Traffic Modeling and Simulation

Background

The problems facing the country’s transportation system are enormous. Over 40,000 fatalities occur each year in traffic accidents. Vehicle emissions are the leading cause of air pollution.

With travel demand expected to increase more than 50% by 2020, it becomes apparent that we can’t just “build our way out” of the problem. We need to operate our existing systems in a more efficient manner. Even modest improvements could produce significant benefits.

Role of High-Performance Computing

High-performance computing can tackle complex regional, metropolitan-scale, multi-modal transportation problems. TRACC employs a holistic approach to diverse and complex processes and to the context-sensitive interactions among

• Multimodal elements
• Human behavior
• Market-driven demand for goods and services
• Traffic control structures
• Weather
• Construction impacts

On its high-performance Linux cluster, TRACC is developing a transportation model for the Chicago Metropolitan Area using the micro-simulation code TRANSIMS. TRANSIMS can simulate each individual traveler’s route throughout the day. It was developed by the U.S. Department of Transportation based on a new paradigm in traffic forecasting.

The road network in the Chicago Business District is being updated based on aerial photography and other public or commercial resources to provide the high- fidelity input required by TRANSIMS. This work is being done by TRACC, Northern Illinois University, and the Chicago Metropolitan Agency for Planning.