LAUSR

Abstract The objective of the current study was to evaluate traffic and geometric features and their influences on the safety performance of roundabouts by developing suitable crash modification factors (CMFs). The cross-sectional method can be applied as an alternative method to estimate the CMFs when before-and-after studies are impractical to apply, e.g., lack of data from the period after implementing treatments. To accomplish the study objective, CMFs were derived from generalised linear models (GLMs), i.e., negative binomial (NB) regression, using data collected on regional roundabouts in Toowoomba City, Australia. Six years of crash data from 49 roundabouts included all recorded crashes as well as traffic and geometric features for the entire roundabouts. Several candidate models were developed using the GLMs. Five models were selected based on statistical significance, goodness-of-fit (GOF) measures, and cumulative residual (CURE) analysis. The results show that increasing the number of entry lanes, entry width, entry radius, traffic volume, circulatory roadway width, weaving width, and speed limit have positive effects on roundabout safety. On the other hand, increasing the number of legs, number of exit lanes, exit width, exit radius, weaving length, central island diameters, and presence of fixed object on a central island have negative effects on roundabout safety. The study shows that quantifying the risk factors can support road safety stakeholders to identify safety improvements at roundabouts more effectively and efficiently.