View slides

Watch video:

Metro's Research and Modeling Services Program is responsible for the development, maintenance, and application of travel demand models for application in long-range planning efforts in the Portland metropolitan region.

Representation of traffic—both vehicular and transit—plays an integral role in the travel demand modeling process. Complex software is required to assign vehicles and transit users to transportation networks to determine viable options available to travelers, costs associated with those options, and sets of routes by which travelers might navigate their trips. 

Metro's current static assignment model has traditionally sufficed for use with Metro's four-step travel demand model. However, static assignments have well-documented limitations that preclude the ability of the analyst to answer complex policy questions, especially those related to greenhouse gas emissions, congestion, and transportation network reliability. In addition, static assignments cannot fulfill a need for small-duration travel time increments required by the next generation activity-based models. 

The shortcomings of the static assignment necessitates Metro’s development and application of regional dynamic traffic assignment (DTA) models. The resolution of these models allows for continuous modeling of traffic over an analysis period, which allows the analyst to capture temporally-based traffic events such as the building and dissipation of queues, measurement of the duration of congestion, and high fidelity speed profiles for use in emissions analysis. 

This presentation will focus on why Metro has developed a DTA, how DTA compares with other models—specifically macro-scale static assignments and micro-simulations—and how DTA has been applied in Metro's modeling process.

Peter Bosa is a Senior Researcher and Modeler with Portland Metro's Transportation Research and Modeling Services program. He has over 10 years of experience in designing and applying transportation models for regional- and corridor-level forecasting of travel behavior, vehicle emissions for air quality analyses, and applications for the Federal Transit Administration's New Starts program. He is one of the primary developers behind Metro's regional dynamic assignment toolset. Prior to working at Metro, Peter received a Master's in Urban and Regional Planning from Portland State University, where he worked in the Intelligent Transportation Systems Lab as a graduate research assistant.