Accident; analysis and prevention最新文献_第7页

Assessing the impact of car-following driving style on traffic conflict risk using asymmetric behavior model and explainable machine learning 使用不对称行为模型和可解释机器学习评估汽车跟随驾驶方式对交通冲突风险的影响。

IF 5.7 1区工程技术 Q1 ERGONOMICS

Accident; analysis and prevention

Pub Date : 2024-12-24 DOI: 10.1016/j.aap.2024.107904

Xiao-chi Ma , Yun-hao Zhou , Jian Lu , Yiik Diew Wong , Jun Zhang , Junde Chen , Chao Gu

To deepen the understanding of the impact of car-following driving style (CFDS) on traffic conflict risk and address the lack of clear CFDS evaluation metrics, this study proposes an improved CFDS metric based on the Asymmetric Behavior (AB) theory. Interpretable machine learning models were utilized for regression analysis to examine the relationship between CFDS and conflict risk. The generalized AB model calculates the difference between vehicle trajectories and the Newell trajectory, constructing the driving style evaluation metric, which quantifies driver aggressiveness in a manner that is both computationally straightforward and easily interpretable. High-precision vehicle trajectory data were collected using radar-camera integrated devices, enabling the use of various interpretable machine learning methods to model and analyze the impact of driving style on conflict risk. The results demonstrate that the proposed car-following driving style evaluation metric consistently shows the highest importance across multiple datasets with different risk levels and sampling windows, indicating a strong correlation with conflict risk. Interpretations using Shapley Additive Explanations reveal a nuanced, yet mostly monotonic impact pattern of driving style across high, medium, and low-risk scenarios, with more aggressive drivers being more prone to high-risk situations. Furthermore, Partial Dependence Plot analysis reveals a complex, saddle-shaped risk curve related to driving style and its interactions, highlighting that aggressive and “pseudo-timid” drivers exhibit higher risks in specific contexts. In summary, this research constructs clear and interpretable CFDS evaluation metrics, validated through case analysis for their rationality and effectiveness, thereby providing new theoretical support for traffic risk prediction and intervention.

为了加深对跟车驾驶方式对交通冲突风险影响的认识，并解决目前缺乏明确的跟车驾驶方式评价指标的问题，本研究提出了一种基于不对称行为理论的改进的跟车驾驶方式评价指标。利用可解释的机器学习模型进行回归分析，检验差价合约与冲突风险之间的关系。广义AB模型计算车辆轨迹与Newell轨迹之间的差异，构建驾驶风格评估指标，以一种计算简单且易于解释的方式量化驾驶员的攻击性。使用雷达-摄像机集成设备收集高精度车辆轨迹数据，使用各种可解释的机器学习方法来建模和分析驾驶风格对冲突风险的影响。结果表明，所提出的跟随汽车驾驶风格评价指标在不同风险水平和采样窗口的多个数据集上始终显示出最高的重要性，表明其与冲突风险有很强的相关性。使用Shapley加性解释的解释揭示了在高、中、低风险场景中，驾驶风格的细微差异，但大多是单调的影响模式，更具攻击性的司机更容易出现高风险情况。此外，部分依赖图分析揭示了与驾驶风格及其相互作用相关的复杂鞍形风险曲线，突出表明攻击性和“伪胆小”司机在特定环境下表现出更高的风险。综上所述，本研究构建了清晰、可解释的cfd评价指标，并通过案例分析验证了其合理性和有效性，为交通风险预测和干预提供了新的理论支持。

{"title":"Assessing the impact of car-following driving style on traffic conflict risk using asymmetric behavior model and explainable machine learning","authors":"Xiao-chi Ma , Yun-hao Zhou , Jian Lu , Yiik Diew Wong , Jun Zhang , Junde Chen , Chao Gu","doi":"10.1016/j.aap.2024.107904","DOIUrl":"10.1016/j.aap.2024.107904","url":null,"abstract":"<div><div>To deepen the understanding of the impact of car-following driving style (CFDS) on traffic conflict risk and address the lack of clear CFDS evaluation metrics, this study proposes an improved CFDS metric based on the Asymmetric Behavior (AB) theory. Interpretable machine learning models were utilized for regression analysis to examine the relationship between CFDS and conflict risk. The generalized AB model calculates the difference between vehicle trajectories and the Newell trajectory, constructing the driving style evaluation metric, which quantifies driver aggressiveness in a manner that is both computationally straightforward and easily interpretable. High-precision vehicle trajectory data were collected using radar-camera integrated devices, enabling the use of various interpretable machine learning methods to model and analyze the impact of driving style on conflict risk. The results demonstrate that the proposed car-following driving style evaluation metric consistently shows the highest importance across multiple datasets with different risk levels and sampling windows, indicating a strong correlation with conflict risk. Interpretations using Shapley Additive Explanations reveal a nuanced, yet mostly monotonic impact pattern of driving style across high, medium, and low-risk scenarios, with more aggressive drivers being more prone to high-risk situations. Furthermore, Partial Dependence Plot analysis reveals a complex, saddle-shaped risk curve related to driving style and its interactions, highlighting that aggressive and “pseudo-timid” drivers exhibit higher risks in specific contexts. In summary, this research constructs clear and interpretable CFDS evaluation metrics, validated through case analysis for their rationality and effectiveness, thereby providing new theoretical support for traffic risk prediction and intervention.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"211 ","pages":"Article 107904"},"PeriodicalIF":5.7,"publicationDate":"2024-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142891209","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Examining the nonlinear effects of traffic and built environment factors on the traffic safety of cyclist from different age groups 考察了交通和建成环境因素对不同年龄段骑车人交通安全的非线性影响。

IF 5.7 1区工程技术 Q1 ERGONOMICS

Accident; analysis and prevention

Pub Date : 2024-12-24 DOI: 10.1016/j.aap.2024.107872

M. Baran Ulak , Mehrnaz Asadi , Karst T. Geurs

In the Netherlands and all over the world, traffic safety problem has been growing particularly for cyclists over the last decades with more people shifting to cycling as a healthy and sustainable mode of transport. Literature shows that age is an important factor in crash involvement and consequences; however, few studies identify the risk factors for cyclists from across different age groups. Therefore, this study aims to identify and understand the effects of traffic, infrastructure, and land use factors on vehicle-to-bike injury and fatal crashes involving cyclists from different age groups. For this purpose, we adopted an approach consisting of resampling and machine learning (XGBoost-Tweedie) techniques to analyse police-reported crashes between the years 2015 and 2019 in the Netherlands. The analysis shows that effects of external variables on crashes widely vary among different age groups and the analysis of total crash rates may not disclose the nature of crashes of cyclist from different age groups. The analysis also shed light on the nonlinear effects of traffic and built environment factors on cyclist crashes, which are usually disregarded in the traffic safety literature. The proposed approach and findings provide a profound understanding of the nature of cyclist crashes and the complex relationships between factors, which can contribute to developing effective crash prevention strategies tailored to different age groups.

在荷兰和世界各地，在过去的几十年里，随着越来越多的人将骑自行车作为一种健康和可持续的交通方式，交通安全问题越来越严重，尤其是对骑自行车的人来说。文献表明，年龄是影响车祸卷入和后果的重要因素；然而，很少有研究确定不同年龄段骑自行车者的危险因素。因此，本研究旨在确定和了解交通、基础设施和土地利用因素对不同年龄段骑自行车者的车辆与自行车碰撞伤害和致命事故的影响。为此，我们采用了一种由重新采样和机器学习（XGBoost-Tweedie）技术组成的方法来分析荷兰警方报告的2015年至2019年之间的撞车事故。分析表明，外部变量对碰撞的影响在不同年龄组之间存在很大差异，总碰撞率的分析可能无法揭示不同年龄组骑车人碰撞的性质。该分析还揭示了交通和建筑环境因素对骑自行车者碰撞的非线性影响，这在交通安全文献中通常被忽视。所提出的方法和研究结果提供了对骑自行车者碰撞的本质和因素之间复杂关系的深刻理解，这有助于制定针对不同年龄组的有效碰撞预防策略。

{"title":"Examining the nonlinear effects of traffic and built environment factors on the traffic safety of cyclist from different age groups","authors":"M. Baran Ulak , Mehrnaz Asadi , Karst T. Geurs","doi":"10.1016/j.aap.2024.107872","DOIUrl":"10.1016/j.aap.2024.107872","url":null,"abstract":"<div><div>In the Netherlands and all over the world, traffic safety problem has been growing particularly for cyclists over the last decades with more people shifting to cycling as a healthy and sustainable mode of transport. Literature shows that age is an important factor in crash involvement and consequences; however, few studies identify the risk factors for cyclists from across different age groups. Therefore, this study aims to identify and understand the effects of traffic, infrastructure, and land use factors on vehicle-to-bike injury and fatal crashes involving cyclists from different age groups. For this purpose, we adopted an approach consisting of resampling and machine learning (XGBoost-Tweedie) techniques to analyse police-reported crashes between the years 2015 and 2019 in the Netherlands. The analysis shows that effects of external variables on crashes widely vary among different age groups and the analysis of total crash rates may not disclose the nature of crashes of cyclist from different age groups. The analysis also shed light on the nonlinear effects of traffic and built environment factors on cyclist crashes, which are usually disregarded in the traffic safety literature. The proposed approach and findings provide a profound understanding of the nature of cyclist crashes and the complex relationships between factors, which can contribute to developing effective crash prevention strategies tailored to different age groups.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"211 ","pages":"Article 107872"},"PeriodicalIF":5.7,"publicationDate":"2024-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142891218","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Cooperative control of self-learning traffic signal and connected automated vehicles for safety and efficiency optimization at intersections 自主学习交通信号与互联自动车辆协同控制，优化交叉口安全与效率。

IF 5.7 1区工程技术 Q1 ERGONOMICS

Accident; analysis and prevention

Pub Date : 2024-12-19 DOI: 10.1016/j.aap.2024.107890

Gongquan Zhang , Fengze Li , Dian Ren , Helai Huang , Zilong Zhou , Fangrong Chang

Cooperative control of intersection signals and connected automated vehicles (CAVs) possess the potential for safety enhancement and congestion alleviation, facilitating the integration of CAVs into urban intelligent transportation systems. This research proposes an innovative deep reinforcement learning-based (DRL) cooperative control framework, including signal and speed modules, to dynamically adapt signal timing and CAV velocities for traffic safety and efficiency optimization. Among the DRL-based signal modules, a traffic state prediction model is merged with the current state to augment characteristics and the agent-learning process. A multi-objective reward function is designed to evaluate safety and efficiency using a traffic conflict prediction model and vehicle waiting time. The double deep Q network (DDQN) model is used to design the agent observing the traffic state, learning the optimal signal control policy, and then inputting the signal phase into the speed module. Based on the green duration analysis and constraints of mixed traffic flow of CAVs and human-driven vehicles, a speed planning model is constructed to optimize CAVs’ speed and alter traffic state, which in turn affects the agent’s next signal decisions. The proposed framework is tested at isolated intersections simulated by two real-world intersections in Changsha, China. The results reveal the superiority of the proposed method over DRL-based traffic signal control (DRL-TSC) in terms of coverage speed and computation time. Compared to actuated signal control, adaptive traffic signal control, and DRL-TSC, the proposed method significantly optimizes traffic safety and efficiency across diverse intersections, temporal spans, and traffic demands. Furthermore, the advantage of the proposed method substantially amplifies with the increased CAV penetration, regardless of the intersection types.

交叉口信号与网联自动驾驶汽车（cav）的协同控制具有增强安全性和缓解拥堵的潜力，促进了cav与城市智能交通系统的融合。本研究提出了一种创新的基于深度强化学习（DRL）的协同控制框架，包括信号和速度模块，以动态适应信号配时和自动驾驶汽车速度，以优化交通安全和效率。在基于drl的信号模块中，将交通状态预测模型与当前状态合并以增强特征和智能体学习过程。利用交通冲突预测模型和车辆等待时间，设计了一个多目标奖励函数来评价安全与效率。采用双深度Q网络（DDQN）模型设计智能体观察交通状态，学习最优信号控制策略，然后将信号相位输入到限速模块。基于对自动驾驶汽车和人类驾驶汽车混合交通流的绿时分析和约束，构建速度规划模型，优化自动驾驶汽车的速度和改变交通状态，进而影响智能体的下一个信号决策。该框架在中国长沙的两个真实交叉口模拟的孤立交叉口上进行了测试。结果表明，该方法在覆盖速度和计算时间上优于基于drl的交通信号控制方法。与驱动信号控制、自适应交通信号控制和DRL-TSC相比，该方法在不同的交叉口、时间跨度和交通需求下显著优化了交通安全和效率。此外，无论交叉口类型如何，该方法的优势都随着CAV穿透的增加而大大增强。

{"title":"Cooperative control of self-learning traffic signal and connected automated vehicles for safety and efficiency optimization at intersections","authors":"Gongquan Zhang , Fengze Li , Dian Ren , Helai Huang , Zilong Zhou , Fangrong Chang","doi":"10.1016/j.aap.2024.107890","DOIUrl":"10.1016/j.aap.2024.107890","url":null,"abstract":"<div><div>Cooperative control of intersection signals and connected automated vehicles (CAVs) possess the potential for safety enhancement and congestion alleviation, facilitating the integration of CAVs into urban intelligent transportation systems. This research proposes an innovative deep reinforcement learning-based (DRL) cooperative control framework, including signal and speed modules, to dynamically adapt signal timing and CAV velocities for traffic safety and efficiency optimization. Among the DRL-based signal modules, a traffic state prediction model is merged with the current state to augment characteristics and the agent-learning process. A multi-objective reward function is designed to evaluate safety and efficiency using a traffic conflict prediction model and vehicle waiting time. The double deep Q network (DDQN) model is used to design the agent observing the traffic state, learning the optimal signal control policy, and then inputting the signal phase into the speed module. Based on the green duration analysis and constraints of mixed traffic flow of CAVs and human-driven vehicles, a speed planning model is constructed to optimize CAVs’ speed and alter traffic state, which in turn affects the agent’s next signal decisions. The proposed framework is tested at isolated intersections simulated by two real-world intersections in Changsha, China. The results reveal the superiority of the proposed method over DRL-based traffic signal control (DRL-TSC) in terms of coverage speed and computation time. Compared to actuated signal control, adaptive traffic signal control, and DRL-TSC, the proposed method significantly optimizes traffic safety and efficiency across diverse intersections, temporal spans, and traffic demands. Furthermore, the advantage of the proposed method substantially amplifies with the increased CAV penetration, regardless of the intersection types.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"211 ","pages":"Article 107890"},"PeriodicalIF":5.7,"publicationDate":"2024-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142871036","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Do automation and digitalization distract drivers? A systematic review 自动化和数字化会分散司机的注意力吗？系统回顾。

IF 5.7 1区工程技术 Q1 ERGONOMICS

Accident; analysis and prevention

Pub Date : 2024-12-19 DOI: 10.1016/j.aap.2024.107888

Neelima C. Vijay , Amit Agarwal , Kamini Gupta

Driving is a multifaceted activity involving a complex interplay of cognitive, perceptual, and motor skills, demanding continuous attention on the road. In recent years, the increased integration of automation and digitalization technologies in vehicles has improved drivers’ convenience and safety. However, the spare attentional capacity available during automation and the prevalence of various infotainment systems in vehicles enable drivers to perform some secondary tasks not related to driving, which may divert their attention away from the road, increasing the chances of accidents. The objective of the present study is to conduct a comprehensive systematic review of existing literature utilizing an eye tracker to analyze driver distraction due to automation and/or digitalization in motorized vehicles, with a focus on identifying the key factors leading to visual distraction. Through a literature search on five databases: Google Scholar, PubMed, ScienceDirect, Scopus, and Web of Science, a total of 4769 articles were initially identified. After a systematic screening, 65 research articles are considered for the review. The findings of the study indicate an increase in the research conducted on driver distraction due to automation and/or digitalization over recent years, with the highest contribution of studies from the United States and China. The lack of studies from other parts of the world like South America, Africa and the limited representation from larger parts of Asia, specifically India, highlights the need for future research in the area. Studies report a diversion in drivers’ visual attention away from the roadway, in terms of long and frequent off-road glances, while engaging in secondary tasks during automation and/or digitalization. Studies also demonstrate changes in the pattern of drivers’ visual attention with respect to different factors like HMI information, type of secondary task, type of input modality, in-vehicle display characteristics, and vehicle automation. Studies have also found success in using feedback to reduce visual distraction and to bring back drivers’ attention on the roads. In light of the findings observed, the review provides a discussion on the effects of automation and/or digitalization technologies on drivers’ visual attention. The study also highlights the areas that are not explored despite the wealth of research available on the topic.

驾驶是一项多方面的活动，涉及认知、知觉和运动技能的复杂相互作用，需要持续关注道路。近年来，自动化和数字化技术在汽车上的日益融合，提高了驾驶员的便利性和安全性。然而，自动化过程中可用的空闲注意力以及各种信息娱乐系统的普及使驾驶员能够执行一些与驾驶无关的次要任务，这可能会分散他们对道路的注意力，增加事故发生的机会。本研究的目的是对现有文献进行全面系统的回顾，利用眼动仪分析机动车辆中由于自动化和/或数字化导致的驾驶员分心，重点是确定导致视觉分心的关键因素。通过b谷歌Scholar、PubMed、ScienceDirect、Scopus和Web of Science五个数据库的文献检索，初步确定了4769篇文章。经过系统筛选，65篇研究论文被纳入综述。研究结果表明，近年来，对自动驾驶和/或数字化导致的驾驶员分心的研究有所增加，其中美国和中国的研究贡献最大。来自世界其他地区（如南美、非洲）的研究不足，以及来自亚洲大部分地区（特别是印度）的研究有限，突显了在该领域开展未来研究的必要性。研究报告称，在自动化和/或数字化过程中，驾驶员在从事次要任务时，会将视觉注意力从道路上转移，比如长时间和频繁的越野扫视。研究还揭示了驾驶员视觉注意模式在人机界面信息、辅助任务类型、输入方式类型、车载显示特征、车辆自动化程度等不同因素下的变化。研究还发现，使用反馈来减少视觉干扰并将驾驶员的注意力重新集中在道路上是成功的。根据观察到的结果，本综述讨论了自动化和/或数字化技术对驾驶员视觉注意力的影响。该研究还强调了尽管有大量关于该主题的研究，但尚未探索的领域。

{"title":"Do automation and digitalization distract drivers? A systematic review","authors":"Neelima C. Vijay , Amit Agarwal , Kamini Gupta","doi":"10.1016/j.aap.2024.107888","DOIUrl":"10.1016/j.aap.2024.107888","url":null,"abstract":"<div><div>Driving is a multifaceted activity involving a complex interplay of cognitive, perceptual, and motor skills, demanding continuous attention on the road. In recent years, the increased integration of automation and digitalization technologies in vehicles has improved drivers’ convenience and safety. However, the spare attentional capacity available during automation and the prevalence of various infotainment systems in vehicles enable drivers to perform some secondary tasks not related to driving, which may divert their attention away from the road, increasing the chances of accidents. The objective of the present study is to conduct a comprehensive systematic review of existing literature utilizing an eye tracker to analyze driver distraction due to automation and/or digitalization in motorized vehicles, with a focus on identifying the key factors leading to visual distraction. Through a literature search on five databases: Google Scholar, PubMed, ScienceDirect, Scopus, and Web of Science, a total of 4769 articles were initially identified. After a systematic screening, 65 research articles are considered for the review. The findings of the study indicate an increase in the research conducted on driver distraction due to automation and/or digitalization over recent years, with the highest contribution of studies from the United States and China. The lack of studies from other parts of the world like South America, Africa and the limited representation from larger parts of Asia, specifically India, highlights the need for future research in the area. Studies report a diversion in drivers’ visual attention away from the roadway, in terms of long and frequent off-road glances, while engaging in secondary tasks during automation and/or digitalization. Studies also demonstrate changes in the pattern of drivers’ visual attention with respect to different factors like HMI information, type of secondary task, type of input modality, in-vehicle display characteristics, and vehicle automation. Studies have also found success in using feedback to reduce visual distraction and to bring back drivers’ attention on the roads. In light of the findings observed, the review provides a discussion on the effects of automation and/or digitalization technologies on drivers’ visual attention. The study also highlights the areas that are not explored despite the wealth of research available on the topic.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"211 ","pages":"Article 107888"},"PeriodicalIF":5.7,"publicationDate":"2024-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142871037","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Influence of road safety policies on the long-term trends in fatal Crashes: A Gaussian Copula-based time series count model with an autoregressive moving average process 道路安全政策对致命碰撞长期趋势的影响：基于高斯copula的自回归移动平均过程时间序列计数模型。

IF 5.7 1区工程技术 Q1 ERGONOMICS

Accident; analysis and prevention

Pub Date : 2024-12-19 DOI: 10.1016/j.aap.2024.107795

Yanqi Lian , Shamsunnahar Yasmin , Md Mazharul Haque

Time series analysis plays a vital role in modeling historical crash trends and predicting the possible changes in future crash trends. In existing safety literature, earlier studies employed multiple approaches to model long-term crash risk profiles, such as integer-valued autoregressive Poisson regression model, integer-valued generalized autoregressive conditional heteroscedastic model, and generalized linear autoregressive and moving average models. However, these modeling frameworks often fail to fully capture several key properties of crash count data, especially negative serial correlation, and nonlinear dependence structures across temporal crash counts. To address these methodological gaps in existing safety literature, this study proposes to use a Gaussian Copula-based model for the long-term crash trend analysis. Specifically, this study proposes to use a Gaussian Copula-based Time Series Count Model with an Autoregressive Moving Average Process for the analysis of long-term trends in fatal crashes. The proposed approach can accommodate several data properties, which include (1) non-negative discrete property of count data, (2) positive and negative serial correlations among time series data, and (3) nonlinear dependence among time-series observations. The performance of the Gaussian Copula-based time series count model is compared with the generalized linear autoregressive and moving average model. The proposed modeling approaches are demonstrated by using yearly fatal crash count data for the years 1986 through 2022 from Queensland, Australia. The major safety interventions implemented in Queensland over those years are also highlighted to assess the possible and plausible impacts of these safety interventions in reducing fatal crash risks. Further, elasticity effects and overall percentage changes in fatal crashes across different time points are computed to demonstrate the implications of the proposed model. The policy analysis exercise shows that the implemented road safety interventions are likely to have diminishing marginal returns, underscoring the need for new and effective road safety policies to achieve the goal of zero fatalities within the set timeframe.

时间序列分析在模拟历史碰撞趋势和预测未来碰撞趋势的可能变化方面起着至关重要的作用。在现有的安全文献中，早期的研究采用了多种方法来模拟长期碰撞风险概况，如整值自回归泊松回归模型、整值广义自回归条件异方差模型、广义线性自回归和移动平均模型。然而，这些建模框架往往不能完全捕获崩溃计数数据的几个关键属性，特别是负序列相关性和跨时间崩溃计数的非线性依赖结构。为了解决现有安全文献中这些方法上的差距，本研究建议使用基于高斯copula的模型进行长期碰撞趋势分析。具体而言，本研究提出使用基于高斯copula的时间序列计数模型和自回归移动平均过程来分析致命事故的长期趋势。该方法可以适应多种数据特性，包括：(1)计数数据的非负离散性，(2)时间序列数据之间的正序列和负序列相关性，以及(3)时间序列观测之间的非线性相关性。比较了基于高斯copula的时间序列计数模型与广义线性自回归模型和移动平均模型的性能。通过使用澳大利亚昆士兰州1986年至2022年的年度致命碰撞计数数据，证明了所提出的建模方法。还强调了这些年来在昆士兰州实施的主要安全干预措施，以评估这些安全干预措施在减少致命碰撞风险方面可能产生的和合理的影响。此外，计算了弹性效应和不同时间点致命碰撞的总体百分比变化，以证明所提出模型的含义。政策分析工作表明，所实施的道路安全干预措施的边际收益可能会递减，这突出表明需要制定新的有效道路安全政策，以便在规定的时间范围内实现零死亡的目标。

{"title":"Influence of road safety policies on the long-term trends in fatal Crashes: A Gaussian Copula-based time series count model with an autoregressive moving average process","authors":"Yanqi Lian , Shamsunnahar Yasmin , Md Mazharul Haque","doi":"10.1016/j.aap.2024.107795","DOIUrl":"10.1016/j.aap.2024.107795","url":null,"abstract":"<div><div>Time series analysis plays a vital role in modeling historical crash trends and predicting the possible changes in future crash trends. In existing safety literature, earlier studies employed multiple approaches to model long-term crash risk profiles, such as integer-valued autoregressive Poisson regression model, integer-valued generalized autoregressive conditional heteroscedastic model, and generalized linear autoregressive and moving average models. However, these modeling frameworks often fail to fully capture several key properties of crash count data, especially negative serial correlation, and nonlinear dependence structures across temporal crash counts. To address these methodological gaps in existing safety literature, this study proposes to use a Gaussian Copula-based model for the long-term crash trend analysis. Specifically, this study proposes to use a Gaussian Copula-based Time Series Count Model with an Autoregressive Moving Average Process for the analysis of long-term trends in fatal crashes. The proposed approach can accommodate several data properties, which include (1) non-negative discrete property of count data, (2) positive and negative serial correlations among time series data, and (3) nonlinear dependence among time-series observations. The performance of the Gaussian Copula-based time series count model is compared with the generalized linear autoregressive and moving average model. The proposed modeling approaches are demonstrated by using yearly fatal crash count data for the years 1986 through 2022 from Queensland, Australia. The major safety interventions implemented in Queensland over those years are also highlighted to assess the possible and plausible impacts of these safety interventions in reducing fatal crash risks. Further, elasticity effects and overall percentage changes in fatal crashes across different time points are computed to demonstrate the implications of the proposed model. The policy analysis exercise shows that the implemented road safety interventions are likely to have diminishing marginal returns, underscoring the need for new and effective road safety policies to achieve the goal of zero fatalities within the set timeframe.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"211 ","pages":"Article 107795"},"PeriodicalIF":5.7,"publicationDate":"2024-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142871038","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Nudges may improve hazard perception in a contextual manner 轻推可以在情境中提高对危险的感知。

IF 5.7 1区工程技术 Q1 ERGONOMICS

Accident; analysis and prevention

Pub Date : 2024-12-19 DOI: 10.1016/j.aap.2024.107899

Shiran Zadka-Peer, Tova Rosenbloom

This research investigates the effectiveness of nudge presentation on Hazard Perception (HP) during a computerized Hazard Perception Test (HPT). Three types of nudges were examined: Reminder, Social Norm, and Negative Reinforcement. Their effects on drivers’ reaction times, hazard misidentifications (errors), and hazard recognition failures (misses) were analyzed. Additionally, the study explored how demographic and personality factors relate to individual differences in nudge responses. Results indicated that nudge presentation, regardless of type, improved reaction times and reduced errors. Reduction in errors was uniquely associated with personal characteristics, showing a positive correlation with age. Specifically, female participants and individuals low in conscientiousness exhibited fewer errors following the Social Norm nudge, while males and highly conscientious individuals showed reduced errors after the Reminder nudge. However, misses were unaffected by nudge presentation. All tested dependent variables were influenced by the order of hazard presentation, reflecting both contextual and nudge presentation effects. To further investigate the order’s impact, a follow-up study examined specific hazards sensitive to nudge presentation. Findings revealed that some hazards were more influenced by nudge/contextual factors, while others were unaffected, highlighting the need to consider complex contextual dynamics in HP research. Overall, the study supports the conclusion that nudge presentation can positively influence HP without distracting drivers, offering a promising strategy for improving road safety.

本研究探讨了在计算机化危险感知测试（HPT）中轻推呈现对危险感知（HP）的有效性。研究了三种类型的助推：提醒、社会规范和负强化。分析了它们对驾驶员反应时间、危险错误识别（错误）和危险识别失败（失误）的影响。此外，研究还探讨了人口统计学和人格因素与轻推反应的个体差异之间的关系。结果表明，轻推的呈现，无论类型，改善反应时间和减少错误。错误的减少与个人特征相关，与年龄呈正相关。具体而言，女性参与者和尽责性低的个体在社会规范推动后出现的错误较少，而男性和高度尽责的个体在提醒推动后出现的错误较少。然而，失误不受轻推的影响。所有被测试的因变量都受到危险呈现顺序的影响，反映了情境和助推呈现效应。为了进一步调查该命令的影响，一项后续研究检查了对轻推表示敏感的特定危害。研究结果显示，一些危害更受推动/环境因素的影响，而其他危害则不受影响，这突出了在HP研究中考虑复杂的环境动力学的必要性。总体而言，该研究支持这样的结论，即轻推可以在不分散驾驶员注意力的情况下对HP产生积极影响，为改善道路安全提供了一个有希望的策略。

{"title":"Nudges may improve hazard perception in a contextual manner","authors":"Shiran Zadka-Peer, Tova Rosenbloom","doi":"10.1016/j.aap.2024.107899","DOIUrl":"10.1016/j.aap.2024.107899","url":null,"abstract":"<div><div>This research investigates the effectiveness of nudge presentation on Hazard Perception (HP) during a computerized Hazard Perception Test (HPT). Three types of nudges were examined: Reminder, Social Norm, and Negative Reinforcement. Their effects on drivers’ reaction times, hazard misidentifications (errors), and hazard recognition failures (misses) were analyzed. Additionally, the study explored how demographic and personality factors relate to individual differences in nudge responses. Results indicated that nudge presentation, regardless of type, improved reaction times and reduced errors. Reduction in errors was uniquely associated with personal characteristics, showing a positive correlation with age. Specifically, female participants and individuals low in conscientiousness exhibited fewer errors following the Social Norm nudge, while males and highly conscientious individuals showed reduced errors after the Reminder nudge. However, misses were unaffected by nudge presentation. All tested dependent variables were influenced by the order of hazard presentation, reflecting both contextual and nudge presentation effects. To further investigate the order’s impact, a follow-up study examined specific hazards sensitive to nudge presentation. Findings revealed that some hazards were more influenced by nudge/contextual factors, while others were unaffected, highlighting the need to consider complex contextual dynamics in HP research. Overall, the study supports the conclusion that nudge presentation can positively influence HP without distracting drivers, offering a promising strategy for improving road safety.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"211 ","pages":"Article 107899"},"PeriodicalIF":5.7,"publicationDate":"2024-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142871039","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Conflict resolution behavior of autonomous vehicles at intersections under mixed traffic environment 混合交通环境下自动驾驶汽车交叉口冲突解决行为研究

IF 5.7 1区工程技术 Q1 ERGONOMICS

Accident; analysis and prevention

Pub Date : 2024-12-18 DOI: 10.1016/j.aap.2024.107897

Md Tanvir Ashraf, Kakan Dey

Navigating intersections is a major challenge for autonomous vehicles (AVs) because of the complex interactions between different roadway user types, conflicting movements, and diverse operational and geometric features. This study investigated intersection-related AV-involved traffic conflicts by analyzing the Arogoverse-2 motion forecasting dataset to understand the driving behavior of AVs at intersections. The conflict scenarios were categorized into AV-involved and no AV conflict scenarios. Depending on whether AVs passed the conflict region first or second in AV-involved scenarios, AV-involved scenarios were further classified into AV-first and AV-second scenarios. An agglomerative hierarchical clustering with t-SNE dimension reduction technique was applied to categorize the driving styles, and a three-layer Bayesian hierarchical model was applied to analyze the effect of driving volatility measures and traffic characteristics on relative crash risks. The clustering result showed that about 29% of the conflict events in the AV-first scenario (human-driven vehicle (HDV) was the following vehicle in passing the conflict region) exhibited high-risk of conflicts. In contrast, all conflicts events in the AV-second category were either low-risk or medium-risk conflicts. Parameter estimates showed that AVs had safer interactions with the other roadway users (i.e., HDVs, pedestrians/cyclists) while maintaining higher speeds and uniform driving profiles. AV’s interaction with vulnerable road users (i.e., pedestrians and cyclists) showed lower crash risk compared to HDVs, indicating AV’s safer driving behavior. AVs also demonstrated safer conflict resolution behavior in performing unprotected left turns compared to HDVs. This study discovered some unique insights into the challenges of introducing AVs in diverse intersection types (i.e., signalized, unsignalized, stop-controlled), which can be used to identify AV technology’s improvement need to better adapt to the mixed traffic driving environment.

由于不同道路用户类型之间的复杂交互、冲突运动以及不同的操作和几何特征，十字路口导航是自动驾驶汽车（AVs）面临的主要挑战。本研究通过分析Arogoverse-2运动预测数据集，研究了与交叉口相关的自动驾驶汽车交通冲突，以了解自动驾驶汽车在交叉口的驾驶行为。冲突场景分为自动驾驶汽车冲突场景和无自动驾驶汽车冲突场景。根据自动驾驶车辆在涉事场景中是先通过冲突区域还是后通过冲突区域，将涉事场景进一步分为自动驾驶车辆先行和自动驾驶车辆后通过冲突区域。采用t-SNE降维技术的聚类层次聚类方法对驾驶风格进行分类，并采用三层贝叶斯层次模型分析驾驶波动性测度和交通特征对相对碰撞风险的影响。聚类结果表明，在自动驾驶汽车优先场景下（人类驾驶汽车是经过冲突区域的尾随车辆），约有29%的冲突事件表现出冲突的高风险。相比之下，AV-second类别中的所有冲突事件要么是低风险冲突，要么是中风险冲突。参数估计表明，自动驾驶汽车与其他道路使用者（即hdv、行人/骑自行车的人）的互动更安全，同时保持更高的速度和统一的驾驶剖面。与hdv相比，自动驾驶汽车与弱势道路使用者（即行人和骑自行车的人）的互动碰撞风险更低，表明自动驾驶汽车的驾驶行为更安全。与hdv相比，自动驾驶汽车在进行无保护左转时也表现出更安全的冲突解决行为。本研究对引入自动驾驶汽车在不同类型交叉口（即有信号、无信号、停车控制）面临的挑战提出了一些独特的见解，可用于确定自动驾驶技术的改进需求，以更好地适应混合交通驾驶环境。

{"title":"Conflict resolution behavior of autonomous vehicles at intersections under mixed traffic environment","authors":"Md Tanvir Ashraf, Kakan Dey","doi":"10.1016/j.aap.2024.107897","DOIUrl":"10.1016/j.aap.2024.107897","url":null,"abstract":"<div><div>Navigating intersections is a major challenge for autonomous vehicles (AVs) because of the complex interactions between different roadway user types, conflicting movements, and diverse operational and geometric features. This study investigated intersection-related AV-involved traffic conflicts by analyzing the Arogoverse-2 motion forecasting dataset to understand the driving behavior of AVs at intersections. The conflict scenarios were categorized into AV-involved and no AV conflict scenarios. Depending on whether AVs passed the conflict region first or second in AV-involved scenarios, AV-involved scenarios were further classified into AV-first and AV-second scenarios. An agglomerative hierarchical clustering with t-SNE dimension reduction technique was applied to categorize the driving styles, and a three-layer Bayesian hierarchical model was applied to analyze the effect of driving volatility measures and traffic characteristics on relative crash risks. The clustering result showed that about 29% of the conflict events in the AV-first scenario (human-driven vehicle (HDV) was the following vehicle in passing the conflict region) exhibited <em>high-risk</em> of conflicts. In contrast, all conflicts events in the AV-second category were either <em>low-risk</em> or <em>medium-risk</em> conflicts. Parameter estimates showed that AVs had safer interactions with the other roadway users (i.e., HDVs, pedestrians/cyclists) while maintaining higher speeds and uniform driving profiles. AV’s interaction with vulnerable road users (i.e., pedestrians and cyclists) showed lower crash risk compared to HDVs, indicating AV’s safer driving behavior. AVs also demonstrated safer conflict resolution behavior in performing unprotected left turns compared to HDVs. This study discovered some unique insights into the challenges of introducing AVs in diverse intersection types (i.e., signalized, unsignalized, stop-controlled), which can be used to identify AV technology’s improvement need to better adapt to the mixed traffic driving environment.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"211 ","pages":"Article 107897"},"PeriodicalIF":5.7,"publicationDate":"2024-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142862749","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Investigating the contributing factors to autonomous Vehicle-Road user Conflicts: A Data-Driven approach 研究自动驾驶车辆与道路使用者冲突的影响因素：数据驱动的方法。

IF 5.7 1区工程技术 Q1 ERGONOMICS

Accident; analysis and prevention

Pub Date : 2024-12-18 DOI: 10.1016/j.aap.2024.107898

Mahdi Gabaire, Haniyeh Ghomi, Mohamed Hussein

With the imminent widespread integration of Autonomous Vehicles (AVs) into our traffic ecosystem, understanding the factors that impact their safety is a vital research area. To that end, this study assessed the impact of a wide range of factors on the frequency of AV-road user conflicts. The study utilized the Woven prediction and validation dataset, which contains over 1000 h of data collected from the onboard sensors of 20 AVs in California. Two Copula-based models were developed to investigate the contributing factors to total and severe AV conflicts in road segments (model M1) and intersections (model M2). For road segments, results indicated that road characteristics (direction, number of lanes, road length, speed limit, the presence of a dividing median) and road infrastructure (presence of bus stops, presence of cycle lanes, and presence of on-street parking) have a significant impact on the hourly conflict rates. Regarding the rate of severe conflicts, road user volume, road characteristics (direction, road type, access point density, the presence of a dividing median), and the presence of cycle lanes were identified as the most influential factors. For intersections, the road user volume and the presence of a physical median were found to be positively associated with the hourly conflict rates, while road user volume, intersection characteristics (posted speed limit, lack of traffic control signals, presence of pedestrian crossing, presence of cycle lane, presence of a dividing median, and truck percentage), and the dominant land use at the intersection area were the most impactful variables on the frequency of severe conflicts.

随着自动驾驶汽车（AVs）即将广泛融入我们的交通生态系统，了解影响其安全性的因素是一个至关重要的研究领域。为此，本研究评估了一系列因素对自动驾驶汽车与道路使用者冲突频率的影响。该研究利用了weave预测和验证数据集，该数据集包含了从加州20辆自动驾驶汽车的车载传感器收集的超过1000小时的数据。建立了两个基于copula的模型，研究了道路（M1模型）和交叉口（M2模型）中自动驾驶汽车总冲突和严重冲突的影响因素。对于路段，研究结果表明，道路特征（方向、车道数、道路长度、速度限制、分隔中位数的存在）和道路基础设施（公交车站、自行车道和路边停车场的存在）对小时冲突率有显著影响。在严重冲突率方面，道路使用者数量、道路特征（方向、道路类型、接入点密度、是否存在分隔中位数）和是否存在自行车道被认为是影响最大的因素。对于十字路口，道路使用者数量和物理中位数的存在被发现与小时冲突率呈正相关，而道路使用者数量、十字路口特征（张贴速度限制、缺乏交通控制信号、存在行人过街、存在自行车道、存在分隔中位数和卡车百分比）以及十字路口区域的主要土地使用是对严重冲突频率影响最大的变量。

{"title":"Investigating the contributing factors to autonomous Vehicle-Road user Conflicts: A Data-Driven approach","authors":"Mahdi Gabaire, Haniyeh Ghomi, Mohamed Hussein","doi":"10.1016/j.aap.2024.107898","DOIUrl":"10.1016/j.aap.2024.107898","url":null,"abstract":"<div><div>With the imminent widespread integration of Autonomous Vehicles (AVs) into our traffic ecosystem, understanding the factors that impact their safety is a vital research area. To that end, this study assessed the impact of a wide range of factors on the frequency of AV-road user conflicts. The study utilized the Woven prediction and validation dataset, which contains over 1000 h of data collected from the onboard sensors of 20 AVs in California. Two Copula-based models were developed to investigate the contributing factors to total and severe AV conflicts in road segments (model M1) and intersections (model M2). For road segments, results indicated that road characteristics (direction, number of lanes, road length, speed limit, the presence of a dividing median) and road infrastructure (presence of bus stops, presence of cycle lanes, and presence of on-street parking) have a significant impact on the hourly conflict rates. Regarding the rate of severe conflicts, road user volume, road characteristics (direction, road type, access point density, the presence of a dividing median), and the presence of cycle lanes were identified as the most influential factors. For intersections, the road user volume and the presence of a physical median were found to be positively associated with the hourly conflict rates, while road user volume, intersection characteristics (posted speed limit, lack of traffic control signals, presence of pedestrian crossing, presence of cycle lane, presence of a dividing median, and truck percentage), and the dominant land use at the intersection area were the most impactful variables on the frequency of severe conflicts.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"211 ","pages":"Article 107898"},"PeriodicalIF":5.7,"publicationDate":"2024-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142862776","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Analysis of factors affecting pedestrian safety for the elderly and identification of vulnerable areas in Seoul 首尔老年人步行安全影响因素分析及脆弱区域识别。

IF 5.7 1区工程技术 Q1 ERGONOMICS

Accident; analysis and prevention

Pub Date : 2024-12-14 DOI: 10.1016/j.aap.2024.107878

Soyoon Kim , Sangwon Choi , Brian H.S. Kim

Walking is the primary means of mobility and a daily activity for the elderly. Despite the need to ensure pedestrian safety given their physical limitations, elderly pedestrian traffic accidents in South Korea occur at a rate 7.7 times higher than in OECD member countries. In preparation for an aging society, there is a growing need to create a safe walking environment for the elderly. This study focuses on Seoul, analyzing the factors that compromise pedestrian safety for the elderly and identifying the characteristics of vulnerable areas. By using elderly pedestrian traffic accident data provided by the Road Traffic Authority and applying factors influencing accident occurrence to the MaxEnt model, the study identified priority elements for ensuring pedestrian safety. Additionally, the study predicted the regional vulnerability of elderly pedestrian accidents with the increasing elderly population in the future and reviewed possible measures to mitigate the risks. The study indicates that areas where elderly pedestrian safety is vulnerable tend to have lower budget allocations for road management, suggesting a need for future policy support. The prediction of elderly pedestrian accident occurrences through this study is expected to be useful in identifying areas with vulnerable pedestrian safety in Seoul, which can be utilized in prioritizing road improvement projects.

步行是老年人的主要行动方式和日常活动。尽管老年人的身体条件有限，需要确保步行安全，但韩国老年人步行交通事故的发生率比经合组织成员国高出 7.7 倍。为迎接老龄化社会的到来，为老年人创造一个安全的步行环境的需求与日俱增。本研究以首尔为重点，分析了影响老年人步行安全的因素，并确定了易受伤害地区的特征。通过使用道路交通管理局提供的老年人行人交通事故数据，并将影响事故发生的因素应用到 MaxEnt 模型中，本研究确定了确保行人安全的优先要素。此外，该研究还预测了随着未来老年人口的增加，老年人行人事故的区域脆弱性，并审查了降低风险的可行措施。研究表明，老年人行人安全易受影响的地区往往在道路管理方面的预算拨款较少，这表明未来需要政策支持。通过本研究对老年人行人事故发生率的预测，预计将有助于确定首尔行人安全易受影响的地区，从而确定道路改善项目的优先次序。

{"title":"Analysis of factors affecting pedestrian safety for the elderly and identification of vulnerable areas in Seoul","authors":"Soyoon Kim , Sangwon Choi , Brian H.S. Kim","doi":"10.1016/j.aap.2024.107878","DOIUrl":"10.1016/j.aap.2024.107878","url":null,"abstract":"<div><div>Walking is the primary means of mobility and a daily activity for the elderly. Despite the need to ensure pedestrian safety given their physical limitations, elderly pedestrian traffic accidents in South Korea occur at a rate 7.7 times higher than in OECD member countries. In preparation for an aging society, there is a growing need to create a safe walking environment for the elderly. This study focuses on Seoul, analyzing the factors that compromise pedestrian safety for the elderly and identifying the characteristics of vulnerable areas. By using elderly pedestrian traffic accident data provided by the Road Traffic Authority and applying factors influencing accident occurrence to the MaxEnt model, the study identified priority elements for ensuring pedestrian safety. Additionally, the study predicted the regional vulnerability of elderly pedestrian accidents with the increasing elderly population in the future and reviewed possible measures to mitigate the risks. The study indicates that areas where elderly pedestrian safety is vulnerable tend to have lower budget allocations for road management, suggesting a need for future policy support. The prediction of elderly pedestrian accident occurrences through this study is expected to be useful in identifying areas with vulnerable pedestrian safety in Seoul, which can be utilized in prioritizing road improvement projects.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"211 ","pages":"Article 107878"},"PeriodicalIF":5.7,"publicationDate":"2024-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142827153","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

How does distraction affect cyclists’ severe crashes? A hybrid CatBoost-SHAP and random parameters binary logit approach 注意力分散如何影响骑车人的严重撞车事故？CatBoost-SHAP 和随机参数二元 Logit 混合方法。

IF 5.7 1区工程技术 Q1 ERGONOMICS

Accident; analysis and prevention

Pub Date : 2024-12-13 DOI: 10.1016/j.aap.2024.107896

Ali Agheli, Kayvan Aghabayk

Cyclists are among the most vulnerable road users, increasingly subject to various sources of distraction, including the use of mobile phones and engagement in other tasks while navigating urban environments. Understanding and mitigating the impact of these distractions on cyclist safety is crucial. Despite the importance of this issue, the effect of distraction on injury severity in cycling crashes has not been extensively studied. This research analyzes four years of U.S. crash data (2019–2022) from the Crash Report Sampling System (CRSS) database, employing a hybrid framework that integrates CatBoost-based SHAP algorithm and the random parameters binary logit model with heterogeneity in means and variances (RPBL-HMV). The proposed approach confirms the significant role of cyclist distraction in crash injury severity. Subsequently, the analysis identifies several factors influencing the likelihood of severe injuries in distracted cyclist crashes. Crashes involving the front of motor vehicles, occurring in rural areas, on two-way roads, at higher speed limits, and during weekends were associated with a higher probability of severe injuries. Conversely, crashes at T-intersections, involving the side or rear of motor vehicles, where cyclists wore helmets, or during rush hour were linked to a reduced likelihood of severe injuries. Notably, interaction effects reveal nuanced patterns. For instance, while crossing roadway actions and rush hour periods individually decrease the likelihood of severe crashes, their combination increases the probability of such outcomes. The findings suggest targeted safety measures and policy interventions aimed at enhancing cyclist safety and promoting safer cycling environments by mitigating distraction-related risks.

骑自行车的人是最易受伤害的道路使用者之一，他们越来越多地受到各种因素的干扰，包括在城市环境中使用手机和从事其他工作。了解并减轻这些分心对骑车人安全的影响至关重要。尽管这一问题非常重要，但分心对骑车撞车事故中受伤严重程度的影响尚未得到广泛研究。本研究分析了来自碰撞报告采样系统（CRSS）数据库的四年（2019-2022 年）美国碰撞数据，采用了一个混合框架，该框架集成了基于 CatBoost 的 SHAP 算法和具有均值和方差异质性的随机参数二元 Logit 模型（RPBL-HMV）。所提出的方法证实了骑车人分心在碰撞伤害严重程度中的重要作用。随后，分析确定了影响分心骑车者撞车严重受伤可能性的几个因素。涉及机动车前部、发生在农村地区、双向道路上、限速较高以及周末的碰撞事故与较高的重伤概率相关。相反，在 T 型交叉路口发生的、涉及机动车侧面或后部的、骑车人戴头盔的或在上下班高峰期发生的撞车事故则与严重受伤的可能性降低有关。值得注意的是，交互效应揭示了细微的模式。例如，虽然横穿马路的行为和上下班高峰期会单独降低发生严重撞车事故的可能性，但两者结合则会增加发生此类事故的可能性。研究结果建议采取有针对性的安全措施和政策干预措施，通过降低与分心有关的风险来提高骑车人的安全，并促进更安全的骑车环境。

{"title":"How does distraction affect cyclists’ severe crashes? A hybrid CatBoost-SHAP and random parameters binary logit approach","authors":"Ali Agheli, Kayvan Aghabayk","doi":"10.1016/j.aap.2024.107896","DOIUrl":"10.1016/j.aap.2024.107896","url":null,"abstract":"<div><div>Cyclists are among the most vulnerable road users, increasingly subject to various sources of distraction, including the use of mobile phones and engagement in other tasks while navigating urban environments. Understanding and mitigating the impact of these distractions on cyclist safety is crucial. Despite the importance of this issue, the effect of distraction on injury severity in cycling crashes has not been extensively studied. This research analyzes four years of U.S. crash data (2019–2022) from the Crash Report Sampling System (CRSS) database, employing a hybrid framework that integrates CatBoost-based SHAP algorithm and the random parameters binary logit model with heterogeneity in means and variances (RPBL-HMV). The proposed approach confirms the significant role of cyclist distraction in crash injury severity. Subsequently, the analysis identifies several factors influencing the likelihood of severe injuries in distracted cyclist crashes. Crashes involving the front of motor vehicles, occurring in rural areas, on two-way roads, at higher speed limits, and during weekends were associated with a higher probability of severe injuries. Conversely, crashes at T-intersections, involving the side or rear of motor vehicles, where cyclists wore helmets, or during rush hour were linked to a reduced likelihood of severe injuries. Notably, interaction effects reveal nuanced patterns. For instance, while crossing roadway actions and rush hour periods individually decrease the likelihood of severe crashes, their combination increases the probability of such outcomes. The findings suggest targeted safety measures and policy interventions aimed at enhancing cyclist safety and promoting safer cycling environments by mitigating distraction-related risks.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"211 ","pages":"Article 107896"},"PeriodicalIF":5.7,"publicationDate":"2024-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142823940","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0