Buses as Probe Vehicles for Travel Time Data Collection on Urban Arterials

Abstract

Floating car method technologies such as global positioning system (GPS) units or data from private sources such as INRIX are used for travel time data collection on urban streets. Time, man-power, and cost deter (limit) collection and purchase of such data for the entire transportation network. An alternate source of the travel time data is using buses as probe vehicles. This is practically feasible and inexpensive as most buses operating in urban areas are equipped with automatic vehicle location (AVL) units. However, the travel time of buses is generally greater than the travel time of a typical vehicle (car). The use or applicability could only be justified if there is strong correlation between car and bus travel times. This paper, therefore, examines the relationship between car and bus travel time. The role of key influential factors (such as the number of signalized intersections, the number of unsignalized intersections, the number of driveways, the number of bus-stops, traffic volume, the number of lanes and the number of turns made by the bus) on the ratio between the two travel times is also evaluated and examined to assess the use of buses as probe vehicles. Results indicate a moderately strong relationship between car and bus travel time. Variables other than the number of signalized intersections per unit distance, the number of lanes, and traffic volume do not seem to play a statistically significant role on the ratio of car to bus travel time irrespective of time-of-the-day. Models by time-of-the-day, developed to estimate car travel time, could be used if bus travel time, the number of signalized intersections per unit distance, the number of lanes, and traffic volume are known.