Raising speed limits on freeways has an impact on both crashes and operational speeds as documented in many previous studies. This study aimed at determining the impacts of changing speed limits on crash frequency and severity, as well as operational speeds in Michigan. Selected sites where the speed limit changed were used to study the impacts of such changes. The analyses included similar sites that did not have a change in the speed limit. Crash data and speed data were collected and analyzed to achieve the goals of the study. A more robust statistical analysis method using the mixed effects negative binomial regression model was used to model crashes and quantify the impact of changing speed limits while controlling for other confounding factors. Speed analysis included estimation of parameters such as the mean, the 85th percentile, the 50th percentile and the standard deviation of speeds in the time before changing the speed limits and compared them with those estimated for the period after changing the speed limits. (Author/publisher)
Abstract