An alternative UGM Paradigm to O-D matrices: the FRETURB model

This paper presents an alternative methodological approach to O-D matrix and analyses its validity on the viewpoint of Bonnafous' operability triangle. The proposed model is able to estimate the impacts of urban goods movements in terms of number of vehicles, total travelled distances and road occupancy rates without generating O-D pairs. The originality of the model arises on two main elements. The first is that the modelling unit is neither the trip nor the quantity of goods, as in many literature works, but the number of movements, i.e. the number of pickup and delivery operations, which is found as the main invariant in urban goods movement generation. The second is that it follows a bottom-up approach, i.e., starting from a rich database of urban goods operations and routes, a set of behavioural functions are defined. The paper is organised with the notion of operability triangle: the model must be at the same time coherent, relevant and measurable. This analysis allows to show how the model FRETURB resolves this problem of magic triangle and also how it was resolved up to here by the various forms of modelling of the urban freight in literature.