Application of GPS and clustering techniques in defining LOS criteria of signalized intersections for Indian cities

Due to the fact that delay ranges of Level of Service (LOS) categories of signalized intersections on urban roads are not well defined for Indian cities, a study is conducted in the city of Mumbai, and the result is validated for Kolkata city. The stopped delay, as derived from speed data collected at second time interval using Global Positioning System (GPS) receiver fitted on mobile devices, is used in this study. Four clustering methods on delay data are used through the application of cluster validation parameters, and it was found that k-means clustering is the most suitable method in defining LOS categories. Consequently, k-means clustering technique is used to define the delay ranges of LOS categories, and cluster validation parameter (silhouettes) is used to analyze the data structure lies within LOS categories. In addition, reliability analysis is applied to check percentage of delay data distributed under each LOS category. The results show that intersection delay ranges for LOS categories valid in Indian context are approximately two times higher - much higher - than those values mentioned in Highway Capacity Manual 2000.

  • Authors:
    • Bhuyan, Prasanta Kumar
    • Rao, K V Krishna
  • Publication Date: 2011


  • English

Media Info

  • Media Type: Print
  • Features: Figures; References; Tables;
  • Pagination: pp 69-75
  • Serial:

Subject/Index Terms

Filing Info

  • Accession Number: 01350450
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Aug 25 2011 11:59AM