{"title":"大流行病中的交通行为和政府干预:针对道路安全的混合可解释机器学习","authors":"","doi":"10.1016/j.tre.2024.103841","DOIUrl":null,"url":null,"abstract":"<div><div>During a pandemic, transportation authorities and policymakers face significant challenges in identifying and validating new travel behavior and how it affects traffic crash patterns to develop effective safety strategies. A timely assessment of an emergency incident’s long-term impact and the development of appropriate response strategies are critical for managing future occurrences. This study investigates to answer these research questions (RQs):</div><div>RQ1: How do various spatio-temporal risk factors influence traffic crash injury severity during the different phases of the COVID-19 pandemic?</div><div>RQ2: What are the key risk factors influencing injury severity in automobile crashes during the pre-pandemic, early pandemic, between the first and second waves of the pandemic, and the post-pandemic era?</div><div>RQ3: How do the implemented government policies and interventions during the pandemic affect transport behavior and road safety?</div><div>This study presents a hybrid explainable machine learning approach based on eXtreme Gradient Boosting (XGBoost) and SHapley Additive exPlanation (SHAP) to identify influential traffic crash-related risk factors for injury severity. Additionally, we propose a statistical learning approach using a nonlinear multinomial logit model to jointly analyze the count of automobile traffic crashes by injury severity and assess the impact of the COVID-19 pandemic across different phases. Our findings include a detailed analysis of system-level taxonomies across feature components, as well as the use of aggregate SHAP scores to classify crash data into high-level contributing variables during the pre-pandemic, intra-pandemic, and post-pandemic phases. The expected outcomes include insights such as identifying the best times to implement travel restrictions to reduce traffic accidents, understanding shifts in traffic flow patterns across pandemic phases, and determining effective public health interventions that can reduce both traffic accidents and congestion. Furthermore, the study reveals that the initial pandemic phase saw a significant decrease in traffic volume and accident rates. In contrast, the subsequent pandemic and post-pandemic phases saw an increase in severe accidents due to risky driving behaviors, emphasizing the importance of adaptive safety measures.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":null,"pages":null},"PeriodicalIF":8.3000,"publicationDate":"2024-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Transport behavior and government interventions in pandemics: A hybrid explainable machine learning for road safety\",\"authors\":\"\",\"doi\":\"10.1016/j.tre.2024.103841\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>During a pandemic, transportation authorities and policymakers face significant challenges in identifying and validating new travel behavior and how it affects traffic crash patterns to develop effective safety strategies. A timely assessment of an emergency incident’s long-term impact and the development of appropriate response strategies are critical for managing future occurrences. This study investigates to answer these research questions (RQs):</div><div>RQ1: How do various spatio-temporal risk factors influence traffic crash injury severity during the different phases of the COVID-19 pandemic?</div><div>RQ2: What are the key risk factors influencing injury severity in automobile crashes during the pre-pandemic, early pandemic, between the first and second waves of the pandemic, and the post-pandemic era?</div><div>RQ3: How do the implemented government policies and interventions during the pandemic affect transport behavior and road safety?</div><div>This study presents a hybrid explainable machine learning approach based on eXtreme Gradient Boosting (XGBoost) and SHapley Additive exPlanation (SHAP) to identify influential traffic crash-related risk factors for injury severity. Additionally, we propose a statistical learning approach using a nonlinear multinomial logit model to jointly analyze the count of automobile traffic crashes by injury severity and assess the impact of the COVID-19 pandemic across different phases. Our findings include a detailed analysis of system-level taxonomies across feature components, as well as the use of aggregate SHAP scores to classify crash data into high-level contributing variables during the pre-pandemic, intra-pandemic, and post-pandemic phases. The expected outcomes include insights such as identifying the best times to implement travel restrictions to reduce traffic accidents, understanding shifts in traffic flow patterns across pandemic phases, and determining effective public health interventions that can reduce both traffic accidents and congestion. Furthermore, the study reveals that the initial pandemic phase saw a significant decrease in traffic volume and accident rates. In contrast, the subsequent pandemic and post-pandemic phases saw an increase in severe accidents due to risky driving behaviors, emphasizing the importance of adaptive safety measures.</div></div>\",\"PeriodicalId\":49418,\"journal\":{\"name\":\"Transportation Research Part E-Logistics and Transportation Review\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":8.3000,\"publicationDate\":\"2024-10-26\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Transportation Research Part E-Logistics and Transportation Review\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S1366554524004320\",\"RegionNum\":1,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"ECONOMICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Transportation Research Part E-Logistics and Transportation Review","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1366554524004320","RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ECONOMICS","Score":null,"Total":0}
Transport behavior and government interventions in pandemics: A hybrid explainable machine learning for road safety
During a pandemic, transportation authorities and policymakers face significant challenges in identifying and validating new travel behavior and how it affects traffic crash patterns to develop effective safety strategies. A timely assessment of an emergency incident’s long-term impact and the development of appropriate response strategies are critical for managing future occurrences. This study investigates to answer these research questions (RQs):
RQ1: How do various spatio-temporal risk factors influence traffic crash injury severity during the different phases of the COVID-19 pandemic?
RQ2: What are the key risk factors influencing injury severity in automobile crashes during the pre-pandemic, early pandemic, between the first and second waves of the pandemic, and the post-pandemic era?
RQ3: How do the implemented government policies and interventions during the pandemic affect transport behavior and road safety?
This study presents a hybrid explainable machine learning approach based on eXtreme Gradient Boosting (XGBoost) and SHapley Additive exPlanation (SHAP) to identify influential traffic crash-related risk factors for injury severity. Additionally, we propose a statistical learning approach using a nonlinear multinomial logit model to jointly analyze the count of automobile traffic crashes by injury severity and assess the impact of the COVID-19 pandemic across different phases. Our findings include a detailed analysis of system-level taxonomies across feature components, as well as the use of aggregate SHAP scores to classify crash data into high-level contributing variables during the pre-pandemic, intra-pandemic, and post-pandemic phases. The expected outcomes include insights such as identifying the best times to implement travel restrictions to reduce traffic accidents, understanding shifts in traffic flow patterns across pandemic phases, and determining effective public health interventions that can reduce both traffic accidents and congestion. Furthermore, the study reveals that the initial pandemic phase saw a significant decrease in traffic volume and accident rates. In contrast, the subsequent pandemic and post-pandemic phases saw an increase in severe accidents due to risky driving behaviors, emphasizing the importance of adaptive safety measures.
期刊介绍:
Transportation Research Part E: Logistics and Transportation Review is a reputable journal that publishes high-quality articles covering a wide range of topics in the field of logistics and transportation research. The journal welcomes submissions on various subjects, including transport economics, transport infrastructure and investment appraisal, evaluation of public policies related to transportation, empirical and analytical studies of logistics management practices and performance, logistics and operations models, and logistics and supply chain management.
Part E aims to provide informative and well-researched articles that contribute to the understanding and advancement of the field. The content of the journal is complementary to other prestigious journals in transportation research, such as Transportation Research Part A: Policy and Practice, Part B: Methodological, Part C: Emerging Technologies, Part D: Transport and Environment, and Part F: Traffic Psychology and Behaviour. Together, these journals form a comprehensive and cohesive reference for current research in transportation science.