Crash risk analysis for Shanghai urban expressways: A Bayesian semi-parametric modeling approach

Abstract

Urban expressway systems have been developed rapidly in recent years in China; it has become one key part of the city roadway networks as carrying large traffic volume and providing high traveling speed. Along with the increase of traffic volume, traffic safety has become a major issue for Chinese urban expressways due to the frequent crash occurrence and the non-recurrent congestions caused by them. For the purpose of unveiling crash occurrence mechanisms and further developing Active Traffic Management (ATM) control strategies to improve traffic safety, this study developed disaggregate crash risk analysis models with loop detector traffic data and historical crash data. Bayesian random effects logistic regression models were utilized as it can account for the unobserved heterogeneity among crashes. However, previous crash risk analysis studies formulated random effects distributions in a parametric approach, which assigned them to follow normal distributions. Due to the limited information known about random effects distributions, subjective parametric setting may be incorrect. In order to construct more flexible and robust random effects to capture the unobserved heterogeneity, Bayesian semi-parametric inference technique was introduced to crash risk analysis in this study. Models with both inference techniques were developed for total crashes; semi-parametric models were proved to provide substantial better model goodness-of-fit, while the two models shared consistent coefficient estimations. Later on, Bayesian semi-parametric random effects logistic regression models were developed for weekday peak hour crashes, weekday non-peak hour crashes, and weekend non-peak hour crashes to investigate different crash occurrence scenarios. Significant factors that affect crash risk have been revealed and crash mechanisms have been concluded.

Approach-level real-time crash risk analysis for signalized intersections

“…Therefore, the significant variables in four time-slice models were investigated. Average speed was found to have significant negative effect on the odds of crash occurrence, which is consistent with previous studies , Ahmed et al 2012a, b, Ahmed and Abdel-Aty 2012, Xu et al 2012, Shi and Abdel-Aty 2015, Yu et al 2016, Yuan et al 2018. The left turn volume was surprisingly found to be negatively correlated with the odds of crash occurrence, which might be explained as the higher left turn volume may increase the driver awareness when approaching the entering approach, which may improve the safety performance.…”

Utilizing bluetooth and adaptive signal control data for real-time safety analysis on urban arterials

“…The results of the slice 2 model indicate that the average speed, upstream left-turn volume, downstream green ratio, and rainy indicator are significantly associated with the crash risk on urban arterials. In general, these finding are consistent with previous studies, in which the average speed was found to have significant negative impact on crash occurrence Ahmed et al, 2012a, b;Ahmed and Abdel-Aty, 2012;Shi and Abdel-Aty, 2015;Xu et al, 2012;Yu et al, 2016), while adverse weather (Ahmed et al, 2012a;Xu et al, 2013a) were found to be positively correlated with crash likelihood. In terms of the effect of traffic volume, only the upstream left-turn volume was found to have significant effect on crash likelihood, which indicates that more vehicles from the intersecting road segment left turning into the subject segment may increase the crash risk on the segment.…”

“…Finally, the slice 2 model was selected to conduct further interpretation and model comparison. Based on the estimation results in the slice 2 model, four variables were found to be significantly associated with the crash occurrence on urban arterials: (1) the negative coefficient (-0.025) of average speed indicates that higher average speed tends to decrease the crash risk, which is consistent with other studies Ahmed et al, 2012a, b;Ahmed and Abdel-Aty, 2012;Shi and Abdel-Aty, 2015;Xu et al, 2012;Yu et al, 2016). This could be explained as the traffic condition with higher average speed, which represents more smooth traffic flow, could have better safety performance.…”

“…Real-time crash risk analysis has been widely adopted to reveal crash occurrence precursors by investigating the differences in traffic conditions between crash and non-crash events. As crash risk analysis is a typical binary classification problem, the most commonly used methods are the matched case-control logistic models (Abdel-Aty and Pande, 2005;Abdel-Aty et al, 2004;Ahmed and Abdel-Aty, 2012;Xu et al, 2012;Zheng et al, 2010), Bayesian logistical models (Ahmed et al, 2012a;Shi and Abdel-Aty, 2015;Wang et al, 2017a;Wang et al, 2015a;, Bayesian random effect logistic models (Shi and Abdel-Aty, 2015;Yu et al, 2016), Bayesian random parameter logistic models (Shi and Abdel-Aty, 2015;Xu et al, 2014;Yu et al, 2017). Besides, several approaches of data mining such as neural networks (Abdel-Aty and Pande, 2005;Abdel-Aty et al, 2008), support vector machines (Yu and Abdel-Aty, 2013;, and Bayesian networks (Hossain and Muromachi, 2012;Sun and Sun, 2015) were also applied to evaluate the real-time crash risk.…”

Predicting Crash Frequency for Urban Expressway considering Collision Types Using Real-Time Traffic Data

“…e crash risk analysis models were developed for total crashes and timerelated crashes to reveal the significant factors that affect crash risk [17]. In addition, typical scenarios leading to an accident were found from the modelling results.…”