Investigating the Relationship between Speed Variability Measures and Road Safety: A Case Study of Residential Urban Streets in the City of Saskatoon

The relationship between vehicle operating speed levels and road safety (i.e., collision frequency) has been the focus of several research studies in recent years. However, more research is needed to understand the relationship between speed variability measures and safety. It is, in fact, important to understand whether vehicles travelling slower or faster than the mean speed of traffic are more often involved in collisions than vehicles travelling at a speed close to the mean speed.

In this study, three speed variability measures based on field free-flow speeds collected for 150 residential urban segments in the City of Saskatoon (Canada) were estimated and their association to collision frequency was explored. The three speed variability measures considered were standard deviation, variance, and coefficient of variation of speeds. Along with speed data, other data related to roadway and traffic features of segments were included in the data collection process.

The coefficient of variation (CV) of speeds was found to be the speed variability measure more associated to crash frequency and, therefore, a path analysis model was built where CV was employed as a mediator variable which is predicted by roadway and traffic characteristics of the sites and can predict crash frequency. A path analysis framework can describe the direct and indirect dependencies among a set of variables. With path analysis, a relationship between an independent and a dependent variable can be direct or mediated by a third factor. The results demonstrated that CV was positively related to crash frequency, i.e., streets with higher speed variability showed lower safety levels, and the relationship was found statistically significant. Moreover, the interrelationship among roadway and traffic factors and crash frequency was analyzed, providing a better understanding of the indirect effect of independent predictors of CV on collision frequency. Overall, the results can be particularly important in the context of using speed-related variables as surrogate measures of safety, which would allow the assessment of safety levels of urban residential streets without waiting for collisions to occur.