Analysis of Bus Stop Delay Variability Using Public Transit GPS Data

Abstract

Travel time is considered as a direct measure of the efficiency and service reliability of a public transit system and increased travel time is contributed by delays at bus stops. Delay at bus stops affects the efficiency of bus operations and the level of service of public transportation. Day-to-day delay variability decreases passenger confidence in perceived reliability, causing uncertainty in making travel decisions. Many research works have been carried out to find delays at bus stops, but there are very few studies about the delay at bus stops in India using automatic vehicle location (AVL) data. In the present study, an attempt has been made to analyze delays at various bus stops in Mysore city using vehicle trajectory-based formulation based on AVL data collected from the Mysore Intelligent Transportation System (ITS). Delay variability has been analyzed using the coefficient of variation (COV) and also by fitting various probability distributions to the data, since distribution fitting helps in estimating the pattern of delay variability. The goodness of fit is tested by the Kolmogorov–Smirnov (KS) test. The results suggest that bus stops in the commercial area face more variability than the bus stops in residential areas. By fitting various distributions to delay data, it was observed that the performance of the generalized extreme value (GEV) distribution in fitting the data is better than other distributions.