Behavioural modelling of public transport passengers using big data sources

The aim of the project is to create methods to combine datasets for analysing travel behaviour in public transport systems, hereby enabling a better foundation for decision-making. This will be achieved by using a combination of I) Smart Card Data (Rejsekort), a large-scale big-data source with strong information on spatial, temporal and mobility choices, and II) representative travel survey data (Transportvaneundersøgelsen), which include socio-demographics and more detailed OD-information, but only samples a small subset of travellers.
The new methods will be based on probabilistic data fusion techniques, generative modelling, and spatio-temporal modelling. The joint dataset will be used for detailed analysis of passenger preferences by use of econometric models, including discrete choice and route choice models.
The project is expected to lead to improved predictive transport models, thus ensuring better public transport systems to the benefit of users, operators and society.
---
DFF International Postdoc project