Disaggregation of aggregate GPS-based cycling data – How to enrich commercial cycling data sets for detailed cycling behaviour analysis

Abstract

In order to investigate cycling behaviour, planners and researchers are increasingly using disaggregate data such as GPS data. However, disaggregate data on cycling behaviour is not available for most cities. At first sight, the data collected by (sport) app providers like Strava could help fill the data gap, as they are available for most cities around the globe. Due to data privacy reasons however, this data is usually aggregated before it is sold commercially by data providers. To use the data for detailed analysis, this article presents a multi-step disaggregation approach to synthesise single routes from aggregate data sets. The approach requires aggregate origin-destination data of cycling demand as the primary input. A double-constrained routing algorithm is subsequently used to derive single bicycle routes from this data. This disaggregate route data can then be enriched with further attribute data and can thereafter be used to estimate bicycle route choice models. This article presents the approach developed as well as a proof of concept using a case study. It further illustrates how the results can estimate a route choice model for a case study area in Germany. The overall results show that the presented approach could easily be used to disaggregate available aggregate cycling data to investigate cycling behaviour.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 7 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.