Numerous efforts have been devoted to investigating the relationship between traffic crashes by severity and explanatory factors. However, traditional methods estimate the crash frequency of each severity level separately, neglecting the correlations of severity levels, which can result in biased model estimations. The primary objective of this study is to develop a multivariate Poisson-lognormal regression model (MVPLN), which can simultaneously model crash counts by severity, to evaluate the effects of contributing factors to crash frequency and to identify the correlations among different severity levels. Crash data were collected at 263 exit ramps on freeways in Florida, United States. The Markov chain Monte Carlo (MCMC) method was used to get the estimation solution of MVPLN model. Significant correlations were identified among crash counts at different severity levels. The effects of crash-related factors of each severity level were explored in the MVPLN model considering the correlation structure.
Using Multivariate Poisson-Lognormal Regression Method for Modeling Crash Frequency by Severity on Freeway Diverge Areas
11th International Conference of Chinese Transportation Professionals (ICCTP) ; 2011 ; Nanjing, China
ICCTP 2011 ; 2385-2394
2011-07-26
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Multivariate Poisson-Lognormal Models for Jointly Modeling Crash Frequency by Severity
Online Contents | 2007
|Analyzing Crash Frequency by Collision Type at Freeway Diverge Areas
Trans Tech Publications | 2013
|Multivariate Poisson-Lognormal Models for Jointly Modeling Crash Frequency by Severity
Transportation Research Record | 2007
|Bayesian Multivariate Poisson Lognormal Models for Crash Severity Modeling and Site Ranking
Transportation Research Record | 2009
|Analysis of Crash Rates at Freeway Diverge Areas using Bayesian Tobit Modeling Framework
Transportation Research Record | 2019
|