Injury severity analysis of motorcycle crashes: A comparison of latent class clustering and latent segmentation based models with unobserved heterogeneity

The latent class clustering and segmentation-based models are employed to account for heterogeneity across different groups. Further, the random parameter variants of these modeling frameworks consider within group. Both approaches have recently gained significant attention in road safety literature. However, similarities differences between two methods seldom explained investigated. To that end, this study proposes compare performance examining crash injury severity outcomes. These been developed based on an ordered logit framework accommodate ordinal nature levels. For outcomes, is first variant structure a segmentation scheme. current also tests incorporates temporal instability exogenous variables multiple years data estimated by using motorcycle Queensland, Australia, from year 2012 through 2016. comparison exercise augmented estimating aggregate level elasticity effects variables. highlights superiority approach compared clustering-based approach. Moreover, both performed better than their fixed-parameter counterparts, which need across- within-group heterogeneity. stability indicate rider year-wise models.