Impacts of traffic barrier types on crash severity

Abstract

Traffic barriers are known as one of the most popular safety countermeasures to minimize severe collisions from crossing medians and run-off-road at locations with non-traversable roadsides. This paper examines the impact of different traffic barrier types on crash severity with five years (2016–2020) crash data from Iowa using a causal inference framework. This includes genetic matching to reduce selection bias and the use of random parameters ordinal logistic regression to estimate the impacts of the barrier types on crash severity. Barriers were classified into three groups for the analysis: guardrails (i.e., semi-rigid), cable barriers (i.e., flexible) and concrete barriers (i.e., rigid). In order to make comparisons across barrier types, genetic matching was performed for paired comparisons (e.g., concrete vs. cable barrier) and then marginal effects were estimated to determine the differences in crash severity. According to the results, the cable barrier is associated with the lowest injury while the concrete barrier is associated with the highest injury levels. Guardrails are safer than concrete barriers and less safe than cable barriers. It is also found that if concrete barrier or guardrail is involved in the crash, there is 12.09 % and 7.69 % increased probability of an injury compared to cable barriers, respectively.