Modelling inter-urban transport flows in Italy: A comparison between neural network analysis and logit analysis

In the present paper a modal split problem is analysed by means of two competing statistical models, the traditional logit model and the new technique for information processing, viz. the feedforward neural network model. This study aims to explore the modal split between rail and road transport modes in Italy in relation to the introduction of a new technological innovation, the new High-Speed Train. The paper is sub-divided into two major parts. The first part offers some general considerations on the use of neural networks in the light of the increasing number of empirical applications in the specific area of transport economics. The second part describes the Italian case study by using the two above mentioned statistical models. The results highlights the fact that the two adopted models, although methodologically different, are both able to provide a reasonable spatial forecasting of the phenomenon studied. In particular, the neural network model turns out to have a slightly better performance, even though there are still critical problems inherent in its application.