ABSTRACT
In recent years transit agencies have been increasingly supported by information technology and telematics to achieve more advanced public transportation systems (APTS), improving real-time transit operations control and supplying real-time traveller information. Operations control variables concern travel speeds and delays, route and link schedule adherence, as well as numbers of passengers on board and waiting at stops. Traveller information variables in the case of individual information (e.g., about specifi c origin-destination and departure time) mainly concern path travel time components (access, waiting, on-board, transfer, egress) and costs, as well as on-board crowding at the single vehicle level. Therefore, real-time operations control and traveller information require real-time estimation and short-term prediction of transit vehicle travel times, on-board loads and other state variables representing system functioning. These variables are obtained through transit simulation systems, where a major role is played by dynamic path choice models, which are able to reproduce the traveller’s choice behavior so as to replicate, in dynamic transit assignment models, the way in which travellers propagate on the network, according to transit service confi gurations.
