Assessing public transport accessibility using GPS data

Abstract

Abstract The transition to sustainable urban mobility emphasises the need for an equitable public transport system that gives users high levels of access to their desired destinations. Critical to this effort is a comprehensive understanding of urban travel patterns and needs. Mobile positioning technologies such as GPS enable the collection of large-scale, detailed data on human mobility, where realised travel behaviour functions as an indicator of urban travel needs. However, GPS data has been underutilised in accessibility studies, either due to unavailability, the conceptual gap between potential and realised travel, or methodological challenges in transforming large-scale datasets into meaningful accessibility indicators. Part of these challenges includes identifying individual trips from continuous GPS data, a process known as trip chain detection. Furthermore, despite accessibility being an inherently dynamic concept, the temporal dimension is often ignored in public transport accessibility studies. The contribution of this paper is twofold. First, we introduce a novel methodological framework for trip chain detection from continuous GPS data without supplementary information. Second, we demonstrate how GPS data can reveal realised travel to help prioritise destinations for accessibility studies, making accessibility metrics more relevant to real-world travel needs. To do so, we incorporate a dynamic accessibility approach to account for the variation in travel time at different times for a comprehensive understanding of public transport accessibility. We model public transport and walking accessibility using General Transit Feed Specification (GTFS) data in a GIS environment. The methodological approach is exemplified in the case of public transport accessibility in Tallinn, Estonia. Our proposed methodology enables transport planners to consider realised travel behaviour and origin-destination locations in public transport-related decision-making to enhance spatiotemporal accessibility of our desired destinations.

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 — 10 citations exist upstream.

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