Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
101 Dynamic Traffic Assignment Model Breakdown James E. Hicks, PB Consult Dynamic traffic assignment (DTA) models arebeing implemented more frequently in practice.An entire Sunday afternoon session at the 2005 TRB Annual Meeting was devoted to presentations on practical experience with DTA models. Furthermore, commercial software vendors are making DTA products available in their traditional transportation planning packages. Several new commercial software releases occurred during the past year. Many presentations of the successes achieved in DTA implementations and demonstrations of the capabilities of DTA models will likely be forthcoming. This paper, however, examines a less successful DTA model. The emphasis will be on describing the development of methods for analyzing DTA model results, understanding the detailed interac- tions in the software, detecting relationships among data elements that produce various results, and synthe- sizing an implementation approach that tries to over- come or avoid obstacles. The goal is not only a set of network simulation results that can be compared with observational data but also evidence that the results are logical and that the model has worked as intended. This paper will discuss the use of one DTA software package, the Vista package (see www.vistatransport.com), and its application in a project in Atlanta, Georgia. First, a brief description of the project identifies the scope for which the DTA model is intended. The data requirements and manipulations for the DTA model are then outlined, and the model results are discussed. An example of the buildup of congestion on a network and its effect on traffic flow patterns is given. Potential underlying causes for the excessive congestion are addressed, which will provide further insight into how the DTA solution algo- rithm functions. Next, the knowledge gained from the analyses performed is described, and an approach to overcome the problems highlighted and the impact of implementing this approach are given. Finally, the find- ings with regard to implementing a large- scale DTA model and the insights gained from breaking down the problems encountered and understanding those prob- lems in the context of the algorithmic steps involved in solving the DTA are summarized. IMPLEMENTATION DETAILS The Georgia Department of Transportation is doing operational planning studies on sections of its freeway system to guide decisions concerning the programming of improvements. Its plan is to use focused microscopic traffic simulation models of sections of its freeway sys- tem to evaluate operational alternatives. A DTA model implementation has been identified as a means to calcu- late realistic time- dependent flows through areas where the DTA model uses input data from the regional travel demand modeling process and produces data required by the microscopic simulation methods. DTA models represent individual vehicle movements, and aggregate link performance characteristics gleaned to determine dynamic route choices and network equilibrium condi- tions. A traffic flow solution determined at such a meso- scopic scale is more appropriate for specifying time- dependent traffic flows through a focused area than a traffic flow solution defined at either a macroscopic