longer than 27 min, suggesting an increased risk due to prolonged queuing. These findings demonstrate that crash propensity in work zones is influenced by roadway, work zone, and queuing characteristics, and support integrating safety metrics into mobility-based lane closure planning tools to enable proactive safety management.