A SYSTEM OF WITHIN-DAY DYNAMIC DEMAND AND ASSIGNMENT MODELS FOR SCHEDULED INTERCITY SERVICES Ennio Cascetta Luigi Biggiero Department of Transportation Engineering, University of Naples "Federico II" Agostino Nuzzolo SINTRA S.r.l. Consulting Francesco Russo Department of Systems Engineering, University of Reggio Calabria ~TRODUCTION Demand and assignment models adopted to simulate intercity transportation services are usually based on a static representation of the system. This implies that scheduled transport services, such as rail or airplane, are modelled as service "lines", following an approach substantially similar to the one used to represent urban transit services. Consistently, mode and service type choice models include in their utility functions only time averaged attributes of the schedule such as frequencies and average travel times and prices. The static approach, though adequate for the long-term planning of new infrastructures, is not satisfactory to support operational, marketing oriented decisions such as the service schedule and time-varying prices. This paper presents a system of within-day dynamic mode/service choice and assignment models which is used as the modelling basis for a large Decision Support System for the operational planning of rail services named SASM, recently developed by the Italian Railways company "Ferrovie dello Stato". The first part of the paper describes the demand model based on a tree-logit service/run/class choice model explicitly taking into account desired departure/arrival time; the supply model based on a diachronic (spatial-temporal) network representation of scheduled services; the assignment model which is a Stochastic Network Loading (SNL) model based on explicit (diaehronic) path enumeration. The second part of the paper describes shortly the basic functionalities of the DSS and some applications of the model system to the Torino-Venezia rail corridor. 1. THE MODEL SYSTEM 1.1. The Demand Model The demand-side of the model system is designed to simulate the impacts of variations in operational service variables (e.g. timetables, travel times, prices) on relevant demand dimensions which include primary mode, service type, run and class. The set of these choices can be seen as the equivalent of "route" choice on the multimodal,