Abstract:Objective: Road crashes are increasing every year in low- and middle-income countries compared to high-income countries, which show decreasing trends. One theory may be because of differences in enforcement of laws, vehicle safety, road standards, and many other factors. A detailed review was made of 5 death trends (total number of deaths/100,000 population and percentages of 4-wheeler, pedestrian, motorized 2/3-wheeler, bicyclist) and the underlying patterns in different countries and regions across the world. This review was done to understand the main reasons for the variances to focus on efficient improvement strategies related to future vehicle and road safety issues. Methods: A self-organizing map (SOM) technique is used to map the nonlinear relationships among different attributes. Overall, 176 countries with 44 attributes were considered from the World Health Organization's (WHO) Global Status Report on Road Traffic Crashes database. Of the 44 attributes, 5 related to accident deaths were considered as response attributes. Results: Very distinct and unique cause-effect patterns for 3 clusters were observed from SOM results. High-income countries were found to have a lower total number of deaths/100,000 population. One theory espouses that this was due to those countries maintaining high vehicle standards and policies, whereas it was quite a different situation for low-income countries. Even though helmet laws were available in Association of South East Asian Nations + 6 (ASEAN + 6) countries, the percentage of 2/3-wheeler deaths may be higher due a lack of enforcement of those laws. Percentage of deaths involving 4-wheeler vehicles was higher in certain countries in the Persian Gulf, the United States, Australia, and New Zealand. This may be due to the fact that these countries have a number of rural areas where drivers drive at highway speeds versus some lower income countries with more urban areas where drivers operate vehicles at slower speeds. Countries with a lack of laws protecting bicyclists saw higher death percentages among bicyclists. The percentage of bicyclist deaths was also higher in areas with no helmet requirement and no investment in infrastructure improvements. The percentage of pedestrian deaths was high when there was no policy to separate road users, especially in low-income African countries. Deaths can be reduced by enforcement of laws and practicing good safety standards related to road traffic. Conclusions: Future vehicle and road safety strategies should consider using advanced statistical tools like SOM to advance safety. Based on a triple-layer (vehicle, infrastructure, and society) safety approach, strict regulations and enforcement are effective measures to reduce fatalities in low- and middle-income countries. On the other hand, introduction of more advanced vehicle technologies will be useful in countries with high gross national incomes (GNIs). Hence, a proper balance of different countermeasures based on economic zones could be effective to reduce total world traffic casualties.

Hierarchical clustering analysis framework of mutually exclusive crash causation parameters for regional road safety strategies

Clustering-Based Roadway Segment Division for the Identification of High-Crash Locations

Factors propelling fatalities during road crashes: A detailed investigation and modelling of historical crash data with field studies

Quantifying the heterogeneity impact of risk factors on regional bicycle crash frequency: A hybrid approach of clustering and random parameter model

Examining Driver Injury Severity in Intersection-Related Crashes Using Cluster Analysis and Hierarchical Bayesian Models.

Exploring the severity of bicycle-vehicle crashes using latent class clustering approach in India

An insight of World Health Organization (WHO) accident database by cluster analysis with self-organizing map (SOM)

Analysis of factors affecting crash under risk scenarios based on driver homogenous clustering

Identification of Significant Factors in Fatal-Injury Highway Crashes Using Genetic Algorithm and Neural Network

Wrong-way driving crash injury analysis on arterial road networks using non-parametric data mining techniques

Identifying accident prone areas and factors influencing the severity of crashes using machine learning and spatial analyses

Pre-crash scenarios for safety testing of autonomous vehicles: A clustering method for in-depth crash data

Investigating risk factors associated with injury severity in highway crashes: A hybrid approach integrating two-step cluster analysis and latent class ordered regression model with covariates

A Cluster-Based Approach for Analysis of Injury Severity in Interstate Crashes Involving Large Trucks

Spatiotemporal Clustering and Analysis of Road Accident Hotspots by Exploiting GIS Technology and Kernel Density Estimation

Phase-wise injury integrated severity modeling of road accidents: a two-stage hybrid multi-criteria decision-making model

Discovering Spatio-Temporal Clusters of Road Collisions Using the Method of Fast Bayesian Model-Based Cluster Detection

Driver injury severity outcome analysis in rural interstate highway crashes: a two-level Bayesian logistic regression interpretation

Decoding the impacts of contributory factors and addressing social disparities in crash frequency analysis

Division and Analysis of Accident-Prone Areas near Highway Ramps Based on Spatial Autocorrelation

Modeling of Low Visibility-Related Rural Single-Vehicle Crashes Considering Unobserved Heterogeneity and Spatial Correlation