Since about 1986 one-lane roundabouts have been introduced in the Netherlands and have become a great success. Circulating traffic has priority, and the design is compact with radial connected legs, leading to low speeds. Safety benefits have been substantial and capacity has been higher than expected. A problem was to choose the right design for cyclists; either no special facility, a cycle lane, or a separate cycle path. In the latter case there has been much controversy regarding whether to give cyclists priority or not. Capacity can be determined using a simple rule or, in critical cases, with a computerised calculation method - the Roundabout Explorer. Two-lane roundabouts are considered to be the next step but only a few have been installed. Their design and analysis of their safety and operation is in its infancy. Theoretical design rules and preliminary safety survey have been carried out. The general idea is that two-lane roundabouts will not be such a big success as the one-lane type, but certainly deserve more study and application. Quite a few designs have been proposed, but to date they have only been evaluated using paper and pencil methods. The research agenda contains proposals to develop a microscopic simulation model for traffic using a roundabout and a consistent procedure to decide which type of intersection - uncontrolled, controlled, or roundabout - is most suitable in a certain situation. (A)
Abstract