Spatial analysis of traffic accidents near and between road intersections in a directed linear network

Road Accident Hotspots on Jordan’s Highway Based on Geometric Designs Using Structural Equation Modeling

One of the primary objectives of transportation engineering is to increase the safety of road infrastructure. This study seeks to determine the relationship between geometric design parameters in relation to road accident criteria based on accident hotspots on Jordan’s Desert Highway. The road accident data (from 2016 to 2019) were collected from the Jordan Traffic Department. The spatial pattern of hotspots was identified using a GIS tool named Getis-Ord Gi* based on the severity index of road accidents. A topographic survey was conducted to investigate the road alignment and intersections at hotspot locations. The study utilized the Structural Equation Modeling (SEM) technique via SmartPLS to highlight the correlation between geometric designs in relation to road accidents. The hotspot analysis (Gits-Ord Gi) discovered 80 road accident hotspots along the highway. The study found that horizontal alignment and road intersections significantly impact road accidents in hotspot locations. Furthermore, vertical alignment has no effect on road accidents in hotspot areas. The study enhanced the comprehension of the factors associated with road geometrics and intersections that affect the occurrence of road accidents.

Black-Spot Analysis in Hungary Based on Kernel Density Estimation

Between 2010 and 2020 in the European Union, 30% of road accidents resulted in the death of a pedestrian or a cyclist. Accidents of unprotected pedestrians and cyclists are the reason why it is essential to introduce road safety measures. In our paper, we identify and rank black spots using an innovative reactive approach based on statistics. We elaborate on the mathematical methodological considerations through the processing of real-life empirical data in a Matlab environment. The applied black-spot analysis is based on a Kernel density estimate method, and the importance of the kernel functions and bandwidth are elaborated. Besides, special attention is devoted to the distorting effect of annual average daily traffic. The result of our research is a new methodology by which the real locations of the examined black spots can be determined. Furthermore, the boundaries of the critical sections and the extent of the formation of black spots can be determined by the introduced mathematical methods. With our innovative model, the black spots can be ranked, and the locations having the highest potential for improvement can be identified. Accordingly, optimal measures can be determined considering social-economic and sustainability aspects.

Taking a mixed-methods approach to collision investigation: AcciMap, STAMP-CAST and PCM

Recently, ergonomics and safety researchers have turned their attention towards applying combinations of sociotechnical methods rather than using single methods in isolation. In the current research, a mixed-method approach combining two systems-based methods, Accimaps and the Systems Theoretic Accident Model and Process - Causal Analysis using Systems Theory (STAMP-CAST), and one cognitive approach, the Perceptual Cycle Model (PCM), were employed in analysing a rail-level crossing incident in Bangladesh. Each method was applied individually to investigate the collision, and interventions were proposed corresponding to incident events at different risk management framework levels. The three methods provided different perspectives of the whole picture, together identifying an array of contributory factors. The complementary nature of these methods aided in proposing a comprehensive set of safety recommendations, thereby demonstrating the benefit of a mixed-method approach for collision investigation in low-income settings.

Crash proximity and equivalent property damage calculation techniques: An investigation using a novel horizontal curve dataset

Despite the numerous breakthroughs in crash analytics, there remains a lack of consensus among safety practitioners as to the optimal method for locating high crash locations. Two critical components in the traffic safety analysis process not agreed upon are 1) how the crash distance to a target location is included in the analysis and 2) how crashes are weighted based on crash-related characteristics. For example, the commonly used buffering technique to determine which crashes are associated with a specific target road segment does not associate crashes that are closer to a target road segment with any additional weight, even though it is likely to be more greatly associated with the characteristics of the target location. Additionally, the commonly used equivalent property damage only (EPDO) crash weight method has been found to weigh fatal crashes significantly more than serious injury crashes, even if the difference between the two outcomes was a single factor. This study proposes more robust crash weighting techniques for use in high-risk location identification using an application of a novel horizontal curve dataset. Specifically, a heteroscedastic censored regression approach was used to investigate the impact of different crash proximity weighting techniques and crash severity weighting methods on model outcomes. The results demonstrate that the use of a linear distance weighting factor used in conjunction with the buffering technique as well as a less precise EPDO weighting factor method results in more robust safety analysis outcomes. The improved results have the potential to improve hot spot identification and resource allocation at both the federal and regional levels by employing models that more accurately link specific crash segments with contributing crash characteristics.

Spatial statistics and random forest approaches for traffic crash hot spot identification and prediction

Crash hot spot identification and prediction using spatial statistics and random forest methods on the interstate of Michigan are evaluated. The Getis-Ord statistics are adopted to identify hot spots using location, frequency, and equivalent property damage only weights computed from the cost and severity of crashes. In the random forest approach, data patterns between 2010 and 2017 are determined to predict hot spots of crashes in 2018. Accordingly, the results indicate that: (i) interstate routes have witnessed 13,089 crashes on significant hot spots, 7,413 on cold spots, and the rest in other locations; (ii) random forest shows 76.7% and 74% accuracy for validation and prediction, respectively. The performance of the model is further affirmed with precision, recall, and F-scores of 75%, 74%, and 70%, respectively; and (iii) clustering of the crashes exhibits spatial dependence of high and low equivalent property damage only crashes. The practical significance of the approach is highlighted in the identification and prediction of hot spots.

Unraveling Urban Form and Collision Risk: The Spatial Distribution of Traffic Accidents in Zanjan, Iran

Official statistics demonstrate the role of traffic accidents in the increasing number of fatalities, especially in emerging countries. In recent decades, the rate of deaths and injuries caused by traffic accidents in Iran, a rapidly growing economy in the Middle East, has risen significantly with respect to that of neighboring countries. The present study illustrates an exploratory spatial analysis' framework aimed at identifying and ranking hazardous locations for traffic accidents in Zanjan, one of the most populous and dense cities in Iran. This framework quantifies the spatiotemporal association among collisions, by comparing the results of different approaches (including Kernel Density Estimation (KDE), Natural Breaks Classification (NBC), and Knox test). Based on descriptive statistics, five distance classes (2-26, 27-57, 58-105, 106-192, and 193-364 meters) were tested when predicting location of the nearest collision within the same temporal unit. The empirical results of our work demonstrate that the largest roads and intersections in Zanjan had a significantly higher frequency of traffic accidents than the other locations. A comparative analysis of distance bandwidths indicates that the first (2-26 m) class concentrated the most intense level of spatiotemporal association among traffic accidents. Prevention (or reduction) of traffic accidents may benefit from automatic identification and classification of the most risky locations in urban areas. Thanks to the larger availability of open-access datasets reporting the location and characteristics of car accidents in both advanced countries and emerging economies, our study demonstrates the potential of an integrated analysis of the level of spatiotemporal association in traffic collisions over metropolitan regions.

Spatial Econometric Analysis of Road Traffic Crashes

Keeping the basic principles of sustainable development, it must be highlighted that decisions about transport safety projects must be made following expert preparation, using reliable, professional methods. A prerequisite for the cost–benefit analysis of investments is to constantly monitor the efficiency of accident forecasting models and to update these continuously. This paper presents an accident forecasting model for urban areas, which handles both the properties of the public road infrastructure and spatial dependency relations. As the aim was to model the urban environment, we focused on the road public transportation modes (bus and trolley) and the vulnerable road users (bicyclist) using shared infrastructure elements. The road accident data from 2016 to 2018 on the whole road network of Budapest, Hungary, is analyzed, focusing on road links (i.e., road segments between junctions) by applying spatial econometric statistical models. As a result of this article, we have developed a model that can be used by decision-makers as well, which is suitable for estimating the expected value of accidents, and thus for the development of the optimal sequence of appropriate road safety interventions.

Modeling the service-route-based crash frequency by a spatiotemporal-random-effect zero-inflated negative binomial model: An empirical analysis for bus-involved crashes

Previous studies related to bus crash frequencies modeling are limited and the statistical models are usually developed at the road segment or zonal level. This study focuses on modeling crash frequencies specifically at the bus-service-route level, which is useful and important to policymakers and bus operation companies toward the improvement of the safety level of bus networks, especially for developing countries where buses are still a major mode of urban travels. Using the observed data adopted from one of the bus operating companies in Beijing, China, we proposed a spatiotemporal-random-effect zero-inflated negative binomial (spatiotemporal ZINB) model to investigate bus crash occurrence and identity key influential factors at the bus-service-route level. The model was motivated to accommodate the special statistical characteristics of the excessive zeros and, more importantly, the potential spatiotemporal correlations of the data. Three degenerated versions of this model were also developed for comparison purposes. Results indicate that the proposed spatiotemporal ZINB model is statistically superior to the others according to a comprehensive judgment based on the EAIC, EBIC, and RMSE criteria. The estimated coefficients reveal the impacts of related factors on the likelihood of bus-involved crashes from bus operation factors including total passengers, number of drivers, and proportion of male drivers as well as planning factors including route length and stop density. On the other hand, the standard deviations of the introduced structured and unstructured spatiotemporal random-effects are statistically significant indicating that the observations are correlated within each route, between neighbor routes and across years. Corresponding policy and practical implications are provided for bus operating companies and planning departments toward the improvement of bus safety.

Development of a Binary Classification Model to Assess Safety in Transportation Systems Using GMDH-Type Neural Network Algorithm

Evaluating road safety is an enduring research topic in Infrastructure and Transportation Engineering. The prediction of crash risk is very important for avoiding other crashes and safeguarding road users. According to this task, awareness of the number of vehicles involved in an accident contributes greatly to safety analysis, hence, it is necessary to predict it. In this study, the main aim is to develop a binary model for predicting the number of vehicles involved in an accident using Neural Networks and the Group Method of Data Handling (GMDH). For this purpose, 775 accident cases were accurately recorded and evaluated from the urban and rural areas of Cosenza in southern Italy and some notable parameters were considered as input data including Daylight, Weekday, Type of accident, Location, Speed limit and Average speed; and the number of vehicles involved in an accident was considered as output. In this study, 581 cases were selected randomly from the dataset to train and the rest were used to test the developed binary model. A confusion matrix and a Receiver Operating Characteristic curve were used to investigate the performance of the proposed model. According to the obtained results, the accuracy values of the prediction model were 83.5% and 85.7% for testing and training, respectively. Finally, it can be concluded that the developed binary model can be applied as a reliable tool for predicting the number of vehicles involved in an accident.