The use of variable message signs (VMSs) is generally considered to be a powerful tool to influence route choice so as to increase safety and comfort during driving and improve network performance and to make optimum use of available capacity. Assessing whether such effects actually occur when VMSs are installed is a difficult task. This paper summarizes an extensive evaluation study carried out on the installation of 14 new VMSs on the Amsterdam orbital motorway. The study involves a number of issues. First, a networkwide analysis of aggregate performance indicators, such as traffic performance, severity of congestion, and travel time delay, is included. Second, the relation between the messages displayed on the VMS and the travel times experienced by drivers is analyzed. Finally, a stimulus-response analysis to find out how drivers actually respond to the messages is described. The study showed that the use of VMSs has a positive impact on network performance in the Amsterdam freeway system. Total congestion has decreased, and traffic performance has increased. Variation in congestion has decreased as well as variation in average travel speed. This finding implies that travel time has become more reliable and that the traffic flow is more homogeneous. The technical assessment showed that the displayed queue lengths, if interpreted correctly by drivers, are a good measure for expected delay. User response analysis showed that information has a significant effect on route choice. In general, the VMSs were found to improve the efficiency and reliability of the freeway system.

Commuters may react to structural congestion on urban roads and freeways by adjusting their departure time. Single-bottleneck models enable direct insight into the mechanisms of departure time choice. The interaction of user equilibrium departure time choice processes for multiple user classes in continuous time is studied in the situation in which all users have to pass a common bottleneck. The study draws on the bottleneck approach introduced by Vickrey and provides a generic algorithm that solves the departure time choice equilibrium problem given heterogeneous departure time preferences, arbitrary origin-bound and destination-bound rescheduling cost functions, and arbitrary queuing cost functions. For all user classes, the (generalized) queuing costs have a travel duration–dependent component and a travel cost–dependent component. First, a solution is provided for the elastic-demand case. It is shown that the equilibrium departure rates per user class can be calculated directly by calculating the acceptable travel times per user class. Second, a solution is provided for the fixed-demand case. The fixed–demand problem is formulated as a minimization problem that is solved using the Nelder–Mead procedure. The algorithm is illustrated with an example showing the effects of a flat peak-hour road pricing scheme on the departure time choice and total incurred costs per user class.

No abstract

No abstract