This study investigates the inclusion of spatio-temporal correlation and interaction in a multivariate random-parameters Tobit model their influence on fitting areal crash rates with different severity outcomes. The spatial is specified via conditional autoregressiv (MCAR) prior, whereas temporal by linear time trend. A formulated as product trend term an MCAR prior. developed for slight injury killed or serious using one year data from 131 traffic analysis zones Hong Kong. proposed...

This study presents an empirical investigation of the impacts real-time weather conditions on freeway crash severity. A Bayesian spatial generalized ordered logit model was developed for modeling severity using hourly wind speed, air temperature, precipitation, visibility, and humidity, as well other observed factors. total 1424 records from Kaiyang Freeway, China in 2014 2015 were collected investigation. The proposed can simultaneously accommodate nature levels correlation across adjacent...

This study develops three temporal multivariate random parameters Tobit models to analyze crash rate by injury severity; these simultaneously accommodate correlation and unobserved heterogeneity across observations correlations severity. The are estimated compared in the Bayesian context with a dataset collected from Hong Kong's Traffic Information System, which contains crash, road geometry, traffic, environmental information on 194 directional segments over five-year period (2002–2006)....

This study presents a joint analysis of daytime and nighttime crash frequencies at the zone level with consideration spatial correlations. Crash data from 131 traffic zones in Hong Kong 2011 are investigated. A Bayesian bivariate conditional autoregressive model is proposed to establish links between attributes, road network characteristics, land use patterns. The allows not only for distinct heterogeneous effects each dependent variable, but also correlations them.The parameter estimates...

With the enormous losses to society that result from highway crashes, gaining a better understanding of risk factors affect traffic crash occurrence has long been prominent focus safety research. In this study, we develop an optimised radial basis function neural network (RBFNN) model approximate nonlinear relationships between frequency and relevant factors. Our case study compares performance RBFNN with traditional negative binomial (NB) back-propagation (BPNN) models for prediction on...

Freeway transportation safety issues have attracted public concern in China for decades. This study aims to identify the factors influencing injury severity of freeway crashes and quantify their effects on likelihood various crash levels, with consideration heterogeneity interactions. The empirical analysis is based three years data from two mountainous freeways Guangdong, China, covering 2021 2023. A random parameters logit model interaction terms developed analysis. Goodness-of-fit...

The safety performance of horizontal and crest vertical curve combinations (also named as or combined curves) is substantially associated with their geometric design. To evaluate accurately, three Bayesian hierarchical negative binomial (NB) models various structures temporal correlation (including linear time trend, quadratic autoregressive-1) are proposed for building a relationship between crash frequency the separated design attributes combination on freeways. An 8 year (2011–2018)...

This paper presents the study on association between in-vehicle music listening, physiological and psychological response, driving performance, using simulator approach, with which personality (temperament) was considered. The performance indicators considered were standard deviation of speed, lane crossing frequency, perceived mental workload, mean variability heart rate. Additionally, effects presence genre (light versus rock music) Twenty participants different personalities (in...

To account for the spatial correlation of crashes that are in close proximity, this study proposes a Bayesian generalized ordered probit (SGOP) model with Leroux conditional autoregressive (CAR) prior crash severity analysis. Proposed can accommodate ordinal nature injury and relax assumption monotonic effects explanatory factors. Additionally, strength is considered. Results indicate proposed SGOP CAR outperforms conventional intrinsic CAR. There moderate crashes. factors including vehicle...