Return to Projects List

[NTC2014-SU-R-02] Combining Different Data Sources to Predict Origin-Destinations and Flow Patterns for Trucks in Large Networks


While other modes are clearly important for freight transportation, trucking is the dominant mode in terms of tons and value. Monitoring freight movement and freight transportation performance is essential in making effective policies and informed decisions to enhance and to efficiently manage the freight transportation system. One of the key aspects of monitoring freight over the highways has to do with determining the flow patterns of trucks, which can be achieved by uniquely identifying trucks at specific points along the roads or by tracking individual trucks using technology such as GPS. However, not all trucks are equipped with tracking devices. While point sensors along the highways allow determining the truck volumes, they do not provide much information about the paths and origin-destinations for trucks. However, by exploiting vehicle-specific attributes (e.g., axle spacings, length) collected by such sensors vehicles can be re-identified (matched) to enable prediction of paths taken by trucks. Data from other infrastructure-based sensors (e.g., Bluetooth readers, AVI sensors) can also be utilized for the same purpose. Furthermore, such data elements can be combined with freight generators in a network (e.g., ports, distribution centers) to better determine origins and destinations.

Developing such a system where data from all these sources are assimilated and synthesized to predict freight patterns will be useful for planning and performance monitoring of the national freight network. In this project, the research team will develop re- identification models to match vehicles between two WIM (weigh-in-motion) stations. At a typical WIM station, total vehicle length, axle spacing, and axle weights are measured per vehicle basis. Such data are then archived for future use. In addition to data from WIM sensors, the team will assume that travel time information (or variation) between the two sites is available. Such information can be obtained from various sources, including private companies (e.g., INRIX) or estimated from point sensors (e.g., loops, radar) installed along the corridor. The travel time information along with WIM data will be incorporated into algorithms to re-identify trucks. In previous models, travel time between the sites is usually assumed to be constant and the variation in travel time is ignored, which is not realistic especially along corridors going through congested urban areas. By incorporating the travel time variation, the re-identification algorithms are expected to produce more accurate matching, which will be explored in this project. This project will be built on the previous re- identification research conducted by the Principal Investigator and will address the following two objectives.

  • Developing vehicle re-identification algorithms that can integrate travel time information with WIM data for matching trucks between two sites
  • Assessing the accuracy of the re-identification algorithms as a function of the reliability of the travel time information