A dynamic prizing scheme for a congestion charging zone based on a network fundamental diagram

Abstract

The continuously rising amount of traffic is becoming more and more of a problem that cities have to deal with it, as it causes congestion, noise and air pollution, among other side effects. As a solution approach for these diverse problems, many cities have already decided to install a toll, which is often designed as a congestion-pricing area. With that, they aim to reduce, if not completely prohibit, traffic in the inner cities, and to provide a source of additional income for the city budget. In order to get the maximum traffic benefits out of such a tolling area and to avoid traffic jams within the tolled area, it is necessary to control the system with an efficient algorithm. However, at the moment there are no dynamic, traffic responsive city tolling systems. The major reason is that such controlling mechanisms, especially if they have to represent a whole network, would require a lot of data. This data is not only difficult to collect and manage, particularly if most of the roads inside the network have to be measured, it also makes it hard to design and calibrate the control algorithm. With the help of a network fundamental diagram (NFD), representing the traffic state of the whole network, this data collection and management problem can be solved. This motivates the idea of using an NFD as a basis for controlling the pricing scheme of a city tolling system. To examine the feasibility of this approach, a microscopically simulated test network was created and a NFD for this network was developed. Using this NFD, a desired occupancy value for the whole network is set, and the toll height is calculated from different control functions of the occupancy inside the tolled network compared to the desired occupancy value. This toll height is subsequently included into the route choice algorithm of the microsimulation, shifting the traffic onto the non-tolled roads. In order to assess the different control approaches, evaluation criteria including a comparison of the resulting network occupancy with the desired one, the stability of the price, and the reaction speed are set. Several approaches to determine the prizing based on the fundamental diagram are developed and tested in the simulated network. The development of the traffic flow and network occupation over time inside the tolled network is compared between the various tolling functions in order to identify the best performing one.