USE OF SEQUNTIAL LEARNING FOR SHORT-TERM TRAFFIC FLOW FORECASTING

This paper examines the potential of dynamic neural networks for forecasting traffic in normal and incident-related conditions. Focus is on the application and performance of an alternative neural computing algorithm which involves "sequential or dynamic learning" of the traffic flow process. Comparisons are drawn regarding forecasting performances and show that the simple dynamic network outperformed the standard Kalman filter neural network.

• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/0968090X
• Supplemental Notes:
  • Publication Date: October 2001
• Corporate Authors:

  University of Leeds

  Institute for Transport Studies
  Leeds, West Yorkshire  United Kingdom  LS2 9JT
• Authors:
  • Chen, Haibo
  • Grant-Muller, Susan
• Publication Date: 2001

Language

• English

Media Info

• Pagination: p. 319-336
• Serial:
  • Transportation Research Part C: Emerging Technologies
  • Publisher: Elsevier
  • ISSN: 0968-090X
  • Serial URL: http://www.sciencedirect.com/science/journal/0968090X

Subject/Index Terms

• TRT Terms: Neural networks; Traffic estimation; Traffic flow
• Subject Areas: Operations and Traffic Management;

Filing Info

• Accession Number: 00823792
• Record Type: Publication
• Source Agency: UC Berkeley Transportation Library
• Files: PATH
• Created Date: Feb 5 2002 12:00AM