A Data-driven Method for Identifying Potential Zones for Airport Shuttle Bus Services

Keywords: airport shuttles, zonal transit service, demand analysis, spatial clustering
Session Type: Virtual Paper Abstract
Day: Tuesday
Session Start / End Time: 3/1/2022 11:20 AM (Eastern Time (US & Canada)) - 3/1/2022 12:40 PM (Eastern Time (US & Canada))
Room: Virtual 26

Abstract

Zonal shuttle services utilize pre-defined service zones to separate the process of boarding and alighting from express transportation. Planning zonal shuttle services is conducive to improving the accessibility of the airport.
The study proposes a three-phase framework to identify the size and boundaries of service zones based on massive travel demand data. Firstly, locate representative links with high potential passenger flow within the maximum walking radius, according to the distribution of demand points. Then, select the detour coefficient as a dissimilarity measure to perform hierarchical clustering of representative links. Finally, establish an integer programming model to allocate the passengers into each cluster to determine service zones.
The rationality and effectiveness of the proposed method are evaluated using taxi data collected in Shanghai, t. The parameter sensitivity analysis shows flexibility consistent with the focus of the planning work. The customized algorithm has advantages in zone sizes and spatial coverage of passengers in comparison to KMeans and DBSCAN. With locations of current shuttle lines in operation, the identification results further clarify the rationality of the service zones generated by the method.
In conclusion, this paper proposes a flexible and fast method to identify service zones. The new data-driven approach can provide insights for airport shuttle planning.