使用基于冲突的实时极值安全模型动态识别短期和长期危险地点

IF 12.5 1区 工程技术 Q1 PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH
Tarek Ghoul , Tarek Sayed , Chuanyun Fu
{"title":"使用基于冲突的实时极值安全模型动态识别短期和长期危险地点","authors":"Tarek Ghoul ,&nbsp;Tarek Sayed ,&nbsp;Chuanyun Fu","doi":"10.1016/j.amar.2022.100262","DOIUrl":null,"url":null,"abstract":"<div><p>A novel and effective approach to safety management requires evaluating the safety of locations over short time periods (e.g. minutes). Unlike traditional methods that are based on aggregate crash records over a few years, crash proneness in this approach reflects short-time durations and is related to dynamic traffic changes and dangerous driving events. This paper proposes a new approach to dynamically assess the crash proneness of traffic conditions within a very short time (e.g., signal cycle length) and to dynamically identify high-risk locations. Using a Bayesian hierarchal Extreme Value Theory (EVT) model, the short-term crash risk metrics, risk of crash (ROC), and return level (RL), are calculated using traffic conflict data. A short-term hazardous location identification and ranking framework is developed based on crash-risk threshold exceedances for every short-term analysis period. By further investigating the variation in short-term crash risk, longer-term hazardous location identification and ranking metrics such as the longer-term crash risk index (LTCRI) and the percent of time exceeding (PTE) were developed. Using these metrics, a framework is proposed by which hazardous intersections can be dynamically classified and ranked in both the short-term and the longer-term. This ranking may be dynamically updated as more data becomes available. The proposed framework was applied to a trajectory dataset consisting of 47 signalized intersections obtained from a UAV-based dataset. Conflicts were identified from vehicle trajectories and were used to compute the proposed short-term and longer-term metrics. The intersections within the network were then ranked based on the proposed framework. This study demonstrates the importance of investigating short-term fluctuations in crash risk that may otherwise be lost to averaging in longer-term analysis and proposes a simple and practical solution.</p></div>","PeriodicalId":47520,"journal":{"name":"Analytic Methods in Accident Research","volume":"37 ","pages":"Article 100262"},"PeriodicalIF":12.5000,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":"{\"title\":\"Dynamic identification of short-term and longer-term hazardous locations using a conflict-based real-time extreme value safety model\",\"authors\":\"Tarek Ghoul ,&nbsp;Tarek Sayed ,&nbsp;Chuanyun Fu\",\"doi\":\"10.1016/j.amar.2022.100262\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>A novel and effective approach to safety management requires evaluating the safety of locations over short time periods (e.g. minutes). Unlike traditional methods that are based on aggregate crash records over a few years, crash proneness in this approach reflects short-time durations and is related to dynamic traffic changes and dangerous driving events. This paper proposes a new approach to dynamically assess the crash proneness of traffic conditions within a very short time (e.g., signal cycle length) and to dynamically identify high-risk locations. Using a Bayesian hierarchal Extreme Value Theory (EVT) model, the short-term crash risk metrics, risk of crash (ROC), and return level (RL), are calculated using traffic conflict data. A short-term hazardous location identification and ranking framework is developed based on crash-risk threshold exceedances for every short-term analysis period. By further investigating the variation in short-term crash risk, longer-term hazardous location identification and ranking metrics such as the longer-term crash risk index (LTCRI) and the percent of time exceeding (PTE) were developed. Using these metrics, a framework is proposed by which hazardous intersections can be dynamically classified and ranked in both the short-term and the longer-term. This ranking may be dynamically updated as more data becomes available. The proposed framework was applied to a trajectory dataset consisting of 47 signalized intersections obtained from a UAV-based dataset. Conflicts were identified from vehicle trajectories and were used to compute the proposed short-term and longer-term metrics. The intersections within the network were then ranked based on the proposed framework. This study demonstrates the importance of investigating short-term fluctuations in crash risk that may otherwise be lost to averaging in longer-term analysis and proposes a simple and practical solution.</p></div>\",\"PeriodicalId\":47520,\"journal\":{\"name\":\"Analytic Methods in Accident Research\",\"volume\":\"37 \",\"pages\":\"Article 100262\"},\"PeriodicalIF\":12.5000,\"publicationDate\":\"2023-03-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"5\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Analytic Methods in Accident Research\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S2213665722000513\",\"RegionNum\":1,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Analytic Methods in Accident Research","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2213665722000513","RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH","Score":null,"Total":0}
引用次数: 5

摘要

一种新颖而有效的安全管理方法需要在短时间内(例如几分钟)评估地点的安全性。与基于几年累积碰撞记录的传统方法不同,该方法中的碰撞倾向反映了短时间持续时间,并且与动态交通变化和危险驾驶事件有关。本文提出了一种在极短时间内(如信号周期长度)动态评估交通状况的碰撞倾向性和动态识别高风险位置的新方法。利用贝叶斯层次极值理论(EVT)模型,利用交通冲突数据计算了短期碰撞风险指标,即碰撞风险(ROC)和回报水平(RL)。根据每个短期分析期的碰撞风险阈值超出情况,制定了短期危险位置识别和排序框架。通过进一步研究短期碰撞风险的变化,开发了长期危险位置识别和排名指标,如长期碰撞风险指数(LTCRI)和超过时间百分比(PTE)。利用这些指标,提出了一个框架,通过该框架可以对危险交叉口进行短期和长期的动态分类和排名。这个排名可能会随着可用数据的增加而动态更新。将该框架应用于从无人机数据集获得的由47个信号交叉口组成的轨迹数据集。从车辆轨迹中识别冲突,并用于计算建议的短期和长期指标。然后根据所提出的框架对网络内的交叉点进行排序。本研究证明了调查短期崩溃风险波动的重要性,否则在长期分析中可能会失去平均,并提出了一个简单而实用的解决方案。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Dynamic identification of short-term and longer-term hazardous locations using a conflict-based real-time extreme value safety model

A novel and effective approach to safety management requires evaluating the safety of locations over short time periods (e.g. minutes). Unlike traditional methods that are based on aggregate crash records over a few years, crash proneness in this approach reflects short-time durations and is related to dynamic traffic changes and dangerous driving events. This paper proposes a new approach to dynamically assess the crash proneness of traffic conditions within a very short time (e.g., signal cycle length) and to dynamically identify high-risk locations. Using a Bayesian hierarchal Extreme Value Theory (EVT) model, the short-term crash risk metrics, risk of crash (ROC), and return level (RL), are calculated using traffic conflict data. A short-term hazardous location identification and ranking framework is developed based on crash-risk threshold exceedances for every short-term analysis period. By further investigating the variation in short-term crash risk, longer-term hazardous location identification and ranking metrics such as the longer-term crash risk index (LTCRI) and the percent of time exceeding (PTE) were developed. Using these metrics, a framework is proposed by which hazardous intersections can be dynamically classified and ranked in both the short-term and the longer-term. This ranking may be dynamically updated as more data becomes available. The proposed framework was applied to a trajectory dataset consisting of 47 signalized intersections obtained from a UAV-based dataset. Conflicts were identified from vehicle trajectories and were used to compute the proposed short-term and longer-term metrics. The intersections within the network were then ranked based on the proposed framework. This study demonstrates the importance of investigating short-term fluctuations in crash risk that may otherwise be lost to averaging in longer-term analysis and proposes a simple and practical solution.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
CiteScore
22.10
自引率
34.10%
发文量
35
审稿时长
24 days
期刊介绍: Analytic Methods in Accident Research is a journal that publishes articles related to the development and application of advanced statistical and econometric methods in studying vehicle crashes and other accidents. The journal aims to demonstrate how these innovative approaches can provide new insights into the factors influencing the occurrence and severity of accidents, thereby offering guidance for implementing appropriate preventive measures. While the journal primarily focuses on the analytic approach, it also accepts articles covering various aspects of transportation safety (such as road, pedestrian, air, rail, and water safety), construction safety, and other areas where human behavior, machine failures, or system failures lead to property damage or bodily harm.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信