Step 4: Trip assignment - What is the route of each trip?
The final step in the forecasting of travel behavior is to determine the routes travelers choose to reach their destinations. This step is known as trip assignment. Only motorized person trips are assigned, which includes both trips made by automobile/car and trips made by public transit. The previous travel model, the Version 2.2 Travel Model, assigned only those made by private vehicles (cars, vans, trucks). The new travel model, the Version 2.3 Travel Model, adds a new capability: the ability to assign transit trips. Private-vehicle trips and truck trips are assigned to the highway network ("traffic assignment") and public transit trips area assigned to the transit network ("transit assignment").
Before conducting traffic assignment and transit assignment, the model must build minimum-impedance paths between all zone pairs in the region. Since the current modeled are has 3,722 transportation analysis zones (TAZs), the number of paths is 3,722 squared, or about 13.9 million zone pairs.
For highway path building and highway assignment, the effects of congestion are taken into account. First of all, each road link has a link performance function, also known as a volume-delay function (VDF), which models the fact that as more traffic uses a link, the effective speed of vehicles on that link decreases. There are different VDFs for each type of road, i.e., freeways, major arterials, minor arterials, collectors, and expressways, thus taking into account the differing carrying capacity of different road types. Second, there is an iterative nature to each traffic assignment, with the iteration stopping when a defined level of convergence is reached. Third, since the congested link speeds coming out of traffic assignment have an effect on trip distribution and mode choice, these congested link speeds are fed back into prior modeling steps (a process known as speed feedback) to ensure that the system is in equilibrium.
For transit network path building and transit assignment, the model is not currently sensitive to congestion effects. Although the Metropolitan Washington Region is arguably a region where crowding on transit is an issue, the current TPB model, like most transit assignment models in the U.S. is not able to account for the effects of crowding on transit (See, for example, page 9-25 of Cambridge Systematics, Inc., 2010. Travel Model Validation and Reasonability Checking Manual, Second Edition, Washington, D.C.: Travel Model Improvement Program, Federal Highway Administration).