Fare Payment Compliance Monitoring and Lost Revenue Estimation using Automatically Collected Data

Abstract: This study investigates fare payment compliance by estimating and analyzing fare system non-interactions within the Massachusetts Bay Transportation Authority (MBTA) bus system. The study leverages automatically collected data from the Automated Fare Collection (AFC) and Automated Passenger Counting (APC) systems. By examining data from over 840,000 passenger boardings across the bus network for various times of the day, patterns of fare non-interaction were identified and the associated lost revenues were estimated. Our findings reveal a fare non-interaction rate of approximately 23%, resulting in significant revenue losses. Since not all fare system non-interactions result in revenue losses, a novel method is used to estimate lost revenue relying entirely on data from the AFC and APC systems, accounting for different fare types. This method allowed us to conduct a comprehensive analysis of four weekdays. Nearly $20,000 in lost revenue per day is estimated from fare system non-interactions, which adds up to approximately $9,500,000 annually. This substantial loss underscores the financial impact of fare non-interactions on the transit system. The analysis of non-interaction rates and lost revenues throughout the day highlights key periods like the AM Peak and PM Peak. Patterns of lost revenue and non-interaction rates are also compared for specific bus stops and routes. Mapping these locations shows where manual observations or fare inspections would be most valuable. These insights can help transit authorities identify where and when fare evasion occurs and create better strategies to improve revenue recovery.