Safety evaluation of raised speed limits on Kansas freeways

Abstract

Setting an appropriate speed limit is necessary to provide safe and efficient traffic operation for all road users. It must also be acceptable to the public and enforceable by police. Lower-than-required speed limits may make the majority of drivers non-compliant, whereas higher speed limits may increase the number of crashes together with related injuries and fatalities. In 2011, the speed limit on a number of freeway segments in the state of Kansas increased from 70 to 75 miles per hour. The objective of this study is to evaluate the safety effects of freeway sections affected by speed limit change in Kansas. Sections where the speed limit changed from 70 mph to 75 mph and other comparable sections where the speed limit remained at 70 mph without any change, were identified. Details of the crashes by severity level for three years before (2008-2010) and three years after (2012-2014) the speed limit change were collected using the state crash database. In order to get a general understanding, characteristics of crashes such as night time versus daytime, number of trucks involved, weather conditions, driver’s gender, and such were considered. Furthermore, several crash contributory causes were also investigated before and after the speed limit changes. In order to evaluate the safety situation, three methods were utilized: 1) Empirical Bayes (EB) observational before-and-after studies, 2) before-and-after method with comparison group, and 3) cross-sectional method using the Negative Binomial (NB) regression model. The evaluation was conducted to see if the speed limit change has caused an increase in total crashes or fatal and injury crashes. In regard to speed analysis, the t-test was applied to see whether significant increases in the 85th percentile speed were observed between before-and-after conditions. Since the sample size was large, the Kolmogorov-Smirnov (K-S) test was also conducted to see if there was any difference between two sets of speed data distributions in the before period compared to the after period. By performing the EB before-and-after study, it was seen that total crashes increased by 16 percent, while using the before-and-after method with the comparison group showed around 27 percent increase in total crashes. Total crash increases were statistically significant according to the EB method, and the before-and-after method with the comparison group. On the other hand, fatal and injury crashes increased by 35 percent based on the before-and-after with the comparison group after the speed limit change. This increase was statistically significant, but the EB method results indicated no increase for fatal and injury crashes when the speed limit was raised to 75 mph. Further, cross-sectional study results showed the speed limit increase had a greatly significant effect on total crashes, an increase of 25 percent; and it was also significant for fatal and injury crashes with those increasing by 62 percent, which is the highest amount of increase compared to the EB method and the before-and-after method with the comparison group. The t-test results showed a five mph increase in the speed limit caused statistically significant increase in 85th percentile speed for the sections affected by speed limit change. However, there was also an increase for the sections without a speed limit change, but this was due to large sample sizes of speed data in the before-and-after period. The K-S test results also depicted the speed distribution of treated sites during the after period was different than the before period. Considering night time crashes versus daytime crashes also showed that there was 1 percent statistically significant increase in the night time crashes compared to daytime crashes after the speed limit increase. There were also more trucks involved in crashes after the speed limit increase. Further, the percentage of adverse weather crashes also decreased by 6 percent for treated sites and 18 percent for non-treated sites. Overall, understanding these results will help with future speed limit adjustments on freeways.