Predicting the future is pretty complicated, especially if it includes models of traffic flows from thousands of commuters in our partner cities. Therefore, the computer model needs to be fed with datasets, which are as accurate and extensive as possible. In the last months, Thessaloniki, Leuven, Madrid and Regensburg compiled possible sources for such data and summarised them in a deliverable. Various sources were used, such as information related to transport supply and demand, travel times, cartography and socio-demographic data, such as income, tourism or house price statistics. 

A closer look into the category “transport supply” provides an insight on the diversity of information gathered. It includes schedules of public transport, general data related to the transport network, as well as information related to shared mobility, including cars, motorcycles or bikes. Additionally, data from other services, such as MaaS and taxis are included.  

The consortium identified various sources, which are often the councils or transport providers in the respective cities or own consortium partners, such as EMT in Madrid or CERTH-HIT in Thessaloniki. Once the data sources were identified, they were tested on their reliability. The vast majority of sources are from 2019, with small exceptions from the year before. Some cities were able to provide datasets on unique topics, such as CERTH, which shared information on taxis and e-scooters or Regensburg with access to a car sharing programme.  

In case you are tempted to learn more about acquiring and assessing data in our four project cities, check out the full publicly available deliverable here.