Mixed logit model-based driver injury severity investigations in single- and multi-vehicle crashes on rural two-lane highways

Crashes occurring on rural two-lane highways are more likely to result in severe driver incapacitating injuries and fatalities. In this study, mixed logit models are developed to analyze driver injury severities in single-vehicle (SV) and multi-vehicle (MV) crashes on rural two-lane highways in New Mexico from 2010 to 2011. A series of significant contributing factors in terms of driver behavior, weather conditions, environmental characteristics, roadway geometric features and traffic compositions, are identified and their impacts on injury severities are quantified for these two types of crashes, respectively. Elasticity analyses and transferability tests were conducted to better understand the models’ specification and generality. The research findings indicate that there are significant differences in causal attributes determining driver injury severities between SV and MV crashes. For example, more severe driver injuries and fatalities can be observed in MV crashes when motorcycles or trucks are involved. Dark lighting conditions and dusty weather conditions are found to significantly increase MV crash injury severities. However, SV crashes demonstrate different characteristics influencing driver injury severities. For example, the probability of having severe injury outcomes is higher when vans are identified in SV crashes. Drivers’ overtaking actions will significantly increase SV crash injury severities. Although some common attributes, such as alcohol impaired driving, are significant in both SV and MV crash severity models, their effects on different injury outcomes vary substantially. This study provides a better understanding of similarities and differences in significant contributing factors and their impacts on driver injury severities between SV and MV crashes on rural two-lane highways. It is also helpful to develop cost-effective solutions or appropriate injury prevention strategies for rural SV and MV crashes.     

Examining driver injury severity outcomes in rural non-interstate roadway crashes using a hierarchical ordered logit model

Rural non-interstate crashes induce a significant amount of severe injuries and fatalities. Examination of such injury patterns and the associated contributing factors is of practical importance. Taking into account the ordinal nature of injury severity levels and the hierarchical feature of crash data, this study employs a hierarchical ordered logit model to examine the significant factors in predicting driver injury severities in rural non-interstate crashes based on two-year New Mexico crash records. Bayesian inference is utilized in model estimation procedure and 95% Bayesian Credible Interval (BCI) is applied to testing variable significance. An ordinary ordered logit model omitting the between-crash variance effect is evaluated as well for model performance comparison. Results indicate that the model employed in this study outperforms ordinary ordered logit model in model fit and parameter estimation. Variables regarding crash features, environment conditions, and driver and vehicle characteristics are found to have significant influence on the predictions of driver injury severities in rural non-interstate crashes. Factors such as road segments far from intersection, wet road surface condition, collision with animals, heavy vehicle drivers, male drivers and driver seatbelt used tend to induce less severe driver injury outcomes than the factors such as multiple-vehicle crashes, severe vehicle damage in a crash, motorcyclists, females, senior drivers, driver with alcohol or drug impairment, and other major collision types. Research limitations regarding crash data and model assumptions are also discussed. Overall, this research provides reasonable results and insight in developing effective road safety measures for crash injury severity reduction and prevention.     

A spatial generalized ordered response model to examine highway crash injury severity

This paper proposes a flexible econometric structure for injury severity analysis at the level of individual crashes that recognizes the ordinal nature of injury severity categories, allows unobserved heterogeneity in the effects of contributing factors, as well as accommodates spatial dependencies in the injury severity levels experienced in crashes that occur close to one another in space. The modeling framework is applied to analyze the injury severity sustained in crashes occurring on highway road segments in Austin, Texas. The sample is drawn from the Texas Department of Transportation (TxDOT) crash incident files from 2009 and includes a variety of crash characteristics, highway design attributes, driver and vehicle characteristics, and environmental factors. The results from our analysis underscore the value of our proposed model for data fit purposes as well as to accurately estimate variable effects. The most important determinants of injury severity on highways, according to our results, are (1) whether any vehicle occupant is ejected, (2) whether collision type is head-on, (3) whether any vehicle involved in the crash overturned, (4) whether any vehicle occupant is unrestrained by a seat-belt, and (5) whether a commercial truck is involved.     

Investigating driver injury severity patterns in rollover crashes using support vector machine models

Rollover crash is one of the major types of traffic crashes that induce fatal injuries. It is important to investigate the factors that affect rollover crashes and their influence on driver injury severity outcomes. This study employs support vector machine (SVM) models to investigate driver injury severity patterns in rollover crashes based on two-year crash data gathered in New Mexico. The impacts of various explanatory variables are examined in terms of crash and environmental information, vehicle features, and driver demographics and behavior characteristics. A classification and regression tree (CART) model is utilized to identify significant variables and SVM models with polynomial and Gaussian radius basis function (RBF) kernels are used for model performance evaluation. It is shown that the SVM models produce reasonable prediction performance and the polynomial kernel outperforms the Gaussian RBF kernel. Variable impact analysis reveals that factors including comfortable driving environment conditions, driver alcohol or drug involvement, seatbelt use, number of travel lanes, driver demographic features, maximum vehicle damages in crashes, crash time, and crash location are significantly associated with driver incapacitating injuries and fatalities. These findings provide insights for better understanding rollover crash causes and the impacts of various explanatory factors on driver injury severity patterns.     

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

There is a high potential of severe injury outcomes in traffic crashes on rural interstate highways due to the significant amount of high speed traffic on these corridors. Hierarchical Bayesian models are capable of incorporating between-crash variance and within-crash correlations into traffic crash data analysis and are increasingly utilized in traffic crash severity analysis. This paper applies a hierarchical Bayesian logistic model to examine the significant factors at crash and vehicle/driver levels and their heterogeneous impacts on driver injury severity in rural interstate highway crashes. Analysis results indicate that the majority of the total variance is induced by the between-crash variance, showing the appropriateness of the utilized hierarchical modeling approach. Three crash-level variables and six vehicle/driver-level variables are found significant in predicting driver injury severities: road curve, maximum vehicle damage in a crash, number of vehicles in a crash, wet road surface, vehicle type, driver age, driver gender, driver seatbelt use and driver alcohol or drug involvement. Among these variables, road curve, functional and disabled vehicle damage in crash, single-vehicle crashes, female drivers, senior drivers, motorcycles and driver alcohol or drug involvement tend to increase the odds of drivers being incapably injured or killed in rural interstate crashes, while wet road surface, male drivers and driver seatbelt use are more likely to decrease the probability of severe driver injuries. The developed methodology and estimation results provide insightful understanding of the internal mechanism of rural interstate crashes and beneficial references for developing effective countermeasures for rural interstate crash prevention.     

Modeling occupant-level injury severity: An application to large-truck crashes

Most of the injury-severity analyses to date have focused primarily on modeling the most-severe injury of any crash, although a substantial fraction of crashes involve multiple vehicles and multiple persons. In this study, we present an extensive exploratory analysis that highlights that the highest injury severity is not necessarily the comprehensive indicator of the overall severity of any crash. Subsequently, we present a panel, hetroskedastic ordered-probit model to simultaneously analyze the injury severities of all persons involved in a crash. The models are estimated in the context of large-truck crashes. The results indicate strong effects of person-, driver-, vehicle-, and crash-characteristics on the injury severities of persons involved in large-truck crashes. For example, several driver behavior characteristics (such as use of illegal drugs, DUI, and inattention) were found to be statistically significant predictors of injury severity. The availability of airbags and the use of seat-belts are also found to be associated with less-severe injuries to car-drivers and car-passengers in the event of crashes with large trucks. Car drivers’ familiarity with the vehicle and the roadway are also important for both the car drivers and passengers. Finally, the models also indicate the strong presence of intra-vehicle correlations (effect of common vehicle-specific unobserved factors) among the injury propensities of all persons within a vehicle.     

An explanatory analysis of driver injury severity in rear-end crashes using a decision table/Naïve Bayes (DTNB) hybrid classifier

Rear-end crashes are a major type of traffic crashes in the U.S. Of practical necessity is a comprehensive examination of its mechanism that results in injuries and fatalities. Decision table (DT) and Naïve Bayes (NB) methods have both been used widely but separately for solving classification problems in multiple areas except for traffic safety research. Based on a two-year rear-end crash dataset, this paper applies a decision table/Naïve Bayes (DTNB) hybrid classifier to select the deterministic attributes and predict driver injury outcomes in rear-end crashes. The test results show that the hybrid classifier performs reasonably well, which was indicated by several performance evaluation measurements, such as accuracy, F-measure, ROC, and AUC. Fifteen significant attributes were found to be significant in predicting driver injury severities, including weather, lighting conditions, road geometry characteristics, driver behavior information, etc. The extracted decision rules demonstrate that heavy vehicle involvement, a comfortable traffic environment, inferior lighting conditions, two-lane rural roadways, vehicle disabled damage, and two-vehicle crashes would increase the likelihood of drivers sustaining fatal injuries. The research limitations on data size, data structure, and result presentation are also summarized. The applied methodology and estimation results provide insights for developing effective countermeasures to alleviate rear-end crash injury severities and improve traffic system safety performance.     

Severity of driver injury and vehicle damage in traffic crashes at intersections: a Bayesian hierarchical analysis

Most crash severity studies ignored severity correlations between driver-vehicle units involved in the same crashes. Models without accounting for these within-crash correlations will result in biased estimates in the factor effects. This study developed a Bayesian hierarchical binomial logistic model to identify the significant factors affecting the severity level of driver injury and vehicle damage in traffic crashes at signalized intersections. Crash data in Singapore were employed to calibrate the model. Model fitness assessment and comparison using intra-class correlation coefficient (ICC) and deviance information criterion (DIC) ensured the suitability of introducing the crash-level random effects. Crashes occurring in peak time and in good street-lighting condition as well as those involving pedestrian injuries tend to be less severe. But crashes that occur in night time, at T/Y type intersections, and on right-most lane, as well as those that occur in intersections where red light cameras are installed tend to be more severe. Moreover, heavy vehicles have a better resistance on severe crash and thus induce less severe injuries, while crashes involving two-wheel vehicles, young or aged drivers, and the involvement of offending party are more likely to result in severe injuries.     

Rollover risk of cars and light trucks after accounting for driver and environmental factors

Characteristics of the driver, roadway environment, and vehicle were associated with the likelihood of rollover occurrence in more than 14000 single-vehicle fatal and 78000 single-vehicle injury crashes during 1995-98. Rollovers were more likely in crashes involving young drivers or occurring on rural curves. After accounting for the effects of driver age and gender, roadway alignment and surface condition, and whether or not the crash occurred in a rural area, light trucks were still twice as likely as cars to experience rollovers. Some light truck models were much more likely than others to experience rollovers. However, while physical differences (e.g. center of gravity height) could explain some of this variability, other factors affecting vehicle stability may be evident only after dynamic testing.     

Relative risk of injury and death in ambulances and other emergency vehicles

This study addresses of the impacts of emergency vehicle (ambulances, police cars and fire trucks) occupant seating position, restraint use and vehicle response status on injuries and fatalities. Multi-way frequency and ordinal logistic regression analyses were performed on two large national databases, the National Highway Traffic Safety Administration’s Fatality Analysis Reporting System (FARS) and the General Estimates System (GES). One model estimated the relative risk ratios for different levels of injury severity to occupants traveling in ambulances. Restrained ambulance occupants involved in a crash were significantly less likely to be killed or seriously injured than unrestrained occupants. Ambulance rear occupants were significantly more likely to be killed than front-seat occupants. Ambulance occupants traveling non-emergency were more likely than occupants traveling emergency to be killed or severely injured. Unrestrained ambulance occupants, occupants riding in the patient compartment and especially unrestrained occupants riding in the patient compartment were at substantially increased risk of injury and death when involved in a crash. A second model incorporated police cars and fire trucks. In the combined ambulance–fire truck–police car model, the likelihood of an occupant fatality for those involved in a crash was higher for routine responses. Relative to police cars and fire trucks, ambulances experienced the highest percentage of fatal crashes where occupants are killed and the highest percentage of crashes where occupants are injured. Lack of restraint use and/or responding with ‘lights and siren’ characterized the vast majority of fatalities among fire truck occupants. A third model incorporated non-special use van and passenger car occupants, which otherwise replicated the second model. Our findings suggest that ambulance crewmembers riding in the back and firefighters in any seating position, should be restrained whenever feasible. Family members accompanying ambulance patients should ride in the front-seat of the ambulance.     

Factors influential in making an injury severity difference to older drivers involved in fixed object-passenger car crashes

To identify factors influencing severity of injury to older drivers in fixed object-passenger car crashes, two sets of sequential binary logistic regression models were developed. The dependent variable in one set of models was driver injury severity, whereas for the other it was the crash severity (most severe injury in the crash). For each set of models, crash or injury severity was varied from the least severity level (no injury) to the highest severity level (fatality) and vice versa. The source of data was police crash reports from the state of Florida. The model with the best fitting and highest predictive capability was used to identify the influence of roadway, environmental, vehicle, and driver related factors on severity. Travel speed, restraint device usage, point of impact, use of alcohol and drugs, personal condition, gender, whether the driver is at fault, urban/rural nature and grade/curve existence at the crash location were identified as the important factors for making an injury severity difference to older drivers involved in single vehicle crashes.     

Analysis of large truck crash severity using heteroskedastic ordered probit models

Long-combination vehicles (LCVs) have significant potential to increase economic productivity for shippers and carriers by decreasing the number of truck trips, thus reducing costs. However, size and weight regulations, triggered by safety concerns and, in some cases, infrastructure investment concerns, have prevented large-scale adoption of such vehicles. Information on actual crash performance is needed. To this end, this work uses standard and heteroskedastic ordered probit models, along with the United States’ Large Truck Crash Causation Study, General Estimates System, and Vehicle Inventory and Use Survey data sets, to study the impact of vehicle, occupant, driver, and environmental characteristics on injury outcomes for those involved in crashes with heavy-duty trucks. Results suggest that the likelihood of fatalities and severe injury is estimated to rise with the number of trailers, but fall with the truck length and gross vehicle weight rating (GVWR). While findings suggest that fatality likelihood for two-trailer LCVs is higher than that of single-trailer non-LCVs and other trucks, controlling for exposure risk suggest that total crash costs of LCVs are lower (per vehicle-mile traveled) than those of other trucks.     

Identifying repeat DUI crash factors using state crash records

The objective of this study is to identify high risk factors that are closely related to repeat DUI crashes using readily available information from the state crash records. Survival analysis was used and a Cox proportional hazards model was developed using the police-reported crash records in the state of Louisiana. A variety of variables were found to be significant in predicting repeat DUI crashes. The factors included the characteristics of the drivers (gender, race, and age), the types of the vehicle (light truck/pick up truck or other), the characteristics of the crash (hit-and-run, driver violations, and whether the driver is arrested), the type of location (residential area or other), and the characteristics of the roadway (highway type and roadway type). This study provides a comprehensive picture of the repeat DUI crashes. The model can quantitatively predict the relative hazards of repeat DUI crashes. It can be used to identify the characteristics of the crash-involved DUI drivers who are at greatest risk of being involved in a subsequent DUI crash, allowing to apply appropriate remedial measures to reduce the risk.     

Vehicle mass and injury risk in two-car crashes: A novel methodology

This paper introduces a novel methodology based on disaggregate analysis of two-car crash data to estimate the partial effects of mass, through the velocity change, on absolute driver injury risk in each of the vehicles involved in the crash when absolute injury risk is defined as the probability of injury when the vehicle is involved in a two-car crash. The novel aspect of the introduced methodology is in providing a solution to the issue of lack of data on the speed of vehicles prior to the crash, which is required to calculate the velocity change, as well as a solution to the issue of lack of information on non-injury two-car crashes in national accident data. These issues have often led to focussing on relative measures of injury risk that are not independent of risk in the colliding cars. Furthermore, the introduced methodology is used to investigate whether there is any effect of vehicle size above and beyond that of mass ratio, and whether there are any effects associated with the gender and age of the drivers. The methodology was used to analyse two-car crashes to investigate the partial effects of vehicle mass and size on absolute driver injury risk. The results confirmed that in a two-car collision, vehicle mass has a protective effect on its own driver injury risk and an aggressive effect on the driver injury risk of the colliding vehicle. The results also confirmed that there is a protective effect of vehicle size above and beyond that of vehicle mass for frontal and front to side collisions.     

Analysis of driver injury severity in single-vehicle crashes on rural and urban roadways

This study analyzes driver injury severities for single-vehicle crashes occurring in rural and urban areas using data collected in New Mexico from 2010 to 2011. Nested logit models and mixed logit models are developed in order to account for the correlation between severity categories (No injury, Possible injury, Visible injury, Incapacitating injury and fatality) and individual heterogeneity among drivers. Various factors, such as crash and environment characteristics, geometric features, and driver behavior are examined in this study. Nested logit model and mixed logit model reveal similar results in terms of identifying contributing factors for driver injury severities. In the analysis of urban crashes, only the nested logit model is presented since no random parameter is found in the mixed logit model. The results indicate that significant differences exist between factors contributing to driver injury severity in single-vehicle crashes in rural and urban areas. There are 5 variables found only significant in the rural model and six significant variables identified only in the urban crash model. These findings can help transportation agencies develop effective policies or appropriate strategies to reduce injury severity resulting from single-vehicle crashes.     

What can we learn about North Dakota's youngest drivers from their crashes?

For North Dakota teens, three of every four deaths are from motor vehicle crashes. Injury crash records for teen drivers were studied to gain insight regarding driver, vehicle, and road factors for public safety policy and program discussions. Results show 14-year-old drivers are three times more likely to die or be disabled in an injury crash than 17-year-old drivers, and that male drivers are 30% less likely to incur severe injury. As expected, seat belt use is a critical factor in severe injury avoidance. The likelihood for death or disablement is 165% greater for unbelted teen drivers than for those who are properly belted. In addition, rural and gravel roads pose a risk. Teens are six times more likely to be severely injured in crashes on rural roads than on urban roads. Findings suggest that an increased licensing age and seat belt emphasis may reduce teen traffic injuries in the state. In addition, more information on exposure should be attained to better understand rural and gravel road as risks.     

Defective equipment and tractor-trailer crash involvement

The role of defective equipment in large truck crashes on interstate highways in Washington State was investigated using a case-control study design. For each large truck involved in a crash, three trucks were randomly selected from the traffic stream at the same time and place as the crash, but one week later both crash and comparison trucks were inspected by Commercial Vehicle Enforcement officers of the Washington State Patrol. The effects of truck equipment condition, truck operating characteristics (carrier type, carrier operation, and truck load), and driver characteristics (driver age, hours of driving) on crash involvement were analyzed by comparing their relative frequency among crash-involved and comparison sample tractor-trailers. A logistic regression model was used to estimate the adjusted odds ratio for each factor. Overall, 77% of tractor-trailers in crashes and 66% of those not involved in crashes had defective equipment warranting a citation. Forty-one percent in crashes had defective equipment warranting taking the truck out of service, and 31% not in crashes had these defects. Brake defects were the most common type and were found in 56% of tractor-trailers in crashes; steering equipment defects were found in 21%. The relative risk of crash involvement for trucks with brake defects was about one and one-half times that for trucks without brake defects. For trucks with steering defects, the relative risk of crash involvement was at least twice that for trucks with steering defects, the relative risk of crash involvement was at least twice that for trucks without defects, and the risk increased substantially for trucks with out-of-service steering defects.     

A note on modeling pedestrian-injury severity in motor-vehicle crashes with the mixed logit model

Pedestrian-injury severity has been traditionally modeled with approaches that have assumed that the effect of each variable is fixed across injury observations. This assumption ignores possible unobserved heterogeneity which is likely to be particularly important in pedestrian injuries because unobserved physical health, strength, and behavior may significantly affect the pedestrians’ ability to absorb collision forces. To address such unobserved heterogeneity, this research applies a mixed logit model to analyze pedestrian-injury severity in pedestrian-vehicle crashes. Using police-reported collision data from 1997 through 2000 from North Carolina, several factors were found to more than double the average probability of fatal injury for pedestrians in motor-vehicle crashes including: darkness without streetlights (400% increase in fatality probability), vehicle is a truck (370% increase), freeway (330% increase), speeding involved (360% increase), and collisions involving a motorist who had been drinking (250% increase). It was also found that the effect of pedestrian age was normally distributed across observations, and that as pedestrians became older the probability of fatal injury increased substantially. Heterogeneity in the mean of the random parameters for the freeway and pedestrian-solely-at-fault collision indicators was related to pedestrian gender, and heterogeneity in the mean of the random parameters for the traffic-sign and motorist-back-up indicators was related to pedestrian age.     

A comprehensive analysis of factors influencing the injury severity of large-truck crashes

Given the importance of trucking to the economic well being of a country and the safety concerns posed by the trucks, a study of large-truck crashes is critical. This paper contributes by undertaking an extensive analysis of the empirical factors affecting injury severity of large-truck crashes. Data from a recent, nationally representative sample of large-truck crashes are examined to determine the factors affecting the overall injury severity of these crashes. The explanatory factors include the characteristics of the crash, vehicle(s), and the driver(s). The injury severity was modeled using two measures. Several similarities and some differences were observed across the two models which underscore the need for improved accuracy in the assessment of injury severity of crashes. The estimated models capture the marginal effects of a variety of explanatory factors simultaneously. In particular, the models indicate the impacts of several driver behavior variables on the severity of the crashes, after controlling for a variety of other factors. For example, driver distraction (truck drivers), alcohol use (car drivers), and emotional factors (car drivers) are found to be associated with higher severity crashes. A further interesting finding is the strong statistical significance of several dummy variables that indicate missing data – these reflect how the nature of the crash itself could affect the completeness of the data. Future efforts should seek to collect such data more comprehensively so that the true effects of these aspects on the crash severity can be determined.