Prediction of Travel Time on an Expressway: Application of a Multivariate Normal Model

Expressway travel time prediction is a vital and fundamental issue in intelligent traffic systems, and its accuracy significantly influences the reliability of travel information services. Most studies give little consideration to information from other roads, which may reduce prediction accuracy in fluctuated traffic conditions or heavy congestion. To fill this gap, this paper presents a novel travel time prediction approach based on a multivariate normal model on an expressway. Large-scale floating car data are used for the model estimation. First, the authors identify that the travel time in a link follows a log-normal distribution. Second, four traffic patterns formed according to the characteristics of the data are divided to establish different spatial-temporal correlation sets; furthermore, according to the sets, real-time travel time and historical data from the given link and surrounding links are combined, and the authors obtain the link travel time distribution using the multivariate normal prediction model. Finally, the authors acquire the path travel time distribution via a nested delay algorithm. To validate the proposed approach, the authors focus on Guangzhou Airport Expressway and take traversed trajectories data from more than 20,000 probe vehicles. The results indicate that the proposed travel time prediction model performs well in real scenarios, especially when traffic conditions have changed accidentally or under conditions of heavy congestion.

• Supplemental Notes:
  • This paper was sponsored by TRB committee ABJ30 Standing Committee on Urban Transportation Data and Information Systems.
• Corporate Authors:

  Transportation Research Board

  500 Fifth Street, NW
  Washington, DC  United States  20001
• Authors:
  • Sun, Wei-Wei
  • He, Zhao-Cheng
  • Chen, Rui-Xiang
• Conference:
  • Transportation Research Board 96th Annual Meeting
  • Location: Washington DC, United States
  • Date: 2017-1-8 to 2017-1-12
• Date: 2017

Language

• English

Media Info

• Media Type: Digital/other
• Features: Figures; Tables;
• Pagination: 20p
• Monograph Title: TRB 96th Annual Meeting Compendium of Papers

Subject/Index Terms

• TRT Terms: Expressways; Mathematical prediction; Real time information; Traffic congestion; Travel patterns; Travel time
• Identifier Terms: Guangzhou Airport Expressway
• Geographic Terms: Guangzhou (China)
• Subject Areas: Data and Information Technology; Highways; Planning and Forecasting;

Filing Info

• Accession Number: 01632484
• Record Type: Publication
• Report/Paper Numbers: 17-04678
• Files: TRIS, TRB, ATRI
• Created Date: Apr 24 2017 9:31AM