Accident Severity Research Articles

Exploring factors influencing pedestrian accident severity: a multi-source study in the city of Berlin

Abstract Background Road safety remains a critical concern in Europe’s transport sector. In urban areas, where 40% of total road fatalities occur, pedestrians are particularly vulnerable, since according to road safety data accidents involving a pedestrian are 2.8 and 15 times more probable to be fatal than in rural areas and motorways respectively. In line with the Vision Zero concept and the European Commission’s directives, many EU members are taking steps to improve road safety. Accordingly, the identification of the key factors behind pedestrian accident occurrence and pedestrian accident severity in urban areas is ever more relevant. Methodology The proposed methodology employs traditional logistic regression models and artificial neural networks, using accident data from Berlin, Germany, sourced from the Berlin Open Data portal. The dataset comprises information on 3,257 accidents involving pedestrians in 2018 or 2019, including details about involved vehicles, accident details, and injury severity information. Additionally, the dataset was augmented with average network speed data from Uber Movement and road network information from Geographic Information System (GIS) applications. Results Formal analysis results indicated several factors as significant to accident severity, such as involvement of bicycle or heavy vehicles, lighting conditions, speed limit and accident type. Additionally, a comparison between modelling approaches shows a clear performance advantage of ANNs over statistical models. Research findings provide insight for various stakeholders working to enhance pedestrian safety in urban areas.

A spatial autocorrelation analysis of road traffic accidents by severity using Moran’s I spatial statistics: a study from Nepal 2019–2022

BackgroundRoad traffic accidents (RTAs) are a significant global public health issue, leading to injuries, fatalities, and substantial economic losses. In Nepal, RTAs are a major concern, with notable regional variations in incidence and severity. This study analyzed the spatial distribution and types of road traffic injuries in Nepal from Fiscal Year (FY) 2019–2020 to 2021–2022, focusing on the impact of road types and geographic factors.MethodsThe study covers all 77 districts of Nepal, incorporating diverse geographic and climatic regions. Data on RTAs, including fatal and non-fatal injuries, were obtained from the Kathmandu Valley Traffic Police Office, Nepal. Spatial analysis was conducted using Quantum GIS (QGIS) and GeoDa software. Univariate and bivariate risk factor analyses were performed using Moran’s I statistics to detect spatial autocorrelation in RTA severity.ResultsThe overall fatality rate increased from 7.70 to 9.89 per 100,000 population from FY 2019–2020 to 2021–2022. However, spatial clustering of crashes showed a decline over the years. In FY 2019–2020 (Moran’s I; 0.276, p-value; 0.001), moderate clustering was observed, which weakened in the subsequent years (Moran’s I; 0.127, p-value; 0.002), with a near-random distribution by FY 2021–2022 (Moran’s I; -0.015, p-value; 0.457). The analysis of crash severity revealed significant variations across districts, with high fatality rates in remote areas like Mustang and Mugu, and low rates in districts such as Manang.Road Network AnalysisThe study examined the impact of different road types on RTA severity. Bitumen (BT) roads showed a negative correlation with RTA rates, while Earthen Roads (ER) were positively associated with higher RTA rates. Hot-spot and cold-spot clusters were identified for each road type, highlighting higher or lower RTA severity areas.ConclusionThis study provides valuable insights into the spatial patterns of RTAs in Nepal and the influence of road types on accident severity. The findings emphasize the need for targeted road safety interventions and infrastructure improvements, particularly in high-risk areas. By understanding spatial distributions and road network impacts, policymakers can better address road safety challenges and reduce the incidence of RTAs in Nepal.

Bibliometric Analysis of Traffic Accident Prediction Studies from 2003 to 2023: Trends, Patterns and Future Directions

Traffic accidents are one of the main causes of fatalities and serious injuries among both adults and children worldwide. Due to the ongoing significant socio-economic losses brought on by traffic accidents, precise estimation of the risk of accidents is crucial to reducing subsequent incidents. For this reason, a significant proportion of the studies in the literature include studies on estimating the risk, severity, frequency, location and duration of accidents. The objective of this article is to identify patterns, gaps and future research trends in traffic accident prediction studies conducted between 2003 and 2023. A bibliometric study is carried out to investigate the links and trends in traffic accident and forecasting studies, with a focus on identifying dominant narratives and networks within the academic community. In the keyword search, 1,566 articles were analysed using the Web of Science main collection and bibliometric indicators such as annual publications and citations, top 10, authors, journals, institutions, most cited articles, and a citation analysis of the articles was presented. The results obtained suggest that the discernible patterns identified in this bibliometric analysis of traffic accidents and their predictions will find a much broader application in new paradigms that are ready to catalyse transformative advances in this field, such as artificial intelligence, machine learning and Industry 4.0 applications.

Data-driven Bayesian risk assessment of factors influencing the severity of marine accidents in port areas

The most prevalent types of ship accidents in port areas are allisions, collisions, and groundings. A comprehensive risk assessment is needed to prevent and mitigate these accidents and their consequences. This study evaluates the risk of such accidents in port areas by presenting a model that elucidates the relationship between risk-identifying factors (RIFs) and accident severity. In this context, the RIFs are determined by analyszing the reports of 528 accidents that occurred in port areas between 1995 and 2023. Subsequently, the model is created by analysing the data derived from these reports using the Tree Augmented Naive Bayes (TAN) algorithm, which is an approach of the data-driven Bayesian network method. The findings of the study indicate that accident type, wind, ship age, and vessel type are the most influential factors in predicting the severity of accidents in port areas. It is thought that the model will assist port authorities in identifying operational risks contributing to accidents and in formulating preventive regulations.

A Machine Learning Approach for Modeling the Spatial-temporal Propagation Pattern of COVID-19

Abstract. Predicting infections resulting from epidemiological contagions involves a complex therapeutic endeavor that necessitates the consideration of various factors within the available data. COVID-19, a globally recognized highly dangerous disease, presents an opportunity to effectively understand its patterns and spatial determinants to combat similar ailments. This study seeks to employ a deep neural network and several machine learning techniques to identify the crucial factors influencing COVID-19 infections and to forecast the spatial-temporal spread of the disease. Previous research has focused on predicting outcomes using a limited set of 8 patient variables. To contribute to this field, the current investigation analyzes a dataset comprising 47,029 records of infected individuals who sought medical attention, with 18,433 of them testing positive for the coronavirus and requiring hospitalization. The study explores the prediction of illness and hospitalization needs based on geographical locations by utilizing machine learning algorithms such as Random Forest, Naive Bayes, Decision Tree, and K-Nearest Neighbor. The machine learning process commences with input data (comprising the 47,029 records), enabling the machine to identify patterns within the dataset and subsequently make informed decisions based on these patterns and insights gained. The primary objective is to allow the machine to autonomously learn and refine its predictions without human intervention. Applying these algorithms to COVID-19 data for spatial-temporal prediction revealed that the Random Forest algorithm achieved the highest accuracy of 0.84, while the Simple Bayes algorithm exhibited the lowest accuracy at 0.50. Additionally, the study compared random forest, decision tree, simple bayes, and K-nearest neighbor algorithms to predict the severity of road accidents.

Boosting Algorithms for the Accident Severity Classification

<div><i>Background:</i> Road accident severity estimation is a critical aspect of road safety analysis and traffic management. Accurate severity estimation contributes to the formulation of effective road safety policies. Knowledge of the potential consequences of certain behaviors or conditions can contribute to safer driving practices. Identifying patterns of high-severity accidents allows for targeted improvements in terms of overall road safety. <i>Objective:</i> This study focuses on analyzing road accidents by utilizing real data, i.e., US road accidents open database called “CRSS.” It employs advanced machine learning models such as boosting algorithms such as LGBM, XGBoost, and CatBoost to predict accident severity classification based on various parameters. The study also aims to contribute to road safety by providing predictive insights for stakeholders, functional safety engineering community, and policymakers using KABCO classification systems. The article includes sections covering theoretical methodology, data analysis, model development, evaluation, performance metrics, and implications for improving road safety measures by comparing the performance of different boosting algorithms on the CRSS dataset. This study aims to identify the most effective machine learning algorithm to integrate into our product line in the near future, enabling accurate prediction of both accident severity and occurrence. <i>Results and Conclusions:</i> This study addresses challenges in evaluating performance metrics for different severity classes within unbalanced datasets, emphasizing the impact of dominant classes like Class O (O = no apparent injury) on overall accuracy. The investigation reveals the limitations and conservatism associated with imbalanced data in boosting models, hinting at a potential ceiling in their performance around 80%. Comparative analysis of algorithms, including CatBoost, XGBoost, and LGBM, demonstrates comparable performance even in the case of applying KNN algorithm for pre-processing, based on various metrics, especially accuracy, <i>F</i><sub>1</sub>-score, ROC-AUC, and PR-AUC for all severity classes. XGBoost with KNN algorithm did not show any significant performance improvement compared to the XGBoost without KNN algorithm. The study includes performance metrics, such as <i>F</i><sub>1</sub>-score, CM upper triangle, ROC-AUC, and PR-AUC applied to an accident analysis case study. Future work directions involve extending the application of CatBoost, XGBoost, and other algorithms to diverse datasets, exploring the capabilities of deep neural networks, refining dataset preparation for accuracy improvement, and creating unified tools for hazard analysis and risk assessment.</div>

Black-Spot Identification from Ramol Cross Road to Kathwada Cross Road

Road accidents are a vital issue that is occurring all over the world. Accident severity is increasing with the exponential increase in vehicle population. Accident leads to disability, pain, suffering, grievous injuries, hospitalization expenditure, damage to health and property, social suffering, general degradation of the environment, and, in extreme cases, death. This study aims to identify black spots in the selected stretch involving four major intersections from Ramol crossroad to Kathwada crossroad of Ahmedabad. Data collection consists of collecting past accident data from traffic police station headquarters. The collected data are categorized into fatal accidents, grievous significant accidents, and minor accidents. The data is analyzed using the accident data collected, the Weighted Severity Index (WSI) method, and the PTV Visum software. WSI is estimated for the accident data for the year 2019-2023. By accident analysis, seven black spots were identified in this study. Inspecting the road stretch noted several issues and probable causes of accidents were identified. Poor road conditions, including cracks and potholes, affect road safety. Improper road markings and signs are one of the causes of accidents. The audited road stretch includes recommendations to prevent the number and severity of accidents.

Identifying High-Risk Patterns in Single-Vehicle, Single-Occupant Road Traffic Accidents: A Novel Pattern Recognition Approach

Despite various interventions in road safety work, fatal and severe road traffic accidents (RTAs) remain a significant challenge, leading to human suffering and economic costs. Understanding the multicausal nature of RTAs, where multiple conditions and factors interact, is crucial for developing effective prevention measures in road safety work. This study investigates the multivariate statistical analysis of co-occurring conditions in RTAs, focusing on single-vehicle accidents with single occupancy and personal injury on Austrian roads outside built-up areas from 2012 to 2019. The aim is to detect recurring combinations of accident-related variables, referred to as blackpatterns (BPs), using the Austrian RTA database. This study proposes Fisher’s exact test to estimate the relationship between an accident-related variable and fatal and severe RTAs (severe casualties). In terms of pattern recognition, this study develops the maximum combination value (MCV) of accident-related variables, a procedure to search through all possible combinations of variables to find the one that has the highest frequency. The accident investigation proceeds with the application of pattern recognition methods, including binomial logistic regression and a newly developed method, the PATTERMAX method, created to accurately detect and analyse variable-specific BPs in RTA data. Findings indicate significant BPs contributing to severe accidents. The combination of binomial logistic regression and the PATTERMAX method appears to be a promising approach to investigate severe accidents, providing both insights into detailed variable combinations and their impact on accident severity.

Feature fusion for a robust traffic accident assistance forecasting model with deep learning

Millions of people are involved in traffic accidents each year. According to the World Health Organization (WHO), approximately 1.19 million people experience fatal outcomes, while 20 to 50 million suffer non-fatal consequences that may affect them for the rest of their lives. Multiple studies suggest that one of the main causes of this high number of affected individuals could be mitigated through the rapid intervention of medical assistance services. This paper thus proposes a general GTAAF model based on deep learning that predicts the severity of traffic accidents in real-time based on the surrounding characteristics, determining whether an accident requires medical assistance, using tabular data as input. This model can be utilized by emergency services to prioritize the allocation of medical resources in any region of the world. The main difference from existing state-of-the-art methods is that these are focused on specific localities and so cannot be applied universally, because each existing method is applied in a specific region of the world due to the individual characteristics of these data. To demonstrate this generalization capability, the GTAAF model was compared with six other state-of-the-art models applied to 8 populations in 3 different countries, showing a significant quality advantage over the other models (up to a 13.8% improvement in the F1-Score metric in classifying accidents that require assistance in the best case, and up to a 6.5% improvement in accidents that do not require assistance).

Анализ влияния нарушений требований безопасности на возникновение происшествий на предприятиях магистрального транспорта газа

The procedure of behavioral safety audit has successfully established itself in the practices of Russian and international companies in various industrial sectors. When conducting a behavioral safety audit, an auditor compares the observed conditions and employee’s actions with the established classifier based on the behavioral safety audit results report. In order to determine the relationship between hazardous actions of employees and dangerous conditions at their workplaces with the potential severity of accident consequences, a retrospective analysis has been conducted for the study, where the root causes of 411 accidents that occurred at main gas transportation enterprises have been compared with violations of safety requirements using 33 criteria of the behavioral safety audit. During the discriminant analysis, canonical discriminant functions and centroids have been determined, the coefficient of canonical correlation and the values of Wilks' lambda criterion and F-criterion have been calculated, and standardized and structural coefficients have been identified. As a result of the study, 10 criteria of the classifier of the behavioral safety audit have been detected, whose violation significantly affects the determination of the potential severity of accidents; the degree and conditions of such impact within the context of the examined selection have been established and substantiated.

Discovering the underground coal mining accident patterns in Spain from 2003 to 2021: Insights through machine learning techniques

The safety of underground coal mining has always been a global concern, involving the stable supply of energy and stakes in miners’ lives. Lessons learned from historical accidents and transforming into practical experience help reduce the quantity and severity of accidents. In this study, six ensemble learning techniques, including AdaBoost, Extra Trees, GBDT, LightGBM, Random Forest, and XGBoost, were used to investigate the correlation between accident-causing factors and severity. Firstly, 39487 underground coal mine accidents data was obtained from Spain, variables were categorized and coded. To address the extreme class imbalance, a new dataset (2468 cases) was obtained by data sampling from the original database. Subsequently, the new dataset was randomly divided into training sets (75% of the data) and test sets (25% of the data), then the hyperparameters of each model were optimized and configured. Thirdly, the models’ performance was evaluated on the test data by five metrics (accuracy, Cohen’s Kappa, precision, recall, and F1). Finally, accident patterns were derived from the identified variables along with preventive strategies. Results show that tree-based ensemble learning model performs better compared to the boosting model, and the relative importance of seven variables were determined, where previous cause (PC) and material agent (MA) are the most important factors, followed by the miner’s physical activity (PA), age (A), and experience (E), scale (S) and preventive organization (PO) are in the third tier. Furthermore, the type of accident and injury caused by PC were confirmed. Working with hand tools, younger age, lack of experience, small-scale coal mines, and unfit preventive organization increased the risk of accidents. This study not only facilitates the prediction of accident severity but also provides strategies for preventing and mitigating accidents.

Analysis of influencing factors of traffic accidents on urban ring road based on the SVM model optimized by Bayesian method.

Based on small scale sample of accident data from specific scenarios, fully exploring the potential influencing factors of the severity of traffic accidents has become a key and effective research method. In order to analyze the factors mentioned above in the scenario of urban ring roads, this paper collected data records of 1250 traffic accidents involving different severity on urban ring road of a central city in northwest China in the past 3 years. Firstly, the Support Vector Machine (SVM) model of non-parametric method is utilized to analyze the data above, and three kernel functions of linear, inhomogeneous polynomial and Gaussian radial basis are constructed respectively. Considering comprehensively 16 potential influencing factors covering the driver-vehicle-road-environment integrated system, the SVM models of above three kernel functions are verified, accuracy reaches 0.771 and F1 reaches 0.841. Then, Bayesian Optimization (BO), Grids Search (GS) and Rough Set (RS) are utilized as optimizer to adjust the parameters of Gaussian radial basis function SVM model, the performance of BO-SVM is further improved and reaches the optimum, with an average accuracy of 0.875 on the test set and a F1 of 0.886, completely outperforming the benchmark models of GS-SVM, RS-SVM, Bilayer-LSTM and BP. Finally, the sensitivity analysis method is utilized to quantify the sensitivity of the potential influencing factors to the severity of road accidents, and the backward selection method is utilized to screen the core influencing factors that influence the severity of accident, concluded that core influencing factors are age, driving mileage and vehicle type. This paper will provide reference for the analysis of the significant influencing factors for road accidents severity, and to provide theoretical support for the precise formulation of accident improvement strategies.

Exploring and identification of road traffic crash hotspots in Kigali-city, Rwanda

Road traffic crashes are on the rise worldwide, particularly in African countries. Rwanda, in this context, faces the challenge of an increasing frequency of crashes each year. Kigali, the capital city of Rwanda, witnesses over three-quarters of the total population as daily road users, leading to high traffic volumes and congestion during peak hours. Identifying dangerous locations and providing insights to transport policymakers is crucial to minimise the severity of crashes in the city. This study focuses on identifying crash hotspots through spatial analysis and explores the various factors contributing to crash severity in the capital city of Rwanda. The study utilises crash data recorded between 2015 and 2022 by the Rwanda National Police, conducting temporal, spatial, and statistical analyses. A multinomial regression model was employed to examine the significance of contributing factors to crash severity. Additionally, six supervised machine learning classifiers were employed to assess their predictive capabilities for accident severity in Kigali city. The findings indicate that the Gasabo district recorded the highest number of crashes among all districts. Time series analysis reveals a decrease in fatal accidents from 2020 to 2022 in Kigali city, while the number of injury and property damage crashes progressively increased from 2019 to 2022. Among the predictive models, the Random Forest model demonstrated strong predictive abilities, achieving an accuracy of 88.11% compared to other models. These results provide valuable insights for road safety policymakers to proactively make decisions and take precautions to prevent crashes before they occur.

Evaluation of the Performance of Neural and Non-Neural Methods to Classify the Severity of Work Accidents Occurring in the Footwear Industry Complex

In the footwear industry, occupational risks are significant, and work accidents are frequent. Professionals in the field prepare documents and reports about these accidents, but the need for more time and resources limits learning based on past incidents. Machine learning (ML) and deep learning (DL) methods have been applied to analyze data from these documents, identifying accident patterns and classifying the damage’s severity. However, evaluating the performance of these methods in different economic sectors is crucial. This study examined neural and non-neural methods for classifying the severity of workplace accidents in the footwear industry complex. The random forest (RF) and extreme gradient boosting (XGBoost) methods were the most effective non-neural methods. The neural methods 1D convolutional neural networks (1D-CNN) and bidirectional long short-term memory (Bi-LSTM) showed superior performance, with parameters above 98% and 99%, respectively, although with a longer training time. It is concluded that using these methods is viable for classifying accidents in the footwear industry. The methods can classify new accidents and simulate scenarios, demonstrating their adaptability and reliability in different economic sectors for accident prevention.

Traffic accident severity prediction based on interpretable deep learning model

ABSTRACT Accurately predicting traffic accident severity is crucial for road safety. However, existing studies lack interpretability in revealing the relationship between accident severity and key factors. To address this issue, we propose a new interpretable analytical framework. The framework utilizes XGBoost and SHAP to select effective factors. Then the AISTGCN model is constructed by improving the STGCN through the local attention mechanism to predict the severity of the accident. Finally, DeepLIFT is used to interpret the forecasts and identify key factors. Our experiments using real-world UK accident data demonstrate that our proposed AISTGCN outperforms baseline models in outcome prediction with an accuracy of 0.8772. The computation time was reduced and the reliability of predictions was enhanced through screening for effective factors. Furthermore, DeepLIFT provides more reasonable explanations when explaining accidents of different severity, indicating that vehicle count significantly impacts. Our framework aids in developing effective safety measures to reduce accidents.

Investigating Factors Influencing Crash Severity on Mountainous Two-Lane Roads: Machine Learning Versus Statistical Models

Due to poor road design, challenging terrain, and difficult geological conditions, traffic accidents on mountainous two-lane roads are more frequent and severe. This study aims to address the lack of understanding of key factors affecting accident severity with the goal of improving mountainous traffic safety, thereby contributing to sustainable transportation systems. The focus of this study is to compare the interpretability of model performances with three statistical models (Ordered Logit, Partial Proportional Odds Model, and Multinomial Logit) and six machine learning models (Decision Tree, Random Forest, Gradient Boosting, Extra Trees, AdaBoost, and XGBoost) on two-lane mountain roads in Yunnan Province, China. Additionally, we assessed the ability of these models to uncover underlying causal relationships, particularly how accident causes affect severity. Using the SHapley Additive exPlanations (SHAP) method, we interpreted the influence of risk factors in the machine learning models. Our findings indicate that machine learning models, especially XGBoost, outperform statistical models in predicting accident severity. The results highlight that accident patterns are the most significant determinants of severity, followed by road-related factors and the type of colliding vehicles. Environmental factors like weather, however, have minimal impact. Notably, vehicle falling, head-on collisions, and longitudinal slope sections are linked to more severe accidents, while minor accidents are more frequent on horizontal curve sections and areas that combine curves and slopes. These insights can help traffic management agencies develop targeted measures to reduce accident rates and enhance road safety, which is critical for promoting sustainable transportation in mountainous regions.

Influence of Road Infrastructure Design over the Traffic Accidents: A Simulated Case Study

The influence of road infrastructure over the severity of road accidents, in particular some specific features of it, represents the subject of this study. Generally, when an accident occurs, its causes are represented by a number of factors such as driver experience, fatigue, driving under the influence of alcohol and other psychoactive substances, road configuration, weather conditions, speeding, distracted driving, and unsafe road infrastructure. Road design is a key factor regarding the safety of all traffic participants. In this paper, the influence of unsafe roadside element designs on the incidence of traffic accidents, the degree of vehicle passenger injury, and the level of car damage were investigated. The present study was inspired by the high number of accidents produced on European route E68 (DN1) in Romania, a significant part of which was generated and accentuated by the effects of improper roadside design.

Quantitative Risk Assessment of an Oil-Gas-Hydrogen-Electricity Integrated Energy Station in China.

This paper presents a quantitative risk assessment (QRA) study of an oil-gas-hydrogen-electricity integrated energy station in China. A comprehensive assessment of the station was made in terms of consequences and risks, respectively. Consequence analysis shows that the severity of hazardous chemical leakage accidents in the station is in the order of natural gas, hydrogen, gasoline, and diesel fuel, especially in the liquefied natural gas (LNG) and compressed natural gas (CNG) storage tank areas, which may cause major accidents in the event of leakage. The results of the risk assessment showed that the individual and societal risks of the integrated oil-gas-hydrogen-electricity energy station exceeded the risk acceptance criteria. Moreover, the catastrophic rupture of LNG and CNG storage tanks is the main cause of the unacceptable risk. Additional mitigation measures for them, mainly including the installation of emergency manual shut-off valves, the installation of combustible gas detection and alarm devices, and pressure relief protection devices, can effectively reduce the risk to an acceptable level.

Analysis of machine learning models for traffic accidents severity classification

In the modern world, traffic accidents frequently result in fatalities and serious injuries. The ability of machine learning to foretell the severity of road traffic accidents has shown great promise. The classification of traffic accidents has shown to be a good application for algorithms like random forest. In this paper, performance on a specific dataset has been evaluated using random forest and other models. The dataset used for the analysis came from a publicly accessible source and contained information on several variables like the type of road, the time of day, and the weather. In order to analyze the severity of accidents, a number of algorithms were applied to the dataset, including decision tree, random forest classifier, and logistic regression algorithm. Each model was evaluated on parameters such as model accuracy, precision and recall of the model, and F1 score. The random forest classifier outperformed the other models, achieving an accuracy of 98.48%. The study concludes that machine learning algorithms can accurately predict the severity of road traffic accidents, which could help to reduce the number of accidents and fatalities on the road.

Does air pollution cause more car accidents? Evidence from auto insurance claims

Using a proprietary data set of auto insurance claims from May 2014 to December 2016, this paper examines the influence of air pollution on the number and severity of traffic accidents in China. Combining an instrumental variable strategy with high-dimensional fixed effects, we find that air pollution significantly increases the occurrence of traffic accidents, with each 1 μg/m3 increase in the particulate matter 2.5 (PM2.5) resulting in a 0.12 % increase in traffic accident probability and a 0.40 % increase in traffic accident number within one day. A different pattern is revealed in our analysis of accident severity, evidenced by a decrease of 1.20 % in the average claim ratio compared to its mean value and a reduction of 26 yuan in the average claim amount made with an increase of 1 μg/m3 in PM2.5. Combining the effect on the number and severity of traffic accidents, for each 1 μg/m3 increase in daily PM2.5, the district daily claim amount decreases by approximately 34 yuan. Further analysis indicates that this may be related to cautious driving behavior resulting from the driver's increased risk aversion. By exercising caution and care on the road, drivers can reduce the negative influence of air pollution on road safety and avoid non-subjective behavioral biases.