Using big data for last mile performance evaluation: An accessibility-based approach

Si Chen,Xiang Yan,Haozhi Pan,Brian Deal

DOI: https://doi.org/10.1016/j.tbs.2021.06.003

IF: 5.85

2021-10-01

Travel Behaviour and Society

Abstract:ObjectiveThe 'last mile' of public transportation describes the final leg of a transit journey. This paper brings an accessibility-based approach to last-mile performance evaluation at the parcel level by measuring desirable destinations reasonably reachable through accessible transit stations.MethodsAn accessibility-based last mile performance measure is developed to include destinations, attractiveness, and transit connectivity. Google Map API data is used to identify potential destinations and further evaluate their popularity.ResultsThe range of last-mile performance scores was 0–91.7954%, with a mean of 49.82% and a standard deviation of 61.61%, indicating high variation of the last mile performance in Chicago. Last mile problem areas in Chicago tend to cluster in more economically challenged areas. Income levels and housing sale price had positive relationships with last mile performance scores.ConclusionAreas with low last-mile accessibility performance are more likely to cluster in communities that have greater economic disadvantages, lower density, and less mixed land use, implying spatial inequality and disparity in overall accessibility.PracticeThe described approach can inform the development of strategic planning interventions to improve transit connectivity and to reduce the disparity of transit connectivity and accessibility across neighborhoods.ImplicationsThe evaluation of last mile connectivity needs to consider both access to transit station and access to potential destinations. The last mile performance score is highly influenced by neighborhood socioeconomic status.

transportation