In this paper we investigate the impact of several weather conditions on the hourly number of crashes in the Netherlands during 2002. The impact of 17 climatic factors, belonging to the categories wind, temperature, sunshine, precipitation, weather image and visibility is quantified and compared with results from other research. The following could be concluded: an increase in maximum wind gust causes an increase in the number of crashes. Global radiation and sunshine duration both have a significant negative impact on road safety. Of all categorical weather indicators, presence of precipitation had the most significant impact. Moreover, the impact of precipitation duration seemed higher than the amount of precipitation. Finally, the direction of the effect of cloudiness on the number of crashes was also found positive. We applied a regression methodology making use of several distributions of the Poisson family. We tested which distribution fitted the actual observations best and found that, on average, the Negative binomial model performed better than the Poisson model, the Zero-inflated Poisson model and the Zero-inflated negative binomial model.
Abstract