Dublin Bus Tracking Service: Design and implementation of a device independent passenger information system

Abstract

Traditionally public transport has been perceived as the less desirable alternative to the car. The environmental argument of using public transport has not led to a significant increase in the usage of public transport. In order to convince people to leave their car at home, public transport must be at least as, if not more desirable than driving a car. There are a number of ways of achieving this, among them reducing fares, introducing quality bus corridors etc. However the most effective strategy to do this must be to reduce the perceived unreliability of public transport. The most effective way of doing this is through a real-time passenger information system Passenger information systems are large-scale capital intensive projects, which means only large organisations and governments can implement them. The most expensive component of the system is the automatic vehicle location (AVL) system. This thesis proposes a new type of AVL system based on third generation mobile positioning technology that significantly reduces the cost of implementing a passenger information system. A best-of-breed architecture for implementing passenger information systems is described. It uses a service chain architecture to aggregate content and services from multiple sources to provide a coherent interface to the end user. The interface can be delivered to multiple devices with different display characteristics. A prototype implementation of this architecture is implemented in Java.