Traffic safety analysis and model updating for freeways using Bayesian method

Wang, Xuesong; Zhang, Qi; Yang, Xin; Pei, Yingying; Yuan, Jinghui

doi:10.1080/19439962.2022.2128957

Cited by 3 publications

(2 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The full Bayesian (FB) method was adopted for statistical inference. In the FB method, the posterior distributions are calculated based on the prior distributions and the observed information ( 26 ). This method can effectively avoid the unreliability in the results that is caused by large deviations of the data.…”

Section: Methodsmentioning

confidence: 99%

Meso Safety Analysis of Major Arterials: The Impact of Intersection Spacing and Network Connection Compatibility

Wang,

Dai,

Hsu

et al. 2023

Transportation Research Record: Journal of the Transportation R

Self Cite

View full text Add to dashboard Cite

With the increase in high-density road network construction in many urban areas, neither macro- nor micro-level safety analysis can fully explain the impact of network characteristics on safety. This study employed a meso-level analysis unit, which combines intersections with adjacent road segments, to investigate the safety impact of intersection spacing and network connection compatibility along 19 major arterials in Suzhou, China. Three main intersection spacing characteristics, the density of intersection spacing (DOIS), the standard deviation of intersection spacing (SDOIS), and the variation of intersection spacing (VOIS), were used to represent the interaction between arterials and other roads in the network. This study introduced the concept of hierarchical differences to quantitatively analyze the safety effects of connections between roads with different accessibility and mobility functions. To manage the hierarchical structure of the data (which has both within-arterial and between-arterial correlations), a two-level hierarchical negative binomial conditional autoregressive model was adopted to examine the safety effects of network characteristics at the meso level. The results showed the existence of unobserved heterogeneities among the selected arterials. The SDOIS and VOIS were found to have positive effects on crash frequency (more crashes on units with higher SDOIS and VOIS), while the DOIS was negatively associated with crash frequency. The meso units with moderately compatible network connections were found to have more crashes than those with compatible connections, while the units with the highest hierarchical differences had fewer crashes. The results from this study are expected to provide suggestions on potential safety hazard screening and crash prevention.

show abstract

Section: Methodsmentioning

confidence: 99%

Meso Safety Analysis of Major Arterials: The Impact of Intersection Spacing and Network Connection Compatibility

Wang,

Dai,

Hsu

et al. 2023

Transportation Research Record: Journal of the Transportation R

Self Cite

View full text Add to dashboard Cite

show abstract

“…In addition, the crash risk of ramp segments is higher than that of general mainline segments: the greater the number of ramps, the higher the crash risk. Different types of ramps have different effects on crashes (13). Vertical alignment is also an important influencing factor on freeway safety, as, for example, crashes increase with the steepness of longitudinal slope (14).…”

Section: Safety Influencing Factors On Freewaysmentioning

confidence: 99%

Analyzing Freeway Safety Influencing Factors Using the CatBoost Model and Interpretable Machine-Learning Framework, SHAP

Li,

Wang,

Yang

et al. 2023

Transportation Research Record: Journal of the Transportation R

Self Cite

View full text Add to dashboard Cite

Exploring and analyzing safety influencing factors can guide targeted traffic safety management. Traditional traffic safety models are aimed at specific data problems and making adjustments to the model structure, which lack focus on predictive ability and have limited information on the analysis of influencing factors. In recent years, machine-learning methods have opened new avenues in modeling that have higher prediction accuracy, can identify complex nonlinear relationships, and can overcome over- and under-dispersion and correlation. Machine-learning methods, however, pose the problem of limited interpretability. The interpretable machine-learning framework SHAP can be an effective solution, which can not only reflect the influence of features in each sample but also generate global interpretation. This study established gradient boosting models including the CatBoost and XGBoost models as traffic safety models, which were compared with a traditional NB regression model and a zero-inflated negative binomial regression model. SHAP was used to analyze several safety influencing factors, including geometric design features, traffic operation characteristics, time of day, and land use. Results confirmed that the CatBoost model has better prediction ability and is a more suitable traffic safety model than the traditional negative binomial regression model. Among the key findings were that ramp type is the most important factor in freeway crash frequency; curve presence has a great positive impact, while truck proportion has a great negative impact; and traffic volume is highly correlated with truck proportion. These findings can provide theoretical support for safety operation management and targeted improvement measures for freeways.

show abstract

Incorporating spatio-temporal correlation and interaction into a generalized ordered logit model for examining injury severities of traffic accidents

Zeng,

Xu,

Liang

et al. 2024

Journal of Transportation Safety & Security

View full text Add to dashboard Cite

Traffic safety analysis and model updating for freeways using Bayesian method

Cited by 3 publications

References 38 publications

Meso Safety Analysis of Major Arterials: The Impact of Intersection Spacing and Network Connection Compatibility

Meso Safety Analysis of Major Arterials: The Impact of Intersection Spacing and Network Connection Compatibility

Analyzing Freeway Safety Influencing Factors Using the CatBoost Model and Interpretable Machine-Learning Framework, SHAP

Incorporating spatio-temporal correlation and interaction into a generalized ordered logit model for examining injury severities of traffic accidents

Contact Info

Product

Resources

About