A profile of HOV lane vehicle characteristics on I-85 prior to HOV-to-HOT conversion

The conversion of high-occupancy vehicle (HOV) lanes to high-occupancy toll (HOT) lanes is currently being implemented in metro Atlanta on a demonstration basis and is under consideration for more widespread adoption throughout the metro region. Further conversion of HOV lanes to HOT lanes is a major policy decision that depends on knowledge of the likely impacts, including the equity of the new HOT lane. Rather than estimating these impacts using modeling or surveys, this study collects revealed preference data in the form of observed vehicle license plate data and vehicle occupancy data from users of the HOV corridor. Building on a methodology created in Spring 2011, researchers created a new methodology for matching license plate data to vehicle occupancy data that required extensive post-processing of the data. The new methodology also presented an opportunity to take an in-depth look at errors in both occupancy and license plate data (in terms of data collection efforts, processing, and the vehicle registration database). \n \nCharacteristics of individual vehicles were determined from vehicle registration records associated with the license plate data collected during AM and PM peak periods immediately prior to the HOV lanes conversion to HOT lanes. More than 70,000 individual vehicle license plates were collected for analysis, and over 3,500 records are matched to occupancy values. Analysis of these data have shown that government and commercial vehicle were more prevalent in the HOV lane, while hybrid and alternative fuel vehicles were much less common in either lane than expected. Vehicle occupancy data from the first four quarters of data collection were used to create the distribution of occupancy on the HOV and general purpose lane, and then the matched occupancy and license plate data were examined. A sensitivity analysis of the occupancy data established that the current use of uncertain occupancy values is acceptable and that bus and vanpool occupancy should be considered when determining the average occupancy of all vehicles on the HOV lane. Using a bootstrap analysis, vehicle values were compared to vehicle occupancy values and the results found that there is no correlation between vehicle value and vehicle occupancy. A conclusions section suggests possible impacts of the findings on policy decisions as Georgia considers expanding the HOT network. Further research using these data, and additional data that will be collected after the HOT lane opens, will include emissions modeling and a study of changes in vehicle characteristics associated with the HOT lane conversion.