Search

The Yangtze River estuary in China is characterized by a complex maritime geographical environment and presents significant challenges to ship manoeuvring and control, thereby increasing the risk of ship collision accidents. Based on the 2013–2022 shipwreck investigation report published by Shanghai and Zhejiang Maritime Safety Administration, this paper analyses the primary factors responsible for ship collision accident severity in the Yangtze River estuary from four aspects, namely ship, environment, human and management. Utilizing accident severity as the dependent variable and 24 factors, including ship type, gross tonnage, wind speed, operational errors and so on, as independent variables, the study employed a stepwise regression approach to filter the variables. Subsequently, an ordered probit regression model was constructed based on the 10 most influential variables, followed by a marginal effect analysis. The findings indicate that a ship's gross tonnage, wind speed, ocean current speed, offshore distance and day/night conditions significantly influence the likelihood of different accident levels. Specifically, wind speed, offshore distance and ocean current speed have a negative impact on minor and general accidents while positively affecting major and severe accidents. Gross tonnage and daytime/nighttime have a positive impact on minor and general accidents but negatively impact major and severe accidents. Moreover, general accidents exhibit the most pronounced marginal effect for each explanatory variable. The findings can help the shipping authorities to identify the causes of ship collision accidents and take effective measures to reduce such accidents, thereby enhancing the safety of ship navigation in the area. [ABSTRACT FROM AUTHOR]

Across the United States, wildlife vehicle crashes (WVCs) are increasing and remain consistently deadly to drivers, despite a downward trend in fatal automobile accidents overall. That said, the factors related to severe WVCs are unclear. With this in mind, we pursued a statistical model to reveal factors associated with WVCs that result in severe injury or death to drivers. We hypothesize that there are statistically significant interactions and non-linear relationships between these factors and severity occurrence. We developed a generalized additive model (GAM) with linear terms, additive terms, and a binary response for severity. We surmise that our fitted model results will quantify the relationship between significant variables and severity occurrence, and ultimately help to develop countermeasures to mitigate serious injury. The model was fitted to WVC records occurring between 2002 and 2019 in the state of New Hampshire. Fitted linear terms revealed: 1) in inclement weather, there is about a 22% increase in the odds of severity for slick surface conditions compared to dry surface conditions; 2) for the warmer months (spring/summer), there is a 42% decrease in the odds of severity for straight roads compared to those with curvature/incline; 3) for highways, the odds of severity decreases by 48% for accidents occurring on NH’s two major intestates highways, and 4) for spring/summer (as compared to the fall/winter), there is more than a 3-fold increase in the odds of severity for two-way traffic. Fitted additive terms revealed: 1) the odds of severity increased in the early hours, between midnight and 6AM, and after 5PM; 2) speeds between 45 and 60 mph are associated with an increase in the odds of a severe accident, while both lower and higher speeds (those below 45 and above 60 mph) are associated with a decrease in the odds of a severe accident; and 3) low, mid-range, and high human population densities are associated with decreases, increases, and decreases in odds of severity, respectively. Cross validation and resulting ROC curves gave evidence that our model is well specified and an effective predictor. Results could be used to inform drivers of potentially dangerous roadways/conditions/times.

Background: Road accidents claim around 1.35 million lives annually, with countries like India facing a significant impact. In 2019, India reported 449,002 road accidents, causing 151,113 deaths and 451,361 injuries. Accident severity modeling helps understand contributing factors and develop preventive strategies. AI models, such as random forest, offer adaptability and higher predictive accuracy compared to traditional statistical models. This study aims to develop a predictive model for traffic accident severity on Indian highways using the random forest algorithm. Methods: A multi-step methodology was employed, involving data collection and preparation, feature selection, training a random forest model, tuning parameters, and evaluating the model using accuracy and F1 score. Data sources included MoRTH and NHAI. Results: The classification model had hyperparameters ‘max depth’: 10, ‘max features’: ‘sqrt’, and ‘n estimators’: 100. The model achieved an overall accuracy of 67% and a weighted average F1-score of 0.64 on the training set, with a macro average F1-score of 0.53. Using grid search, a random forest Classifier was fitted with optimal parameters, resulting in 41.47% accuracy on test data. Conclusions: The random forest classifier model predicted traffic accident severity with 67% accuracy on the training set and 41.47% on the test set, suggesting possible bias or imbalance in the dataset. No clear patterns were found between the day of the week and accident occurrence or severity. Performance can be improved by addressing dataset imbalance and refining model hyperparameters. The model often underestimated accident severity, highlighting the influence of external factors. Adopting a sophisticated data recording system in line with MoRTH and IRC guidelines and integrating machine learning techniques can enhance road safety modeling, decision-making, and accident prevention efforts.

Background: Road accidents claim around 1.35 million lives annually, with countries like India facing a significant impact. In 2019, India reported 449,002 road accidents, causing 151,113 deaths and 451,361 injuries. Accident severity modeling helps understand contributing factors and develop preventive strategies. AI models, such as random forest, offer adaptability and higher predictive accuracy compared to traditional statistical models. This study aims to develop a predictive model for traffic accident severity on Indian highways using the random forest algorithm. Methods: A multi-step methodology was employed, involving data collection and preparation, feature selection, training a random forest model, tuning parameters, and evaluating the model using accuracy and F1 score. Data sources included MoRTH and NHAI. Results: The classification model had hyperparameters ‘max depth’: 10, ‘max features’: ‘sqrt’, and ‘n estimators’: 100. The model achieved an overall accuracy of 67% and a weighted average F1-score of 0.64 on the training set, with a macro average F1-score of 0.53. Using grid search, a random forest Classifier was fitted with optimal parameters, resulting in 41.47% accuracy on test data. Conclusions: The random forest classifier model predicted traffic accident severity with 67% accuracy on the training set and 41.47% on the test set, suggesting possible bias or imbalance in the dataset. No clear patterns were found between the day of the week and accident occurrence or severity. Performance can be improved by addressing dataset imbalance and refining model hyperparameters. The model often underestimated accident severity, highlighting the influence of external factors. Adopting a sophisticated data recording system in line with MoRTH and IRC guidelines and integrating machine learning techniques can enhance road safety modeling, decision-making, and accident prevention efforts.

Occupational accidents in the construction industry occur more frequently when compared with other industries. Construction occupational accidents still have not been prevented at the desired level. Several studies in the literature have been conducted to predict the occurrence frequency of these accidents using classical statistical and machine-learning techniques. However, some challenges regarding imbalanced and multicollinearity problems present in the dataset are not considered while analyzing data with a large size and a large number of categorical variables. This study aims to predict the severity of nonfatal construction accidents considering mentioned challenges to obtain more accurate results. In this study, standard binary logistic regression, Firth, Ridge, Lasso, and Elastic Net Regularized logistic regression models were used for the prediction of lost workdays in the construction industry and results were compared. The data used were classified into five groups: victim, workplace, accident time, accident and sequence of events, and postaccident state-related variables. The results showed that Firth's logistic model is the best-performing model and age, education, vocational education, workplace size, project type, working environment, accident month and year, general and specific activities, material agent, type of injury, and part of body injured are the most significant variables. This study, by providing interpretable machine learning tools, is the first attempt to use proposed models in the area of construction safety in the literature. [ABSTRACT FROM AUTHOR]

Access to healthcare services is more challenging in remote northern regions due to higher travel costs associated with longer distances and harsh environments. Emergency response to road accidents in remote regions can take significantly longer than in more easily accessible locations, and potentially lead to more severe health outcomes. Accordingly, it is important to have insights on the factors that influence road accident severity in remote regions. This paper uses police accident data from Canada’s Northwest Territories between 1989 and 2019 to assess the influence of various factors on accident severity, including environmental, infrastructure-specific, geographical and accident-specific characteristics. Using multinomial logistic regression, we find that remoteness, off-road vehicle involvement and alcohol involvement increase the odds of a road accident being in a higher severity category. Overall, we find that risk factors that are more prevalent in Canada’s northern, remote regions may increase the severity of accidents in comparison to less remote regions.

According to the World Health Organisation (WHO), the majority of victims of road traffic accidents are vulnerable road users, including pedestrians. However, underreporting of such accidents leads to inadequate availability of reliable data in this area of research. The present study was an attempt to investigate the severity of this problem of underreporting of accidents involving pedestrians. Primary data were collected for the purpose, and statistical techniques such as univariate and bivariate analyses, Chisquare testing, and Binary Logistic Regression were used to examine the association of several diverse factors including demographic features, severity of injury, lost productivity, and receipt of compensation, with accident reporting behaviour. The results revealed that the most common accidents (30%) involved motorcycles, while accidents due to slipping on roads or pavements and roadside structures together accounted for 33% of all reported accidents. These, according to this study, appeared to largely go under-reported, precluding further action. The study found a strong relationship between reporting behaviour and the nature and severity of injuries sustained in the accident. The lowest chance of reporting an accident was found in regard to slipping on a pavement while the injuries treated without visiting a hospital had the lowest likelihood of being reported to the Police.

Background: Road accidents claim around 1.35 million lives annually, with countries like India facing a significant impact. In 2019, India reported 449,002 road accidents, causing 151,113 deaths and 451,361 injuries. Accident severity modeling helps understand contributing factors and develop preventive strategies. AI models, such as random forest, offer adaptability and higher predictive accuracy compared to traditional statistical models. This study aims to develop a predictive model for traffic accident severity on Indian highways using the random forest algorithm. Methods: A multi-step methodology was employed, involving data collection and preparation, feature selection, training a random forest model, tuning parameters, and evaluating the model using accuracy and F1 score. Data sources included MoRTH and NHAI. Results: The classification model had hyperparameters ’max depth’: 10, ’max features’: ’sqrt’, and ’n estimators’: 100. The model achieved an overall accuracy of 67% and a weighted average F1-score of 0.64 on the training set, with a macro average F1-score of 0.53. Using grid search, a random forest Classifier was fitted with optimal parameters, resulting in 41.47% accuracy on test data. Conclusions: The random forest classifier model predicted traffic accident severity with 67% accuracy on the training set and 41.47% on the test set, suggesting possible bias or imbalance in the dataset. No clear patterns were found between the day of the week and accident occurrence or severity. Performance can be improved by addressing dataset imbalance and refining model hyperparameters. The model often underestimated accident severity, highlighting the influence of external factors. Adopting a sophisticated data recording system in line with MoRTH and IRC guidelines and integrating machine learning techniques can enhance road safety modeling, decision-making, and accident prevention efforts.

In this paper, the SPSS program (version 25) and Binary Logistic Regression Model were used to implement and identify the risk factors that affect traffic accidents on Baghdad highways. Due to the increase in the number of traffic accidents that led to injuries and deaths in Iraq during the past years and the lack of specialized studies in traffic accidents, especially on highways, this required the preparation of a study to know the causes of accidents and to explore the factors that have a relative impact on (the severity of the accident). Four highways in the capital, Baghdad, were chosen in this study, major and vital in terms of the number of drivers who use them daily, which are (Mohamed Al-Qasim Expressway, Army Canal Expressway, Salah Al-Din Street (Expressway), and Baghdad International Airport Street (Expressway)). Three hundred and forty-nine traffic accident forms were collected from the traffic directorates on both sides of Al-Karkh and Al-Rusafa for the years from 2006 to 2019. After the analysis by Binary Logistic Regression, the results showed that (contributing factors, road condition, cause of an accident like (parking on highway, loss of control, lack of attention, sudden stopping and lack of attention), vehicle body type, speed). Resulting from the BLR model.

Despite the measures put in place in different countries, road traffic fatalities are still considered one of the leading causes of death worldwide. Thus, the reduction of traffic fatalities or accidents is one of the contributing factors to attaining sustainability goals. Different factors such as the geometric structure of the road, a non-signalized road network, the mechanical failure of vehicles, inexperienced drivers, a lack of communication skills, distraction and the visual or cognitive impairment of road users have led to this increase in traffic accidents. These factors can be categorized under four headings that are: human, road, vehicle factors and environmental road conditions. The advent of machine learning algorithms is of great importance in analysing the data, extracting hidden patterns, predicting the severity level of accidents and summarizing the information in a useful format. In this study, three machine learning algorithms for classification, such as Decision Tree, LightGBM and XGBoost, were used to model the accuracy of road traffic accidents in the UK for the year 2020 using their default and hyper-tuning parameters. The results show that the high performance of the Decision Tree algorithm with default parameters can predict traffic accident severity and provide reference to the critical variables that need to be monitored to reduce accidents on the roads. This study suggests that preventative strategies such as regular vehicle technical inspection, traffic policy strengthening and the redesign of vehicle protective equipment be implemented to reduce the severity of road accidents caused by vehicle characteristics. [ABSTRACT FROM AUTHOR]

This paper focuses on the aspect of road safety based on the impact of pavement conditions on accident severities. Four models of binomial and multinomial logistic regression were produced using the R software and utilizing two years of accident and pavement conditions data on Malaysian highways. The surface characteristics analyzed included the International Roughness Index (IRI), rut depth (RD) and mean texture depth (MTD). The accident severity assessed ranged from damage to death. Results indicated that IRI has the highest tendency to affect accident severity for all models. At the thresholds identified for all independent variables, the chances of death had increased significantly. As such, efforts must be driven to ensure the thresholds are not reached in the maintenance of pavements for better road safety. [ABSTRACT FROM AUTHOR]

To understand the contributory factors to rear-end accident severity on mountainous expressways, a total of 1039 rear-end accidents, occurring on G5 Jingkun Expressway from Hechizhai to Qipanguan in Shaanxi, China over the period of 2012 to 2017, were collected, and a non-parametric Classification and Regression Tree (CART) model was used to explore the relationship between severity outcomes and driver factors, vehicle characteristics, roadway geometry and environmental conditions. Then the random forest model was introduced to examine the accuracy of variable selection and rank their importance. The results show that driver’s risky driving behaviours, vehicle type, radius of curve, angle of deflection, type of vertical curve, time, season, and weather are significantly associated with rear-end accident severity. Speeding and driving while drunk and fatigued are more prone to result in severe consequences for such accidents and driving while fatigued is found to have the highest fatality probability, especially during the night period (18:00–24:00). The involvement of heavy trucks increases the injury probability significantly, but decreases the fatality probability. In addition, adverse weather and sharp curve with radius less than 1000mare the most risk combination of factors. These findings can help agencies more effectively establish stricter regulations, adopt technical measures and strengthen safety education to ensure driver’s driving safety on mountainous expressways for today and tomorrow.

BackgroundRoad traffic safety has considerably improved in China. However, the changes may differ in the economically backward and altitude higher western region. This study aims to investigate changes in the occurrence and severity of traffic casualties in western China and illuminate several key causal factors.Materials and methodsTraffic accident data from the Annual Traffic Accident Statistics Report combined with population and vehicle data from the China Statistics Bureau between 2009 and 2019, were retrospectively analyzed. Traffic accident numbers, fatalities, human injury (HI), case fatality rates (CFR), mortality per 100,000 population (MRP), and mortality per 10,000 vehicles (MRV) were compared between the western and eastern regions. The HI, CFR, MRV, and MRP between the four groups based on the altitude of cities, below 500 meters, 500 to 1,500 meters, 1,500 to 3,000 meters, and over 3,000 meters, were compared using one-way analysis of variance. One hundred and seventy-eight cases of extremely serious traffic accidents were further analyzed in terms of accident occurrence time, vehicle type, road grade, road shape, accident pattern, and accident reason. The differences of accident characteristics between the eastern and western regions were compared using the chi-square test.ResultsThe number of traffic accidents and fatalities decreased in low-altitude areas in western China. However, there was a significant increasing trend in the high altitude area. The HI, CFR, MRV, and MRP were higher in the western region than that in the eastern and national. Those accident indicators tended to increase with increasing altitude. And there were statistically significant differences (p < 0.05) among groups from different altitudes. Chi-square test results show that there are statistically significant differences (p < 0.05) in term of road grade, road shape, accident pattern between eastern and western. Low-grade roads, combined curved and sloping roads, and rollover were significant features associated with traffic accidents in the western region. Bad roads were the main cause of rollover accidents in western China, which will lead to more serious casualties. Over speeding, overloading, bad weather, vehicle failure, and driver error were the top five accident causes.ConclusionTraffic accidents are increasing in high-altitude areas of western China, and they lead to more severe casualties. The characteristics of serious traffic accidents in this part of the country differ from those of the eastern regions. Improving road safety facilities, restrictions of speed, and improving medical treatment at accident scenes may be effective measures to reduce traffic accidents related injuries in the western region.

[Display omitted] • Analyzed 75 biogas plant accidents (occurrences) globally from 1990 to 2023. • Identified occurrences' common causes and results and suggested preventive measures. • Correlation screened factors related to occurrences, their likelihood, and severity. • Causes are component fail > maintenance > NaTech > equipment > operation > No PPE. • Explosions are the most common accident type, forming 69.3% of all occurrences. Biogas plants numbers are increasing worldwide, but their safety record is rarely investigated. This paper analyzes 75 occurrences of various types of accidents in biogas plants worldwide between 1990 and 2023. The study comprehensively reviewed accident reports and research literature with input from plant operators and safety experts. We aim to identify the common causes and consequences of accidents (occurrences) and suggest preventive measures to improve safety. The occurrences' primary causes were component failure > maintenance error > natural and technological disasters (NaTech) > equipment failure > operational error > no personal protective equipment (PPE). The most common occurrences were gas explosions 69.3%, toxic gas releases (biohazard) 21.3%, asphyxia (biohazard) 4%, malfunctioning (electric and mechanical hazard) 2.7%, and fires 2.7%. The accident consequences ranged from minor injuries (76) to fatalities (51) and extensive property damage. Lack of PPE and gas pipelines (mechanical and biohazards) correlated positively and significantly (R2 = 0.70), while operational errors and asphyxia (biohazard) scenarios correlated positively and moderately (R2 = 0.55). The plant design, operating procedures, and maintenance practices strongly influence the occurrences' likelihood and severity. This study provides valuable insights for stakeholders, researchers, and policymakers interested in promoting biogas' safe and sustainable development. Future studies should investigate the relationship between plant size and accident frequency and assess the effectiveness of safety management and risk assessment methodologies in mitigating such occurrences. [ABSTRACT FROM AUTHOR]

The traffic characteristics of Sichuan and Guizhou differ from those of other regions due to its unique geographical features. In addition, accident studies in China mainly focus on urban roads in the eastern and central regions. However, studies on western regions, especially non-urban roads, are scarce. Thus, this study aims to explore the factors that influence the severity of accidents on non-urban roads in typical regions of Sichuan and Guizhou. A total of 541 cases from 2014 to 2020 were selected from the database of the China In-Depth Accident Study, where 18 variables, which may exert an impact on accident severity, were extracted after screening. First, heterogeneity of data was eliminated through latent class analysis (LCA). The ordered probit (OP) model was then conducted for each class to obtain significant variables that exert an impact on accident severity. The study quantified the degree of influence of the significant variables using marginal effect analysis. The LCA results demonstrate that data were categorized into the following classes, namely, (a) two-vehicle accidents involving trucks, (b) pedestrian and multiple-vehicle accidents, (c) two-wheeler accidents, and (d) single-vehicle accidents. The OP results show that most variables could exert impact on accident severity, and some of them exerted varying levels of influence on the severity of different classes, whereas others only influence a specific class. According to this study, we obtained the accident characteristics of these regions and put forward some targeted suggestions to further improve the level of road traffic safety. The findings can provide support for the construction of transportation in line with the regional characteristics in China. [ABSTRACT FROM AUTHOR]

Purpose – The operation safety of the high-speed railway has been widely concerned. Due to the joint influence of the environment, equipment, personnel and other factors, accidents are inevitable in the operation process. However, few studies focused on identifying contributing factors affecting the severity of high-speed railway accidents because of the difficulty in obtaining field data. This study aims to investigate the impact factors affecting the severity of the general high-speed railway. Design/methodology/approach – A total of 14 potential factors were examined from 475 data. The severity level is categorized into four levels by delay time and the number of subsequent trains that are affected by the accident. The partial proportional odds model was constructed to relax the constraint of the parallel line assumption. Findings – The results show that 10 factors are found to significantly affect accident severity. Moreover, the factors including automation train protection (ATP) system fault, platform screen door and train door fault, traction converter fault and railway clearance intrusion by objects have an effect on reducing the severity level. On the contrary, the accidents caused by objects hanging on the catenary, pantograph fault, passenger misconducting or sudden illness, personnel intrusion of railway clearance, driving on heavy rain or snow and train collision against objects tend to be more severe. Originality/value – The research results are very useful for mitigating the consequences of high-speed rail accidents.

The paper deals with the task of modeling discrete questionnaire data with a reduced dimension of the model. The discrete model dimension is reduced using the construction of local models based on independent binomial mixtures estimated with the help of recursive Bayesian algorithms in the combination with the naive Bayes technique. The main contribution of the paper is a three-phase algorithm of the discrete model dimension reduction, which allows to model highdimensional questionnaire data with high number of explanatory variables and their possible realizations. The proposed general solution is applied to the traffic accident questionnaire analysis, where it takes the form of the classification of the accident circumstances and prediction of the traffic accident severity using the currently measured discrete data. Results of testing the obtained model on real data and comparison with theoretical counterparts are demonstrated. [ABSTRACT FROM AUTHOR]

This paper aims at assessing and identifying a fatal injury risk model with an exponential structure for the accidents that occur in rainy conditions, based on highly disaggregated level of data. A probabilistic approach is used in order to assess the fatal risk as a function of probability and accidents' consequences, based on 14 types of predictors that relate to road environment, time of accidents and lighting conditions. The influence of circadian rhythms on the drivers' behavior is taken into consideration by setting up the accidents' time intervals based on the "black time". The Adjusted Bayesian Information Criterion (BIC') is used to further test the fit of probability model in a comprehensive way and the risk distributions are described using their descriptive parameters. The highways are most dangerous during rains, since the odds for a fatal accident are 3.340 times higher than on streets. Also, the odds for those accidents that occurred between 02:00 ∼ 04:59 are 1.624 times higher than between 16:00 ∼ 20:59. This paper demonstrates that the quality of the roadways is important since there are 2.574 times higher odds for an accident to be fatal where the roadway has potholes compared with a normal pavement. [ABSTRACT FROM AUTHOR]

Traffic accidents are emerging as a serious social problem in modern society but if the severity of an accident is quickly grasped, countermeasures can be organized efficiently. To solve this problem, the method proposed in this paper derives the MDG (Mean Decrease Gini) coefficient between variables to assess the severity of traffic accidents. Single models are designed to use coefficient, independent variables to determine and predict accident severity. The generated single models are fused using a weighted-voting-based bagging method ensemble to consider various characteristics and avoid overfitting. The variables used for predicting accidents are classified as dependent or independent and the variables that affect the severity of traffic accidents are predicted using the characteristics of causal relationships. Independent variables are classified as categorical and numerical variables. For this reason, a problem arises when the variation among dependent variables is imbalanced. Therefore, a harmonic average is applied to the weights to maintain the variables' balance and determine the average rate of change. Through this, it is possible to establish objective criteria for determining the severity of traffic accidents, thereby improving reliability. [ABSTRACT FROM AUTHOR]

As tugboats interact very closely with ships in restricted waters, the possibility of accidents increases in these operations. Despite the high accident possibility, there is a gap in studies on tugboat accidents. This study aims to analyse accidents involving tugboats using data mining. For this purpose, a tugboat accidents dataset consisting of a total of 496 accident records for the period from 2008 to 2019 was collected. Logistic regression and decision tree algorithms were implemented to the dataset. The results revealed that tugboat propulsion type is the most important and influential factor in the severity of tugboat accidents. The inferences drawn from these results could be beneficial for tugboat operators and port authorities in enhancing their awareness of the factors affecting tugboat accidents. In addition, the outputs of this study can be a reference for management units in developing strategies for preventing tugboat accidents and can also be used in effective planning for practicable prevention programmes and practices. [ABSTRACT FROM AUTHOR]

Background : In China, despite the decrease in average road traffic fatalities per capita, the fatality rate and injury rate have been increasing until 2015. Purpose : This study aims to analyze the road traffic accident severity in China from a macro viewpoint and various aspects and illuminate several key causal factors. From these analyses, we propose possible countermeasures to reduce accident severity. Method : The severity of traffic accidents is measured by human damage (HD) and case fatality rate (CFR). Different categorizations of national road traffic census data are analyzed to evaluate the severity of different types of accidents and further to demonstrate the key factors that contribute to the increase in accident severity. Regional data from selected major municipalities and provinces are also compared with national traffic census data to verify data consistency. Results : From 2000 to 2016, the overall CFR and HD of road accidents in China have increased by 19.0% and 63.7%, respectively. In 2016, CFR of freight vehicles is 33.5% higher than average; late-night accidents are more fatal than those that occur at other periods. The speeding issue is severely becoming worse. In 2000, its CFR is only 5.3% higher than average, while in 2016, the number is 42.0%. Conclusion and practical implementation : A growing trend of accident severity was found to be contrasting to the decline of road traffic accidents. From the analysis of casual factors, it was confirmed that the release way of the impact energy and the protection worn by the victims are key variables contributing to the severity of road traffic accidents. • We adopted two indicators, Human Damage and Case Fatality Rate, to describe the severity of accidents. • The two indicators reveal the increasing traffic accident severity in China from a new perspective. • Our analysis on accident severity provides insights and countermeasures to reduce fatalities in road traffic accidents. [ABSTRACT FROM AUTHOR]

Traffic accidents are emerging as a serious social problem in modern society but if the severity of an accident is quickly grasped, countermeasures can be organized efficiently. To solve this problem, the method proposed in this paper derives the MDG (Mean Decrease Gini) coefficient between variables to assess the severity of traffic accidents. Single models are designed to use coefficient, independent variables to determine and predict accident severity. The generated single models are fused using a weighted-voting-based bagging method ensemble to consider various characteristics and avoid overfitting. The variables used for predicting accidents are classified as dependent or independent and the variables that affect the severity of traffic accidents are predicted using the characteristics of causal relationships. Independent variables are classified as categorical and numerical variables. For this reason, a problem arises when the variation among dependent variables is imbalanced. Therefore, a harmonic average is applied to the weights to maintain the variables’ balance and determine the average rate of change. Through this, it is possible to establish objective criteria for determining the severity of traffic accidents, thereby improving reliability.

Safety of any particular Road way facility cannot be attributed to set of parameters specific to a certain domain. Unsignalized intersections are no exceptions, thus, making them an important area of study. This paper presents the results of the analysis of four parameters, namely road width, traffic control, lane marking and landuse; and their sub-class on the safety of unsignalized intersections. The raw accident data was obtained from MIROS (Malaysian Institute of Road Safety Research). It was then reduced for descriptive analysis. Hypothesis testing was performed to assess the significance of all parameters and severity analysis was done to accomplish micro scale examination of each sub-class. The results show that landuse and lane marking are statistically significant. They are important variables to predict accidents whereas traffic control and road width are not significant. Intersections located in city with single line lane marking having no control and major road width greater than 9 meters were found to have the highest severity indices.

This paper focuses on the probability of crashes with severe and mild injuries in motorcyclists. The probability of crashes took human, road and environment, and vehicle factors into consideration. From July to December, 2015, 70.93% of the crashes that occurred in Indonesia involved motorcycles. The research took place in Bekasi City, Indonesia. The samples consisted of 184 respondents who had experienced crashes. The results indicated that the probability of severe injuries from the crashes was 13% and the probability of mild injuries was 87%. The mean absolute deviation of the model was 20.20%. Female drivers were more likely to be severely injured than males. Driving on roads which have road side variability and driving on curvy roads would be able to decrease the level of monotonous driving from 41% to 21%. Motorcycles which have engine capacity above 125 cm3 were 14% more likely to experience crashes with severely injuries. [ABSTRACT FROM AUTHOR]