Source-linked AI summary
Empirical assessment of the impact of highway design exceptions on the frequency and severity of vehicle accidents
Nataliya V. Malyshkina, Fred L. Mannering
TL;DR
The safety effects of approved highway design exceptions are not well understood. This study models accident frequency and severity at Indiana roadway sites with and without exceptions and finds no statistically significant effect on either outcome.
Problem
The safety impact of highway design exceptions granted when projects do not meet current standards is not well understood.
Method
The study compares design-exception and control roadway sites using statistical models and likelihood-ratio tests for accident frequency and severity.
Results
Approved design exceptions had no statistically significant impact on accident severity or accident frequency.
Takeaways & Limitations
The results suggest that Indiana’s process for granting design exceptions avoided adverse overall safety consequences during the study period.
Takeaways & Limitations
The limited number of available design exceptions prevents broad policy statements, requiring case-by-case comparison with previously granted exceptions.
Abstract
from arXiv · showhide
Compliance to standardized highway design criteria is considered essential to ensure the roadway safety. However, for a variety of reasons, situations arise where exceptions to standard-design criteria are requested and accepted after review. This research explores the impact that design exceptions have on the accident severity and accident frequency in Indiana. Data on accidents at roadway sites with and without design exceptions are used to estimate appropriate statistical models for the frequency and severity accidents at these sites using some of the most recent statistical advances with mixing distributions. The results of the modeling process show that presence of approved design exceptions has not had a statistically significant effect on the average frequency or severity of accidents -- suggesting that current procedures for granting design exceptions have been sufficiently rigorous to avoid adverse safety impacts.
Abstract
The study finds no statistically significant effect of approved design exceptions on accident severity or frequency. Overall, the results suggest that the current approval process has avoided adverse safety consequences, although frequency-model differences warrant concern.
- Accident severity: Design exceptions were statistically insignificant in the accident-severity model, indicating no significant impact on injury severity.The mixed multinomial logit severity model had a McFadden ρ2 statistic above 0.5.
- Accident frequency: Design exceptions were also statistically insignificant in the accident-frequency model, indicating no significant impact on accident frequency.The negative binomial model had a McFadden ρ2 statistic above 0.75.
- Accident frequency: The likelihood-ratio test for separate frequency models produced X^2 = 23.00 with 10 degrees of freedom and a p-value of 0.0107.The test used 122 accident-frequency observations, limiting confidence in the separate frequency-model estimates.
- Overall conclusion: Overall, the current design-exception approval process appears sufficiently strict to avoid adverse safety consequences.The authors caution that potentially different processes generating accident frequencies across segment types may affect future design-exception decisions.
Parameter (t-ratio) Averaged elasticities*
The averaged-elasticity results report effects for design-exception and segment-type indicators, alongside driver and roadway characteristics. The design-exception indicator has a coefficient of .460 (.752) and reported elasticity values ranging from -.0149 to .0038.
- Driver characteristics: The female-fault indicator is reported as -.308 (2.06), with elasticity values of -.0155, .0543, and -.0155.The indicator equals 1 when the at-fault driver's gender is female and 0 otherwise.
- Driver characteristics: The middle-age indicator is reported as -.577 (3.12), with elasticity values of -.0162, .0501, and -.0162.The indicator equals 1 when the oldest driver's age is between 30 and 39 years old and 0 otherwise.
- Design exception and segment-type parameters: The design exception indicator is reported as .460 (.752), with elasticity values of -.0004, -.0007, .0038, -.0149, and .0038.The indicator equals 1 when the road segment had a design exception and 0 otherwise.
- Design exception and segment-type parameters: The bridge-segment indicator is reported as -.244 (-.395), with elasticity values of .0003, .0005, -.0119, .0443, and -.0119.The indicator equals 1 when the road segment is a bridge segment and 0 otherwise.