Mode choice and spatial distribution in long-distance passenger transport – Does mobile network data deliver similar results to other transportation models?

Abstract

Sound knowledge of travel behavior is important for many findings and decisions in science, business and transport policy. This is not only about spatial distribution, but also and above all about the choice of transport mode. Anonymized mobile network data (MND) can provide valuable information here. This relatively new method is fundamentally superior to other approaches in many respects, but a comprehensive scientific evaluation is still pending. In this context, the present study should help to close research gaps by comparing the results of an MND analysis with the output of an existing transportation model based on System Dynamics. The comparison includes the number of trips within one year, the choice of means of transport and the spatial distribution in long-distance passenger transport within Germany (for all trips with a distance of 50 km or more). We found that the results for travel frequency and modal split were very similar at high aggregation (total annual values for the study area). Also, for distance categories (with an interval of 100 km each) the deviations were very small. However, there were significant differences for origin-destination pairs. In order to be able to make a well-founded comparison here, more in-depth research is necessary in the future.