Title:
Using automated passenger counter data to understand ridership change on a zonal basis
Using automated passenger counter data to understand ridership change on a zonal basis
Author(s)
Joshi, Sanskruti
Advisor(s)
Watkins, Kari E.
Van Hentenryck, Pascal
Berrebi, Simon J.
Van Hentenryck, Pascal
Berrebi, Simon J.
Editor(s)
Collections
Supplementary to
Permanent Link
Abstract
This study aims at understanding various issues or gaps in working with Automated Passenger Counter (APC) data and suggests a step-by-step process of cleaning the data in a format that can be used to answer research questions regarding transit ridership on a zonal-basis. To check discrepancies in APC data, this study uses General Transit Feed Specification (GTFS) which is a standard format for transit information, including schedules, stops, number of trips, fares and location of stops. APC was compared with GTFS data to check for gaps, such as missing trips, naming convention of routes and locations of stops. By using GTFS data for checking and cleaning, stops with missing trips were weighted and outliers were removed from the dataset. Apart from removing outliers, GTFS was also used to match up route names in APC similar to that in GTFS to streamline the process of comparing the two datasets. This research provides transit agencies with a methodology to check, clean and store APC data in a standardized format. Best practice of APC data management enables more in-depth analysis of ridership trends at all levels of aggregation.
Sponsor
Date Issued
2019-04-30
Extent
Resource Type
Text
Resource Subtype
Thesis