Operations planning for complex urban transportation systems require simulation models that could be relied upon to accurately mimic real world traffic behavior. Neither the traditional Travel Demand Model (TDM) nor the typical microscopic traffic simulation model is considered an adequate tool for evaluating operational strategies that could have system wide impacts. One of the key challenges faced by these modeling tools is how to replicate the decision making process of every individual driver regarding which path to follow from his/her origin to destination. A realistic representation of how individuals make these decisions is critical, because the aggregation of these decisions/paths result in the time and space dependent volumes on the transportation network and the resulting time and space dependent speeds and travel times. Unless a realistic simulation model, that adequately addresses this path finding problem, is used as the tool, all large scale operations planning efforts will not generate a lot of confidence.

The Maricopa Association of Governments (MAG) has identified the Dynus-T mesoscopic simulation model with Dynamic Traffic Assignment as the tool of choice for operations planning. In-house verification of Dynamic Traffic Assignment (DTA) model has been performed through a simulation involving a real-life crash in 2011. The DTA model significantly improves the reliability, efficiency and scale of operational strategies and their evaluation by introducing reasonable human decision making process and mesoscopic simulation. A 24-hour regional DTA model is being developed to serve as a tool for regional operations planning. The goal of this effort is to enable the region to conduct operations planning and analysis of strategies across the region. The model development effort is expected to be completed by the end of 2012.

The challenges and solutions encountered during the MAG 24-hour regional model development will be presented. The following topics will be covered: Network Refinement, Demand Loading, Data Gathering, Signal Timing Conversion, Model Calibration and Verification. Initial findings from the application of the model to the I-10 Integrated Corridor Management System will also be presented. Other potential applications of the model will also be discussed.