Pub Date : 2025-02-01DOI: 10.1016/j.jsr.2024.12.004
Carla Schiemer , Mark S. Horswill , Andrew Hill , Mathew J. Summers , Kayla B. Stefanidis
Introduction: As the medically prescribed use of cannabis flower continues to increase, there is a need to understand how vaporized cannabis can acutely affect driving-related skills and risk-taking behaviors in medicinal populations. Method: Given this, the present study examined the acute effects of vaporized cannabis flower on measures of hazard perception, driving-related risk-taking behaviors, and subjective perceptions of driving skills in a sample of adult medicinal cannabis patients. Participants (N = 38, M age = 43) attended both a baseline (no cannabis) and intervention appointment (with cannabis consumption), where they completed video-based tasks and self-report measures of driving ability. Results: After vaporizing one dose of their prescribed cannabis flower, participants exhibited no significant changes in performance on any of the video-based tasks (hazard perception skill, gap acceptance, following distance or speed) compared to baseline. However, cannabis consumption resulted in significant reductions in perceived hazard perception task performance and on-road traffic conflict prediction ability. Furthermore, there was a lack of association between objective and subjective hazard perception performance at both time points. Practical applications: These results suggest that while acute prescribed cannabis consumption may reduce appraisals of selected skills, overall hazard perception ability and driving-related risk-taking behavior may remain unchanged.
{"title":"The acute effects of vaporized cannabis on drivers’ hazard perception and risk-taking behaviors in medicinal patients: A within-subjects experiment","authors":"Carla Schiemer , Mark S. Horswill , Andrew Hill , Mathew J. Summers , Kayla B. Stefanidis","doi":"10.1016/j.jsr.2024.12.004","DOIUrl":"10.1016/j.jsr.2024.12.004","url":null,"abstract":"<div><div><em>Introduction</em>: As the medically prescribed use of cannabis flower continues to increase, there is a need to understand how vaporized cannabis can acutely affect driving-related skills and risk-taking behaviors in medicinal populations. <em>Method:</em> Given this, the present study examined the acute effects of vaporized cannabis flower on measures of hazard perception, driving-related risk-taking behaviors, and subjective perceptions of driving skills in a sample of adult medicinal cannabis patients. Participants (<em>N</em> = 38, <em>M</em> age = 43) attended both a baseline (no cannabis) and intervention appointment (with cannabis consumption), where they completed video-based tasks and self-report measures of driving ability. <em>Results:</em> After vaporizing one dose of their prescribed cannabis flower, participants exhibited no significant changes in performance on any of the video-based tasks (hazard perception skill, gap acceptance, following distance or speed) compared to baseline. However, cannabis consumption resulted in significant reductions in perceived hazard perception task performance and on-road traffic conflict prediction ability. Furthermore, there was a lack of association between objective and subjective hazard perception performance at both time points. <em>Practical applications:</em> These results suggest that while acute prescribed cannabis consumption may reduce appraisals of selected skills, overall hazard perception ability and driving-related risk-taking behavior may remain unchanged.</div></div>","PeriodicalId":48224,"journal":{"name":"Journal of Safety Research","volume":"92 ","pages":"Pages 385-392"},"PeriodicalIF":3.9,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143100948","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.jsr.2025.01.004
Noah J. Goodall
Introduction: This paper reviewed current driving automation (DA) and baseline human-driven crash databases and evaluated their comparability. Method: Five sources of DA crash data and three sources of human-driven crash data were reviewed for consistency of inclusion criteria, scope of coverage, and potential sources of bias. Alternative methods to determine vehicle automation capability using vehicle identification number (VIN) from state-maintained crash records were also explored. Conclusions: Evaluated data sets used incompatible or nonstandard minimum crash severity thresholds, complicating crash rate comparisons. The most widely-used standard was “police-reportable crash,” which itself has different reporting thresholds among jurisdictions. Although low- and no-damage crashes occur at greater frequencies and have more statistical power, they were not consistently reported for automated vehicles. Crash data collection can be improved through collection of driving automation exposure data, widespread collection of crash data form electronic data recorders, and standardization of crash definitions. Practical applications: Researchers and DA developers may use this analysis to conduct more thorough and accurate evaluations of driving automation crash rates. Lawmakers and regulators may use these findings as evidence to enhance data collection efforts, both internally and via new rules regarding electronic data recorders.
{"title":"Comparability of driving automation crash databases","authors":"Noah J. Goodall","doi":"10.1016/j.jsr.2025.01.004","DOIUrl":"10.1016/j.jsr.2025.01.004","url":null,"abstract":"<div><div><em>Introduction:</em> This paper reviewed current driving automation (DA) and baseline human-driven crash databases and evaluated their comparability. <em>Method</em>: Five sources of DA crash data and three sources of human-driven crash data were reviewed for consistency of inclusion criteria, scope of coverage, and potential sources of bias. Alternative methods to determine vehicle automation capability using vehicle identification number (VIN) from state-maintained crash records were also explored. <em>Conclusions</em>: Evaluated data sets used incompatible or nonstandard minimum crash severity thresholds, complicating crash rate comparisons. The most widely-used standard was “police-reportable crash,” which itself has different reporting thresholds among jurisdictions. Although low- and no-damage crashes occur at greater frequencies and have more statistical power, they were not consistently reported for automated vehicles. Crash data collection can be improved through collection of driving automation exposure data, widespread collection of crash data form electronic data recorders, and standardization of crash definitions. <em>Practical applications</em>: Researchers and DA developers may use this analysis to conduct more thorough and accurate evaluations of driving automation crash rates. Lawmakers and regulators may use these findings as evidence to enhance data collection efforts, both internally and via new rules regarding electronic data recorders.</div></div>","PeriodicalId":48224,"journal":{"name":"Journal of Safety Research","volume":"92 ","pages":"Pages 473-481"},"PeriodicalIF":3.9,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143239645","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.jsr.2024.11.028
Robyn Gerhard , Belinda J Gabbe , Peter Cameron , Stuart Newstead , Christopher N Morrison , Nyssa Clarke , Ben Beck
Introduction: Road traffic crashes globally cause 1.3 million deaths yearly and the rate of nonfatal crashes is increasing. Nonfatal injuries impact long-term quality of life, which is often overlooked in evaluations. The preferred method for using health-related quality of life and disability for evaluating road safety interventions have not been established. Method: A scoping review of peer-reviewed and grey literature was undertaken to understand health-related quality of life and disability measures currently used to evaluate road safety interventions. We included English language studies that used any health-related quality of life or disability measure to evaluate any real-world intervention aimed at reducing the number or severity of road traffic crashes. Results: Nine different health-related quality of life measures were used in the 18 included studies. The most commonly used measure was a quality-adjusted life year, which was used by seven studies, followed by the Glasgow Outcome Scale used by five studies. Two studies used two different health-related quality of life or disability measures. Five studies used primary data (collected directly for the purpose of the study) and 13 studies used existing data sources not explicitly collected for the reported evaluation. Of these 13 studies, 5 used an injury registry as the data source. Six different methods of deriving utility weights for calculating quality-adjusted life years were used. Conclusions: This review found that evaluations of road safety interventions using health-related quality of life or disability measures were rare. There was a lack of consistency in the measures used which prevented comparisons across evaluations. Further, inconsistent methods were used to derive utility weights for quality-adjusted life years. Practical Applications: Future evaluations of roads safety interventions need to consider longer-term outcomes. Consistent methods for measuring health-related quality of life and disability burden are needed, as are empirically derived utility weights for quality-adjusted life years.
{"title":"A scoping review on the methods used to assess health-related quality of life and disability burden in evaluations of road safety interventions","authors":"Robyn Gerhard , Belinda J Gabbe , Peter Cameron , Stuart Newstead , Christopher N Morrison , Nyssa Clarke , Ben Beck","doi":"10.1016/j.jsr.2024.11.028","DOIUrl":"10.1016/j.jsr.2024.11.028","url":null,"abstract":"<div><div><em>Introduction:</em> Road traffic crashes globally cause 1.3 million deaths yearly and the rate of nonfatal crashes is increasing. Nonfatal injuries impact long-term quality of life, which is often overlooked in evaluations. The preferred method for using health-related quality of life and disability for evaluating road safety interventions have not been established. <em>Method</em>: A scoping review of peer-reviewed and grey literature was undertaken to understand health-related quality of life and disability measures currently used to evaluate road safety interventions. We included English language studies that used any health-related quality of life or disability measure to evaluate any real-world intervention aimed at reducing the number or severity of road traffic crashes. <em>Results</em>: Nine different health-related quality of life measures were used in the 18 included studies. The most commonly used measure was a quality-adjusted life year, which was used by seven studies, followed by the Glasgow Outcome Scale used by five studies. Two studies used two different health-related quality of life or disability measures. Five studies used primary data (collected directly for the purpose of the study) and 13 studies used existing data sources not explicitly collected for the reported evaluation. Of these 13 studies, 5 used an injury registry as the data source. Six different methods of deriving utility weights for calculating quality-adjusted life years were used. <em>Conclusions</em>: This review found that evaluations of road safety interventions using health-related quality of life or disability measures were rare. There was a lack of consistency in the measures used which prevented comparisons across evaluations. Further, inconsistent methods were used to derive utility weights for quality-adjusted life years. <em>Practical Applications</em>: Future evaluations of roads safety interventions need to consider longer-term outcomes. Consistent methods for measuring health-related quality of life and disability burden are needed, as are empirically derived utility weights for quality-adjusted life years.</div></div>","PeriodicalId":48224,"journal":{"name":"Journal of Safety Research","volume":"92 ","pages":"Pages 459-472"},"PeriodicalIF":3.9,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143100590","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.jsr.2024.11.021
Margareth Gutiérrez , Raúl Ramos , Jose J. Soto , Felisa Córdova
Introduction: Traffic crashes remain a leading cause of fatalities worldwide, with higher fatality and injury rates in non-developed countries. Understanding the relationship among variables influencing traffic crashes and its outcome, measured as crash severity, is crucial for developing effective and targeted countermeasures to mitigate this problem. Method: In this study, we analyze traffic crashes involving pedestrians in Chile from 2022 to 2023. This allowed us to consider the entire country rather than a specific urban area, which is the first of its kind for a Latin American country. A Hierarchical Ordered Probit (HOPIT) model was estimated to model both risk propensity and severity of pedestrian and vehicle crashes while maintaining an ordered threshold structure. Findings reveal that pedestrian and driver characteristics significantly influence crash severity. Results: Male drivers have a higher probability of being involved in more severe crashes. Meanwhile, older pedestrians present a higher risk of severe and fatal injuries. Crash severity is significantly influenced by variables related to vehicle type and environmental factors. Pedestrians hit by heavy-duty vehicles have a 60% and 30% higher chance of suffering fatal or severe injuries, respectively. Highways exhibit a 421% higher chance of fatal injuries, followed by crashes at night and crashes in rural areas with 380% and 267%, respectively. Practical Applications: This research indicates the need for targeted safety measures addressing pedestrian and driver demographics and behavior, vehicle types, and environmental factors to effectively reduce pedestrian injury severity.
{"title":"Factors influencing pedestrian injury severity in Chile: A hierarchical probit ordered model approach","authors":"Margareth Gutiérrez , Raúl Ramos , Jose J. Soto , Felisa Córdova","doi":"10.1016/j.jsr.2024.11.021","DOIUrl":"10.1016/j.jsr.2024.11.021","url":null,"abstract":"<div><div>Introduction: Traffic crashes remain a leading cause of fatalities worldwide, with higher fatality and injury rates in non-developed countries. Understanding the relationship among variables influencing traffic crashes and its outcome, measured as crash severity, is crucial for developing effective and targeted countermeasures to mitigate this problem. Method: In this study, we analyze traffic crashes involving pedestrians in Chile from 2022 to 2023. This allowed us to consider the entire country rather than a specific urban area, which is the first of its kind for a Latin American country. A Hierarchical Ordered Probit (HOPIT) model was estimated to model both risk propensity and severity of pedestrian and vehicle crashes while maintaining an ordered threshold structure. Findings reveal that pedestrian and driver characteristics significantly influence crash severity. Results: Male drivers have a higher probability of being involved in more severe crashes. Meanwhile, older pedestrians present a higher risk of severe and fatal injuries. Crash severity is significantly influenced by variables related to vehicle type and environmental factors. Pedestrians hit by heavy-duty vehicles have a 60% and 30% higher chance of suffering fatal or severe injuries, respectively. Highways exhibit a 421% higher chance of fatal injuries, followed by crashes at night and crashes in rural areas with 380% and 267%, respectively. Practical Applications: This research indicates the need for targeted safety measures addressing pedestrian and driver demographics and behavior, vehicle types, and environmental factors to effectively reduce pedestrian injury severity.</div></div>","PeriodicalId":48224,"journal":{"name":"Journal of Safety Research","volume":"92 ","pages":"Pages 272-282"},"PeriodicalIF":3.9,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143100947","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.jsr.2024.12.008
Cara J. Hamann , Stephanie Jansson , Linder Wendt , Michelle Reyes , Jon Davis , Joseph E. Cavanaugh , Corinne Peek-Asa
Introduction
Alcohol impairment is a major contributor to road traffic crashes and has increased across the United States in recent years. In 2022, over 13,000 people were killed in drunk driving crashes. Enforcement of impaired driving laws is an essential strategy to reduce alcohol-impaired driving and subsequent crashes. However, little is known about conviction outcomes related to alcohol-involved crashes. The aim of this study is to examine the association between charge combinations and conviction rates among alcohol-influenced drivers involved in crashes.
Methods
Data for this study included 2016–2019 Iowa Department of Transportation crash data linked to charges and convictions from the Iowa Court Information System. The study sample included drivers with reported BAC ≥ 0.08 g/dl and/or driver condition reported as under influence of alcohol. Charges were divided into three categories: alcohol, moving, and administrative/miscellaneous. Two logistic regression models were built with any conviction and alcohol conviction as the outcomes. The main predictor was charge combination.
Results
The study sample included 8,238 alcohol-impaired drivers, of whom 6,846 (83.1%) were charged with any type of traffic offense and 6,253 (75.8%) were charged with alcohol-related traffic offenses. Among charged drivers, 96.2% were convicted on any traffic charge and 87.7% were convicted on an alcohol charge. Drivers with a combination of alcohol, administrative, and moving violation charges had higher odds of any conviction (OR = 2.6, 95% CI = 1.7–4.3) compared to drivers with only alcohol charges.
Conclusions
Charging impaired drivers with multiple types of charges was associated with increased odds of conviction on any charge but not on alcohol charges, which had high conviction rates overall.
Practical Applications
Results from this study can help guide law enforcement to ensure appropriate charges are made in all relevant categories and optimal combinations of charges are administered to impaired drivers to increase odds of conviction.
{"title":"Charge combinations and conviction rates among alcohol-influenced drivers involved in motor vehicle crashes in Iowa","authors":"Cara J. Hamann , Stephanie Jansson , Linder Wendt , Michelle Reyes , Jon Davis , Joseph E. Cavanaugh , Corinne Peek-Asa","doi":"10.1016/j.jsr.2024.12.008","DOIUrl":"10.1016/j.jsr.2024.12.008","url":null,"abstract":"<div><h3>Introduction</h3><div>Alcohol impairment is a major contributor to road traffic crashes and has increased across the United States in recent years. In 2022, over 13,000 people were killed in drunk driving crashes. Enforcement of impaired driving laws is an essential strategy to reduce alcohol-impaired driving and subsequent crashes. However, little is known about conviction outcomes related to alcohol-involved crashes. The aim of this study is to examine the association between charge combinations and conviction rates among alcohol-influenced drivers involved in crashes.</div></div><div><h3>Methods</h3><div>Data for this study included 2016–2019 Iowa Department of Transportation crash data linked to charges and convictions from the Iowa Court Information System. The study sample included drivers with reported BAC ≥ 0.08 g/dl and/or driver condition reported as under influence of alcohol. Charges were divided into three categories: alcohol, moving, and administrative/miscellaneous. Two logistic regression models were built with any conviction and alcohol conviction as the outcomes. The main predictor was charge combination.</div></div><div><h3>Results</h3><div>The study sample included 8,238 alcohol-impaired drivers, of whom 6,846 (83.1%) were charged with any type of traffic offense and 6,253 (75.8%) were charged with alcohol-related traffic offenses. Among charged drivers, 96.2% were convicted on any traffic charge and 87.7% were convicted on an alcohol charge. Drivers with a combination of alcohol, administrative, and moving violation charges had higher odds of any conviction (OR = 2.6, 95% CI = 1.7–4.3) compared to drivers with only alcohol charges.</div></div><div><h3>Conclusions</h3><div>Charging impaired drivers with multiple types of charges was associated with increased odds of conviction on any charge but not on alcohol charges, which had high conviction rates overall.</div></div><div><h3>Practical Applications</h3><div>Results from this study can help guide law enforcement to ensure appropriate charges are made in all relevant categories and optimal combinations of charges are administered to impaired drivers to increase odds of conviction.</div></div>","PeriodicalId":48224,"journal":{"name":"Journal of Safety Research","volume":"92 ","pages":"Pages 375-384"},"PeriodicalIF":3.9,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143100949","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.jsr.2024.12.002
Boshra Besharatian, Sattar Dorafshan
Introduction: Animal vehicle collisions (AVCs) are a global safety concern, requiring analysis and predictive models for understanding and mitigation. Police crash report data are one of the main sources of AVC data globally. However, they are prone to reporting policy change and other inconsistencies, particularly in rural areas, hindering the development of predictive models. Through development of a robust approach for data cleaning, quality control, feature selection, and contribution level identification, this study proposes a pipeline to address this shortcoming. Method: North Dakota crash data set is used as a case study due to high rates on AVC in this rural region and its diverse wildlife ecosystem. Theil’s U association index, and chi-square tests were implemented in the pipeline to evaluate the proposed pipeline effectiveness. The pipeline detects and removes skewed proportion samples, while addressing data collection inconsistency, low variance, and duplicated features. Results: Pipeline imposed 3.5% sample size and 88.9% feature size reduction on the original crash data over 20 years. Observation on the modified dataset revealed year, day, and driver features had the lowest while hour, county, and speed limit had the highest statistical contribution to the AVC. Light, hour, and month were lumped in daily solar cycle and represented as a single temporal feature that can be used effectively to develop predictive model. Finally, presented pipeline increased spatiotemporal integrity while reducing the runtime by 92.46% for the association analysis.
{"title":"A pipeline to enhance animal vehicle collision analysis in crash report dataset","authors":"Boshra Besharatian, Sattar Dorafshan","doi":"10.1016/j.jsr.2024.12.002","DOIUrl":"10.1016/j.jsr.2024.12.002","url":null,"abstract":"<div><div><em>Introduction</em>: Animal vehicle collisions (AVCs) are a global safety concern, requiring analysis and predictive models for understanding and mitigation. Police crash report data are one of the main sources of AVC data globally. However, they are prone to reporting policy change and other inconsistencies, particularly in rural areas, hindering the development of predictive models. Through development of a robust approach for data cleaning, quality control, feature selection, and contribution level identification, this study proposes a pipeline to address this shortcoming. <em>Method:</em> North Dakota crash data set is used as a case study due to high rates on AVC in this rural region and its diverse wildlife ecosystem. Theil’s U association index, and chi-square tests were implemented in the pipeline to evaluate the proposed pipeline effectiveness. The pipeline detects and removes skewed proportion samples, while addressing data collection inconsistency, low variance, and duplicated features. <em>Results:</em> Pipeline imposed 3.5% sample size and 88.9% feature size reduction on the original crash data over 20 years. Observation on the modified dataset revealed year, day, and driver features had the lowest while hour, county, and speed limit had the highest statistical contribution to the AVC. Light, hour, and month were lumped in daily solar cycle and represented as a single temporal feature that can be used effectively to develop predictive model. Finally, presented pipeline increased spatiotemporal integrity while reducing the runtime by 92.46% for the association analysis.</div></div>","PeriodicalId":48224,"journal":{"name":"Journal of Safety Research","volume":"92 ","pages":"Pages 245-261"},"PeriodicalIF":3.9,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143096643","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.jsr.2024.12.001
Nuri Park , Juneyoung Park , Chris Lee
Introduction: Roadway crash data are very rare and occur randomly, therefore there are several challenges to developing a crash prediction model for real-time traffic safety management. Recently, to resolve the problem of crash data sample size, researchers have conducted studies on crash data augmentation using machine learning techniques for developing safety evaluation models. However, it’s important to incorporate the specific characteristics of crash data into augmentation and crash risk assessment, as these characteristics vary depending on spatial and temporal conditions. Method: Therefore, this study developed a real-time crash risk model in three stages. First, crash data were clustered to define heterogeneous crash risk situations and then, key variables were derived by the ensemble and explainable artificial intelligence techniques, Boruta-SHAP. Second, augmentation of each clustered crash data was performed using oversampling techniques including Conditional Generative Adversarial Network (CGAN), which can consider each crash risk cluster’s characteristics. Finally, crash risk models were developed and compared with other crash risk models developed by using binary logistic regression model (BLM), Random Forest (RF), extreme gradient boosting (XGBoost), and Support Vector Machine (SVM). Results: The results showed that the CGAN-based XGBoost model has the best performance and the variable of the temporal speed difference at 10-minute intervals and the precipitation variable have a large impact on crash risk prediction. This paper emphasizes that crash risk characteristics must be distinguished in crash risk prediction and provides new insights into addressing the imbalance data issue within crash and non-crash datasets.
{"title":"Conditional Generative Adversarial Network-Based roadway crash risk prediction considering heterogeneity with dynamic data","authors":"Nuri Park , Juneyoung Park , Chris Lee","doi":"10.1016/j.jsr.2024.12.001","DOIUrl":"10.1016/j.jsr.2024.12.001","url":null,"abstract":"<div><div><em>Introduction</em>: Roadway crash data are very rare and occur randomly, therefore there are several challenges to developing a crash prediction model for real-time traffic safety management. Recently, to resolve the problem of crash data sample size, researchers have conducted studies on crash data augmentation using machine learning techniques for developing safety evaluation models. However, it’s important to incorporate the specific characteristics of crash data into augmentation and crash risk assessment, as these characteristics vary depending on spatial and temporal conditions. <em>Method:</em> Therefore, this study developed a real-time crash risk model in three stages. First, crash data were clustered to define heterogeneous crash risk situations and then, key variables were derived by the ensemble and explainable artificial intelligence techniques, Boruta-SHAP. Second, augmentation of each clustered crash data was performed using oversampling techniques including Conditional Generative Adversarial Network (CGAN), which can consider each crash risk cluster’s characteristics. Finally, crash risk models were developed and compared with other crash risk models developed by using binary logistic regression model (BLM), Random Forest (RF), extreme gradient boosting (XGBoost), and Support Vector Machine (SVM). <em>Results:</em> The results showed that the CGAN-based XGBoost model has the best performance and the variable of the temporal speed difference at 10-minute intervals and the precipitation variable have a large impact on crash risk prediction. This paper emphasizes that crash risk characteristics must be distinguished in crash risk prediction and provides new insights into addressing the imbalance data issue within crash and non-crash datasets.</div></div>","PeriodicalId":48224,"journal":{"name":"Journal of Safety Research","volume":"92 ","pages":"Pages 217-229"},"PeriodicalIF":3.9,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143100944","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.jsr.2024.11.022
Fletcher J. Howell, Azhaginiyal Arularasu, David B. Logan, Sjaan Koppel
Introduction:
Naturalistic driving studies (NDS) have great potential to characterize the road infrastructure factors influencing everyday driving. A systematic review was undertaken to evaluate the objectives, data processing, and analyses in best-practice applications of NDS data to road infrastructure. Method: Following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines, a systematic search of seven databases was conducted on 27 June 2023 (PROSPERO CRD42023434948). Fifty-three English-language, peer-reviewed studies were analyzed on the basis of the primary infrastructure category reflected in the research aims. Results: Studies described curves (14), turns at intersections (8), intersections (6), multi-modal treatments (6), ramps (4), work zones (4), charging (2), and other factors (9). Each study was assessed for the risk of methodological bias using amended National Heart, Lung, and Blood Institute templates for Quality Assurance. 74% of studies were assessed to be of ’Good’ quality, 13% of ‘Fair’ quality, and 13% of ‘Poor’ quality. Road infrastructure was characterized by external video (38%) complemented by non-NDS sources including satellite imagery (21%) and government data (19%). Data preparation was required in 91% of studies to extract meaningful variables (e.g. manual video coding) and/or link multiple datasets. Analysis predominantly determined correlations between aspects of driver behavior (speed, trajectory, etc.) and infrastructure factors (geometry, lane configuration, etc.). Conclusions: The methods employed were broadly applicable, but required considerable subject-specific adaptation for non-NDS datasets and/or time-consuming video coding. The incorporation of road infrastructure factors in NDS research can continue to be improved by reducing the computational cost of sample processing.Practical Applications: Encouraged by the adaptability of the identified methods, NDS research has the potential to benefit from the consideration of road infrastructure factors in a Safe System context. The analytical requirements for all components of the Safe System should be considered when planning future NDS data collections and/or analysis.
{"title":"Naturalistic driving study data applied to road infrastructure: A systematic review","authors":"Fletcher J. Howell, Azhaginiyal Arularasu, David B. Logan, Sjaan Koppel","doi":"10.1016/j.jsr.2024.11.022","DOIUrl":"10.1016/j.jsr.2024.11.022","url":null,"abstract":"<div><h3>Introduction:</h3><div>Naturalistic driving studies (NDS) have great potential to characterize the road infrastructure factors influencing everyday driving. A systematic review was undertaken to evaluate the objectives, data processing, and analyses in best-practice applications of NDS data to road infrastructure. <em>Method:</em> Following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines, a systematic search of seven databases was conducted on 27 June 2023 (PROSPERO CRD42023434948). Fifty-three English-language, peer-reviewed studies were analyzed on the basis of the primary infrastructure category reflected in the research aims. <em>Results:</em> Studies described curves (14), turns at intersections (8), intersections (6), multi-modal treatments (6), ramps (4), work zones (4), charging (2), and other factors (9). Each study was assessed for the risk of methodological bias using amended National Heart, Lung, and Blood Institute templates for Quality Assurance. 74% of studies were assessed to be of ’Good’ quality, 13% of ‘Fair’ quality, and 13% of ‘Poor’ quality. Road infrastructure was characterized by external video (38%) complemented by non-NDS sources including satellite imagery (21%) and government data (19%). Data preparation was required in 91% of studies to extract meaningful variables (e.g. manual video coding) and/or link multiple datasets. Analysis predominantly determined correlations between aspects of driver behavior (speed, trajectory, etc.) and infrastructure factors (geometry, lane configuration, etc.). Conclusions: The methods employed were broadly applicable, but required considerable subject-specific adaptation for non-NDS datasets and/or time-consuming video coding. The incorporation of road infrastructure factors in NDS research can continue to be improved by reducing the computational cost of sample processing.Practical Applications: Encouraged by the adaptability of the identified methods, NDS research has the potential to benefit from the consideration of road infrastructure factors in a Safe System context. The analytical requirements for all components of the Safe System should be considered when planning future NDS data collections and/or analysis.</div></div>","PeriodicalId":48224,"journal":{"name":"Journal of Safety Research","volume":"92 ","pages":"Pages 346-374"},"PeriodicalIF":3.9,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143100951","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.jsr.2024.12.007
Alysha R. Meyers , Tara N. Schrader , Edward Krieg , Steven J. Naber , Chih-Yu Tseng , Michael P. Lampl , Brian Chin , Steven J. Wurzelbacher
Introduction: This study aimed to develop a set of broad clinical diagnosis (ClinDx) groups relevant to occupational safety and health. The ClinDx groups are necessary for analysis and interpretation of longitudinal health data that include injury and disease codes from the Ninth and Tenth Revision of the International Classification of Disease, Clinical Modification (ICD-9-CM, ICD-10-CM). Methods: Claims data were analyzed for Ohio Bureau of Workers’ Compensation insured employers from 2011 to 2018. We used interrupted time series regression models to estimate level (frequency) and slope (trend) changes to the percentage of each ClinDx group in October 2015. We created ClinDx groups aligned with ICD-10-CM structure and coding principles. Each ClinDx group was counted once per claim (distinct groups). Monthly percentages were calculated based on the injury date. When present, seasonality was assessed separately for each outcome using an autoregressive-moving average model. Results: The final set of ClinDx groups included 57 mutually exclusive and exhaustive groups. The study population included 661,684 claims, with 959,322 distinct ClinDx groups. Among all claims, 96.27% included injury code(s) and 11.77% included disease(s) codes. At the transition to ICD-10-CM, 33 ClinDx groups lacked any statistically significant (P < 0.05) changes between periods. We observed level changes for 17 ClinDx groups and slope changes for nine groups. Eight ClinDx groups had ≥ 20% (+/-) level changes. Conclusion: While the transition to ICD-10-CM is a break in series, about two-thirds of disease groups and half of injury groups were relatively stable across the transition. These findings also underscore the need for characterizing both injury and disease outcomes when analyzing workers’ compensation data. Practical Applications: The 57 ClinDx groups created in this study may be a practical starting point for other occupational epidemiologic analyses that include a mixture of ICD-9-CM and ICD-10-CM data.
{"title":"Clinical diagnosis groups developed to bridge the ICD-9-CM to ICD-10-CM coding transition and monitor trends in workers’ compensation claims — Ohio, 2011–2018","authors":"Alysha R. Meyers , Tara N. Schrader , Edward Krieg , Steven J. Naber , Chih-Yu Tseng , Michael P. Lampl , Brian Chin , Steven J. Wurzelbacher","doi":"10.1016/j.jsr.2024.12.007","DOIUrl":"10.1016/j.jsr.2024.12.007","url":null,"abstract":"<div><div><em>Introduction:</em> This study aimed to develop a set of broad clinical diagnosis (ClinDx) groups relevant to occupational safety and health. The ClinDx groups are necessary for analysis and interpretation of longitudinal health data that include injury and disease codes from the Ninth and Tenth Revision of the International Classification of Disease, Clinical Modification (ICD-9-CM, ICD-10-CM). <em>Methods:</em> Claims data were analyzed for Ohio Bureau of Workers’ Compensation insured employers from 2011 to 2018. We used interrupted time series regression models to estimate level (frequency) and slope (trend) changes to the percentage of each ClinDx group in October 2015. We created ClinDx groups aligned with ICD-10-CM structure and coding principles. Each ClinDx group was counted once per claim (distinct groups). Monthly percentages were calculated based on the injury date. When present, seasonality was assessed separately for each outcome using an autoregressive-moving average model. <em>Results:</em> The final set of ClinDx groups included 57 mutually exclusive and exhaustive groups. The study population included 661,684 claims, with 959,322 distinct ClinDx groups. Among all claims, 96.27% included injury code(s) and 11.77% included disease(s) codes. At the transition to ICD-10-CM, 33 ClinDx groups lacked any statistically significant (P < 0.05) changes between periods. We observed level changes for 17 ClinDx groups and slope changes for nine groups. Eight ClinDx groups had ≥ 20% (+/-) level changes. <em>Conclusion:</em> While the transition to ICD-10-CM is a break in series, about two-thirds of disease groups and half of injury groups were relatively stable across the transition. These findings also underscore the need for characterizing both injury and disease outcomes when analyzing workers’ compensation data. <em>Practical Applications:</em> The 57 ClinDx groups created in this study may be a practical starting point for other occupational epidemiologic analyses that include a mixture of ICD-9-CM and ICD-10-CM data.</div></div>","PeriodicalId":48224,"journal":{"name":"Journal of Safety Research","volume":"92 ","pages":"Pages 408-419"},"PeriodicalIF":3.9,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143100885","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.jsr.2024.12.006
Tianyi Li , Joshua Klavins , Te Xu , Niaz Mahmud Zafri , Raphael Stern
Introduction: Many factors influence the yielding result of driver–pedestrian interactions, including traffic, vehicle, roadway, pedestrian attributes, and more. While researchers have examined the individual influence of these factors on interaction outcomes, there is a noticeable absence of comprehensive, naturalistic studies in current literature, particularly those investigating the impact of the built environment on driver-yielding behavior. Method: To address this gap, our study introduces an extensive open-source dataset, compiled from video data at 18 unsignalized intersections across Minnesota. Documenting more than 3000 interactions, this dataset provides a detailed view of driver–pedestrian interactions and over 50 distinct contextual variables. The data, which covers individual driver–pedestrian interactions and contextual factors, is made publicly available at https://hdl.handle.net/11299/254556. Using logistic regression, we developed a classification model that predicts driver yielding based on the identified variables. Results: Our analysis indicates that vehicle speed, the presence of parking lots, proximity to parks or schools, and the width of major road crossings significantly influence driver yielding at unsignalized intersections. Conclusions: Through our findings and by publishing one of the most comprehensive driver–pedestrian datasets in the United States, our study will support communities across Minnesota and the United States in their ongoing efforts to improve road safety for pedestrians and be helpful for automated vehicle design. Practical Applications: We have compiled a dataset on driver–pedestrian interactions at 18 unsignalized intersections in Minnesota, making it one of the most extensive datasets available in the United States. This dataset can be utilized by researchers and local agencies to enhance intersection safety and walkability. Furthermore, our study proposes recommendations for increasing pedestrian safety at intersections, providing valuable insights that local governments can use as guidance for designing future intersections.
{"title":"Identifying built environment factors influencing driver yielding behavior at unsignalized intersections: A naturalistic open-source dataset collected in Minnesota","authors":"Tianyi Li , Joshua Klavins , Te Xu , Niaz Mahmud Zafri , Raphael Stern","doi":"10.1016/j.jsr.2024.12.006","DOIUrl":"10.1016/j.jsr.2024.12.006","url":null,"abstract":"<div><div><em>Introduction</em>: Many factors influence the yielding result of driver–pedestrian interactions, including traffic, vehicle, roadway, pedestrian attributes, and more. While researchers have examined the individual influence of these factors on interaction outcomes, there is a noticeable absence of comprehensive, naturalistic studies in current literature, particularly those investigating the impact of the built environment on driver-yielding behavior. <em>Method</em>: To address this gap, our study introduces an extensive open-source dataset, compiled from video data at 18 unsignalized intersections across Minnesota. Documenting more than 3000 interactions, this dataset provides a detailed view of driver–pedestrian interactions and over 50 distinct contextual variables. The data, which covers individual driver–pedestrian interactions and contextual factors, is made publicly available at <span><span>https://hdl.handle.net/11299/254556</span><svg><path></path></svg></span>. Using logistic regression, we developed a classification model that predicts driver yielding based on the identified variables. <em>Results</em>: Our analysis indicates that vehicle speed, the presence of parking lots, proximity to parks or schools, and the width of major road crossings significantly influence driver yielding at unsignalized intersections. <em>Conclusions</em>: Through our findings and by publishing one of the most comprehensive driver–pedestrian datasets in the United States, our study will support communities across Minnesota and the United States in their ongoing efforts to improve road safety for pedestrians and be helpful for automated vehicle design. <em>Practical Applications</em>: We have compiled a dataset on driver–pedestrian interactions at 18 unsignalized intersections in Minnesota, making it one of the most extensive datasets available in the United States. This dataset can be utilized by researchers and local agencies to enhance intersection safety and walkability. Furthermore, our study proposes recommendations for increasing pedestrian safety at intersections, providing valuable insights that local governments can use as guidance for designing future intersections.</div></div>","PeriodicalId":48224,"journal":{"name":"Journal of Safety Research","volume":"92 ","pages":"Pages 331-345"},"PeriodicalIF":3.9,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143100950","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号