Bus transit service reliability: Understanding the impacts of overlapping bus service on headway delays and determinants of bus bunching

To retain and attract new riders, transit agencies are frequently in search for ways to improve system reliability. Transit agencies typically operate several routes that partially or fully overlay (or overlap) one another in order to offer a better service coverage to reach various destinations. While previous research has focused on understanding the general factors that impact headway adherence and delay, there has been little effort to address the effects of overlapping bus routes on service headway adherence and service bunching. This research investigates the impacts of bus route overlapping on service headway delay and probability of bunching at the stop-level of analysis. The study uses automatic vehicle location (AVL) and automatic passenger count (APC) systems data collected from TriMet, the public transit provider for Portland, Oregon, USA, along one of its heavily utilized bus corridors, the Barbur bus corridor. It is shown that service overlapping can increase headway delay by 3.8 seconds, with no impacts on service bunching. It is also shown that headway delay is a function of scheduled headway between trips. Thus, scheduling more time between trips decreases the service delay, with a minimum of delay occurring at 20 minutes. Trips starting late at the beginning of a route increase the odds of bunching for the following trip on schedule more than its delay. This study offers transit agencies and schedulers a better understanding of the effects of service overlapping on service headway delays from schedule and the determinants of us bunching, which are important components of transit service reliability.