Since 12 years ramp metering is applied in The Netherlands with currently 30 installations in operation at several locations throughout the country. Given the discussion, like in the United States, on (the lack of) effectiveness and detrimental effects in the urban areas, the decision was taken to do a study on the network effects of ramp metering. In this study a synthetic approach is chosen, that is: the study deals with an artificial network of a motorway with two on-ramps and two intersections in an urban network. A fully actuated controller controls the intersections, so they adapt automatically to any traffic load in the urban area. The on-ramp controllers act on the basis of the so called RWS-strategy which is a demand capacity approach as normally applied in the United States, but in real life other strategies like ALINEA and FUZZY control have been assessed. In this paper four different scenarios will be tested. No-control, control at the downstream on-ramp only, control at the upstream on-ramp only and control at both on-ramps. The traffic streams in the network are adjusted according to rules with respect to equalising travel time along different routes. First the authors will describe the tool used for this study: FLEXSYT. They will then describe the network and the approach being followed. At the end they will draw the conclusions. It will be shown that a positive effect on capacity of the motorway, also is very beneficial for the urban network. However, a major effect of ramp metering could be the rerouting of vehicles to their original routes, travelling less kilometres in the urban area.
Abstract