Applying interpretable machine learning to classify tree and utility pole related crash injury types

In spite of enormous improvements in vehicle safety, roadway design, and operations, there is still an excessive amount of traffic crashes resulting in injuries and major productivity losses. Despite the many studies on factors of crash frequency and injury severity, there is still further research to be conducted. Tree and utility pole/other pole related (TUOP) crashes present approximately 12 to 15% of all roadway departure (RwD) fatal crashes in the U.S. The count of TUOP crashes comprise nearly 22% of all fatal crashes in Louisiana. From 2010 to 2016, there were 55,857 TUOP crashes reported in Louisiana. Individually examining each of these crash reports is not a realistic option to investigate crash factors. Therefore, this study employed text mining and interpretable machine learning (IML) techniques to analyze all TUOP crashes (with available crash narratives) that occurred in Louisiana from 2010 to 2016. This study has two major goals: 1) to develop a framework for applying machine learning models to classify injury levels from unstructured textual content, and 2) to apply an IML framework that provides probability measures of keywords and their association with the injury classification. The present study employed three machine learning algorithms in the classification of injury levels based on the crash narrative data. Of the used modeling techniques, the eXtreme gradient boosting (XGBoost) model shows better performance, with accuracy ranging from 0.70 to 24% for the training data and from 0.30% to 16% for the test data.     

Modeling two-way stop-controlled intersection crashes with zero-inflated models on Louisiana rural two-lane highways

Intersection safety continues to be a crucial issue throughout the United States. In 2016, 27% of the 37,461 traffic fatalities on U.S. roadways occurred at or near intersections. Nearly 70% of intersection-related fatalities occurred at unsignalized intersections. At such intersections, vehicles stopping or slowing to turn create speed differentials between vehicles traveling in the same direction. This is particularly problematic on two-lane highways. Research was performed to analyze safety performance for intersections on rural, two-lane roadways, with stop control on the minor roadway. Roadway, traffic, and crash data were collected from 4148 stop-controlled intersections of all 64 Parishes (counties) statewide in Louisiana, for the period of 2013 to 2017. Four count approaches, Poisson, Negative Binomial (NB), Zero-inflated Poisson (ZIP) and Zero-inflated Negative Binomial (ZINB) were used to model the number of intersection crashes for different severity levels. The results indicate that ZIP models provide a better fit than all other models. In addition to traffic volume, larger curve radii of major and minor roads and wider lane widths of major roads led to significantly smaller crash occurrences. However, higher speed limits of major roads led to significantly greater crash occurrences. Four-leg stop-controlled intersections have 35% greater total crashes, 49% greater fatal and injury crashes, and 25% greater property damage only (PDO) crashes, relative to three-leg intersections.     

Road characteristics and environment factors associated with motorcycle fatal crashes in Malaysia

This study aims to determine risk factors contributing to traffic crashes in 9,176 fatal cases involving motorcycle in Malaysia between 2010 and 2012. For this purpose, both multinomial and mixed models of motorcycle fatal crash outcome based on the number of vehicle involved are estimated. The corresponding model predicts the probability of three fatal crash outcomes: motorcycle single-vehicle fatal crash, motorcycle fatal crash involving another vehicle and motorcycle fatal crash involving two or more vehicles. Several road characteristic and environmental factors are considered including type of road in the hierarchy, location, road geometry, posted speed limit, road marking type, lighting, time of day and weather conditions during the fatal crash. The estimation results suggest that curve road sections, no road marking, smooth, rut and corrugation of road surface and wee hours, i.e. between 00.00 am to 6 am, increase the probability of motorcycle single-vehicle fatal crashes. As for the motorcycle fatal crashes involving multiple vehicles, factors such as expressway, primary and secondary roads, speed limit more than 70 km/h, roads with non-permissible marking, i.e. double lane line and daylight condition are found to cause an increase the probability of their occurrence. The estimation results also suggest that time of day (between 7 pm to 12 pm) has an increasing impact on the probability of motorcycle single-vehicle fatal crashes and motorcycle fatal crashes involving two or more vehicles. Whilst the multinomial logit model was found as more parsimonious, the mixed logit model is likely to capture the unobserved heterogeneity in fatal motorcycle crashes based on the number of vehicles involved due to the underreporting data with two random effect parameters including 70 km/h speed limit and double lane line road marking.     

Socio-economic costs of road crashes in middle-income countries: Applying a hybrid approach to Kazakhstan

Information about road crash costs is a valuable input for road safety policy making and it is essential for conducting cost-benefit analysis of road safety interventions. This paper presents a methodology for assessing the socio-economic costs of road crashes as well as an estimate of the volume of these costs in Kazakhstan. Five costs components have been taken into account: medical costs, production loss, human costs, vehicle damage and administrative costs. A hybrid methodological approach has been used, which implies that three different types of methods have been applied to capture all costs: the human capital method (production loss), willingness to pay (human costs) and restitution costs method (other components). Input data were retrieved from existing databases from a variety of road safety stakeholders and other organizations. A household survey was conducted to collect additional information, including the willingness to pay for fatal crash risk reductions. Remaining data gaps have been bridged by using data from other countries. The socio-economic costs of road crashes in Kazakhstan are estimated at $6.8 billion in 2012, which corresponds to 3.3% of GDP. Human costs account for 81% of the total costs, vehicle damage for 11% and production loss for 6%. Administrative and medical costs are relatively very small cost components. More than half of the costs is related to injuries, while fatalities account for about a third of the total costs and property damage only accounts for approximately 10%.     

Investigating fatal and injury crash patterns of teen drivers with unsupervised learning algorithms

Teenagers have been emphasized as a critical driver population class because of their overrepresentation in fatal and injury crashes. The conventional parametric approaches rest on few predefined assumptions, which might not always be valid considering the complicated nature of teen drivers' crash characteristics that are reflected by multidimensional crash datasets. Also, individual attributes may be more speculative when combined with other factors. This research employed joint correspondence analysis (JCA) and association rule mining (ARM) to investigate the fatal and injury crash patterns of at-fault teen drivers (aged 15 to 19 years) in Louisiana. The unsupervised learning algorithms can explore meaningful associations among crash categories without restricting the nature of variables. The analyses discover intriguing associations to understand the potential causes and effects of crashes. For example, alcohol impairment results in fatal crashes with passengers, daytimes severe collisions occur to unrestrained drivers who have exceeded the posted speed limits, and adverse weather conditions are associated with moderate injury crashes. The findings also reveal how the behavior patterns connected with teen driver crashes, such as distracted driving in the morning hours, alcohol intoxication or using cellphone in pickup trucks, and so on. The research results can lead to effectively targeted teen driver education programs to mitigate risky driving maneuvers. Also, prioritizing crash attributes of key interconnections can help to develop practical safety countermeasures. Strategy that covers multiple interventions could be more effective in curtailing teenagers' crash risk.     

Extremely serious crashes on urban roadway networks: Patterns and trends

Extremely serious traffic crashes, defined as having a death toll of two and greater than two, have become a serious safety concern on urban roadways in Louisiana. This study examined the different contributing factors of these crashes to determine significant trends and patterns. We collected traffic crash data from Louisiana during the period of 2013 to 2017 and found that a total of 72 extremely serious crashes (around 2% of all traffic fatalities) occurred on Louisiana urban roadway networks. As crash data contain an enormous list of contributing factors, there was an issue of ‘more features than data points’ in solving the research problem. Most of these variables are categorial in nature. We selected a dimension reduction tool called Taxicab Correspondence Analysis (TCA) to investigate the complex interaction between multiple factors under a two-dimensional map. Findings of the study reveal several key clusters of attributes that show patterns of association between different crash attributes. The conclusions of this study are exploratory, and the results can help in better visualizing the association between key attributes of crashes. The findings have potentials in designing suitable countermeasures to reduce extremely serious crashes.     

Using on-board video data for safety analysis – An analysis of right hook crashes involving large buses at signalized intersections

Bus right hook (BRH) crashes at intersections are one of the most common types of crashes for bus carriers, which accounted for as high as 16% of fatal and injury crashes involving large buses at intersections in Taiwan. A BRH crash occurs when a bus and another vehicle traveling in the same direction head into an intersection, but the bus driver makes a right turn across the path of the through-moving vehicle, and both vehicles collide. This study responds to the research needs to identity factors associated with BRH crashes by utilizing in-vehicle data recorder (IVDR) data. A four step analysis procedure was developed, including (1) video data coding, (2) crash sequence analysis to identify crash contributing factors, (3) a case-control study to examine the relationship between the crash contributing factors and crash occurrence, and (4) modeling crash risk in terms of the crash contributing factors to better understand the crash generating process. This study first identified the existence of driver unattended time as the time between when the driver last checked the right back mirror to finally steering for a right turn, indicating the time period wherein the driver did not track the through vehicle on the right side using the right back mirror. It was found that BRH crashes could be attributed to the concurrence of unattended time and the speed difference between the bus and through vehicle. Several recommendations are discussed based on the results to further develop countermeasures to reduce this type of crash.     

Effects of the autonomous vehicle crashes on public perception of the technology

In March 2018, an Uber-pedestrian crash and a Tesla's Model X crash attracted a lot of media attention because the vehicles were operating under self-driving and autopilot mode respectively at the time of the crash. This study aims to conduct before-and-after sentiment analysis to examine how these two fatal crashes have affected people's perceptions of self-driving and autonomous vehicle technology using Twitter data. Five different and relevant keywords were used to extract tweets. Over 1.7 million tweets were found within 15 days before and after the incidents with the specific keywords, which were eventually analyzed in this study. The results indicate that after the two incidents, the negative tweets on “self-driving/autonomous” technology increased by 32 percentage points (from 14% to 46%). The compound scores of “pedestrian crash”, “Uber”, and “Tesla” keywords saw a 6% decrease while “self-driving/autonomous” recorded the highest change with an 11% decrease. Before the Uber-incident, 19% of the tweets on Uber were negative and 27% were positive. With the Uber-pedestrian crash, these percentages have changed to 30% negative and 23% positive. Overall, the negativity in the tweets and the percentage of negative tweets on self-driving/autonomous technology have increased after their involvement in fatal crashes. Providing opportunities to interact with this developing technology has shown to positively influence peoples' perception.     

Run-off-road and recovery – state estimation and vehicle control strategies

Despite many advances in vehicle safety technology, traffic fatalities remain a devastating burden on society. With over two-thirds of all fatal single-vehicle crashes occurring off the roadway, run-off-road (ROR) crashes have become the focus of much roadway safety research. Current countermeasures, including roadway infrastructure modifications and some on-board vehicle safety systems, remain limited in their approach as they do not directly address the critical factor of driver behaviour. It has been shown that ROR crashes are often the result of poor driver performance leading up to the crash. In this study, the performance of two control algorithms, sliding control and linear quadratic control, was investigated for use in an autonomous ROR vehicle recovery system. The two controllers were simulated amongst a variety of ROR conditions where typical driver performance was inadequate to safely operate the vehicle. The sliding controller recovered the fastest within the nominal conditions but exhibited large variability in performance amongst the more extreme ROR scenarios. Despite some small sacrifices in lateral error and yaw rate, the linear quadratic controller demonstrated a higher level of consistency and stability amongst the various conditions examined. Overall, the linear quadratic controller recovered the vehicle 25% faster than the sliding controller while using 70% less steering, which combined with its robust performance, indicates its high potential as an autonomous ROR countermeasure.     

Using machine leaning techniques for evaluation of motorcycle injury severity

There is a growing interest in the application of the machine learning techniques in predicting the motorcycle crash severity. This is partly due to a progress in autonomous vehicles technology, and machine learning technique, which as a main component of autonomous vehicle could be implemented for traffic safety enhancement. Wyoming's motorcycle crash fatalities constitute a concern since the count of riders being killed in motorcycle crashes in 2014 was 11% of the total road fatalities in the state. The first step of crash reduction could be achieved through identification of contributory factors to crashes. This could be accomplished by using a right model with high accuracy in predicting crashes. Thus, this study adopted random forest, support vector machine, multivariate adaptive regression splines and binary logistic regression techniques to predict the injury severity outcomes of motorcycle crashes. Even though researchers applied all the aforementioned techniques to model motorcycle injury severities, a comparative analysis to assess the predictive power of such modeling frameworks is limited. Hence, this study contributes to the road safety literature by comparing the performance of the discussed techniques. In this study, Wyoming's motorcycle crash injury severities are modeled as functions of the characteristics that give rise to crashes. Before conducting any analyses, feature reduction was used to identify a best number of predictors to be included in the model. Also to have an unbiased estimation of the performance of different machine learning techniques, 5-fold cross-validation was used for model performance evaluation. Two measure, Area under the curve (AUC), and confusion matrix were used to compare different models' performance. The machine learning results indicate that random forest model outperformed the other models with the least misclassification and higher AUC. It was also revealed that a dichotomous response variable, with fatality and incapacitation injury in one category, along with all other categories in another group would result in a lower misclassification rate than a polychotomous response variable. This might result from the nature of motorcycle crashes, lacking a protection compared with passenger cars, preventing machine learning technique to get trained properly. Moreover, the most important variables identified by the random forest model are those related to the operating speed, resentful other party, traffic volume, truck traffic volume, riding under the influence, horizontal curvature, wide roadway with more than two lanes and rider's age.     

Safety evaluation of unsignalized and signalized restricted crossing U-turn (RCUT) intersections in rural and suburban areas based on prior control type

The focus of this paper is on evaluating the safety effectiveness of restricted crossing U-turn (RCUT) intersections in rural and suburban areas based on prior control type. Both, unsignalized and signalized RCUT intersections were evaluated using the Empirical Bayes (EB) before-after evaluation method. The 42 RCUT intersections selected for this research were converted from a two-way stop-controlled (TWSC) intersection or signalized intersection in the rural and suburban areas. The results show a 70.63% reduction in the total number of crashes and a 76.10% reduction in the number of fatal and injury crashes at unsignalized stop-controlled RCUT intersections in the rural area. Also, an 89.25% reduction in the total number of crashes and a 94.42% reduction in the number of fatal and injury crashes was observed at offset three-legged unsignalized stop-controlled RCUT intersections converted from four-legged TWSC intersections in rural areas. In the suburban areas, a 64.86% reduction in the total number of crashes and a 73.39% reduction in the number of fatal and injury crashes was observed at unsignalized stop-controlled RCUT intersections. Further, a 10.15% and a 31.08% reduction in the total number of crashes, and an 84.26% and 41.31% reduction in the number of fatal and injury crashes was observed at a signalized RCUT intersection in the rural and suburban areas, respectively. The safety effectiveness of unsignalized RCUT intersections in the rural areas with a larger sample size was found to be higher than was observed by researchers in the past. While unsignalized RCUT intersections in the suburban areas are effective, there is not enough evidence to support the installation of signalized RCUT intersections. These findings help researchers and practitioners in making informed decisions and installing RCUT intersections from a safety perspective.     

Sampling bias and weight factors for in-depth motorcycle crash data in Thailand

Motorcycle crashes are documented in Thailand's national records but are underreported and lacking detail. In-depth motorcycle crash data, collected by Thailand Accident Research Center (TARC), contains a smaller number of motorcycle crashes but more detail. However, to draw conclusions at a national level, representativeness of the TARC in-depth data is currently unknown, and the correction of sampling biases may be required. In this study, the Capture-recapture method was used to examine the underreporting in the national crash data (from the government insurance company). It was found that 69% of fatal and 70% of non-fatal injuries were underreported, respectively. The in-depth crash data was found to be biased. The weighting methods post-stratification and iterative proportional fitting were applied to compensate for the bias and are shown to improve the representativeness of the in-depth motorcycle crash data. Weighted in-depth crash data appears to be suitable to draw conclusions on motorcyclist safety in Thailand.     

Modeling lane-change-related crashes with lane-specific real-time traffic and weather data

This study investigates the relationship between lane-change-related crashes and lane-specific, real-time traffic factors. It is anticipated that the real-time traffic data for the two lanes—the vehicle's lane (subject lane) and the lane to which that a vehicle intends to change (target lane)—are more closely related to lane-change-related crashes, as opposed to congregated traffic data for all lanes. Lane-change-related crash data were obtained from a 62-mile long freeway in Southeast Wisconsin in 2012 and 2013. One-minute traffic data from the 5- to 10-minute interval prior to the crashes were extracted from an immediately upstream detector station and two immediately downstream stations from the crash location. Weather information was collected from a major historical weather database. A matched case-control logistic regression was used for analysis. Results show that the following factors significantly affect the probability of a lane-change-related crash: average flow into the target lane at the first downstream station, the flow ratio at the second downstream station, and snow conditions. Additionally, the average speed in the target lane at the first downstream station contributes to the occurrence of lane-change crashes during snowy conditions. According to the model, the probability of a lane-change-related crash under real-time traffic conditions can aid in flagging potential crash-prone conditions. The identified contributing factors can help traffic operators select traffic control and management countermeasures to proactively mitigate lane-change-related crashes.     

Modeling wrong-way crashes and fatalities on arterials and freeways

Wrong way driving (WWD) research and mitigation measures have primarily focused on limited access facilities. This is most likely due to the higher incidence of fatal WWD crashes with dramatic consequences on freeways, media attention, and a call for innovative solutions to address the problem. While public agencies and published literature address WWD incidence on freeway systems, the crash analyses on non-limited access facilities, i.e., arterial corridors, remains untouched. This research extends previous works and attempts to provide many new perspectives on arterial WWD incidence. In particular, one work showed that while WWD fatalities are more likely to occur on freeways, the likelihood of these crashes is higher on arterials. Hence this work with univariate and multivariate analyses of WWD and non-WWD crashes, and fatal and non-fatal WWD incidents. Results show the impressive negative impacts of alcohol use, driver defect, nighttime and weekend incidence, poor street lighting, low traffic volumes, rural geography, and median and shoulder widths. The objective here is to highlight the need for paying greater attention to WWD crashes on arterial corridors as is done with fatal WWD incidents on freeway systems. It suffices to say that while engineering countermeasures should evolve from the traditional signing and pavement markings to connected vehicle technology applications, there is a clear and compelling need to focus on educational campaigns specifically targeting drunken driving, and enforcement initiatives with an objective to mitigate WWD in the most efficient manner possible.     

Predicting downgrade crash frequency with the random-parameters negative binomial model: Insights into the impacts of geometric variables on downgrade crashes in Wyoming

Road deaths, injuries and property damage place a huge burden on the economy of most nations. Wyoming has a high crash rate on mountain passes. The crash rates observed in the state is as a result of many factors mainly related to the challenging mountainous terrain in the state, which places extra burden on drivers in terms of requiring higher levels of alertness and driving skill. This study was conducted to investigate factors leading to crashes on Wyoming downgrades, with a focus on geometric variables. Traditionally, crash frequency analysis is conducted using count models such as Poisson or negative binomial models. However, factors that affect crash frequency are known to vary across observations. The use of a methodology that fails to take into account heterogeneity in observed and unobserved effects relating to roadway characteristics can lead to biased and inconsistent estimates. Inferences made from such parameter estimates may be misleading. This study employed the random-parameters negative binomial regression models to evaluate the impact of geometric variables on crash frequency. Five separate models were estimated for total, fatal/injury, property damage only (PDO), truck, and non-truck crash frequencies. Several geometric and traffic variables were found to influence the frequency of crashes on downgrades. These included segment length, vertical grade, shoulder width, lane width, presence of downgrade warning sign, vertical curve length, presence of a passing lane, percentage of trucks, number of lanes and AADT. The results suggest that segment length, lane width, presence of a passing lane, presence of a downgrade warning sign, vertical grade, and percentage of trucks are best modeled as random parameters. The findings of this study will provide transportation agencies with a better understanding of the impact of geometric variables on downgrade crashes.     

Macroscopic analysis on the frequency and severity of pedestrian crashes on National Roads in Metro Manila,Philippines

Pedestrians are the most vulnerable road users; thus, understanding the primary factors that lead to pedestrian crashes is a chief concern in road safety. However, owing to the limitations of crash data in developing countries, only a few studies have evaluated the comprehensive characteristics of pedestrian crashes, specifically on different road types. This study attempted to develop pedestrian crash frequency and severity models on national roads by using the road characteristics and built environment parameters, based on the road crash data (2016–2018) that involved pedestrians in Metro Manila, Philippines. Remarkable findings included primary roads, presence of footbridges, road sections with bad surface conditions, and increased fractions of commercial, residential, and industrial roads, which exhibited a greater likelihood of pedestrian crashes. Crashes involving elderly pedestrians, heavier vehicles, late-night hours, fair surface conditions, and open spaces were associated with increased likelihoods of fatal outcomes. Essentially, this study provides a macroscopic perspective in understanding the factors associated with the severity and frequency of pedestrian crashes, and it would aid the authorities in identifying proper countermeasures.     

Modeling injury severity of crashes involving trucks: Capturing and exploring risk factors associated with land use and demographic in addition to crash,driver, and on-network characteristics

Nearly 499,000 motor vehicle crashes involving trucks were reported across the United States in 2018, out of which 22% resulted in fatalities and injuries. Given the growing economy and demand for trucking in the future, it is crucial to identify the risk factors to understand where and why the likelihood of getting involved in a severe or moderate injury crash with a truck is higher. The focus of this research, therefore, is on developing a methodology, capturing and integrating data, exploring, and identifying risk factors associated with surrounding land use and demographic characteristics in addition to crash, driver, and on-network characteristics by modeling injury severity of crashes involving trucks. Crash data for Mecklenburg County in North Carolina from 2013 to 2017 was used to develop partial proportional odds model and identify risk factors influencing injury severity of crashes involving trucks. The findings indicate that dark lighting condition, inclement weather condition, the presence of double yellow or no-passing zone, road sections with speed limit >40 mph and curves, and driver fatigue, impairment, and inattention have a significant influence on injury severity of crashes involving trucks. These outcomes indicate the need for effective geometric design and improved visibility to reduce the injury severity of crashes involving trucks. The likelihood of a severe or moderate injury crash involving a truck is also high in areas with high employment, government, light commercial, and light industrial land uses. The findings can be used to identify potential risk areas, proactively plan and prioritize the allocation of resources to improve safety of transportation system users in these areas.     

What affects pedestrian crossing difficulty at urban intersections in a developing country?

In developing countries, road traffic crashes involving pedestrians have become a foremost concern. At present, road safety assessment plans and selection of interventions are primarily restricted to traditional approaches that depend on the investigations of historical crash data. However, in developing countries such as India, the availability, consistency, and accuracy of crash data are major concerns. In contrast, proactive approaches such as studying road users' risk perception have emerged as a substitute method of examining potential risk factors. An individual's risk perception offers vital information on probable crash risk, which may be beneficial in detecting high-risk locations and major causes of crashes. Since the pedestrian fatality risk is not uniform across the urban road network level, it may be expected that pedestrians' perceived risk measured in terms of “crossing difficulty” would also vary across the sites. In this perspective, the present paper establishes a mathematical association between the pedestrians' perceived “crossing difficulty” and actual crashes. The model outcome confirms that pedestrians' perceived crossing difficulty is a good surrogate of fatal pedestrian crashes at the intersection level in Kolkata City, India. Subsequently, to examine the impact of traffic exposures, road infrastructure, land use, spatial factors, and pedestrian-level attributes on pedestrians' “crossing difficulty”; a set of Ordered Logit models are developed. The model outcomes show that high vehicle and pedestrian volume, vehicular speed, absence of designated bus stop, the presence of inaccessible pedestrian crosswalk, on-street parking, lack of signalized control (for both vehicle and pedestrian), inadequate sight distance, land use pattern, slum population, pedestrian-vehicular post encroachment time, waiting time before crossing, road width, and absence of police enforcement at an intersection significantly and positively increase pedestrian's crossing difficulty at urban intersections. To end, the model findings are advantageously utilized to develop a set of countermeasures across 3E's of road safety.     

Influence of segment length on the fitness of multivariate crash prediction models applied to a Brazilian multilane highway

Road safety modeling enables the development of crash prediction models and the investigation of which factors contribute to crash occurrence. Developing multivariate response models is also valuable, but such models are currently under-exploited. Machine learning techniques, especially artificial neural networks (ANN), have been presented as possible alternatives. Furthermore, selecting a proper roadway segmentation is one of the first tasks in the standard crash modeling workflow. However, this is a challenging task, especially in terms of choosing a segment length. This article presents a study of the influence of segment length on the development of multivariate response models (i.e., three response variables: property damage only crashes, injured victims crashes, and fatal crashes). The models use ANN for a road segment of a Brazilian divided multilane highway. The highway to be modeled was divided into segments with 10 different fixed lengths. The model characterization included geometric and operational data available for the years from 2011 to 2017. The models were evaluated in terms of errors and by residual plot analysis. The 5-km segment of the northbound carriageway and the 4.5-km segment of the southbound carriageway presented the smallest errors and the highest values of R². The residual analyses confirmed the trend to improve the model with the greater segment lengths. This was clear by the residues' distribution around zero, except for the output “Fatal crashes”. The better performance of the longer segments models was expected because these models aggregate more crashes into one segment. The reduction of no crash observations also facilitated the improvement of the models' goodness-of-fit. The use of ANNs also revealed its potential value. However, it is still important to seek strategies to deal with the excess of zeros in fatal crashes; a problem that also occurs in the traditional statistical modeling process.     

Factors influencing crash involvement during early years of owning a new vehicle

Though automobile manufacturers are investing efforts to make newer vehicles safer to drive, an element of uncertainty with the new vehicle seems to persist with the drivers during the early years of ownership. This could be due to a lack of familiarity of the vehicle's power, dimensions or available technologies/features. While the uncertainty in itself is a potential cause of a crash, it is important for the policy-makers, practitioners, and automobile manufacturers to understand the factors that could further aggravate the problem. This research focuses on identifying the factors influencing the likelihood of getting involved in a crash and its severity when driving a new vehicle. Crash data for North Carolina for the years 2013 to 2018 (six years) was used develop partial proportionality odds models, compute the odds ratios, analyze the effects of explanatory variables, and identify factors influencing crashes by the age of the vehicle. The likelihood of getting involved in a severe or moderate injury crash when driving a new vehicle is less for drivers in the age group ≤19 years. Erratic driving behavior (like making wide turns, weaving and swerving in traffic, driving with headlights off, driving on center-line or lane-line, etc.) and speeding increase the risk of getting involved in a moderate injury crash when driving a new vehicle. Likewise, the odds of getting involved in a crash are high on weekends and in adverse weather conditions when driving a new vehicle. They are higher when driving a new motorcycle, heavy vehicle or farm machinery. The findings help policy-makers and practitioners formulate strategies to educate drivers on factors influencing crash risk when driving a new vehicle. Further, automobile manufacturers can establish guidance programs and documentation that explain what to expect when buying and driving a new vehicle.