Prepared by: Parsons Brinckerhoff for SHRP-2
American motorists experience more hours of congested conditions every year, but planning models rely on relatively thin behavioral information to take congestion into account. Highway operations and road pricing strategies are being employed to address congestion, but the planning process is not well equipped to describe the effective capacity available when roads are congested or to describe the relief obtained by improvement strategies. Variable tolls are being considered to encourage motorists to shift travel time out of congested periods, to use less congested roads, or to change mode. Current travel models are not capable of simulating all of the factors needed to calculate the effect of tolls on congestion because they do not include disaggregated models of choice behavior for a range of users under various choice conditions. A disaggregate approach is required that deals directly with the decisions faced by individuals rather than large groups of people.
The objective of this project was to develop mathematical descriptions of the full range of highway user behavioral responses to congestion, travel time reliability, and pricing. This included formatting the mathematical descriptions of behavior so that they could be incorporated into various travel demand modeling systems in use or being developed. Another objective was to examine network assignment practices needed to support models that simulate behavioral responses to congestion, travel time reliability, and pricing.
Over the past 30 years, research has advanced the understanding and prediction of travelers’ behavior choices in response to changes in traffic congestion and changes in the price of travel. This project synthesized that research to select the important and well-founded behavioral hypotheses, and it tested those hypotheses statistically on the most suitable data sets available in the United States. The project identified the most important contextual influences on behavioral sensitivity to highway congestion and pricing, and it provided guidance on the relative magnitude of those influences. Travel demand modeling studies are typically limited in scope and time frame, and they are usually tied to a single survey data set. This project had the unique opportunity to carry out a comprehensive series of hypotheses tests on data sets that were available from multiple areas across the United States.