Analyzing travel behavior differences across population groups: An explainable machine learning approach with big mobility data

Yingrui Zhao and Kathleen Stewart

Journal of Transport Geography, 2025, vol. 128, issue C

Abstract: Understanding differences in travel behavior across population groups is fundamental for fostering more comprehensive transportation systems. Big mobile device location data offers new opportunities for larger-scale finer-grained spatial analyses of human mobility. In this study, we analyzed over 3 million vehicle trips in Maryland over a two-month period in 2018 to explore travel behavior differences across income and racial groups. We employed Random Forest models combined with SHapley Additive exPlanations (SHAP) to interpret how trip distribution, distance, and duration varied with demographic, socioeconomic, and built-environment features. The findings revealed important socio-spatial differences in travel behavior. Racial differences in mobility were associated with population density and indicated the presence of possible mobility segregation. Specifically, trips related to home tracts that were majority Hispanic and Black were concentrated in higher-density urban areas, while trips related to home tracts that were majority White were more dispersed across the state. Trip distance and duration models showed higher variation across income, age, and racial groups. Tracts with lower median incomes and a higher percentage of older age groups were associated with shorter trips, and tracts with a higher percentage of White population were associated with longer trip distances, particularly in rural census tracts. Additional county-level analysis showed that tracts with a higher percentage of Black population in counties with higher socio-economic status were also associated with longer travel distances. These findings demonstrate the potential of mobile device location data to identify travel behavior differences and reveal important mobility patterns, including mobility segregation, which can help guide local transportation planners in developing transportation systems that improve or expand the activity spaces of different population groups.

Keywords: Travel behavior; Mobile device data; Explainable machine learning; Trip distance; Human mobility (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0966692325002595

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:jotrge:v:128:y:2025:i:c:s0966692325002595

DOI: 10.1016/j.jtrangeo.2025.104368

Access Statistics for this article

Journal of Transport Geography is currently edited by Frank Witlox

More articles in Journal of Transport Geography from Elsevier
Bibliographic data for series maintained by Catherine Liu ().