Effects of dynamic speed limits on a Dutch freeway

The paper presents the effects of Dynamic Speed Limit control (DSL) in a 3 lane section on the A13 freeway, in Rotterdam. The main objective is to contrast the latest empirical findings in Barcelona [1], where the mainline metering capability (i.e. gating, meaning mainline flow restriction) due to sub-critical speed limits (e.g. down to 40 km/h) was questioned. Moreover, according to the Barcelona results, the validity of the current fundamental diagram models accounting for DSL could be strongly compromised. This investigation takes advantage of the huge empirical traffic database recorded by the Dutch government and of DSL strategies being present on most of their freeways. Data is treated in order to identify stationary periods of traffic. A method, reproducible elsewhere, and computerized into an algorithm has been developed to this end. The fundamental diagram is used as a graphical tool to assess the results. The main findings do not contradict the Barcelona observations. Flows of 1850 veh/h per lane were steadily observed for the sub-critical speed limit of 50 km/h. The reactive nature of the DSL-control implemented in the Netherlands, implies that very low speed limits do not affect a wide range of traffic states. This hampered the possibility of extending this conclusion. Further research with less limiting DSL control strategies would be necessary to clarify the extent of the phenomenon considered.Peer Reviewe