ON-LINE TRAFFIC ASSIGNMENT AND NETWORK LOADING. IN: URBAN AND REGIONAL TRANSPORTATION MODELING. ESSAYS IN HONOR OF DAVID BOYCE

This chapter on on-line traffic assignment and network loading is from a book of essays published in honor of David Boyce for his contributions to the fields of transportation modeling and regional science. The authors first consider the use of equilibrium modeling to estimate current and future use of traffic networks; static traffic assignment (STA) is a model where, for given origin-destination predicted flow volumes, an equilibrium concept is utilized to assign routes and load flows on these routes. To model the route travel times and the speed profiles during the trip actually experienced by the traveler, researchers have developed the dynamic traffic assignment (DTA) approaches. The authors then propose a model that captures the dynamics of route-choice equilibration, provide an analytical traffic assignment approach to manage an impacted neighborhood network (INN) when count detectors and traffic probes are available, and demonstrate the feasibility and applicability of the approach. The model considers the dynamics of travel through the network, as well as day-to-day congestion-forming processes, rather than the use of explicit Bureau of Public Roads (BPR) functions. Analysis of the results from the assignment/simulation model indicate that the dynamic equilibrium process does behave as expected in the field: the process exhibits dynamic travel times and day-to-day convergence to a steady state; the routes most used for an origin-destination pair, at steady state, have nearly equal travel times; and if a long route is chosen for an origin-destination pair then only a very small fraction of travelers choose it.