Effects of wet weather on drivers' risk accident perception during motorist-following behaviour on two-way two-lane highway

Abstract

Motorist-following behaviour is expected to be influenced by driving environment such as weather conditions. In road accident studies, close or unsafe motorist-following behaviour is considered as one of the main factors of rear-end collisions. Therefore, it is imperative that such a fundamental aspect of motorists’ behaviour is clearly understood so that effective accident mitigation strategies can be formulated. Driven by this requirement, this study examines the effect of wet weather conditions on drivers’ perception of safe following distance and to predict dynamically its effect on potential accident risk. Motorist-following behaviour was analysed based on different types of vehicles driven, i.e. car following car, car following heavy good vehicle (HGV), HGV following car, and HGV following HGV at two types of highway geometry i.e. passing and no-passing zones. Traffic data for more than 200,000 vehicles were collected using automatic traffic counter at 12 selected sites in Johor and Pahang States, Malaysia, during dry and wet weather conditions. Five-minute interval rainfall data were obtained during the study period, from the rain gauge stations located within 2 km of the study sites. Regression techniques were then used to develop empirical models of motorist-following behaviour. The developed models on time to collision (TTC), accident probability index (API) and chosen risk index (CRI) were used to predict and evaluate the effect of wet weather on traffic conflict, accident probability and severity during motorist-following behaviour. The analysis revealed that the shift from dry to wet weather showed an increase in time gap and a reduction in mean speed irrespective of highway geometry and vehicles following category. This is supported by the observed increase of time gap in the range of 12.15% to 17.88% and 7.33% to 17.61%; and the decrease of mean speed in range of 9.23% to 10.74% and 11.44% to 12.14% at passing and no-passing zones, respectively. The results revealed that traffic conflict occurred at lower speeds during wet weather compared to dry weather irrespective of highway geometry and vehicles following category. The analysis showed that the API values for wet weather conditions are lower than the API values for dry weather conditions. Such a result suggests that the wet weather condition increases the accident probability when compared with the dry weather conditions. Based on API, the percentage of vehicles travelling at relatively safe speed during dry weather was found to be 51% and 53 % compared to 25% and 22% during wet weather at passing and no-passing zones, respectively. The results of the analysis also showed that the potential accident severity under wet weather conditions is lower than that under dry weather conditions. This is indicated by lower CRI values during wet weather compared to dry weather. Based on CRI, the percentage of vehicles travelling at relatively safe speeds during wet weather was found to be 29.5% and 27% at passing and no-passing zone, respectively. In summary, all objectives of the study were achieved. The study provides new information that is essential in traffic safety management which includes the effects of wet weather conditions on potential accident risk for motorist-following situation.