Research | 2013
Effects of Increased U-Turns at Intersections on Divided Facilities and Median Divided Versus Five-Lane Undivided Benefits
Phillips, S., D. Carter, J. Hummer, R. Foyle. Effects of Increased U-Turns at Intersections on Divided Facilities and Median Divided Versus Five-Lane Undivided Benefits, prepared for North Carolina Department of Transportation, FHWA/NC/2004-07 (2004) [See also Carter, D., J. Hummer, R. Foyle, and S. Phillips. Operational and Safety Effects of U-turns at Signalized Intersections, TRR 1912, Transportation Research Board of the National Academies, Washington D.C., 2005, pp. 11-18.]
The choice of midblock left turn treatments is a controversial issue. The two main competitors for midblock left-turn treatment on four-lane arterials are raised medians and two-way left-turn lanes (TWLTL). This research focused on determining the effects of median installation on midblock road segments and the adjacent signalized intersections. The areas of focus were vehicular safety and operational impacts. For the segment safety study, predictive collision models were calibrated using geometric, volume, land use, and collision data for 143 midblock segments.
Analysis showed that collisions were significantly related to cross-section type, average daily traffic (AADT), segment length, predominant land use, and approach density (two-way total). For predominantly residential and industrial land uses, the raised median design was always associated with fewer collisions than the TWLTL design. For predominantly business and office land uses, the raised median design had a safety advantage for low approach densities. For higher driveway densities, the raised median was slightly safer at high traffic volumes and the TWLTL was slightly safer at lower traffic volumes.
The signalized intersection study dealt with the effects of U-turns in exclusive left turn lanes. This included analyses of the safety of U-turns and the operational impacts of U-turns on saturation flow rate. The safety study examined a set of 78 intersections in North Carolina, one-third of which were chosen because they were known to be U-turn problem sites. Although the group of study sites was purposely biased toward sites with high U-turn percentages, the study found that 65 of the 78 sites did not have any collisions involving U-turns in the three-year study period, and the U-turn collisions at the remaining 13 sites ranged from 0.33 to 3.0 collisions per year.
The intersection operational analysis involved measurements of vehicle headways in exclusive left turn lanes at 14 intersections. Regression analysis of U-turn percentage versus saturation flow rate indicates a 1.8% saturation flow rate loss in the left turn lane for every 10% increase in U-turn percentage and an additional 1.5% loss for every 10% U-turns if the U-turning movement is opposed by protected right turn overlap from the cross street.
Overall, this research found that many of the typically cited drawbacks to median-oriented designs are not justified. Raised medians may increase U-turns at adjacent intersections, but this was found to have minimal effects on safety and operational performance. Additionally, raised medians are generally safer than TWLTLs on midblock segments.