Examining Credit Score as a Surrogate Measure of Risk to Improve Traffic Crash Prediction—California Case Study

This study investigated an idea of using a location-based surrogate measure of risk (credit rating) to improve crash prediction. The work presented here is the initial step in this investigation specifically examining if any relationship exists between location-based credit rating and crash frequencies. Because traffic volume is highly correlated with traffic crashes, the sites examined were grouped into bins by annual average daily traffic (AADT). Over 1,300 intersections in California were examined with 5 years of traffic and crash data. Credit scores were obtained for the zip codes where these intersections were located to determine if there is any relationship between credit score and number of crashes. The analysis showed positive and negative trends between credit score and crash risk for differing levels of credit rating and AADT. Future work is recommended to further examine the trends revealed in this research and determine if the use of driver-related risk surrogate measures to account for the heterogeneity of driver populations across jurisdictions could improve traffic crash prediction models through the development of a crash modification factor (CMF).

• Record URL:
  https://doi.org/10.1061/9780784483534.012
• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/isbn/9780784483534
• Supplemental Notes:
  • © 2021 American Society of Civil Engineers.
• Corporate Authors:

  American Society of Civil Engineers

  1801 Alexander Bell Drive
  Reston, VA  United States  20191-4400
• Authors:
  • Hyde, A
  • Park, S
  • McFadden, J
  • Graettinger, A
• Publication Date: 2021

Language

• English

Media Info

• Pagination: pp 134 - 142
• Monograph Title: International Conference on Transportation and Development 2021: Transportation Operations, Technologies, and Safety

Subject/Index Terms

• TRT Terms: Annual average daily traffic; Case studies; Crash data; Crash modification factors; Crash risk forecasting; Credit; Heterogeneity; Intersections; Location data; Predictive models; Traffic volume
• Identifier Terms: ZIP codes
• Geographic Terms: California
• Subject Areas: Economics; Highways; Safety and Human Factors; Society;

Filing Info

• Accession Number: 01777539
• Record Type: Publication
• ISBN: 9780784483534
• Files: TRIS, ASCE
• Created Date: Jul 23 2021 3:26PM