Application of the MFD: Subnetwork Perimeter Control

As welfare increases in a region or a country, it is common for the mobility of its inhabitants to increase as well, most often resulting in an increase of the number of vehicles. At increasing welfare we also see that cities are growing and the number of activities employed in these cities grow as well. This then results in more urban traffic and harder to reach city centres. To alleviate these newfound traffic problems, improvements in infrastructure and traffic measures, such as traffic lights and roundabouts, are made. Although urban traffic flow improves by these measures, it is mostly just a matter of time before new traffic problems arise, as a result of the increasing number of vehicles. As the space for construction of new infrastructure becomes more and more limited, it is the traffic that needs to be managed in order to increase city mobility. Using network-wide traffic management strategies, traffic can be distributed more evenly over the network or in time, increasing the overall network performance.

It has recently been proposed that the performance of a complete network, or a part thereof (subnetwork) can be graphically shown using aggregated, averaged data for flow and density. This resulting graph is the so-called macroscopic fundamental diagram. In this diagram aggregated data for the flow and densities are used, creating a relation between the average flow (production) and the average density (accumulation) in a network. The resulting relation is concave, which means that for any network there is an accumulation at which the production is maximal. When this optimum accumulation is transgressed, spillback occurs and the network gets into a state of gridlock. It is than most evident that when one wants to optimise traffic flow in a network, one should strive to keep the amount of vehicles in the (sub)network at this optimal level. To this end, the principal of perimeter control has been introduced, which aims to regulate the inflow at the boundary of a (sub)network in such a way that the number of vehicles in it is at this optimum. The best way to achieve this, is to use the available traffic signals and prolong the red times for inbound traffic and increase green time for outbound traffic.

The objective of this research is to create suitable perimeter control strategies using information of the macroscopic fundamental diagram. With a case study of The Hague, it will be investigated if these strategies can be properly implemented in an existing city and to see if perimeter control can indeed improve traffic flow.


Research Topic: