This thesis studies the safety effects of section control systems on Dutch motorways. Section control is a speed enforcement method that measures the mean speed of vehicles passing a specific road section. When this mean speed is, after measurement correction, above the lower ticket threshold, the driver receives a fine. The detection probability of speed offenders at section control is nearly 100%. Therefore, section control is a very effective measure to prevent speeding and therefore reduce the mean speed and speed variance of traffic. Both mean speed and speed variance are important factors in the occurrence of crashes. The risk a crash risk increases with a higher speed or a larger speed variance. Traffic safety is generally expressed in the number of occurring crashes on the road. Foreign studies already showed promising results concerning the traffic safety effects of a section control system. However, well designed studies are scarce. A well-designed study into the safety effects of section control in the Netherlands has not yet been carried out. The study design for this research is a pre-post study design with a matched comparison group. The section control systems within the scope of this research were statistically tested for significant changes in relation to their matched comparison groups, between the pre- and post-study period on the mean speed, speed variance, crash severity and the sum of these crash severities. The statistical tests included a repeated measures ANOVA for the speed data analysis and the chi-square or Fischer’s exact test and the advanced chi-square test for the crash data analysis. A crash analysis was also done for the proportional distribution of crashes within the treatment group, to test the influence of section control on the surrounding road sections. Section control systems with a peak-hour lane and/or with a variable speed limit were excluded from this research, because these systems pose large difficulties for obtaining comparable road statistical road data. Furthermore, section control systems implemented after the summer of 2012 were excluded from this research too, because both pre- and post-period preferable need a minimal length of one year. As a result, there were seven section control systems on fourteen different road sections within the scope of this research. On eleven of these road sections, section control systems were implemented together with a speed limit reduction. Because a speed limit reduction also influences the traffic safety of a road, an effort has been made to predict the expected number of crashes in the post-period situation with a section control system but without a speed limit reduction. Firstly a speed analysis, based on minute speed data from the MoniCa database, was done to test whether section control is able to reduce the mean speed and speed variance of the traffic significantly. This analysis showed section control decreased the mean speed and the speed variance significantly on roads where section control was activated simultaneously with a speed limit reduction. Compared to their matched comparison groups, the mean speed decreased on average by ± 16% and the speed variance decreased on average by ± 41%. The section control systems implemented in the existing road design did not reduce the mean speed and speed variance significantly compared to their matched comparison groups. Compared to their matched comparison groups, the mean speed decreased on average by ± 3% and the speed variance decreased on average by ± 17%. After the speed analysis it was decided to also perform an crash analysis to study whether the decreased mean speeds and speed variances also provided significant reductions on the number of crashes. This analysis was done with the available registered casualties and property damage only crashes from the BRON database. On roads where section control was activated simultaneously with a speed limit reduction, all crash severities and the sum of these crash severities showed significant reductions. Compared to their matched comparison groups, the number of serious injuries decreased by 37%, the number of slight injuries by 13% and the number of property damage only crashes by 11%. The total number of casualties decreased, compared to their matched comparison groups, by 15% and the total number of casualties plus the property damage only crashes by 12%. The proportional distribution of casualties and crashes on the upstream and downstream road section of section control slightly increased over time, compared to the road sections with section control. On the section control systems implemented in the existing road design, only serious injuries did not obtained a significant reduction. Compared to their matched comparison groups, the number of slight injuries decreased by 34% and the number of property damage only crashes by 4%. The total number of casualties decreased, compared to their matched comparison groups, by 15% and the total number of casualties plus property damage only crashes by 18%. The proportional distribution of casualties and crashes on the upstream and downstream road section of section control changed only slightly and on a different way for every crash severity. After the section control systems with speed limit reduction were corrected for the safety effects of a speed limit reduction, the number of serious injuries decreased, compared to their matched comparison groups, by 15%. The number of slight injuries and the number of property damage only crashes increased instead of decreased. The total number of casualties and the total number of casualties plus property damage only crashes, also show an increase compared to their matched comparison groups. Based on the research question ‘What are the safety effects of section control systems, installed before summer 2012, on motorways in the Netherlands?’ and on the available data and section control systems within the scope of this research, the conclusion can be drawn that section control together with a speed limit reduction of 20 [km/h] improves the traffic safety on the road sections with section control. Based on this research it can also be concluded that only a section control system can improve traffic safety but only based on crash rates and not on mean speed and speed variance. The last conclusion that can be drawn based on this research, is that section control does not influence the traffic safety on the adjacent upstream and downstream road section of section control. (Author/publisher)
Abstract