Final report

Mid-Atlantic Universities Transportation Center

PDF

Urban traffic congestion is a problem that plagues many cities in the United States. Testing strategies to alleviate this

congestion is especially challenging due to the difficulty of modeling complex urban traffic networks. However, recent work has

shown that these complicated systems can be modeled in relatively simple ways by leveraging consistent relationships that

exist between network-wide averages of pertinent traffic properties, such as average network flow, network density and the

rate at which trips are completed. Using these “macroscopic” traffic models, various control strategies can be developed to

mitigate congestion and improve network performance. However, the effectiveness of many of these strategies depends on the

ability to estimate traffic conditions on the network in real-time. This jointly proposed research between Penn State and Virginia

Tech investigated how real-time mobile vehicle probes can be combined with macroscopic urban traffic models to inform more

efficient network-wide traffic control strategies. Additionally, this work will examine how the effectiveness of these strategies

can be directly measured in the field using only mobile vehicle probe data. These two efforts can lead to more efficient control

of downtown traffic networks and a reduction in vehicular delay during rush hour periods.

US Transportation Collection

DRTR12-G-UTC03