Route choice models used in the construction of dynamic travel matrices

One of the most important stages of traffic modeling in cities is mapping traveler behavior related to the decision of choosing a route. This problem is particularly significant in dense urban networks, where effective traffic management is often associated with determining so-called dynamic travel matrices calculated based on instantaneous traffic flow values on network segments. The article identifies the fundamental problems associated with modeling decision-making processes regarding route choice in urban networks. Against this background, a classification of the models used is presented. The special role of information in the decision-making process is also emphasized.INTRODUCTION
Modeling travel behavior and accurately mapping the decisions made by travelers is a very complex issue and has been the subject of intensive research for many years. The choice of route by individual travelers depends on many factors, both objective and subjective, and their identification requires appropriate datasets containing, among other things, data on spatial development of the area, traffic conditions in the transport network, organization and management systems, and the availability of specific types of information. These data can be characterized by varying levels of detail, which has a significant impact on the accuracy and reliability of the mapping. Furthermore, more complex models require taking into account the variable and random nature of travelers’ decision-making processes and phenomena occurring in the network.
Route choice models for travelers are based on determining the relationships between demand and supply characteristics in the transport system. To describe the variability of demand over time, so-called dynamic travel matrices are used, which are updated at specific intervals [13, 14, 15]. Such matrices can be used in systems for effective traffic management when, in critical situations (e.g., accidents, road works, breakdowns, congestion), knowledge of the origins and destinations of individual road users becomes necessary. Only on this basis can optimal detour routes be determined in urban networks.
In dense urban networks, modeling traveler behavior is much more complex. These networks are characterized by, among other things, (…)

Route choice models used in the construction of dynamic travel matrices – PDF file for download