Dynamic assignment models based on the availability of on-line traffic information

Ref. prof. Riccardo Minciardi


The objective of this research activity/thesis is that of developing new models capable of representing the dynamic behavior of drivers in a congested urban traffic network, in the assumption that significant information is available to each driver, concerning the present traffic conditions and possibly the incoming ones.
In the literature concerning dynamic traffic assignment, the assumptions about the availability of on-line information are generally quite restrictive. Instead, the development and the application of technologies able to support bi-directional communication between car (or truck) drivers and a network center determine the necessity to consider new models in which the behavior of the drivers is carefully represented. In particular, the network center is assumed to receive, possibly through area concentrators, the information relevant to each single vehicle (including its final destination), to elaborate this information (possibly though the use of traffic dynamic models), and to provide it individually to the drivers.
The result of this activity should be the development of new (dynamic) models for route choice, based on the User Equilibrium principle, but in time-varying conditions, and on the basis of the availability of detailed information of interest for the single user.


Traffic models, dynamic traffic assignment, user equilibrium

Required Background:

no previous experience in vehicle traffic model is needed; basic knowledge in system theory and optimization is required. Some experience in computer programming (general purpose languages, like, for instance C++)

Collaborations with external company: The development of the research requires the interaction with important companies within the framework of a 3-years Project funded by the Italian Ministry of University and Research.