Study Details
Study Title: Evaluating the Impact of Raising Speed Limit on Urban Freeways Using Mixed Effects Negative Binomial Regression
Authors: Kwayu et al.
Publication Date:JAN, 2018
Abstract: Numerous statistical approaches have been used in establishing the safety impacts associated with raising posted speed limit, especially on highways. These approaches range from simple naïve before and after study to more advanced statistical approaches which control for exposure and other confounding factors. In this study, mixed effects negative binomial regression was used in quantifying the changes in fatal, incapacitating and non-incapacitating (KAB) crashes, total 6 crashes and road departure crashes after raising the posted speed limit in some of Michigan urban freeways. This method was preferred as it offers the ability to control for individual random effects which vary across the freeway corridors, intra-cluster correlation of crashes between corridors or segments that are nested in the same corridor, overdispersion in crash data and time effect. The importance of these factors was demonstrated by comparing the estimation results of mixed effects negative binomial model and standard negative binomial model. The standard negative binomial model underestimated the impact of speed limit on KAB and total crashes while compensating for the missing variables, namely time effect and random effects. The results from mixed effects negative binomial regression showed a net increase in KAB crashes, total crashes and road departure crashes after raising the speed limit. The effect of raising speed limit was more pronounced on curved freeway segments compared to straight freeway segments. Therefore, the design standards for horizontal curve, vertical curve and other geometric features should be thoroughly assessed to ensure that they meet required standards for the proposed speed limit changes.
Study Citation: Kwayu, K. M., V. Kwigizile, J. Oh, H. Cho, and M. Bott. "Evaluating the Impact of Raising Speed Limit on Urban Freeways Using Mixed Effects Negative Binomial Regression". Presented at the 97th Annual Meeting of the Transportation Research Board, Paper No. 18-04815, Washington, D.C., (2018).
CMFs Associated With This Study
Category: Speed management
Countermeasure: Raise posted speed
| CMF | CRF(%) | Quality | Crash Type | Crash Severity | Roadway Type | Area Type |
|---|---|---|---|---|---|---|
| 1.081 | -8.1 |  | All | All | Principal Arterial Other Freeways and Expressways | Urban |
| 1.247 | -24.7 | | All | All | Principal Arterial Other Freeways and Expressways | Urban |
| 1.058 | -5.8 | | All | All | Principal Arterial Other Freeways and Expressways | Urban |
| 1.132 | -13.2 | | Run off road | All | Principal Arterial Other Freeways and Expressways | Urban |
| 1.21 | -21 | | Run off road | All | Principal Arterial Other Freeways and Expressways | Urban |
| 1.102 | -10.2 |  | All | K,A,B | Principal Arterial Other Freeways and Expressways | Urban |
