Comparisons to previous intersection treatment

Exhibit 5-9 shows the crash frequencies (average annual crashes per roundabout) experienced at eleven intersections in the U.S. that were converted to roundabouts. As the exhibit shows, both types of roundabouts showed a reduction in both injury and property-damage crashes after installation of a roundabout. It should be noted that due to the small size of the data sample, the only result that is statistically significant is the injury crash reduction for small and moderate roundabouts.

Exhibit 5-9. Average annual crash frequencies at 11 U.S. intersections converted to roundabouts.

Compared to results from Australia, France, and the United Kingdom, these crash frequencies are quite high. Annual crash frequencies in France, Australia, and United Kingdom of 0.15, 0.6, and 3.31 injury crashes per roundabout, respectively, have been reported (1, 10). The reader should note that the UK has many high-volume, multilane roundabouts.

In spite of the higher frequencies, injury crash rates, which account for traffic volume exposure, are significantly lower at U.S. roundabout sites. In a recent study of eight single-lane roundabouts in Maryland and Florida, the injury crash rate was found to be 0.08 crashes per million entering vehicles (5). By comparison, the injury crash rate was reported to be 0.045 crashes per million entering vehicles in France and 0.275 crashes per million entering vehicles in the United Kingdom (1, 10).

Experiences in the United States show a reduction in crashes after building a roundabout of about 37 percent for all crashes and 51 percent for injury crashes. These values correspond with international studies with much larger sample sizes, as shown in Exhibit 5-10.

Exhibit 5-10. Mean crash reductions in various countries.

The findings of these studies show that injury crashes are reduced more dramatically than crashes involving property damage only. This again is in part due to the configuration of roundabouts, which eliminates severe crashes such as left turn, head-on, and right angle collisions. Most of these studies also show that crash reduction in rural areas is much higher than in urban areas.

Note that the geometry of many studied sites may not necessarily conform to good roundabout design. Improved design principles, such as an emphasis on achieving consistent speeds, may result in better safety performance. It should also be noted that these crash reductions are generally for sites where roundabouts were selected to replace problem intersections. Therefore, they do not necessarily represent a universal safety comparison with all other intersection types.

Collisions at roundabouts tend to be less severe than at conventional intersections. Most crashes reported at roundabouts are a result of drivers failing to yield on entry, referred to as entering-circulating crashes. In addition, rear-end collisions and single vehicle crashes have been reported in many studies. Exhibit 5-11 shows the percentage of the three main crash types reported in different countries.

Exhibit 5-11. Reported proportions of major crash types at roundabouts.
Scroll to Top