Finite mixture modeling approach for developing crash modification factors in highway safety analysis
Conference Paper
Overview
Identity
Additional Document Info
View All
Overview
abstract
This study aimed to investigate the relative performance of two alternative models (negative binomial (NB) model and two-component finite mixture of negative binomial models (FMNB-2)) in terms of crash modification factors (CFMs). For a modeling dataset, rural multilane segment crash data for divided highways in California and Texas were used. The resultant CMF function from the FMNB-2 model showed several good properties over the NB model. First, the safety effect of a covariate was better reflected by the CMF function from the FMNB-2 model, since the model takes into account the differential responsiveness of crash frequency to the covariate. Second, the safety effect did not increase continuously without a limit, but leveled off after a certain value of the covariate, which makes more logical sense. However, there was a possibility that a U-shaped relationship could be found, which opens up a new avenue for discussion among highway safety community.
