Memo

Date: April 17, 2007

To: Tom Welch, Steve Gent, Dave Matulac, and Michael Pawlovich

From: Shauna Hallmark, Tom McDonald, and Eric Fitzsimmons

Subject: Statewide Estimate of Red Light Running Crashes

The following reflects an analysis to estimate the magnitude of red light running crashes statewide in Iowa.

The Federal Highway Administration (FHWA) estimates that red light running causes more than 100,000 crashes and 1,000 fatalities annually in the US (FHWA, 2007). Retting et al (1995) indicated that occupant injuries occurred in 45% of red light running crashes as compared to other urban crashes and account for 16 to 20% of total crashes at urban signalized intersections.

The number of red light running type crashes for the state of Iowa was assessed for 2001 to 2006. Location of signalized intersections in the state was determined using a database based on the Iowa DOT GIMS database which is described in Appendix A. The Iowa DOT crash database (version “crashes (2001 2005) Jan 2006” for 2001 to 2005 and “statewide20012006(April2007)” for 2006) was used. Crashes were spatially located and those crashes within 25 meters of each signalized intersection were extracted for each year. Red light running and red light running type crashes were then summarized from those signalized intersection crashes.

Red light running crashes were defined as crashes where the officer indicated “ran traffic signal” or “failure to yield right of way on right turn on red” as the major cause. Since the officer may not always indicate cause, the number of crashes with collision types which are indicative of red light running were also evaluated. These include angle—on-coming left turn, broadside, head-on, and sideswipe—opposite direction. These crashes are referred to as “red light running type crashes”.

The number of red light running crashes for each year from 2001 to 2006 and the number red light running type crashes for each year from 2001 to 2006 are shown in Figure 1. As indicated, from 1,290 to 1,994 red light running crashes per year occur. This results in an annual average of 1,764 red light running crashes. Red light running type crashes range from 3,120 to 4,708 with an annual average of 4,055 crashes. Additionally, an annual average of 68 fatal and major injury red light running crashes and 123 fatal or major injury red light running type crashes occur.

An average of 8,162 total crashes occurs at signalized intersections in Iowa per year. Red light running crashes account for 21.6% of all crashes at signalized intersections and red light running type crashes account for 49.7% of all signalized intersection crashes annually. Additionally, red light running crashes account for 35.0% of the fatal and major injury crashes at signalized intersections and red light running type crashes account for 63.6% of fatal and major injury crashes at signalized intersections.

Figure 1: Red Light Running and Red Light Running Type Crashes in Iowa from 2001 to 2006

References

Federal Highway Administration. Stop Red Light Running Facts and Statistics, U.S. Department of Transportation. FHWA Safety. Accessed April 2007.

Retting, R.A., A.F. Williams, and M.A. Greene. 1998. “Red Light Running and Sensible Countermeasures: Summary of Research Findings.” Journal of the Transportation Research Record 1640. pp. 23-26.

Retting, R.A, A.F. Williams, D.F. Presusser, and H.B. Weinstein. 1995. “Classifying Urban Crashes for Countermeasure Development.” Accident Analysis and Prevention. Vol. 27. pp. 283-294.

Appendix A

A statewide intersection database was developed for Iowa based on the 1999 GIMS snapshot linework from the Iowa DOT (see report on SICL for description of how the intersection database was created). This database represents the intersection of any location where 3 or more approaches join. No statewide inventory of signalized intersections exists. In order to determine where signalized intersections were located, a database created by Mr. Michael Pawlovich was used. Crashes from 2001 to 2005 which were within 25 meters of an intersection node in the statewide intersection database were extracted. Total number of crashes was summed for each location. The number crash cases that have some indication that a traffic signal was present from the Traffic Controls crash option were recorded in one field. The total number of crash cases where a traffic signal was indicated in either the Traffic Controls crash option or Contributing Circumstances, Driver crash option were summarized. This indicated that the reporting officer had indicated presence of a traffic signal. Officers may not always fill in either option when reporting a crash. The point ID was related to the point ID from the statewide intersection database. An intersection was determined to be signalized if the following conditions were met:

  • 2 to 4 crashes had occurred and a signal was indicated for at least 2 crashes
  • 6 or 7 crashes had occurred and a signal was indicated for at least 3 crashes
  • 8 or more crashes had occurred and a signal was indicated for at least 4 crashes