(08/2012 – 03/2015)

The objective of SEMAPHORE was to provide transportation planning and forecasting systems with knowledge about individual mobility behavior extracted from cell phone communication data. This was achieved by statistical inference algorithms which semantically analyze and enrich cell phone trajectories by combining them with additional data sources and background knowledge. Mobility behavioral patterns were discovered and exploited for inference of missing data and statistical projection. Translation of the data into the domain ontology of travel demand modelling and suitable interfaces allowed interoperability with the transportation planning software PTV VISUM™. This opened up cell phone data as a novel and rich source of information for transportation planning, which was shown by empirical evaluation to have a huge potential to improve the quality of the traffic models and to reduce the costs for mobility data surveys.