• UTRC II SUBMISSION SYSTEM
  • Careers
  • Contact
  • Login / Register

Search form

Home
  • Home
  • About
    • Welcome to the UTRC Site
    • Theme
    • Staff
    • Board of Directors
    • Press
    • Annual Report
    • Program Progress Performance Report
    • Newsletter
  • Research
    • Projects
    • RFPs
    • Submit Your Proposal
    • Funding Categories
      • UTRC Research Initiative
      • UTRC Advanced Technology Initiative
      • UTRC Faculty Development Mini-grants
      • UTRC Best Transportation Paper Competition
      • News
  • Publications
  • Directory
    • Consortium Universities
    • Partners
    • Principal Investigators
    • Staff
    • Board of Directors
  • Education
    • Where to Study
    • Transportation and Planning Doctoral Series
    • AITE Scholarships
    • UTRC Dissertation Grants
    • Summer Institute
    • September 11th Memorial Program
    • Technology Transfer and Training
    • Online Graduate Certificate Program
    • UTRC Travel Grants
    • Student Award Recipients
    • Apply For Scholarships
  • Events
    • Upcoming Events
    • Past Events
    • Visiting Scholar Seminar Series
  • Resources

Estimating Multi-class Truck Origin-Destination Flows Through Data Fusion from Multiple Sources

Efforts to manage truck flows in congested urban areas have important implications not only for congestion relief, but also for air quality improvement and reductions in energy use. A vital input to these flow management efforts is knowledge of the origin-destination (O-D) movement patterns for various classes of trucks, and this presents a substantial challenge. The problem of estimating O-D tables from observed data (usually link counts) has been studied since the 1970’s, and there are effective methods for a single-class (i.e., passenger cars) problem under assumptions of deterministic user equilibrium flow patterns in the network. There has also been some work on extending the ideas to conditions reflecting stochastic user equilibrium.

Relatively little work has been done on the multi-class problem, which is of particular concern for estimating truck flows in different size classes. Furthermore, as more advanced traffic surveillance technologies become available, data beyond simple link counts can be used in the estimation process. Finally, the transition from estimating a static O-D pattern to estimating dynamic O-D flows creates additional challenges.

The purpose of this project is to test a newly developed method for estimating multi-class O-D tables for trucks, using more comprehensive observable data than link counts, and reflecting the uncertainty inherent in network flows by using a stochastic equilibrium formulation, rather than a deterministic one. The model formulation uses a bi-level optimization, for which a specialized solution method has been developed. In this project, the solution method will be tested using a set of networks of varying character and data of varying types. The result of this project will be a validated method for improving the estimation of multi-class truck origin-destination flows, based on a fusion of data from various sources and reflecting the uncertainty in network flow patterns. This is a key step in moving toward the ability to better manage truck flows in real time to reduce congestion, reduce energy consumption, and improve air quality in urban areas.

Project Details

Project Type: 
UTRC Research Initiative
Project Dates: 
January 1, 2012 to July 31, 2013
Principal Investigators: 
Dr. Mark A. Turnquist (Late)
Institution: 
Cornell University
Sponsor(s): 
Research and Innovative Technology Administration / USDOT (RITA)
Publications: 
Final Report
Project Status: 
Complete
Research Categories: 
Vehicle and Equipment
Please subscribe to our Newsletter:

Get our newsletter

Please enter your email address to subscribe to our newsletter:

Contact Us

University Transportation Research Center
Marshak Hall - Science Building, Suite 910 
The City College of New York
138th Street & Convent Avenue ,New York, NY 10031