Author(s)

L. D’Acierno, M. Gallo & B. Montella

Abstract

Urban road networks are generally shared by at least three different kinds of vehicles: private cars, buses and freight vehicles. These vehicle categories influence their travel times reciprocally (the crossed congestion phenomenon). Hence, in order to simulate the performance of urban road networks, it is worth developing a multimodal assignment model that takes into account the effects of reciprocal influences among vehicles on user travel choices (at least on mode and path choices). In this paper, we extend the model proposed in a previous paper to the case of urban freight distribution simulation, considering crossed congestion among all considered transportation systems. The proposed model is tested on a real dimension network adopting three algorithmic approaches (external, internal and hyper-network) for its solution and comparing their performances in terms of calculation times and result reliability. Finally, we discuss the effects of crossed congestion on freight path choices and the influences of freight vehicle presence on (road and transit) user choices. Keywords: multimodal traffic assignment, crossed congestion, urban freight distribution, non-separable cost functions. 1 Introduction Generally, a multimodal assignment problem has to be adopted when it is necessary to estimate user flows on different transportation systems under two assumptions: that modal split, as well path choice, is affected by network costs

Keywords

multimodal traffic assignment, crossed congestion, urban freight distribution, non-separable cost functions.