An automated system for city-scale travel time calculation

Abstract

Traffic authorities need to provide travel time information over and above the level of service information. Publishing travel time information is one of the key aspects of traffic management [1]. An accurate travel time does help reduce congestion and decrease driver frustration. The focus of this paper is to describe a highly accurate and automated system for travel time calculation. This system utilizes the network information to extract the frequency used routes. It helps the traffic authority to easily create registry of the most popular routes. This system uses real time and predicted traffic information to improve the accuracy of travel time information. It implements a classifier model in case the real-time traffic information and historic delay information is available and probabilistic model to generate the wait time at stop points like traffic signal and toll booths. This system integrates with the incident and event systems to incorporate the impact of various events such as accident and public-gathering on travel time.