Analysis of traffic accident causes based on data augmentation and ensemble learning with high-dimensional small-sample data

This article presents a comprehensive investigation into the causes of road transport accidents in Ukraine, focusing on the application of risk analysis to minimize economic and human losses. The study begins with a detailed examination of the current road safety scenario in Ukraine, highlighting the alarming statistics of road accidents and their consequences. The central aim is to develop effective strategies for controlling and managing the causes of road accidents. The research identifies the primary factors contributing to road accidents, such as exceeding safe speed limits, driving under the influence of alcohol, etc. A risk management system is proposed, encompassing the creation of a structured risk management framework, development of methodologies, and implementation of information and analytical systems. Th e article emphasizes the importance of continuous risk monitoring and assessment, including risk identification, analysis, and management. Statistical observations and analysis play a crucial role in diagnosing risks, which are then used to devise risk minimization measures and improve road safety. The study concludes with recommendations for risk-based decision-making and the implementation of targeted interventions to reduce road accidents and associated risks in Ukraine. First published online 11 February 2025

ObjectivesTo design a comprehensive road traffic injury surveillance system for National Highways & Motorway Police of Pakistan on the basis of WHO Injury Surveillance Guidelines and keeping in view several...

The high impact of traffic accidents makes it imperative to formulate public policies to reduce their occurrence. In this task, knowing the cause of accidents is of paramount importance. The use of data mining and big data adapts to the complexity of the phenomenon under study. In order to classify some possible causes of traffic accidents, we built a data model to describe the behavior and dynamic of the participant agents in the traffic accident event in Colombia. This paper presents the application of MLP and Naive Bayes algorithms to identify the possible immediate cause and the rules decision algorithm PART for the root cause of traffic accidents. Models have been tested aiming to obtain the goodness of fit by increasing metrics like Recall, Precision, ROC and Kappa index, and minimize the RMSE.

Nowadays, congestion and accidents are creating major risks to cities, including delays, higher fuel usage, and compromised safety. Effective traffic modelling and accident data analysis are critical for identifying high risk identifying accident-prone locations, understanding the causes of accidents and creating focused actions to enhance traffic flow and safety. GIS is an effective tool for integrating, analysing and visualizing different geographical data relevant to transportation networks such as, traffic flow, infrastructure, and safety. It enables geographical analysis and visualization of accident hotspots by integrating accident data, road conditions, traffic numbers, and environmental factors. The use of GIS in traffic modelling and accident data analysis provides considerable benefits in urban transportation planning and management. The aim of the paper is to provide an overview of the application of GIS in traffic modelling and accidental data analysis, highlighting the methodologies, advancements, and challenges in this field. The review shall provide a comprehensive assessment of the current state of traffic modelling and accidental data analysis using GIS. It will highlight the significant contributions of GIS technology, identify key research gaps, and offer insights into future directions for enhancing transportation planning and decision-making processes.

Road traffic accident databases provide the basis for road traffic accident analysis, the data inside which usually has a radial, multidimensional, and multilayered structure. Traditional data mining algorithms such as association rules, when applied alone, often yield uncertain and unreliable results. An improved association rule algorithm based on Particle Swarm Optimization (PSO) put forward by this paper can be used to analyze the correlation between accident attributes and causes. The new algorithm focuses on characteristics of the hyperstereo structure of road traffic accident data, and the association rules of accident causes can be calculated more accurately and in higher rates. A new concept of Association Entropy is also defined to help compare the importance between different accident attributes. T-test model and Delphi method were deployed to test and verify the accuracy of the improved algorithm, the result of which was a ten times faster speed for random traffic accident data sampling analyses on average. In the paper, the algorithms were tested on a sample database of more than twenty thousand items, each with 56 accident attributes. And the final result proves that the improved algorithm was accurate and stable.

This study analyzes how the number of traffic accidents can be effectively reduced by detailed study over the documented accidents. In this paper, we propose a new method called the traffic accident causes analysis based on dominance rough set theory. This method improvises on the limitations of Pawlak's rough set model in analyzing the causes of accidents, such as using discrete continuous attributes to obtain unreasonable rules. Furthermore, the paper combines facts about traffic accidents and the objective data processing ability of rough set theory. Specifically, every single cause of accidents is sorted according based on our experience and then relevant algorithms are applied to illustrate the relationship. Lastly, an example is demonstrated that shows that this method reduces the uncertainty of discretization of continuous attributes and excludes the causes that rule against human empirical knowledge.

The features of the educational process in training personnel for the road industry at Altai State Technical University are considered, taking into account the development of the digital economy. Much attention has been paid to key technologies that provide digitalization for the road industry. The set of disciplines necessary for obtaining knowledge in digital technologies for the further activities of university graduates is analyzed. When studying bachelor’s and master’s programmes, future specialists acquire knowledge of modern methods in computer-aided design of highways, geographic information systems in construction, information modeling and intelligent transportation systems. During the educational process, an understanding of the importance of implementing an automated control of equipment in road construction works using onboard information management system is also formed. The directions of scientific research are described in which students studying Master’s Programme in Highway Design Construction and Management work. These studies make extensive use of digital technology. The analysis of road traffic accidents for identifying hazardous sections is given, as well as the analysis of the current road state with the development of measures for improving the transport and operational road condition. Current research areas using digital technologies are in demand in modern conditions.

Explanatory and prediction power of two macro models. An application to van-involved accidents in Spain

IntroductionViolence has acquired an endemic character in society; traffic violence stands out, particularly considering the upward trend in deaths among motorcyclists.ObjectivesTo describe the sociodemographic profile and analyze the vulnerabilities of motorcyclists from their experiences in traffic.MethodsThis is an exploratory descriptive study with a quantitative approach, performed through interviews with three groups of motorcyclists: those who used motorcycles for recreation, transportation, and for work.ResultsMost participants were male (> 80%), mainly young adults in the transportation and work groups (p < 0.001); with higher schooling and income levels in the recreation group (p < 0.001); and with lightweight motorcycles in the transportation and work groups (p < 0.001). Similar experiences fragilizing all three groups include alcohol consumption, disregard of traffic rules, and acts of aggression. With a different experience from the transportation and work groups, the recreation group experienced circumstances that were favorable to traffic safety. Overall, the pattern of answers in the transportation and work groups associated them with higher susceptibility; the work group, represented by motorcycle couriers, was marked by accidents and poor working conditions.ConclusionsContextual variables inflict a higher traffic vulnerability on motorcyclists of the transportation and work groups. We advocate that further exploration of the analysis of road traffic accidents be a social construct, and advances in the knowledge of risks and margins for implementing safety measures will only be possible when the approaches consider social, economic, and cultural aspects.

Safety on road networks is the utmost important factor to consider for public well-being and transportation efficiency.This study introduces a new approach that combines Getis-Ord spatial statistics and crash rate analysis to identify significant road traffic accidents (RTAs) characterized by hotspots on road segments in Addis Ababa's road network.The study's results visually portray the crash locations, associating them with the underlying road network, which demonstrates a notable concentration of accident hotspots, between the years from 2014 to 2019 on Addis Ababa's roads network.The RTAs spatial analysis resulted in the identification of hotspots on 33 road segments, 3 intersections, and 10 roundabouts/squares.Among the identified hotspots, the road segment recognized as Djibouti Street, extending from Bole Edna Mall to the 22 "mazoriya" roundabouts, stands out as the most significant accident hotspot.It exhibits an average of 37.5 crashes per kilometer per year, encompassing a road segment length of 1141 m.Using both methods in this study is crucial for validating findings by identifying high-crash segments and enhancing their reliability and hotspot accuracy.This unique validation method links each traffic accident's spatial data with the road network using both crash rate and spatial statistical analysis, effectively pinpointing accident hotspots.Given the limited resources, this approach enhances awareness of accident-prone locations, enabling the prioritization of safety measures to improve road safety.It effectively addresses spatial analysis gaps related to RTAs in Ethiopia and holds practical significance by identifying and prioritizing safety measures aimed at reducing accidents within Addis Ababa's road network.

In recent years there has been serious concern on the increasing rate of road traffic accidents in Nigeria. Geographic Information System (GIS), a high performance computer based tool is useful in road traffic management and vehicular movement study. In this work, the GIS software was used to analyze and depicts the trend and characteristics of road traffic accidents in Nigeria. The data used in this study were obtained from Federal Road Safety Commission Dashboard (FRSC Planning Advisory Unit Analysis) for the year 2012 and 2013 and Nigerian Tribune Publication of Tuesday 3 rd April , 2012. The data obtained was entered and use to developed road traffic accidents information system (RTAIS). Analysis and capability of the developed system was illustrated and shown graphically. A total of 1,060,507 cases of road traffic crashes occurred between the year 1960 and 2012 of which 170,844 were fatal, 456,055 were serious and 428.790 were minor, claiming 322,427 lives. This study further revealed that Federal Capital Territory (FCT) Abuja ranked first with 6.71% based on road traffic crash fatalities per 100,000 human populations on state by state with Kebbi state being the least (0.25%). The GIS –based analysis shows that category of vehicles involved in road traffic crashes in year 2012 on state basis varies with cars having the highest cases(4,632) followed by buses (2,239), motorcycles (1,807),truck(1,037) and tankers/trailers having the least(641). It is recommended that the Nigerian Police Force (NPF), Federal Road Safety Commission (FRSC), Vehicle Inspection Office (VIO) and other agencies linked with road traffic management and safety will find this study useful in identifying remedial measures for the improvement of safety on Nigerian roads. Key Words: Road Traffic, Accident, GIS, Vehicles Analysis, Nigeria

BackgroundRoad traffic accidents (RTAs) continue to pose a significant menace to global public health in the form of a high incidence of mortality, disability and economic expense. Their space-time trends are of importance for policy decision-making. This particular study employed spatial analysis to identify high-risk zones and found significant clustering of accidents in urban centers as well as increasing semi-urban and rural vulnerabilities, supporting the need for safety interventions and road infrastructure improvements in Nepal. This paper aims to determine and analyze the incidence of RTAs in Nepal from 2019 to 2022, primarily focusing on vehicle-types and spatial distribution.MethodsData from all seven provinces and Kathmandu Valley Traffic Police Office were analyzed to examine RTAs patterns across 77 districts of Nepal. The data were processed and visualized using Quantum GIS (QGIS), and spatial analysis performed using Global and Local Moran’s I statistics, along with Local Indicators of Spatial Association (LISA), to identify spatial clusters of accidents.ResultsThis study identified statistically significant spatial clustering of vehicle types involved in RTAs. High-High (HH) clusters, indicating areas with elevated accident rates surrounded by similarly high-risk zones were concentrated in urban centers particularly Kathmandu, Lalitpur, and Bhaktapur. Conversely, Low-Low (LL) clusters, reflecting lower accident rates in sparsely populated regions, were observed in rural areas. Temporal analysis revealed a steady rise in RTAs incidence, with rates increasing from 63.35 per 100,000 population in 2019–2020 (Moran’s I = 0.741) to 94.46 in 2020–2021 (Moran’s I = 0.595) and 123.05 in 2021–2022 (Moran’s I = 0.556).ConclusionThis present study observed the growing incidence of RTAs in Nepal. The results highlight the critical need for geographically tailored road safety interventions with priority given to urban and semi-urban zones. Effective strategies should emphasize enhanced road traffic law enforcement, strict regulation of commercial and two-wheeled vehicles as well as targeted infrastructure upgrades in an effective manner.

Analysis of traffic accidents on rural highways using Latent Class Clustering and Bayesian Networks

This research develops and tests a computerized model for intersection accident data visualization and analysis. The model implemented a rule-based visualization module that automatically constructs the collision diagram according to traffic flow conflicts, where accident characteristics can be analyzed either individually or collectively. A data acquisition module was developed for accessing accident databases maintained in state transportation and public safety departments. Using a local database management system as part of the model, the user can select to analyze any accident data through Structured Query Language statements. In addition, this model has implemented a data analysis and graphing module for comparing similarities and identifying differences in accident data and for presenting the results of the analysis through charts and tables. The model was tested with accident data obtained from the Ohio Department of Public Safety in a number of scenario studies, and some of them were presented in this paper as examples. Because the model is very time efficient and user friendly, it can be developed into an effective engineering tool for field applications.

Worldwide, road accidents have become more of a problem due to the exponential growth in vehicle traffic, which has led to a significant increase in both human mortality and economic losses. Traditional methods of accident analysis, which rely on static statistical models and clustering techniques, have limitations when it comes to accurately predicting future accidents because they fail to account for the nonlinear, dynamic, and multivariate aspects of accident causation. In order to address these constraints, this study presents a methodology for road traffic accident causation analysis and predictive modeling that is based on machine learning. After implementing thorough preprocessing steps like data cleaning, normalization, and class balancing using the Synthetic Minority Over-sampling Technique (SMOTE), the framework deploys classifiers like Decision Tree (DT), and Random Forest (RF). Analyses of comparative performance show that the Random Forest model maintains interpretability and routinely produces superior predictive accuracy, allowing for robust accident prediction. In addition, policymakers and traffic safety authorities can benefit from feature importance analysis's identification of significant causal determinants, such as weather conditions, roadway characteristics, and illumination factors. In addition to improving forecast accuracy, the results show that the suggested method connects black-box learning models with practical decision-making tools. Ultimately, our study aids in the creation of solutions that are scalable, interpretable, and deployable for intelligent transportation system design, proactive accident avoidance, and real-time traffic risk assessment.