Run-off-road Research Articles

Real-Time Run-Off-Road Risk Prediction Based on Deep Learning Sequence Forecasting Approach

Driving risk prediction is crucial for advanced driving technologies, with deep learning approaches leading the way in driving safety analysis. Current driving risk prediction methods typically establish a mapping between driving features and risk statuses. However, status prediction fails to provide detailed risk sequence information, and existing driving safety analyses seldom focus on run-off-road (ROR) risk. This study extracted 660 near-roadside lane-changing samples from the high-D natural driving dataset. The performance of sequence and status prediction for ROR risk was compared across five mainstream deep learning models: LSTM, CNN, LSTM-CNN, CNN-LSTM-MA, and Transformer. The results indicate the following: (1) The deep learning approach effectively predicts ROR risk. The Macro F1 Score of sequence prediction significantly surpasses that of status prediction, with no notable difference in efficiency; (2) Sequence prediction captures risk evolution trends, such as increases, turns, and declines, providing more comprehensive safety information; (3) The presence of surrounding vehicles significantly impacts lane change duration and ROR risk. This study offers new insights into the quantitative research of ROR risk, demonstrating that risk sequence prediction is superior to status prediction in multiple aspects and can provide theoretical support for the development of roadside safety.

Investigating factors affecting injury severity of single-vehicle run-off-road crashes

This study aimed to identify and investigate the contributing factors influencing injury severity in single-vehicle run-off-road (ROR) crashes, which are known for their high severity. The primary objective was to analyze and compare the impact of these factors across three distinct vehicle classes: passenger cars, sport utility vehicles (SUVs), and pickups. A mixed logit model with heterogeneity in mean and variance was developed to analyze the injury severity outcomes in ROR crashes for the three vehicle classes. The model accounted for the potential variations in the impact of contributing factors across different vehicle types. The study revealed several significant variables consistently influencing injury severity across all three vehicle classes. These included driver age, alcohol or drug usage, seatbelt utilization, airbag deployment, higher travel speeds, and the vehicle model year post-2010. Notably, as driver age increased, the impact on changes in injury severity outcomes was more pronounced for pickup drivers compared to those operating passenger cars and SUVs. Among the common findings was the highly effective role of seatbelt usage in mitigating injury severity in ROR crashes. Additionally, passenger cars were associated with increased injury severity, particularly at relatively higher travel speeds exceeding 75 mph when contrasted with SUVs and pickups traveling between 61 and 75 mph. The study highlights the importance of considering vehicle class-specific factors in analyzing injury severity in ROR crashes. Recommendations include further in-depth investigations into distinct factors contributing to injury severity within each vehicle class and utilizing more extensive crash datasets to gain additional insights for enhancing road safety.

Exploring interacting effects of risk factors on run-off-road crash severity: An interpretable machine learning model joint with latent class clustering

Run-Off-Road (ROR) crashes are frequent and pose a significant risk of injury and fatality. Given the complexity of their mechanisms and the interaction of multiple factors, this study aims to comprehensively investigate the factors influencing the severity of ROR crashes, which have been understudied. Furthermore, addressing the current methodological challenge in machine learning (ML) crash modeling, this study proposes an approach to tackle unobserved heterogeneity in ML. An interpretable ML joint with prior latent class clustering is implemented. The significant risk factors and interactions are interpreted using SHAP (SHapley Additive exPlanations) method across clusters. This study utilizes ROR crash records, traffic, and geometric data from main suburban freeways in Iran collected over a 5-year period. The key interacting factors associated with severe ROR during adverse weather (cluster 1) are: co-occurrence of low congestion and higher speed variation; low congestion, nighttime darkness and rollover; roadway departure by buses and mini-buses and rollover occurrence; vehicle departure and collision with fixed objects. Moreover, the critical interactions for nighttime condition (cluster 2) are: curve sections combined with longitudinal slope; inside shoulder width <1.5 m and a hit median concrete NewJersey barrier. The risky interactions for crashes occurred in curves (cluster 3) are: departing in two-lane sections in low congestion conditions; vehicles collisions with the median concrete NewJersey barriers. The findings of this study enhance comprehension of the significant effects of interactions under various conditions, offering valuable insights for policymakers. Additionally, recommendations are offered to mitigate the risk of severe ROR crashes.

Estimating impact of pavement surface condition and geometrics design on two-wheeler run-off road crashes on horizontal curves

Introduction-Motorized Two-Wheeler (MTW) is a convenient and affordable mode of transportation despite being highly vulnerable to rollover crashes. The possible fact may be that MTW is a single-track vehicle and thus lack longitudinal stability, particularly on horizontal curves (HC). This stability depends on external factors such as curve geometrics and contact surface properties. Therefore, this research aims to identify the impact of pavement surface conditions and roadway geometrics on MTW run-off road crashes on HC. Method-To consider the temporal effect of pavement surface conditions, this study analyzed different combinations of light and surface moisture conditions for pre- and post-pavement surfacing conditions. Thus, Generalized Estimating Equation (GEE) linked with the log function was used to account for repeated observations. Sample data were collected bi-annually from 76 HC for two years of pre- and post-resurfacing. Pavement surface condition parameter includes skid resistance, rut depth, and international roughness index (IRI). However, to restrain the impact of pavement surface conditions, a separate analysis was carried out to investigate the impact of geometrical design attributes. Results-The analysis concluded that surface moisture condition, light condition, rut depth, and their interactions with AADT were found significant for pre-resurfacing. For post-resurfacing, surface moisture condition, and light condition, IRI and its interaction with AADT were found significant. Similarly, roadway geometrics such as curve radius and their interaction with the presence of sight distance, public activity centre and cross-slope were found significant. Conclusion-Overall, the study insights and confirms how and to what extent pavement surface properties and geometric attributes impact the subject crashes. The results will enhance the understanding of roadway architecture and may help highway designers to implement, either at the intervention or treatment level, to improve safety and comfort.

Application of Bayesian Semi-Parametric and Hierarchical Models for Analyzing Dispersed Traffic Barriers Crash Data

Introduction: Despite the traffic barriers effectiveness in reduction of the severity of run-off road crashes, the severity of barrier crashes still accounts for a significant fraction of road fatalities. Although extensive research has already been conducted in studying traffic barrier crashes, those studies mostly either consider the severity or frequency of crashes. Here, the equivalent property damage only (EPDO) was used to account for both aspects of crashes. While modeling EPDO crashes, there are challenges associated with that type of dataset including its sparse distribution, and the presence of heterogeneity in the dataset due to aggregation of various crash types. Methods: Ignoring the sparse nature of the data might result in biased or even erroneous results. Thus, in this study we identify factors to barriers EPDO crashes while considering the discussed challenges. Those consideration are especially important as in the next step we will employ the modeling results for conducting the cost-benefit analysis. Two main methods were considered in this study to address the discussed challenges including parametric and non-parametric Bayesian hierarchical models. A semiparametric Bayesian approach was used to relax the normality assumption by using a mixture of multivariate Dirichlet prior, defining a flexible nonparametric model for the random effects’ distribution, and using grouping to account for the heterogeneity due to the structure of the dataset. On the other hand, Bayesian hierarchical models with two distributions of Poisson and negative binomial with similar levels of hierarchy were considered. These models were chosen as closest models to the Bayesian semiparametric model. The incorporated models were compared in terms of deviance information criterion (DIC). Results and Discussion: The results highlighted that although the semi-parametric method outperforms the Bayesian hierarchical model with Poisson distribution, the Bayesian hierarchical model with negative binomial (NB) distribution outperform the semi-parametric model. The findings might be related to the severe sparse nature of the EPDO, which cannot optimally be accounted by semiparametric approach, and the model needs more flexibility. Conclusion: It was found that being unrestrained, driving in interstate system, driving in clear weather, light conditions, and driving in a higher traffic all increase the likelihood of EPDO crashes. Also, while some predictors were significant in less accommodative models of semi-parametric or Poisson models, they were not for Negative binomial model.

Single-Vehicle Run-Off Road Crashes Because of Cellphone Distraction: Finding Patterns with Rule Mining

A wide array of literature strongly indicates a higher likelihood of roadway departure because of cellphone use; however, patterns of associative attributes in single-vehicle run-off-road (SVROR) crashes because of cellphone distraction remained unexplored. Using the association rule mining (ARM) method, this study aimed to identify the variable categories that concurrently occur in such crashes, visualize the structures representing those concurrent associations, and discuss the crash patterns associated with different severity types. The SVROR crashes with cellphone use by the driver at fault were highlighted to be strongly connected to non-usage of safety restraints, weekends, both lighted and unlighted dark conditions, two-lane highways without physical separation, roadway curves, and so forth. Alongside several attributes, non-usage of restraints was strongly associated with fatal and both incapacitating and non-incapacitating injury crashes. The findings of this study can benefit the determination of suitable countermeasures to prevent cellphone-related SVROR crashes.

Approach for Quantifying Roadside Hazard Risk on Two-Lane Highways in Low- and Middle-Income Countries

With the adaptation of a “Safe system” approach and design for “forgiving roads” in low- and middle-income countries (LMICs), there is a need to develop new strategies to minimize run-off-road (ROR) crash severity of errant vehicles with potential roadside hazards specially on two-lane highways. While a common practice among highway designers or road safety auditors have always been to rate the roadside condition to prioritize hazardous road segment, the method has mostly been based on subjective judgement rather than systematic investigation. Moreover, the overall roadside environment of a LMIC such as India is different from a developed country, and needs more in-depth quantification. However, in most LMICs, details of roadside hazards are often not available from the crash records. To address this gap, an attempt has been made to develop a hierarchical roadside risk assessment index through a combined reactive–proactive approach for typical two-lane highways in LMICs. The proposed methodology was illustrated and validated on two-lane highways in India. The results obtained from the study indicated that the roadside hazard types in LMICs are almost similar to the developed countries, however their associated risk levels are different. Moreover, the predominant risk factors related to roadside hazard observed on two-lane highways in such countries like India is also very diverse. Thus, the insights obtained from the study can aid organizations and agencies responsible for road safety management, reach more flexible decisions in planning and designing safer roads, especially in context of LMICs.

COVID-19 and Injury Severity of Drivers Involved in Run-Off-Road Crashes: Analyzing the Impact of Contributing Factors

This study examined the relationship between the lockdown during the COVID-19 pandemic and the severity of injuries sustained by drivers involved in run-off-road (ROR) crashes. A random parameter ordered logit (RPOL) modeling framework was utilized to account for the ordinal nature of severity outcome and capture the potential unobserved heterogeneity. The data used in this study contained ROR crashes that occurred in the state of Florida from April to September for 2019 and 2020 representing non-pandemic and pandemic time periods, respectively. Separate driver injury severity models were developed across the two time periods, and the overall stability of the model estimates was examined through likelihood ratio tests. The impacts of various potential contributing factors, including crash-, driver-, and vehicle-related variables, roadway geometric characteristics, environmental conditions, and traffic-specific factors, were assessed. Although the developed models share some common features, the analysis results showed that the model specifications indicated a strong temporal instability among the estimated parameters. Compared to the non-pandemic period, the following variables resulted in increased driver injury severity in ROR crashes during the pandemic: drivers 65 years or older, careless driving, and absence of traffic control devices.

Corridor-Based Procedure for Determining Longitudinal Barrier Length of Need

Determining the optimal length of need (LON) to prevent vehicles from striking roadside hazards is an important part of roadside design. Although barriers can provide significant safety benefits by preventing run-off-road (ROR) crashes, the barrier itself presents a small but nonzero risk of serious injury or fatality. The purpose of this study is to present a method for quantifying the protection of barrier LON using departure corridors developed with NCHRP 17-43 data. Based on real-world ROR crash trajectories from the NCHRP 17-43 database, departure corridors were constructed. This study focused on a subset of 171 crashes on roads with a speed limit between 60 and 80 km/h with a reconstructed impact speed. In these crashes, the vehicle trajectory was interrupted by either a collision or a rollover. Our approach was to extrapolate the trajectory that could have been followed by the vehicle had the collision not occurred. The original trajectory data were retained before impact and extrapolated after impact using a linear fit. These corridors incorporated the effects of hazard location, departure side, and barrier location on the runout length. Left-side departures required a longer LON to intercept the same proportion of crashes. The runout length needed for objects close to the road based on the corridor method was much smaller than the length recommended by the 2011 Roadside Design Guide. The benefit of this corridor-based approach to barrier placement is that each agency can balance barrier length and the proportion of encroachments intercepted depending on their priorities.

Finite element simulation and failure analysis of fixed bollard system according to the PAS 68:2013 standard

Fixed bollards are very significant elements in providing urban roadside safety. These structures, called passive systems, allow pedestrians and disabled people to use sidewalks effectively and safely, as well as they have a crucial role in protecting the above-ground assets of infrastructure facilities against run-off-road (ROR) crashes. In this study, structural and safety performance analyses of fixed bollards that protect natural gas above ground Reducing and Metering Station-B (RMS-B) type regulating stations used in many points in Istanbul were made. In this regard, finite element (FE) models of the steel pipe of the fixed bollard, the concrete of the foundation and the soil where the bollard was mounted were created. Existing fixed bollards have been verified with accident data. In the light of the calibration and validation, numerical models were created and combined with the vehicle models specified in PAS 68:2013 standard to simulate dynamic behaviours. The combined FE models were analysed applying LS-DYNA software. As a result of the numerical analysis, the fixed bollard system cannot safely stop the 30000 kg N3, 18000 kg N3 and 7500 kg N2 class vehicles, and cannot prevent them from damaging the RMS-B. Besides, the fixed bollard system cannot provide safety at the speed level of 48 km/h, considering the urban speed limit for 3500 kg N1 and 2500 kg N1G class vehicles. However, the fixed bollard system can safely stop 1500 kg and 900 kg M1 class vehicles up to 64 km/h speed limit. These analyses revealed that considering the regional traffic characteristics of Istanbul, the existing fixed bollard systems are not sufficient for the protection of hazardous roadside facilities and need to be improved. Full-scale crash tests should be carried out in future studies to validate the FE simulation models. Furthermore, various bollard designs should be developed, especially urban road requirements.

Prediction of Run-Off Road Crash Severity in South Korea’s Highway through Tree Augmented Naïve Bayes Learning

The run-off road crash (RORC) is a representative type of lethal crash. The severity of RORC has increased owing to a combination of factors, such as roadside geometry, traffic conditions, and weather/climatic conditions. In this study, a model for estimating the RORC severity was developed based on various factors, including fixed objects, roadway geometry, traffic conditions, and road traffic environment. To develop the model, the accident data of crashes with roadside fixed objects on highways, as well as information on fixed object-related variables and roadway geometry-related variables, were collected. To improve the model in terms of implementing a close reflection of the real world, a learning method with tree augmented naïve Bayes (TAN), which takes into account the causal links between variables, was applied. The results of the analysis showed that the severity of crashes with roadside fixed objects increased sharply when the vertical slope was ≥4%, the radius of the curve was ≥250 m, the distance between the fixed object and the roadway was less than 3 m, or the density of fixed objects installation was greater than 2 for every 10 m. The proposed model allows for an analysis of sections with a high RORC severity on the roadways in operation and provides improvement measures to reduce the severity of RORC.

Bayesian extreme value analysis of kinematic-based surrogate measure of safety to detect crash-prone conditions in connected vehicles environment: A driving simulator experiment

Connected Vehicles (CVs) technology has provided large-scale driving database embedded in Basic Safety Messages (BSMs). This valuable data source can shed more light on tracking individual driving behaviors to detect crashes. This study delves into the possibility of detecting crashes using BSMs in controlled high-fidelity driving simulator experiments.To this end, two driving simulator scenarios were designed to simulate Run-off-Road (ROR), and Rear-End (RE) crashes. Twenty-four professional truck drivers were recruited to drive the scenarios. In each scenario, crash and non-crash cases were identified from vehicles’ trajectories, resulting in four study cases. Drivers’ behaviors were quantified by characterizing two Kinematic-based Surrogate Measures of Safety (K-SMoS), namely Absolute value of Derivative of Instantaneous Acceleration (ADInstAccel) and Absolute value of Derivative of Steering (ADSteering). Extreme defensive driving volatilities under crash and non-crash cases were modeled by extreme value analysis of K-SMoS and fitting their associated Generalized Extreme Value (GEV) distributions under Bayesian inference. Accordingly, for each K-SMoS, the crash detection was formulated as a binary classification between two K-SMoS GEV continuous distributions under crash and corresponding non-crash conditions.Qualitative uncertainty analysis of joint posterior density distributions of GEVs’ parameters revealed a higher uncertainty of extreme driving behaviors in crash conditions. Regardless, notable relative increases in the central tendency of extreme K-SMoS in crash compared to non-crash conditions were found, implying the possibility of crash detection by tracking extreme drivers’ behaviors using trajectory-level observations. This visual inference was affirmed by the result of binary classification of GEV distributions associated with K-SMoS. Depending on the crash type and K-SMoS, 71% to 81% accuracy in crash detection was obtained, where ADSteering outperformed ADInstAccel in terms of the discriminative ability. Besides, using sensitivity–specificity analysis, the optimal threshold of 1.24 (rad/s) and 1.31 (m/s3), respectively, for ADSteering and ADInstAccel, were identified to detect crashes. These findings can potentially enhance CVs' automation level in spatiotemporally identifying crash-prone conditions to disseminate distress notifications. Furthermore, the introduced methodology can be a complementary one to what has been followed in the crash detection domain.

Reducing Vehicle Submersions and Consequent Fatalities on Highway I-75 in Florida (Alligator Alley): Effectiveness of a Cable Safety Barrier System

Objective:We evaluated the effectiveness of a Cable Safety Barrier (CSB) system in preventing Run-Off-Road (ROR) Vehicle Immersions (VIs) and fatalities in canals along the I-75 freeway (Alligator Alley) in Collier County, Florida. The CSB system, which runs along both sides of the 80-km stretch of freeway and was installed between 2003 and 2004.Methods:Data from the Fatal Analysis Reporting System (FARS) were used to compare annual VIs and VI fatalities between pre-installation of the CSB system (1995-2002) to post-installation (2005-2012). As well, post-installation data from the Florida Department of Transport (FDOT) (2007-2011) and police reports were reviewed to determine the number of, and manner in which, vehicles were either contained by, or crossed, the CSB by either penetrating or overriding the barriers.Results:Pre- to post-installation, total accidents increased from 81.4/y to 106.2/y, accidents resulting in VIs decreased from 13.8% to 2.4%, and accidents resulting in VI fatalities decreased from 3.4% to 0.4% (FDOT). Fatal vehicle immersions decreased from 2.4/y to 0.9/y (P&lt;0.01) and vehicle immersion fatalities decreased from 3.3/y to 1.4/y (P&lt;0.05) (FARS). Post-installation, 531 accidents occurred with 110 ROR vehicles travelling towards the canals; 91 vehicles contacted the CSB with only 14 vehicles (15.4%) penetrating the barrier, and 7 (7.7%) overriding the barrier (FDOT).Conclusion:The CSB system along I-75 in Collier County dramatically decreased ROR vehicles from reaching the parallel canals, and consequent vehicle immersion fatalities. Results support the installation of lateral CSB systems on other high-risk roadways to reduce ROR crashes into water, or with other secondary hazards.

Assessment of commercial truck driver injury severity based on truck configuration along a mountainous roadway using hierarchical Bayesian random intercept approach

For the last decade, disaggregate modeling approach has been frequently practiced to analyze truck-involved crash injury severity. This included truck-involved crashes based on single and multi-vehicles, rural and urban locations, time of day variations, roadway classification, lighting, and weather conditions. However, analyzing commercial truck driver injury severity based on truck configuration is still missing. This paper aims to fill this knowledge gap by undertaking an extensive assessment of truck driver injury severity in truck-involved crashes based on various truck configurations (i.e. single-unit truck with two or more axles, single-unit truck pulling a trailer, semi-trailer/tractor, and double trailer/tractor) using ten years (2007–2016) of Wyoming crash data through hierarchical Bayesian random intercept approach. The log-likelihood ratio tests were conducted to justify that separate models by various truck configurations are warranted. The results obtained from the individual models demonstrate considerable differences among the four truck configuration models. The age, gender, and residency of the truck driver, multi-vehicles involvement, license restriction, runoff road, work zones, presence of junctions, and median type were found to have significantly different impacts on the driver injury severity. These differences in both the combination and the magnitude of the impact of variables justified the importance of examining truck driver injury severity for different truck configuration types. With the incorporation of the random intercept in the modeling procedure, the analysis found a strong presence (24%–42%) of intra-crash correlation (effects of the common crash-specific unobserved factors) in driver injury severity within the same crash. Finally, based on the findings of this study, several potential countermeasures are suggested.

Quantifying and comparing the effects of key risk factors on various types of roadway segment crashes with LightGBM and SHAP

Understanding and quantifying the effects of risk factors on crash frequency is of great importance for developing cost-effective safety countermeasures. In this paper, the effects of key crash contributing factors on total crashes and crashes of different collision types are analyzed separately and compared. A novel Machine Learning (ML) method, Light Gradient Boosting Machine (LightGBM), is introduced to model a Texas dataset consisting of vehicle crashes occurred from 2015 to 2017. Compared with other commonly used ML methods such as eXtreme Gradient Boosting (XGBoost), LightGBM performs significantly better in terms of mean absolute error (MAE) and root mean squared error (RMSE). In addition, the SHapley Additive explanation (SHAP) approach is employed to interpret the LightGBM outputs. Significant risk factors are identified, including speed limits, area type, number of lanes, roadway functional class, shoulder width and shoulder type. With the SHAP method, the importance, total effects, and main and interaction effects of risk factors are quantified. The results suggest that the importance of risk factors vary across collision types. Speed limit is a more important risk factor than right/left shoulder width, lane width, and median width for Rear-End (RE) crashes, while the opposite relationship is found for Run-Off-Road (ROR) crashes. Also, it is found that narrow lanes (8ft to 11ft) increase the risk for all types of crashes (i.e., Total, ROR, and RE) in this study. For road segments with 5 or 6 lanes in both directions combined, a lane width greater than or equal to 12ft may help reduce the risk of all types of crashes. These results have important implications for developing accurate crash modification factors and cost-effective safety countermeasures.

Investigation of injury severities in single-vehicle crashes in North Carolina using mixed logit models

IntroductionRoadway departure (RwD) crashes, comprising run-off-road (ROR) and cross-median/centerline head-on collisions, are one of the most lethal crash types. According to the FHWA, between 2015 and 2017, an average of 52 percent of motor vehicle traffic fatalities occurred each year due to roadway departure crashes. An avoidance maneuver, inattention or fatigue, or traveling too fast with respect to weather or geometric road conditions are among the most common reasons a driver leaves the travel lane. Roadway and roadside geometric design features such as clear zones play a significant role in whether human error results in a crash. Method: In this paper, we used mixed-logit models to investigate the contributing factors on injury severity of single-vehicle ROR crashes. To that end, we obtained five years' (2010–2014) of crash data related to roadway departures (i.e., overturn and fixed-object crashes) from the Federal Highway Administration's Highway Safety Information System Database. Results: The results indicate that factors such as driver conditions (e.g., age), environmental conditions (e.g., weather conditions), roadway geometric design features (e.g., shoulder width), and vehicle conditions significantly contributed to the severity of ROR crashes. Conclusions: Our results provide valuable information for traffic design and management agencies to improve roadside design policies and implementing appropriately forgiving roadsides for errant vehicles. Practical applications: Our results show that increasing shoulder width and keeping fences at the road can reduce ROR crash severity significantly. Also, increasing road friction by innovative materials and raising awareness campaigns for careful driving at daylight can decrease the ROR crash severity.

Evaluation of Centerline Audible Roadway Delineators (CARDs) in New York State

This study evaluated the safety effectiveness of the systemic implementation of centerline rumble strips on New York State-owned roads. In New York State, rumble strips installed on the centerline are referred to as centerline audible roadway delineators (CARDs). The effect of CARDs on head-on sideswipe (HO/SS), run-off-road (ROR), injury crashes, and fatal crashes was analyzed. This was an observational, naïve before–after with comparison group study. The treatment group included all segments where CARDs were installed in 2013, 2014, and 2015 in New York State Department of Transportation Regions 1 and 4, and Dutchess County in Region 8. Eligible segments without CARDs in those regions constituted the control group. The crashes comprised 2009 to 2017 HO/SS and ROR fatal, all injury, and property damage only crashes. The study showed that CARDs are an effective treatment to reduce ROR and HO/SS crashes. There was a decrease of 24 ROR and 9 HO/SS crashes on CARDs segments installed in 2013, a decrease of 12 ROR and 27 HO/SS crashes on CARDs segments installed in 2014, and a decrease of 40 ROR and 35 HO/SS crashes on CARDs segments installed in 2015. Crash modification factors (CMFs) developed for CARDs included ROR crashes (a CMF of 0.774 [expected reduction of 22.6%] in all ROR crashes); HO/SS crashes (a CMF of 0.623 [expected reduction of 37.7%] in HO/SS total crashes); and HO/SS injury crashes (a CMF of 0.572 [expected reduction of 42.8%] in injury HO/SS crashes).

Variable categories influencing single-vehicle run-off-road crashes and their severity

Single-vehicle run-off-road (ROR) crashes account for a significant number of all the single-vehicle crashes and all the fatalities in the United States. The variable categories influencing single-vehicle ROR crashes and their severity could be different from other single-vehicle crashes. This paper, therefore, focuses on investigating single-vehicle ROR crashes to identify and compare variable categories influencing their occurrence and severity. A logistic regression model was developed to evaluate and identify variable categories associated with single-vehicle ROR crashes compared to other single-vehicle crashes. The results indicate that variables such as road characteristics (number of lanes, road surface condition, road terrain, functional classification, and geometric condition), weather condition, road lighting condition, vehicle type, and, driver characteristics (driver physical condition, age, and gender) are observed to have a statistically significant influence on the occurrence of single-vehicle ROR crashes. A partial proportionality odds model was developed to evaluate and identify variable categories associated with injury severity of single-vehicle ROR crashes. The results indicate that variables and variable categories such as weather condition (cloudy and rain), speed limit (<= 40.2 kmph [25 mph] and > 88.5 kmph [55 mph]), vehicle type (primarily buses), drivers’ age (<= 18 years), road surface condition (presence of water, ice, snow, and slush), and, mountainous terrain condition are observed to have a statistically significant influence on single-vehicle ROR crash severity.

Risk Factors Affecting Driver Severity of Single- Vehicle Run Off Road Crash for Thailand Highway

Single-Vehicle Run Off Road (ROR) crash has been the leading crash type in terms of frequency and severity in Thailand. In this study, multinomial logit analysis was applied to identify the risk factors potentially influencing driver injury severity of single-vehicle ROR crash using accident records between 2011 and 2017 which were extracted from Highway Accident Information Management System (HAIMS) database. The analysis results show that the age of driver older than 55 years old, male driver, driver under influence of alcohol, drowsiness, ROR to left/right on straight roadway increase the probability of fatal crash, while other factors are found to mitigate severity such as the age of driver between 26-35 years old, using seatbelt, ROR and hit fixed object on straight and curve segment of roadway, mounted traffic island, intersection-related and accident in April. This study recommends the need to improve road safety campaign, law enforcement, and roadside safety features that potentially reduce level of severity of driver involving in single-vehicle ROR crash.

Rule-based safety prediction models for rural two-lane run-off-road crashes

• Analyzed rural two-lane run-off-road crashes in Louisiana. • Applied both statistical and machine learning models to determine the best-fit model. • Cubist provides the highest accuracies with rules-based regression models. • Rules-based models are capable in addressing localized issues by improving prediction accuracies. During 2015–2017, the yearly average of fatalities caused by roadway departure (RwD) crashes was 19,233 in the U.S. Roadway departure crashes, or crashes in which a vehicle crosses an edge line, centerline, or otherwise leaves the traveled way, account for 52 percent of all traffic fatalities in the U.S. during this time. A majority of RwD crashes are run-off-road (ROR) crashes; these are crashes that result in a vehicle crossing the edge line on either side of the roadway. In this study, the research team analyzed seven years (2010–2016) of rural two-lane ROR crash data from Louisiana to better comprehend ROR crashes in refining safety predictions for rural two-lane highways. Statistical model (negative binomial model) and three separate machine learning models (random forest, support vector machine, and Cubist) were applied to determine the best fit models. Overall, Cubist is characterized by a better performance in estimating ROR crashes on rural two-lane roadways. The Cubist approach introduced rules-based safety performance functions (SPFs) for total and fatal and injury crashes. This approach will be beneficial for the safety practitioners in tackling localized issues in crash data with an emphasis on prediction accuracy.