Factors that affect highway-related crash frequency and injury severity vary across observations. Using a methodology that does not account nor correct for heterogeneity in observed and unobserved crash factors across highway segments may lead to biased and inconsistent estimated coefficients, thus resulting in erroneous inferences. The present paper demonstrates the use of random-parameters models to facilitate and enhance how crash factors affect crash frequency and injury severity along a highway segment. The results indicate that a unit increase in the presence of stop sign along a highway segment reduces crash frequency by 2.471 for 87.24% of the roadway segments. For the remaining 12.76% of the roadway segments, crash frequency is increased by the same margin. Using the random-parameters multinomial logit model, the result indicates that, for 90.89% of the observations, the presence of a stop sign on a highway segment increases the probability of the injury outcome. For 9.11% of the observations, the presence of a stop sign on a highway segment reduces the probability of the injury outcome, and the marginal effect value across observations is 0.0017. Vertical grades greater than 5% increase crash frequency for 58.46% of the highway segments, and decrease for 41.54% of the highway segments by 0.121 for one unit increase in vertical grades.
Random-parameters analysis of highway characteristics on crash frequency and injury severity
2016
Article (Journal)
Electronic Resource
Unknown
Metadata by DOAJ is licensed under CC BY-SA 1.0
A multivariate random-parameters Tobit model for analyzing highway crash rates by injury severity
Online Contents | 2017
|A spatial generalized ordered response model to examine highway crash injury severity
Online Contents | 2013
|Cluster-Based Approach to Analyzing Crash Injury Severity at Highway–Rail Grade Crossings
Transportation Research Record | 2017
|Injury Related Parameters and Crash Severity in Frontal Impact Barrier Tests
SAE Technical Papers | 1990
|