PRINCIPAL COMPONENT ANALYSIS OF URBAN TRAFFIC CHARACTERISTICS AND METEOROLOGICAL DATA

This study uses principal component analysis (PCA) in order to determine underlying components, physical interpretations and interrelationships among traffic, emission and meteorological variables. One-year (1997) meteorological and traffic data from an urban intersection in Delhi is used. In urban intersections, the complexities of site, traffic and meteorological characteristics may result in a high cross-correlation among variables. PCA can provide an independent linear combination of the variables in such situations. In this paper, PCA is used to analyze average emission, traffic and meteorological data from 1-, 8- and 24-hour time intervals. Results show that four principal components for the 24-hour average have the highest loading for traffic and emission variables, with a strong correlation between them. Principal component loading for the 1- and 8-hour data indicate the least variation among them. Findings also indicate that there is a weak correlation of traffic and emission variables with meteorological variables for all three time average periods. This research demonstrates the usefulness of PCA in analyzing large multivariate data sets.

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 00960774
  • Record Type: Publication
  • Files: TRIS, ATRI
  • Created Date: Jul 14 2003 12:00AM