PRINCIPLE COMPONENTS ANALYSIS BASED INCIDENT DETECTION

The problems of incident detection and short term forecasting were originally closely connected. A development phase has followed where the problems have been largely considered in isolation. These stages of development are critically reviewed. It seems that short term forecasting as a method of incident detection has largely fallen out of fashion, mainly because other algorithms of superior performance have been developed. This is unfortunate because algorithms which define an incident as any event which cannot be forecast have several useful properties. For example, it is not necessary to carry out an extensive data collection procedure in order to build up a data base of real incidents for training. A self-bootstrapping algorithm based on principal component analysis is presented which is not a forecasting based method, but which shares the same definition of an incident. Initial results are very promising.

• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/isbn/0784403333
• Corporate Authors:

  American Society of Civil Engineers

  1801 Alexander Bell Drive
  Reston, VA  United States  20191-4400
• Authors:
  • Dougherty, M
  • Chen, Huanlei
  • Montgomery, F
• Conference:
  • Fifth International Conference on Applications of Advanced Technologies in Transportation Engineering
  • Location: Newport Beach, California
  • Date: 1998-4-26 to 1998-4-29
• Publication Date: 1998

Language

• English

Media Info

• Features: Figures; References; Tables;
• Pagination: p. 51-58

Subject/Index Terms

• TRT Terms: Algorithms; Analysis; Components; Forecasting; Incident detection
• Subject Areas: Highways; Operations and Traffic Management; I73: Traffic Control;

Filing Info

• Accession Number: 00769581
• Record Type: Publication
• ISBN: 0784403333
• Files: TRIS
• Created Date: Sep 16 1999 12:00AM