Crash Events Research Articles

Crash Contributing Factors and Patterns Associated with Fatal Truck-Involved Crashes in Bangladesh: Findings from the Text Mining Approach

Despite extensive research on traffic injury severities, relatively little is known about the factors contributing to truck-involved crashes in developing countries, especially in the context of Bangladesh. Because of the unavailability of authentic crash data sources, this study collected data from alternative sources such as online English news media reports. The current study prepared a database of 144 truck-involved fatal crash reports during the period of 12 months (January 2021 to December 2021). The crash reports contain a bag of 15,300 words. Several state-of-the-art text mining tools were utilized to identify crash patterns, including word cloud analysis, word frequency analysis, word co-occurrence network analysis, rapid automatic keyword extraction, and topic modeling. The analysis revealed several important crash contributing factors, such as the type of vehicle involved (auto-rickshaw, bus, van, motorcycle), the manner of collision (head-on), the time of the day (morning, night), driver behavior (speeding, overtaking, wrong-way driving), and environmental factors (dense fog). In addition, “coming from opposite direction” and “head-on collision” are two important sequences of events in truck-involved crashes. Truck drivers are also involved in crashes with trains at rail crossings. The findings of this research can assist policymakers in identifying crash avoidance strategies to lower truck-related crashes in Bangladesh.

Point-process modeling of secondary crashes

Secondary crashes or crashes that occur in the wake of a preceding or primary crash are among the most critical incidents occurring on highways, due to the exceptional danger they present to the first responders and victims of the primary crash. In this work, we developed a self-exciting temporal point process to analyze crash events data and classify it into primary and secondary crashes. Our model uses a self-exciting function to describe secondary crashes while primary crashes are modeled using a background rate function. We fit the model to crash incidents data from the Florida Department of Transportation, on Interstate-4 (I-4) highway for the years 2015–2017, to determine the model parameters. These are used to estimate the probability that a given crash is secondary crash and to find queue times. To represent the periodically varying traffic levels and crash incidents, we model the background rate, as a stationary function, a sinusoidal non-stationary function, and a piecewise non-stationary function. We show that the sinusoidal non-stationary background rate fits the traffic data better and replicates the daily and weekly peaks in crash events due to traffic rush hours. Secondary crashes are found to account for up to 15.09% of traffic incidents, depending on the city on the I-4 Highway.

Crashworthiness of Foam-Filled and Reinforced Honeycomb Crash Absorbers in Transverse Direction

AbstractHoneycomb crash absorbers have been widely studied as energy absorption devices for use in automotive industries. However, none of these investigations have studied the side impact of empty and foam-filled honeycomb absorbers and adding stiffeners between the different layers of the corrugated sheets which are composing the honeycomb structure to analyse the structure under transverse (L-direction) impacts. In this paper, the foam-filled and reinforced honeycomb crash absorbers are investigated under axial (T) and transverse (L) loading directions. Experimental results for both empty and foam-filled specimens under quasi-static and impact loads were implemented to validate the developed finite element model. Finite element analysis (FEA) was performed to find out the crashworthiness behaviour of the structure under axial and transverse impacts according to road conditions. Finally, a new design of stiffened honeycomb crash absorber was developed and investigated to reduce the level of acceleration experienced by the passengers during the crash event. In this regard, it is concluded that all the requirements related to the energy absorption capabilities and generated deceleration under impact loading can be met by introducing an advanced method to reinforce honeycomb absorbers using stiffeners. It is also proven that the thickness of these stiffeners will not significantly influence the force levels. Due to increase of wall thickness from 1 to 3 mm, the mean crushing force increased from 129 kN to 148 kN. This growth is not sufficient as the goal is to obtain a mean crushing force of 300 kN. Thickening the stiffeners would lead to a loss of efficiency of the structure, as the small increase in mean force would not make up for the gain in mass. Thus, increasing the corrugated sheet’ thickness becomes necessary.

The effect of geomagnetic storms on aircraft accidents between the years 1919–2023 in civil aviation

Detailed research on the effect of geomagnetic storms (GSs) on air crash investigations, which is a very important space weather event and could be the main reason for most aircraft accidents, especially unexplained accidents, has not been done by this time. In this study, we investigated the effect of GSs depending on seasonal changes, solar activity, and geomagnetic activity (GA) over 115 years of aircraft accidents from the decreasing phase of the Solar Cycle (SC) 15 to the increasing phase of SC 25. In our analysis, the accident rates of 1959 air crash events correspond to 68 %, 22 %, 8 %, and 2 % due to mechanical and pilotage reasons, unknown reasons, bad weather conditions, and turbulence that might be related to GSs are obtained for latitudes between ± 10⁰ ≤ φ ≤ ±90⁰ in the northern and southern hemispheres. The MATLAB R2018a Programming Language is used to determine and draw all data in our statistical analyses. Nevertheless, we reconstructed the GA indices such as Dst, Kp, Ap, aa, and am for 1919–2023 by using the Autoregressive Integrated Moving Average (ARIMA) Forecasting Model for years without observations and therefore missing data. The first important result is that the number of aircraft accidents increased in the decreasing phases of solar activity cycles (SACs) depending on the effective structure of the coronal holes rather than increasing phases. The second important result is that the highest number of accidents occurred in the equinoxes in March and September (and mostly at the beginning of October) while the lower number of accidents was seen in the solstices such as June and December. The third important result is that the rate of accidents depends on how many storms were seen in the decreasing phase of SACs. The SC maximum intensity value that reaches the higher number, or even the duration of the SC will not affect aircraft accidents directly. The GA indices do not follow a regular pattern according to the SC shows that we should look at the geomagnetic structure of the intensity of three elements of a magnetic field of Earth as D (Declination), H (Horizontal), and z (Vertical), and therefore their intensities.

Review on Automated Accident Detection System

Abstract: Major accidents on highways, freeways and local roads can lead to huge social and economic impacts. Minor accidents may be resolved by the passengers themselves and do not require escorting to hospitals whereas major accidents where airbags are deployed require the immediate attention of authorities. Automated Accident Detection System [AADS] is an auto-detection unit system that immediately notifies an Emergency Contact through an alert alarm when someone presses the trigger button present on poles having CCTV cameras in the end vehicle is detected by the system, detailing the location and time of the accident. The idea is that as soon as an accident is detected by the system, the authorities should immediately be notified to prevent further car congestion as well as allow the passengers to be escorted to the hospital within time. A complete system would automatically detect and record traffic conditions associated with accidents such as the time of the accident, video of the accident, and the traffic light signal controller parameters. The basic research required to develop the system is considered. This involves developing methods for processing acoustic signals and recognizing accident events from the background traffic events. A database of vehicle crashes, car braking, construction, and traffic sounds was created. The meal frequency cepstral coefficients were computed as a feature vector for input to the classification system. A neural network was used to classify these features of the crash and non-crash events.

Imbalance-learning road crash assessment under reduced visibility settings: A proactive multicriteria decision-making system

Road crash prediction is a fundamental key in designing efficient intelligent transportation systems. There has been a pronounced progress in the use of machine learning models for crash events assessment by the transportation safety research community in recent years. However, little attention has been paid so far to evaluating reduced-visibility crash occurrences within a heuristic ensemble system. This study presents a proactive multicriteria decision-making system that can predict crash occurrences based on real-time roadway properties, land zones’ characteristics, vehicle telemetry, driver inputs and weather conditions collected using a desktop driving simulator. A key novelty of this work is implementing a genetic algorithm-based feature selection approach along with ensemble modeling strategies using AdaBoost, XGBoost and RF techniques to establish effective crash predictions. Furthermore, since crash events occur in rare instances tending to be underrepresented in the dataset, an imbalance-learning methodology to overcome the issue was adopted on the basis of several data resampling approaches to increase the predictive performance namely SMOTE, Borderline-SMOTE, SMOTE-Tomek Links and ADASYN strategies. To our knowledge, there has been a limited interest at adopting an ensemble-based imbalance-learning strategy examining the impact of real-time features’ combinations on the prediction of road crash events under reduced visibility settings.

Analisis Faktor-Faktor yang Berpengaruh terhadap Sikap Pengguna Aset Digital Non-Fungible Token Berbasis Blockchain pada Komunitas NFT Indonesia

Blockchain is an inspiring new technology that requires a lot of attention from various researchers and companies. This technology offers various benefits, where the newest derivative of this technology is Non-Fungible Tokens (NFT). In NFT, various digital assets such as images, videos, music or virtual creations are traded. The relatively new NFT creates a lack of user knowledge to be able to utilize this technology as a digital asset for trading. On the one hand, users welcome the use of NFTs as a tool to gain financial gain, but on the other hand, there are users who doubt the long-term success of NFTs as a form of investment and see them as a temporary trend that may be detrimental in the event of a market crash. The purpose of this research is to explore various factors that can impact the attitudes of NFT users. The research used a quantitative method which was carried out at the Indonesian NFT Community, involving 377 respondents as a sample. Questionnaires were distributed online using the Google form on the Telegram platform. The data analysis used in this research is by using Structural Equation Modeling (SEM). The results of the study show that there are five factors that significantly influence the attitude of NFT users. Perceived ease of use, perceived usefulness, perceived trust, and personal innovativeness have a positive and significant effect on user attitudes, while perceived risk has a negative effect and is one of the factors that has the most impact on user attitudes than other factors, followed by perceived trust in the second position as the most dominant variable. So like any new technology, NFTs have risks and trust issues that most users consider.

Geopolitical risk and stock price crash risk: The mitigating role of ESG performance

We study the effect of geopolitical risk (GPR) on stock price crash risk and we investigate the mediating role of the ESG factors in this relationship. Using a large international sample of publicly listed firms, we find that higher GPR causes stock price crashes to occur more frequently. This result holds to several robustness checks and to the use of different measures of stock price crash risk and we rule out any potential endogeneity concern using an Instrumental Variables (IV) approach. We also find that the causal effect of GPR on crash events is mainly driven by the expectations and threats of geopolitical tensions (geopolitical threats), rather than their effective realization and escalation (geopolitical acts). However, when exploring the potential mitigating role of ESG factors, we observe these negative implications to be less severe for high ESG-rated issuers and, specifically, for firms scoring high in the Environmental and Social dimensions. Our study demonstrates that firms more engaged in ESG practices are more resilient to the GPR's adverse effect on stock price crash risk.

Influence of pre-strain on tensile response of extra deep drawing (EDD) steel under varying strain rates and crash performance

This study explores the impact of pre-strain on the uniaxial tensile stress-strain behavior of extra deep drawing (EDD) steel at varying strain rates. The EDD steel blanks were subjected to a uniaxial pre-strain of 15 % in either the transverse or rolling direction. Subsequently, uniaxial tensile tests were conducted at different strain rates, both orthogonal and parallel to the initial pre-straining direction. The study also investigates the alterations in yield stress (YS), ultimate tensile strength (UTS), uniform elongation (UEL), total elongation (TEL), and strain energy absorption (SED, represented by the area under the tensile stress-strain curve) at diverse strain rates after pre-straining. The experimental findings indicate that pre-straining is advantageous for enhancing YS, UTS, and SED. However, this improvement comes at the expense of UEL or TEL. A square crush tube, which shares similarities with automotive components, was examined to assess the crash performance of several automotive structural parts during crash events. A validated constitutive model was used to predict the square tube's energy absorption and crush performance under high-velocity conditions for different pre-straining scenarios.

Regional brain strain dependance on direction of head rotation

Brain injuries in automated vehicles during crash events are likely to include mechanisms of head impact in non-standard positions and postures (i.e., occupants not facing forward in an upright position). Federal regulations currently focus on impact conditions in primary planes of motion, such as frontal or rear impacts (sagittal plane of motion) or side impact (coronal plane of motion) and do not account for out of position occupants or non-standard postures. The objective of the present study was to develop and use the anatomically accurate brain finite element model to parametrically determine the injury metrics under different vectors with head rotation. A custom developed brain finite element model with anatomical accuracy and several anatomical regions defined was used to evaluate whole-brain strain as well as regional brain strain. Cumulative Strain Damage Measure (CSDM) at a threshold of 20% strain and the 95th percentile of the maximum principal strain (MPS95) were calculated for the whole brain and each brain region under multiple rotational directions. The model was exposed to a sinusoidal angular acceleration pulse of 5000 rad per second squared (rad/s2–) over 12.5 ms. The same pulse was used in the primary axes of motion and (lateral bending, flexion, extension, axial rotation) and combined axes representing oblique flexion and oblique extension. Whole brain CSDM20 was highest for lateral bending. Whole brain MPS95 was highest for axial rotation. The rCSDM20 was more susceptible to impact direction, with several brain regions having substantial accumulation of strain for oblique flexion and lateral bending. Comparatively, rMPS95 was more consistent across all rotation directions. The present study quantified the regional brain strain response under multiple rotational vectors identifying a high amount of variability in the accumulation of strain (i.e., CSDM20) in the hypothalamus, hippocampus, and midbrain specifically. While there was a high amount of variability in the accumulation of strain for multiple regions, the maximum strain measured (i.e., MPS95) in the regions was more consistent.

Efficient Fusion Decision System for Predicting Road Crash Events: A Comparative Simulator Study for Imbalance Class Handling

Road crash events are a fact of life. Although significant progress have been made in adopting machine learning techniques for analyzing road crashes, there has been limited emphasis on evaluating crash events within data fusion systems. The primary purpose of this study is to outline and validate a comparative safety analysis of an ensemble fusion system founded on the use of different base classifiers and a meta-classifier to procure more efficient crash prediction. Three categories of features namely vehicle-telemetry, driver-inputs and environmental-conditions have been collected using a driving-simulator in order to identify the crash strongest precursors through feature extraction technique. Furthermore, optimized strategies using AdaBoost, XGBoost, RF, GBM, LightGBM, CatBoost and KNN techniques were implemented to establish effective predictions within a fusion-based system. To ensure that the proposed system provide superior decisions given the infrequent nature of crash events, an imbalance-learning approach was conducted based on three resampling strategies: over-sampling, under-sampling and SMOTE-Tomek-Links. The findings depict that the superior performance has been attained when adopting LightGBM, CatBoost and KNN as base classifiers along with SMOTE-TL as balancing technique and XGBoost as meta-classifier with 89.19% precision, 96.77% recall and 92.83% f1-score. To our knowledge, there has been a limited interest, if not at all, at endorsing a fusion-based system examining the impact of real-time features' combinations on the prediction of road crashes while providing a critical analysis of class-imbalance. Overall, the findings emphasized the relevance of the explanatory features and can be endorsed in designing efficient intelligent transportation systems.

Crash injury severity prediction considering data imbalance: A Wasserstein generative adversarial network with gradient penalty approach

For each road crash event, it is necessary to predict its injury severity. However, predicting crash injury severity with the imbalanced data frequently results in ineffective classifier. Due to the rarity of severe injuries in road traffic crashes, the crash data is extremely imbalanced among injury severity classes, making it challenging to the training of prediction models. To achieve interclass balance, it is possible to generate certain minority class samples using data augmentation techniques. Aiming to address the imbalance issue of crash injury severity data, this study applies a novel deep learning method, the Wasserstein generative adversarial network with gradient penalty (WGAN-GP), to investigate a massive amount of crash data, which can generate synthetic injury severity data linked to traffic crashes to rebalance the dataset. To evaluate the effectiveness of the WGAN-GP model, we systematically compare performances of various commonly-used sampling techniques (random under-sampling, random over-sampling, synthetic minority over-sampling technique and adaptive synthetic sampling) with respect to dataset balance and crash injury severity prediction. After rebalancing the dataset, this study categorizes the crash injury severity using logistic regression, multilayer perceptron, random forest, AdaBoost and XGBoost. The AUC, specificity and sensitivity are employed as evaluation indicators to compare the prediction performances. Results demonstrate that sampling techniques can considerably improve the prediction performance of minority classes in an imbalanced dataset, and the combination of XGBoost and WGAN-GP performs best with an AUC of 0.794 and a sensitivity of 0.698. Finally, the interpretability of the model is improved by the explainable machine learning technique SHAP (SHapley Additive exPlanation), allowing for a deeper understanding of the effects of each variable on crash injury severity. Findings of this study shed light on the prediction of crash injury severity with data imbalance using data-driven approaches.

Assessment of finite element human body and ATD models in estimating injury risk in far-side impacts using field-based injury risk

The objective of this study was to assess the ability of finite element human body models (FEHBMs) and Anthropometric Test Device (ATD) models to estimate occupant injury risk by comparing it with field-based injury risk in far-side impacts. The study used the Global Human Body Models Consortium midsize male (M50-OS+B) and small female (F05-OS+B) simplified occupant models with a modular detailed brain, and the ES-2Re and SID-IIs ATD models in the simulated far-side crashes. A design of experiments (DOE) with a total of 252 simulations was conducted by varying lateral ΔV (10-50kph; 5kph increments), the principal direction of force (PDOF 50°, 60°, 65°, 70°, 75°, 80°, 90°), and occupant models. Models were gravity-settled and belted into a simplified vehicle model (SVM) modified for far-side impact simulations. Acceleration pulses and vehicle intrusion profiles used for the DOE were generated by impacting a 2012 Camry vehicle model with a mobile deformable barrier model across the 7 PDOFs and 9 lateral ΔV’s in the DOE for a total of 63 additional simulations. Injury risks were estimated for the head, chest, lower extremity, pelvis (AIS 2+; AIS 3+), and abdomen (AIS 3+) using logistic regression models. Combined AIS 3+ injury risk for each occupant was calculated using AIS 3+ injury risk estimations for the head, chest, abdomen, and lower extremities. The injury risk calculated using computational models was compared with field-based injury risk derived from NASS-CDS by calculating their correlation coefficient. The field-based injury risk was calculated using risk curves that were created based on real-world crash data in a previous study (Hostetler et al., 2020). Occupant age (40 years), seatbelt use (belted occupant), collision deformation classification, lateral ΔV, and PDOF of the crash event were used in these curves to estimate field injury risk. Large differences in the kinematics were observed between HBM and ATD models. ATD models tended to overestimate risk in almost every case whereas HBMs yielded better risk estimates overall. Chest and lower extremity risks were the least correlated with field injury risk estimates. The overall risk of AIS 3+ injury risk was the strongest comparison to the field data-based risk curves. The HBMs were still not able to capture all the variance but future studies can be carried out that are focused on investigating their shortfalls and improving them to estimate injury risk closer to field injury risk in far-side crashes.

New Addressable Applications for Rheocasting to Escape from an Oversupplied Market

The current market change of aluminium HPDC castings started with the “Dieselgate”. First, there was a shift from diesel to petrol engines. In the second step, hybrid and battery-powered cars gained significant market shares in sales statistics. Therefore, lucrative powertrain components are falling away. As powertrain foundries still want to utilize their machines to capacity, they are pushing into the structural castings market. As a result, there is an oversupply of casting machines here, which massively depresses prices and, thus, margins for tenders. With rising energy costs, these declining margins were eaten up, bringing foundries into crisis. Implementing Rheocasting at the existing die-casting cells is the solution to move into new market shares that are now not accessible in conventional HPDC.One of the new applications is electronic housings, high wall thickness parts, and fatigue bearing parts. These parts are commonly manufactured in sand or gravity castings because of their high wall thickness and low tolerance of porosity. Rheocasting is the perfect process for high-wall thickness components. Because of the semisolid melt preparation and the lamellar filling behavior, these components can be manufactured from the same alloy without pores or voids.This flow behavior of the semisolid slurry also results in a longer flow length. Slower casting speeds and lower pressure settings result in lower clamping forces. This gives an advantage in production costs and targets battery constructions made of castings and sheet metal. Structural battery housings must be leak-tight, even in a crash event. Having it in one casting instead of an assembly reduces the leakage area and improves crash performance.Another industry that relies on Rheocasting is the telecommunications industry. The power electronics in these 5G modules are significantly larger and generate much more waste heat. Until now, many antennas have been actively cooled or milled from a block of aluminum. The milled housings are significantly too expensive to enable series production. Therefore, the goal is to reproduce the passively cooled modules in die casting. Due to the process, the thermal conductivity in conventional HPDC is around 120 to 130 W/m*K. Similarly, no slim cooling fins can be formed. Only the Rheocasting process makes it possible to cast other alloys with low proportions of alloying elements, such as AlSi2Mn. This allows fins with a wall thickness of down to 0.4 mm and thermal conductivity of up to 190 W/m*K.Rheocasting enables access to market segments out of reach. These bring an unbeatable cost advantage against the current suppliers: gravity and sand casters. The low cycle time in Rheocasting brings back the high margins needed to sustain the business. Also, these products can be delivered with even better properties on smaller casting machines.

Integrating biomechanics with stakeholder perspectives to inform safety in grassroots dirt track racing

Grassroots dirt track racing is a foundational part of motorsports with a high risk of severe injury. This study aimed to gather perspectives and experiences of motorsports drivers surrounding safety and head acceleration events experienced during grassroots dirt track racing to inform strategies to improve driver safety. Thirteen drivers (n=9 who primarily race on dirt tracks; n=4 who primarily race on pavement tracks) with prior dirt track racing experience participated in separate, group-specific focus groups and/or one-on-one interviews where video, simulations of head motion, and head acceleration data were shared. Peak kinematics of laps and crash contact scenarios were recorded, and head perturbations (i.e., deviations in head motion relative to its moving-average trajectory) were quantified for each lap and presented through guided discussion. Responses were summarized using Rapid Assessment Process. Audio recordings and field notes were collected from focus groups and interviews and analyzed across 25 domains. Drivers described dirt track racing as short, fast bursts of racing. Benefits of dirt track racing for driver development were described, including learning car control. Drivers acknowledged risks of racing and expressed confidence in safety equipment but identified areas for improvement. Drivers observed lateral bouncing of the head in video and simulations but recognized that such motions were not noticed while racing. Track conditions and track type were identified as factors influencing head perturbations. Mean PLA (5.5 g) and PRV (3.07 rad/s) of perturbations experienced during racing laps and perturbation frequencies of 5 and 7 perturbations per second were reported. Generally, drivers accurately estimated the head acceleration magnitudes but were surprised by the frequency and maximum magnitude of perturbations. Maximum perturbation magnitudes (26.8 g and 19.0 rad/s) were attributed to hitting a “rut” in the dirt. Drivers described sudden stops, vertical loads due to landing from a large height, and impacts to the vehicle frame as crash events they physically feel the most. Summary statistics for crashes (medians = 7.30 g, 6.94 rad/s) were reported. Typical impact magnitudes measured in other sports (e.g., football) were provided for context. Upon reviewing the biomechanics, drivers were surprised that crash accelerations were relatively low compared to other contact/collision sports. Pavement drivers noted limited safety features in dirt track racing compared to pavement, including rigidity of vehicle frames, seat structure, seatbelt integration, and lack of oversight from sanctioning bodies. Most drivers felt seat inserts and head and neck restraints are important for injury prevention; however, usage of seat inserts and preferred head and neck restraint system differed among drivers. Drivers described their perspectives and experiences related to safety and identified strategies to improve safety in grassroots dirt track racing. Drivers expressed support for future safety research.

Real-Time Big Data Analytics and Proactive Traffic Safety Management Visualization System

Big data and data-driven analysis could be utilized for traffic management to improve road safety and the performance of transportation systems. This paper introduces a web-based proactive traffic safety management (PATM) and real-time big data visualization tool, which is based on an award-winning system that won the US Department of Transportation (USDOT) Solving for Safety Visualization Challenge and was selected as one of the USDOT Safety Data Initiative (SDI) Beta Tools. State-of-the-art research, especially for real-time crash prediction and PATM, are deployed in this study. A significant amount of real-time data is accessed by the system in order to conduct data-driven analysis, such as traffic data, weather data, and video data from closed-circuit television (CCTV) live streams. Based on the data, multiple modules have been developed, including real-time crash/secondary crash prediction, CCTV-based expedited detection, PATM recommendation, data sharing, and report generation. Both real-time data and the system outputs are visualized at the front end using interactive maps and various types of figures to represent the data distribution and efficiently reveal hidden patterns. Evaluation of the real-time crash prediction outputs is conducted based on one-month real-world crash data and the prediction results from the system. The comparison results indicate excellent prediction performance. When considering spatial-temporal tolerance, the sensitivity and false alarm rate of the prediction results [i.e., high crash potential event (HCPE)] are 0.802 and 0.252, respectively. Current and potential implementation are also discussed in this paper.

VARVO: a Novel Method for the Fast Detection of Vehicle Crash Events from Video Only Data

Around 1,35 million people worldwide die each year because of traffic incidents, and it is estimated that another 50 million suffers serious injuries. This picture is particularly dramatic in the Andean Region where the death toll due to traffic accidents is as high as 127 deaths per million inhabitants. In recent years the deployment of the so-called Intelligent Transport Systems (ITS) across several developed countries has helped to reduce the number of deaths due to traffic accidents. An integral part of an ITS is the automatic detection of traffic incidents from video and sensor data. However, the scarcity of curated datasets, especially those that contained a reasonable number of positive instances of traffic incidents is hampering the development of artificial intelligence applications for the domain of traffic research. Given this scenario, we pursued answering the following research question: is it possible to detect car crashes through supervised machine learning based on the estimated speeds of cars from video only-data? Here we present VARVO, a novel algorithm for the detection of traffic incidents that does not rely on sensors for cars speed detection. VARVO performs a supervised classification task based on the sequential use of convolutional network-based object detection and bi-directional tracking. We also describe how the models implemented in VARVO improved their classification accuracy by applying an oversampling algorithm to deal with class imbalance. We believe that the deployment of VARVO could be linked to static traffic video cameras and could be part of the Intelligent Transport Systems foundations in other Andean countries.

Driver head kinematics in grassroots dirt track racing crashes: A pilot analysis

Motorsport athletes experience head acceleration loading during crashes; however, there is limited literature quantifying the frequency and magnitude of these loads, particularly at the grassroots level of the sport. Understanding head motion experienced during crash events in motorsport is necessary to inform interventions to improve driver safety. This study aimed to quantify and characterize driver head and vehicle kinematics during crashes in open-wheel grassroots dirt track racing. Seven drivers (ages 16–22, n = 2 female) competing in a national midget car series were enrolled in this study over two racing seasons and were instrumented with custom mouthpiece sensors. Drivers’ vehicles were outfitted with an incident data recorder (IDR) to measure vehicle acceleration. Forty-one crash events were verified and segmented into 139 individual contact scenarios via film review. Peak resultant linear acceleration (PLA) of the vehicle and PLA, peak rotational acceleration (PRA), and peak rotational velocity (PRV) of the head were quantified and compared across the part of the vehicle contacted (i.e., tires or chassis), the vehicle location contacted (e.g., front, left, bottom), the external object contacted (i.e., another vehicle, wall, or the track), and the principal direction of force (PDOF). The median (95th percentile) PLA, PRA, and PRV of the head and PLA of the vehicle were 12.3 (37.3) g, 626 (1799) rad/s2, 8.92 (18.6) rad/s, and 23.2 (88.1) g, respectively. Contacts with a non-horizontal PDOF (n = 98, 71%) and contact with the track (n = 96, 70%) were common in the data set. Contact to the left side of the vehicle, with the track, and with a non-horizontal PDOF tended to have the greatest head kinematics compared to other factors in each sub-analysis. Results from this pilot study can inform larger studies of head acceleration exposure during crashes in the grassroots motorsports environment and may ultimately support evidence-based driver safety interventions.

Volatility Spillovers and Contagion During Major Crises: An Early Warning Approach Based on a Deep Learning Model

This paper contributes to the ongoing debate on the nature and characteristics of the volatility transmission channels of major crash events in international stock markets between 03 July 1997 and 09 March 2021. Using dynamic conditional correlations (DCC) for conditional correlations and volatility clustering, GARCH-BEKK for the direction of transmission of disturbances, and the Diebold-Yilmaz spillover index for the level of volatility contagion, the paper finds that the climbs in external shock transmissions have long-lasting impacts in domestic markets due to the contagion effect during crisis periods. The findings also reveal that the heavier magnitude of financial stress is transmitted between Asian countries via the Hong Kong stock market. Additionally, the degree of volatility spillovers between advanced and emerging equity markets is smaller compared to the pure spillovers between advanced markets or emerging markets, offering a window of opportunity for international market participants in terms of portfolio diversification and risk management applications. Furthermore, the study introduces a novel early warning system created by integrating DCC correlations with a state-of-the-art deep learning model to predict the global financial crisis and COVID-19 crisis. The experimental analysis of long short-term memory network finds evidence of contagion risk by verifying bursts in volatility spillovers and generating signals with high accuracy before the 12-month crisis period. This provides supplementary information that contributes to the decision-making process of practitioners, as well as offering indicative evidence that facilitates the assessment of market vulnerability for policymakers.

Balancing Digital Forensic Investigation with Cybersecurity for Heavy Vehicle Traffic Crashes

AbstractAfter a traffic crash event, traffic crash investigators collect evidence and data to assist in reconstructing the events to determine crash causation. Some of the data collected in a crash investigation is in the form of digital data from event data recorders built into the electronic control units in the vehicles. Occasionally, traffic crashes are severe enough to destroy the typical network‐based communications protocols to extract the digital forensic data. In these cases, more invasive techniques of gathering forensic data through in‐circuit programming ports or direct reading of data bearing memory chips is needed. While a digital forensic investigation satisfies a virtuous need for society in determining the truth of a traffic crash, the same techniques can be applied by nefarious actors interested in stealing intellectual property (IP) from the same data bearing chips. The exposure of the executable binary containing the IP of the manufacturer has prompted auto makers and suppliers to eliminate access to these sources of digital forensic data by disabling the Joint Task Action Group (JTAG) instrumentation and obfuscating or encrypting the binary data. Herein lies the purpose of this paper, which is to take a systems engineering approach to balance the needs and requirements for a manufacturer to provide sufficient forensic artifacts in the case of an investigation while improving their cybersecurity posture and limiting their exposure to the theft of intellectual property or cyberat‐tack.An activity diagram is presented to show a system model for responding to and investigating a crash event. These activities inform the needs of an improved event data recorder technologies that contain information necessary to reconstruct the crash. Some proposed top level system requirements are presented with a discussion of how they satisfy the needs of the manufacturer and the crash investigator. Specific requirements of recorded data give a notion of a minimal set of recorded data to help investigators. These requirements will improve both the availability and adequacy of forensic data needed for crash event reconstruction. In addition, a separate requirement governing the preservation of Original Equipment Manufacturer (OEM) proprietary software is made, such that their intellectual property is protected to encourage the requirement compliance. Finally, a discussion of how the proposed requirements help determine if a crash event was a result of a cyber‐attack demonstrates the important nature of addressing these needs in future systems.