Estimating running speed on urban roads

摘要

This article proposes a statistical model for the prediction of the running speed on urban roads, developed and calibrated on the basis of an experimental survey on a sample of 17 roads localised in two different Italian cities (Parma and Reggio Calabria). The model is designed to determine the average running speed of the vehicles according to different parameters, such as the length of the road link, the road width, the degree of disturbance related to lateral parking and the degree of saturation of the road. A new development brought from the model concerning the formulae developed in the literature is the explicit relationship between the running speed and infrastructure length. After a literature review of cost functions applicable in city environment, the paper presents database that was used, the adopted procedure for specification model and the relative outcomes of statistical validation.