Project R-2479

Improving travel time estimation in the context of congestion by taking into account human decision behavior and road network dynamics (Research)

Traffic jams are a major problem in Belgium. They cause an annual economic loss of approximately 114 million euro. Estimating a reliable travel time can be problematic, both for end users and for policy makers. One of the reasons is that current travel time estimation does not sufficiently take human decision behavior in case of congestion into account, although it has a clear impact. A second reason is that congestion is not static since road networks constitute dynamic and stochastic systems. If a methodology can be developed that is able to account for both issues and predict travel time in a more reliable way, it would be possible to determine both structural and incidental queuing locations and quantify their corresponding time loss profiles, which would be beneficial both for end users and for policy makers. A suitable way of describing systems in a state of congestion is the use of micro-simulation models, which describe the behavior of individual decision makers and the interaction between the system level and the individual. While the model is able to account for some simple decision behavior, it is certainly not able yet to account for behavioral adaptation in case of congestion.

01 October 2010 - 30 September 2014