The growing urban population, along with the complexity of our transportation systems, is posing serious challenges to urban mobility. In this context, the low-cost monitoring of large transportation networks through different technologies is essential to improve the efficiency of transportation services. Traffic data can be obtained from several sources such as smart cards, license plate recognition from video-based technologies, Media Access Control (MAC) addresses captured from Bluetooth and/or Wifi devices, magnetic sensors, Light Detection And Ranging (LIDAR) systems or global positioning systems (GPS). All these data collected by cities or companies carry important information on traffic, congestion, average speeds, variations of speed on a network level and also about driving behaviours on a driver level. The general objective of this session is to present problems at the interface of statistics and transportation engineering to tackle statistical and methodological challenges related to the analysis of urban-mobility data collected from three sources: GPS and telematics data, traffic counting devices and smart card systems. Commonly, such high-dimensional data are collected over both space and time. This poses important modelling challenges related to the large volume of data, the inherent spatial and temporal correlations, the semi-continuity of the road travel network structure, and the high frequency of missing values.