Uncertainty analysis of accident notification time and emergency medical service response time in work zone traffic accidents.

ABSTRACT

OBJECTIVE: Taking into account the uncertainty caused by exogenous factors, the accident notification time (ANT) and emergency medical service (EMS) response time were modeled as 2 random variables following the lognormal distribution. METHOD: Their mean values and standard deviations were respectively formulated as the functions of environmental variables including crash time, road type, weekend, holiday, light condition, weather, and work zone type. Work zone traffic accident data from the Fatality Analysis Report System between 2002 and 2009 were utilized to determine the distributions of the ANT and the EMS arrival time in the United States. A mixed logistic regression model, taking into account the uncertainty associated with the ANT and the EMS response time, was developed to estimate the risk of death. RESULTS: The results showed that the uncertainty of the ANT was primarily influenced by crash time and road type, whereas the uncertainty of EMS response time is greatly affected by road type, weather, and light conditions. In addition, work zone accidents occurring during a holiday and in poor light conditions were found to be statistically associated with a longer mean ANT and longer EMS response time. The results also show that shortening the ANT was a more effective approach in reducing the risk of death than the EMS response time in work zones. CONCLUSIONS: To shorten the ANT and the EMS response time, work zone activities are suggested to be undertaken during non-holidays, during the daytime, and in good weather and light conditions.