Rain Reconstruction from Various Weather-Related Data Sets Using Logistic Regression: Methodology and Applications

Knowledge of rain exposure is necessary for computing the effect of rainfalls on both the injury accident’s occurrence and on the driver’s behavior. Different meteorological data sources, each of them having its advantages and drawbacks, have to be matched before being used for that purpose. A statistical approach (the logistic regression technique) has been retained, which aims at reconstituting the relevant information related to rainfall, even at places which can be remote from the meteorological measurement stations. It consists in combining various meteorological sources, such as both human and sensor-based data collection. The analysis is based on a 6 minutes time scale, rather than on the usual hour time scale. The available weather information is used, after a learning phase, to model the probability rainfalls occur during each 6 minutes period. This methodology is applied to estimate the risk of injury accident due to rain, on the French main and country roads in Haute-Normandie. This method allows to compute the risk during rain and to compare it with the risk during clement weather conditions. The risk during rainfalls is estimated at 21.9 accidents per 100 millions vehicle whereas the risk during normal weather conditions is estimated at 10.4. Therefore, the average added risk due to rain is estimated at 2.1 and at 2.4 in case of bends, and these results are consistent with other related results. At last, the modification of the driver’s time gaps due to rain is also considered.

Language

  • English

Media Info

  • Media Type: CD-ROM
  • Features: Figures; References; Tables;
  • Pagination: 16p
  • Monograph Title: TRB 86th Annual Meeting Compendium of Papers CD-ROM

Subject/Index Terms

Filing Info

  • Accession Number: 01044833
  • Record Type: Publication
  • Report/Paper Numbers: 07-0916
  • Files: BTRIS, TRIS, TRB
  • Created Date: Feb 8 2007 5:29PM