A Novel Transit Bus Number Identification Approach for Frictionless Fare Collection Using Passenger Location Data

Automated Fare Collection (AFC) systems are essential for advancing public transportation infrastructure. This study introduces a Be-In Be-Out (BIBO) framework that uses smartphone location data to facilitate frictionless fare collection by identifying public bus trips. The framework consists of two main components: a Travel Mode Detector (TMD), which operates on mobile devices to identify transportation modes and collect location data during transit, and a Journey Recognition Module (JRM), which processes the collected data in the cloud to identify bus trips. By combining GPS and General Transit Feed Specification (GTFS) data, the system enables efficient real-time bus trip recognition. The framework incorporates trajectory similarity algorithms such as Dynamic Time Warping (DTW) and Longest Common Subsequence (LCSS) within the JRM to enhance the accuracy of user-to-bus matching. Experimental evaluations demonstrate the effectiveness of the approach, highlighting the superior performance of DTW in accurately identifying transit bus IDs.