Entrepôts, Représentation et Ingénierie des Connaissances
Publications of the ERIC lab


by Year
by Author
by Topic
by Type
- A dynamic bayesian network approach to forecast short-term urban rail passenger flows with incomplete data

Author(s): Roos J.(Corresp.), Gavin G., Bonnevay S.

Conference: European Transport Conference (Barcelone, ES, 2016-10-05)
Proceedings: , vol. p. (2016)


n this paper, we first present a brief state of the art of short-term traffic forecasting. After introducing Bayesian networks and their temporal extension, we then detail our modelling approach. First, we focus on a single RER station with a complete set of ticket validation and count data. Then the application area is extended to an entire metro line, using an incomplete dataset and integrating transport service data in the modelling. Finally, we conclude the paper and discuss the limitations of our approach.