Use of Roundabouts as Alternatives to All-Way-Stop Controls

This report evaluates the performance of modern roundabouts in Delaware. Delaware Department of Transportation is planning to install modern roundabouts to replace some other types of intersection control. The primary purpose of this report is to obtain Delaware-specific parameters for design of the modern roundabout such as the critical headway and follow-up headway. Three modern roundabouts in Delaware and two in Maryland were the basis of this study. Maryland is one of the leading states in the use of modern roundabouts in the United States. The roundabouts in Maryland were included to enable some comparisons to be made between the critical headway and follow-up headway obtained from each state. Comparing the parameters obtained in Maryland to a similar research carried out in summer of 2005, it became apparent that the critical headway had reduced. While a statistically rigorous approach was not used a decrease in critical headway significantly impacts the performance of the roundabout. The capacity increases and queue lengths and delay will decrease. It stands to reason that as drivers get more and more comfortable driving the roundabouts, they will accept smaller but safe gaps to merge into the roundabout thus reducing the critical headway. Computer simulations were used to assess the capacity (v/c ratios), delay, and queue lengths. The simulations were based on HCM, aaSIDRA, and the NCHRP approach described in Report 572. Preliminary results indicated that the roundabouts are performing at an acceptable level of service. The capacity at Rehoboth however is exceeded leading to long queue lengths and delays from the simulation. This is the result based on the volumes generated with Delaware Department of Transportation (DelDOT) Penninsula Model. No significant delays or queuing were observed at the site during data collection. Crash analysis was conducted based on incidents reported before construction of the roundabout and after construction where appropriate. Sufficient data is not available to make sound conclusions on safety. As time elapses sufficient post roundabout data can be collected for a statistically accurate estimate. There was virtually no pedestrian activity at all the locations studied except Rehoboth. Vehicle-pedestrian (and bicyclist) observations did not reveal any alarming issues. No excessive delays imposed by one group on the other were observed. Motorists yielded as needed. A few pedestrians at this location were not using the crosswalks provided to cross the streets. A few crossings were being initiated at the yield line which imposed further delay on motorist waiting to merge. No audible pedestrian signals (APS) were on site to assist any blind pedestrians that may cross the streets at this location. It appears there is a push to make this provision mandatory and future roundabouts may have to provide APS. These observations from this research will enhance the planning and design of roundabouts in Delaware. Continued research will enable a more complete and reliable assessment of modern roundabouts in the state.