Impact of operating speed measures on traffic crashes: Annual and daily level models for rural two-lane and rural multilane roadways

A significant association between crash severity and operating speed is known to exist. However, the findings related to the speed-crash association are inconclusive. Some studies found that higher speed is associated with a higher number of crashes, whereas other studies found the opposite result. Some of the critical issues in this research problem result from study design, the definition of operating speed measures, types and granularity of operating speed measures, spatial correlation, and design standards of different roadway facilities. The road safety profession will benefit greatly from informative research on the impact of vehicle operating speed, roadway design elements, and traffic volume on crash outcomes. This study investigated the speed-crash association in both annual and daily level datasets to determine how roadway characteristics interact with various speed measures to impact the likelihood of crash occurrences on both annual and daily levels. For annual models, the average operating speed is positively associated with both fatal and injury and property damage only (PDO) crashes. However, for daily models, this association is mostly negative and insignificant. The standard deviation of operating speed is positively associated with crash occurrences for both daily and annual models. The findings of this study can provide additional insights into the speed-crash association literature.