Image-based activity pattern segmentation using longitudinal data of the German Mobility Panel. (November 2020)

• Record Type: Journal Article
• Authors: von Behren, Sascha; Hilgert, Tim; Kirchner, Sophia; Chlond, Bastian; Vortisch, Peter
• Is Part Of: Transportation research interdisciplinary perspectives. Volume 8(2020)
• Publication Date: 2020-11
• Subjects: Clustering -- Activity pattern -- Visualization -- GraDiV -- German mobility panel -- Germany; Transportation -- Periodicals; 388.05
• DOI: 10.1016/j.trip.2020.100264
• Languages: English
• ISSNs: 2590-1982

Abstract

Highlights: Images of week activity schedules are powerful to cluster mobility patterns. Clustering show seven activity patterns with differences in intra-week stability of activities. Images offer a visual representation of travel behavior with an added value for interpretation. Two activity patterns are influenced by long-distance travel at weekends (7%) or weekdays (3%). Besides Pensioners and Pupils/students, people tend to switch into other patterns between years. Abstract: In this paper, we present an approach to segment people based on a visualization of the longitudinal week activity data from the German Mobility Panel. In order to perform segmentations, different clustering methods are commonly used. Most of the approaches require comprehensive prior knowledge about the input data, e.g., condensing information to cluster-forming variables. As this may influence the method itself, we used images with a high degree of freedom. These images show week activity schedules of people, including all trips and activities with their purposes, modes as well as their duration or their temporal position within the week. Thus, we answer the question whether using only this type of image data as input will produce reasonable clustering results as well. For the clustering, we extracted the images from an existing tool, processed them for the method and finally used them again to select the final cluster solution based on the visual impression of cluster assignments. Our results are meaningful as we identified seven activity patterns (clusters) using this visual validation. The approach is confirmed by the data-based analysis of the cluster solution showing also interpretable key figures for all patterns. Thus, we show an approach taking into account many aspects of travel behavior as an input to clustering, while ensuring the interpretability of solutions. Usually, key figures from the data are used for validation, but this practice may obscure some aspects of the longitudinal data, which are visible when looking on the images as validation.