TRB 2016 Blue Ribbon Committee
16th National Transportation Planning Applications Conference

A Regional Wide Microsimulation Model: Validation and Applications

Corresponding Author: Lan Jiang, Maricopa Association of Governments

Presented By: Lan Jiang, Daniel Morgan, Maricopa Association of Governments, Caliper Corporation


Motivated by the increasing interest in modeling dynamics of traffic conditions and analyzing the operation of complex transportation system, Maricopa Association and Governments (MAG) cooperated with Caliper Corporation to develop a regional wide microsimulation model covering MAG’s entire service region. The regional microsimulation network represents a detailed street geography and lane geometry including over 3,600 signalized intersections. The model includes auto and truck travel, scheduled bus services and light rail operations on more than 100 local and express routes.

We begin with a brief introduction of the MAG regional microsimulation model highlighting its core input data and challenges in the course of development. The OD trip tables for the region are derived from the MAG regional travel demand model and subsequently refined using dynamic traffic assignment (DTA) and dynamic origin-destination matrix estimation (D-ODME) techniques. A combination of various data resources is used for the model calibration and validation including 15-minute directional count data, 15-minute NPMRDS speed data and vehicle trajectory data obtained from a time-lapse aerial photo survey. We also discuss the best way to leverage the vehicle trajectory data in calibrating driver behavioral parameters in the regional microsimulation model. The validation process and results are then presented. The microsimulation model is not only providing us a useful traffic analysis tool but also can be used as a visualization tool which facilitates communications with the general public. Lastly, we demonstrate model applications in analyzing project alternatives, impacts and visualization of future scenarios.


Discuss This Abstract