Using Transit Vehicles to Measure Freeway Traffic Conditions

Recently, many public transit systems have equipped their vehicles with automatic vehicle location (AVL) systems, which periodically provide position and speed for each transit vehicle in the fleet. Although AVL is deployed for transit operations, the vehicles also provide valuable information about the traffic stream throughout the road network. In this study we develop a methodology to mine the transit AVL data and find all trips that use any portion of a pre-specified portion of a freeway. These trips are then used to measure travel time and average speed over the freeway, thereby quantifying conditions on the facility. The results are validated against concurrent loop detector data from the corridor, though the greatest benefit would be expected in areas without fixed vehicle detection.