Urban Bus Arrival Time Prediction: A Review of Computational Models

Abstract

— Traffic flow in major urban roads is affected by several factors. It is often interrupted by stochastic conditions, such as traffic lights, road conditions, number of vehicles on the road, time of travel, weather conditions, driving style of vehicles. The provision of timely and accurate travel time information of transit vehicles is valuable for both operators and passengers, especially when dispatching is based on estimation of potential passengers waiting along the route rather than the predefined time schedule. Operators manage their dispatches in real time, and passengers can form travel preferences dynamically. Arrival time estimation for time scheduled public transport busses have been studied by many researchers using various paradigms. However, dynamic prediction on some type of transit vehicles, which do not follow any dispatch time schedule, or stop station constrains introduces extra complexities. In this paper, a survey on the recent studies using historical data, statistical methods, Kalman Filters and Artificial Neural Networks (ANN) have been applied to GPS data collected from transit vehicles, are collected with an emphasis on their model and architecture.