Analytical models can help policy-makers to identify the most promising directions for changes in transport policy, before any costly experiments are launched. Various models exist at national level, but these are not consistent between Member States. Therefore, models are needed to support the development of integrated policy at a European level. Moreover, modelling has to cover all modes as well as cross-border trade, in order to address the objectives of the Common Transport Policy.
The main goal of the STREAMS project was to develop a prototype model to predict the demand for transport across the EU and its broad distribution across the transport network. In addition, STREAMS aimed to produce a 'reference scenario' for the year 2020, providing an initial forecast for selected transport variables.
STREAMS developed and validated a model covering all travel within EU Member States and across its borders (all modes, passenger and freight transport, including short trips and walking). There are two main modules - one determines transport demand, and the other assigns the demand to the transport network. The model is disaggregated at the NUTS2 level used in transport statistics, dividing the EU into some 200 zones and thousands of links representing physical connections and terminals (such as roads and airports). The model is therefore suitable for forecasting aggregate transport statistics for the EU as a whole.
The model was calibrated for a base year of 1994. The resulting output data were shown to match well with various sets of observed transport statistics. Projections for 2020 were reasonably comparable with those from other smaller-scale models.
The reference scenario for 2020 gave the following key results for passenger transport:
- an increase in passenger-kilometres of 50% from 1994 levels, due to increases in trip distances (primarily) and in the number of trips;
- a 5% overall increase in the modal share of the car (e.g. due to increases in business, shopping and leisure trips), largely at the expense of walking and cycling;
- significant increases in the modal shares of rail and air for longer trips.
For freight, the 2020 projection indicated:
- a doubling of freight tonne-km, and a 60% increase in tonnage moved, with the greatest increases in the international sector (especially exports/imports to the EU);
- only minor changes in overall modal shares, with trucks gaining share in the intra-EU freight market but losing share to rail transport in the international market.
The STREAMS model can make rough estimates of the proportion of the road network that is congested. Between 1994 and 2020, the number of kilometres of 'overloaded' links is predicted to double to 9% of the network.
The STREAMS model has been used by the European Commission to produce traffic forecasts for the whole of the European Union, as part of a pilot Strategic Environmental Assessment of the Trans-European Transport Network. The model has since been developed further in the SCENES project.
The model is best suited to applications which require an overview forecast of transport in the EU or Member States as a whole. For example, it can be used to look at policies aimed at modal shift and demand management, or the impacts of socio-economic change. On the other hand, it is less well suited to studying local policy options or assessing the effect of a particular piece of infrastructure.