Heterogeneous ensemble learning for enhanced crash forecasts - A frequentist and machine learning based stacking framework.

INTRODUCTION: This study aims to increase the prediction accuracy of crash frequency on roadway segments that can forecast future safety on roadway facilities. A variety of statistical and machine learning (ML) methods are used to model crash frequency with ML methods generally having a higher prediction accuracy. Recently, heterogeneous ensemble methods (HEM), including "stacking," have emerged as more accurate and robust intelligent techniques providing more reliable and accurate predictions. METHODS: This study applies "Stacking" to model crash frequency on five-lane undivided (5 T) segments of urban and suburban arterials. The prediction performance of "Stacking" is compared with parametric statistical models (Poisson and negative binomial) and three state-of-the-art ML techniques (Decision tree, random forest, and gradient boosting), each of which is termed as the base-learner. By employing an optimal weight scheme to combine individual base-learners through stacking, the problem of biased predictions in individual base-learners due to differences in specifications and prediction accuracies is avoided. Data including crash, traffic, and roadway inventory were collected and integrated from 2013 to 2017. The data are split into training (2013-2015), validation (2016), and testing (2017) datasets. After training five individual base-learners using training data, prediction outcomes are obtained for the five base-learners using validation data that are then used to train a meta-learner. RESULTS: Results of statistical models reveal that crashes increase with the density (number per mile) of commercial driveways whereas decrease with average offset distance to fixed objects. Individual ML methods show similar results - in terms of variable importance. A comparison of out-of-sample predictions of various models or methods confirms the superiority of "Stacking" over the alternative methods considered. CONCLUSIONS AND PRACTICAL APPLICATIONS: From a practical standpoint, "stacking" can enhance prediction accuracy (compared to only one base-learner with a particular specification). When applied systemically, stacking can help identify more appropriate countermeasures.

Interactive impacts of walkability, social vulnerability, & travel behavior on COVID-19 mortality: A hierarchical Bayesian spatial random parameter approach.

While existing research highlights the built and social environment impacts on COVID-19 mortality, no empirical evidence exists on how the built and social environments may interact to influence COVID-19 mortality. This study presents a rigorous empirical assessment of the interactive impacts of social vulnerability and walkability on neighborhood-level COVID-19 mortality rates. Based in King County, WA, a unique data infrastructure is created by spatially integrating diverse census tract-level data on COVID-19 mortalities, walkability characteristics, social vulnerability, and travel behavior measures. Advanced Markov Chain Monte Carlo (MCMC) based Full Bayes hierarchical spatial random parameter models are developed to simultaneously capture spatial and unobserved random heterogeneity. Around 46% of the neighborhoods had opposite levels of walkability and social vulnerability. Compared to low walkability and high social vulnerability, neighborhoods with high walkability and low social vulnerability (i.e., best case scenario) had on average 20.2% (95% Bayesian CI: -37.2% to -3.3%) lower COVID-19 mortality rates. Analysis of the interactive impacts when only one of the social and built environment metrics was in a healthful direction revealed significant offsetting effects - suggesting that the underlying structural social vulnerability issues faced by our communities should be addressed first for the infectious disease-related health impacts of walkable urban design to be observed. Concerning travel behavior, the findings indicate that COVID-19 mortality rates may be reduced by discouraging auto use and encouraging active transportation. The study methodologically contributes by simultaneously capturing spatial and unobserved heterogeneity in a holistic Full Bayesian framework.

Toward better measurement of traffic injuries - Comparison of anatomical injury measures in predicting the clinical outcomes in motorcycle crashes.

INTRODUCTION: Little evidence exists in the literature regarding the discrimination power of better anatomical injury measures in differentiating clinical outcomes in motorcycle crashes. Furthermore, multiple injuries to different body parts of the rider are seldom analyzed. This study focuses on comparing anatomical injury measures such as the injury severity score (ISS) and the new injury severity score (NISS) in capturing injuries of multiple injured riders and examining the discriminatory capabilities of the ISS and NISS in predicting clinical outcomes post motorcycle crash. METHODS: The study harnessed unique and comprehensive injury data on 322 riders from the US DOT Federal Highway Administration's Motorcycle Crash Causation Study (MCCS). Detailed exploratory analysis is performed and discrete/ordered statistical models are estimated for three clinical outcomes: mortality risk, trauma risk, and trauma status. RESULTS: Around 9% of the riders died and 45% of the riders had injuries. Around 36% of the riders were hospitalized, disabled, or institutionalized. While a very strong dependence was found between ISS and NISS, ISS underestimated injuries sustained by riders. Statistical models for mortality risk revealed that a unit increase in the ISS and NISS was correlated with a 1.18 and 1.17 times increase in the odds of mortality, respectively. Moreover, a unit increase in ISS and NISS values was correlated with a higher trauma risk by 1.48 and 1.36 times, respectively. Our analysis reveals that the probability of a rider being hospitalized or disabled/institutionalized increases with an increase in the NISS. Conclusions and practical applications: The NISS exhibits significantly better calibration and discriminatory ability in differentiating survivors and non-survivors and in predicting trauma status - underscoring the importance of accounting for microscopic body-part-level injury data in motorcycle crashes. We consider that compared with the KABCO scale, the ISS and NISS are more nuanced scores that can better measure the overall injury intensity and can lead to more targeted countermeasures.

Causal evaluation of the health effects of light rail line: A Natural Experiment.

BACKGROUND AND OBJECTIVE: No research to date has causally linked built environment data with health care costs derived from clinically assessed health outcomes within the framework of longitudinal intervention design. This study examined the impact of light rail transit (LRT) line intervention on health care costs after controlling for mode-specific objectively assessed moderateto-vigorous physical activity (MVPA), participant-level neighborhood environmental measures, demographics, attitudinal predispositions, and residential choices. DATA AND METHODS: Based on a natural experiment related to a new LRT line in Portland - 282 individuals divided into treatment and control groups were prospectively followed during the pre- and post-intervention periods. For each individual, we harness high-resolution data on Electronic Medical Record (EMR) based health care costs, mode-specific MVPA, survey-based travel behavior, attitudinal/perception information, and objectively assessed built environment measures. Simulation-assisted longitudinal grouped random parameter models are developed to gain more accurate insights into the effects of LRT line intervention. RESULTS: Regarding the "average effect" of the LRT line intervention, no statistically significant reductions in health care costs were observed for the treated individuals over time. However, substantial heterogeneity was observed not only in the magnitude of effects but its direction as well after controlling for the within- and between-individual variations. For a subgroup of treated individuals, the LRT line opening decreased health care costs over time relative to the control group. Further comparative analysis based on the findings of heterogeneity-based models revealed that the effect of LRT intervention for the treated individuals differed by individual characteristics, attitudes/perceptions, and neighborhood level environmental features. CONCLUSIONS: The study revealed the presence of significant effect modifiers and distinct subgroup structures in the data related to the effects of LRT line intervention on health care costs. Severe implications of ignoring unobserved heterogeneity are highlighted. Limitations and potential avenues for future research are discussed.

The 2019 Conference on Health and Active Transportation: Research Needs and Opportunities.

Active transportation (AT) is widely viewed as an important target for increasing participation in aerobic physical activity and improving health, while simultaneously addressing pollution and climate change through reductions in motor vehicular emissions. In recent years, progress in increasing AT has stalled in some countries and, furthermore, the coronavirus (COVID-19) pandemic has created new AT opportunities while also exposing the barriers and health inequities related to AT for some populations. This paper describes the results of the December 2019 Conference on Health and Active Transportation (CHAT) which brought together leaders from the transportation and health disciplines. Attendees charted a course for the future around three themes: Reflecting on Innovative Practices, Building Strategic Institutional Relationships, and Identifying Research Needs and Opportunities. This paper focuses on conclusions of the Research Needs and Opportunities theme. We present a conceptual model derived from the conference sessions that considers how economic and systems analysis, evaluation of emerging technologies and policies, efforts to address inclusivity, disparities and equity along with renewed attention to messaging and communication could contribute to overcoming barriers to development and use of AT infrastructure. Specific research gaps concerning these themes are presented. We further discuss the relevance of these themes considering the pandemic. Renewed efforts at research, dissemination and implementation are needed to achieve the potential health and environmental benefits of AT and to preserve positive changes associated with the pandemic while mitigating negative ones.

Neighborhood-level COVID-19 hospitalizations and mortality relationships with built environment, active and sedentary travel.

Most of the existing literature concerning the links between built environment and COVID-19 outcomes is based on aggregate spatial data averaged across entire cities or counties. We present neighborhood level results linking census tract-level built environment and active/sedentary travel measures with COVID-19 hospitalization and mortality rates in King County Washington. Substantial variations in COVID-19 outcomes and built environment features existed across neighborhoods. Using rigorous simulation-assisted discrete outcome random parameter models, the results shed new lights on the direct and indirect connections between built environment, travel behavior, positivity, hospitalization, and mortality rates. More mixed land use and greater pedestrian-oriented street connectivity is correlated with lower COVID-19 hospitalization/fatality rates. Greater participation in sedentary travel correlates with higher COVID-19 hospitalization and mortality whereas the reverse is true for greater participation in active travel. COVID-19 hospitalizations strongly mediate the relationships between built environment, active travel, and COVID-19 survival. Ignoring unobserved heterogeneity even when higher resolution smaller area spatial data are harnessed leads to inaccurate conclusions.

Treating two pandemics for the price of one: Chronic and infectious disease impacts of the built and natural environment.

Compact walkable environments with greenspace support physical activity and reduce the risk for depression and several obesity-related chronic diseases, including diabetes and heart disease. Recent evidence confirms that these chronic diseases increase the severity of COVID-19 infection and mortality risk. Conversely, denser transit supportive environments may increase risk of exposure to COVID-19 suggesting the potential for contrasting chronic versus infectious disease impacts of community design. A handful of recent studies have examined links between density and COVID-19 mortality rates reporting conflicting results. Population density has been used as a surrogate of urban form to capture the degree of walkability and public transit versus private vehicle travel demand. The current study employs a broader range of built environment features (density, design, and destination accessibility) and assesses how chronic disease mediates the relationship between built and natural environment and COVID-19 mortality. Negative and significant relationships are observed between built and natural environment features and COVID-19 mortality when accounting for the mediating effect of chronic disease. Findings underscore the importance of chronic disease when assessing relationships between COVID-19 mortality and community design. Based on a rigorous simulation-assisted random parameter path analysis framework, we further find that the relationships between COVID-19 mortality, obesity, and key correlates exhibit significant heterogeneity. Ignoring this heterogeneity in highly aggregate spatial data can lead to incorrect conclusions with regards to the relationship between built environment and COVID-19 transmission. Results presented here suggest that creating walkable environments with greenspace is associated with reduced risk of chronic disease and/or COVID-19 infection and mortality.

Built environment, driving errors and violations, and crashes in naturalistic driving environment.

Driving errors and violations are highly relevant to the safe systems approach as human errors tend to be a predominant cause of crash occurrence. In this study, we harness highly detailed pre-crash Naturalistic Driving Study (NDS) data 1) to understand errors and violations in crash, near-crash, and baseline (no event) driving situations, and 2) to explore pathways that lead to crashes in diverse built environments by applying rigorous modeling techniques. The "locality" factor in the NDS data provides information on various types of roadway and environmental surroundings that could influence traffic flow when a precipitating event is observed. Coded by the data reductionists, this variable is used to quantify the associations of diverse environments with crash outcomes both directly and indirectly through mediating driving errors and violations. While the most prevalent errors in crashes were recognition errors such as failing to recognize a situation (39 %) and decision errors such as not braking to avoid a hazard (34 %), performance errors such as poor lateral or longitudinal control or weak judgement (8 %) were most strongly correlated with crash occurrence. Path analysis uncovered direct and indirect relationships between key built-environment factors, errors and violations, and crash propensity. Possibly due to their complexity for drivers, urban environments are associated with higher chances of crashes (by 6.44 %). They can also induce more recognition errors which correlate with an even higher chances of crashes (by 2.16 % with the "total effect" amounting to 8.60 %). Similar statistically significant mediating contributions of recognition errors and decision errors near school zones, business or industrial areas, and moderate residential areas were also observed. From practical applications standpoint, multiple vehicle technologies (e.g., collision warning systems, cruise control, and lane tracking system) and built-environment (roadway) changes have the potential to reduce driving errors and violations which are discussed in the paper.

A taxonomy of driving errors and violations: Evidence from the naturalistic driving study.

Driving errors and violations are identified as contributing factors in most crash events. To examine the role of human factors and improve crash investigations, a systematic taxonomy of driver errors and violations (TDEV) is developed. The TDEV classifies driver errors and violations based on their occurrence during the theoretically based perception-reaction process and analyzes their contributions in safety critical events. To empirically explore errors and violations, made by drivers of instrumented vehicles, in diverse built environments, this study harnesses unique and highly detailed pre-crash sensor data collected in the Naturalistic Driving Study (NDS), containing 673 crashes, 1,331 near-crashes and 7,589 baselines (no-event). Human factors are categorized into recognition errors, decision errors, performance errors, and errors due to the drivers' physical condition or their lack of contextual experience/familiarity, and intentional violations. In the NDS data, built environments (measured by roadway localities) are classified based on roadway functional classification and land uses, e.g., residential areas, school zones, and church zones. Based on the crash percentage to baseline percentage in a specific locality, interstates and open country/open residential (rural and semi-rural settings) may pose lower risks, while urban, business/industrial, and school zone locations showed higher crash risk. Human errors and violations by instrumented vehicle drivers contributed to 93% of the observed crashes, while roadway factors contributed to 17%, vehicle factors contributed in 1%, and 4% of crashes contained unknown factors. The most common human errors were recognition and decision errors, which occurred in 39% and 34% of crashes, respectively. These two error types occurred more frequently (each contributing to nearly 39% of crashes) in business or industrial land use environments (but not in dense urban localities). The findings of this study reveal continued prevalence of human factors in crashes. The distribution of driving errors and violations across different roadway environments can aid in the implementation of driver assistance systems and place-based interventions that can potentially reduce these driving errors and violations.

Injury severity analysis of pedestrian and bicyclist trespassing crashes at non-crossings: A hybrid predictive text analytics and heterogeneity-based statistical modeling approach.

Non-motorists involved in rail-trespassing crashes are usually more vulnerable to receiving major or fatal injuries. Previous research has used traditional quantitative crash data for understanding factors contributing to injury outcomes of non-motorists in train involved collisions. However, usually overlooked crash narratives can provide useful and unique contextual crash-specific information regarding factors associated with injury outcomes. The main objective of this study is to harness the rapid advancements in more sophisticated qualitative analysis procedures for identifying thematic concepts in unstructured crash narrative data. A two-staged hybrid approach is proposed where text mining is applied first to extract valuable information from crash narratives followed by inclusion of the new variables derived from text mining in formulation of advanced statistical models for injury outcomes. By using ten-year (2006-2015) non-motorist non-crossing trespassing injury data obtained from the Federal Railroad Administration, statistical procedures and advanced machine learning text analytics are applied to extract unique information on contributory factors of trespassers' injury outcomes. The key concepts are systematically categorized into trespasser, injury, train, medical, and location related factors. A total of 13 unique variables are extracted from the thematic concepts that are not present in traditional tabular crash data. The analysis reveals a positive statistically significant association between presence of crash narrative and trespasser's injury outcome (coded as minor, major, and fatal injury). Compared to crashes with minor injuries, crashes involving major and fatal injuries are more likely to be reported with crash narratives. A crosstabulation of new variables derived from text mining with injury outcomes revealed that trespassers with confirmed suicide attempts, trespassers wearing headphones, or talking on cell phones are more likely to receive fatal injuries. Among other factors identified, trespassers under alcohol influence, trespasser hit by commuter train, and advance warnings by engineer are associated with more severe (major and fatal) trespasser injury outcomes. Accounting for unobserved heterogeneity and controlling for other factors, fixed and random parameter discrete outcome models are developed to understand the heterogeneous correlations between trespasser injury outcomes and the new crash specific explanatory variables derived from text mining - providing deeper insights. Practical implications and future research directions are discussed.

Exploratory analysis of automated vehicle crashes in California: A text analytics & hierarchical Bayesian heterogeneity-based approach.

Automated vehicles (AVs) represent an opportunity to reduce crash frequency by eliminating driver error, as safety studies reveal human error contributes to the majority of crashes. To provide insights into the contributing factors of AV crashes, this study created a unique database from the California Department of Motor Vehicles 124 manufacturer-reported Traffic Collision Reports and was linked with detailed data on roadway and built-environment attributes. A novel text analysis was first conducted to extract useful information from crash report narratives. Of the crashes that could be geocoded (N = 113), results indicate the most frequent AV crash type was rear-end collisions (61.1%; N = 69) and 13.3% (N = 15) were injury crashes. These noteworthy outcomes and a small sample size motivated us to rigorously analyze rear-end and injury crashes in a Full Bayesian empirical setup. Owing to the potential issue of unobserved heterogeneity, hierarchical-Bayes fixed and random parameter logit models are estimated. Results reveal that when the automated driving system is engaged and remains engaged, the likelihood of an AV-involved rear-end crash is substantially higher compared to a conventionally-driven AV or when the driver disengages the automated driving system prior to a crash. Given the AV-involved crashes, the likelihood of an AV-involved rear-end crash was significantly higher in mixed land-use settings compared to other land-use types, and was significantly lower near public/private schools. Correlations of other roadway attributes and environmental factors with AV-involved rear-end and injury crash propensities are discussed. This study aids in understanding the interactions of AVs and human-driven conventional vehicles in complex urban environments.

Injury Severity and Contributing Driver Actions in Passenger Vehicle-Truck Collisions.

Large-scale truck-involved crashes attract great attention due to their increasingly severe injuries. The majority of those crashes are passenger vehicle-truck collisions. This study intends to investigate the critical relationship between truck/passenger vehicle driver's intentional or unintentional actions and the associated injury severity in passenger vehicle-truck crashes. A random-parameter model was developed to estimate the complicated associations between the risk factors and injury severity by using a comprehensive Virginia crash dataset. The model explored the unobserved heterogeneity while controlling for the driver, vehicle, and roadway factors. Compared with truck passengers, occupants in passenger vehicles are six times and ten times more likely to suffer minor injuries and serious/fatal injuries, respectively. Importantly, regardless of whether passenger vehicle drivers undertook intentional or unintentional actions, the crashes are more likely to associate with more severe injury outcomes. In addition, crashes occurring late at night and in early mornings are often correlated with more severe injuries. Such associations between explanatory factors and injury severity are found to vary across the passenger vehicle-truck crashes, and such significant variations of estimated parameters further confirmed the validity of applying the random-parameter model. More implications based on the results and suggestions in terms of safe driving are discussed.

Exploring microscopic driving volatility in naturalistic driving environment prior to involvement in safety critical events-Concept of event-based driving volatility.

The sequence of instantaneous driving decisions and its variations, known as driving volatility, prior to involvement in safety critical events can be a leading indicator of safety. This study focuses on the component of "driving volatility matrix" related to specific normal and safety-critical events, named "event-based volatility." The research issue is characterizing volatility in instantaneous driving decisions in the longitudinal and lateral directions, and how it varies across drivers involved in normal driving, crash, and/or near-crash events. To explore the issue, a rigorous quasi-experimental study design is adopted to help compare driving behaviors in normal vs unsafe outcomes. Using a unique real-world naturalistic driving database from the 2nd Strategic Highway Research Program (SHRP), a test set of 9593 driving events featuring 2.2 million temporal samples of real-world driving are analyzed. This study features a plethora of kinematic sensors, video, and radar spatiotemporal data about vehicle movement and therefore offers the opportunity to initiate such exploration. By using information related to longitudinal and lateral accelerations and vehicular jerk, 24 different aggregate and segmented measures of driving volatility are proposed that captures variations in extreme instantaneous driving decisions. In doing so, careful attention is given to the issue of intentional vs. unintentional volatility. The volatility indices, as leading indicators of near-crash and crash events, are then linked with safety critical events, crash propensity, and other event specific explanatory variables. Owing to the presence of unobserved heterogeneity and omitted variable bias, fixed- and random-parameter discrete choice models are developed that relate crash propensity to unintentional driving volatility and other factors. Statistically significant evidence is found that driver volatilities in near-crash and crash events are significantly greater than volatility in normal driving events. After controlling for traffic, roadway, and unobserved factors, the results suggest that greater intentional volatility increases the likelihood of both crash and near-crash events. A one-unit increase in intentional volatility is associated with positive vehicular jerk in longitudinal direction increases the chance of crash and near-crash outcome by 15.79 and 12.52 percentage points, respectively. Importantly, intentional volatility in positive vehicular jerk in lateral direction has more negative consequences than intentional volatility in positive vehicular jerk in longitudinal direction. Compared to acceleration/deceleration, vehicular jerk can better characterize the volatility in microscopic instantaneous driving decisions prior to involvement in safety critical events. Finally, the magnitudes of correlations exhibit significant heterogeneity, and that accounting for the heterogeneous effects in the modeling framework can provide more reliable and accurate results. The study demonstrates the value of quasi-experimental study design and big data analytics for understanding extreme driving behaviors in safe vs. unsafe driving outcomes.

Examining correlations between motorcyclist's conspicuity, apparel related factors and injury severity score: Evidence from new motorcycle crash causation study.

Motorcyclists are vulnerable road users at a particularly high risk of serious injury or death when involved in a crash. In order to evaluate key risk factors in motorcycle crashes, this study quantifies how different "policy-sensitive" factors correlate with injury severity, while controlling for rider and crash specific factors as well as other observed/unobserved factors. The study analyzes data from 321 motorcycle injury crashes from a comprehensive US DOT FHWA's Motorcycle Crash Causation Study (MCCS). These were all non-fatal injury crashes that are representative of the vast majority (82%) of motorcycle crashes. An anatomical injury severity scoring system, termed as Injury Severity Score (ISS), is analyzed providing an overall score by accounting for the possibility of multiple injuries to different body parts of a rider. An ISS ranges from 1 to 75, averaging at 10.32 for this sample (above 9 is considered serious injury), with a spike at 1 (very minor injury). Preliminary cross-tabulation analysis mapped ISS to the Abbreviated Injury Scale (AIS) injury classification and examined the strength of associations between the two. While the study finds a strong correlation between AIS and ISS classification (Kendall's tau of 0.911), significant contrasts are observed in that, when compared to ISS, AIS tends to underestimate the severity of an injury sustained by a rider. For modeling, fixed and random parameter Tobit modeling frameworks were used in a corner-solution setting to account for the left-tail spike in the distribution of ISS and to account for unobserved heterogeneity. The developed random parameters Tobit framework additionally accounts for the interactive effects of key risk factors, allowing for possible correlations among random parameters. A correlated random parameter Tobit model significantly out-performed the uncorrelated random parameter Tobit and fixed parameter Tobit models. While controlling for various other factors, we found that motorcycle-specific shoes and retroreflective upper body clothing correlate with lower ISS on-average by 5.94 and 1.88 units respectively. Riders with only partial helmet coverage on-average sustained more severe injuries, whereas, riders with acceptable helmet fit had lower ISS Methodologically, not only do the individual effects of several key risk factors vary significantly across observations in the form of random parameters, but the interactions between unobserved factors characterizing random parameters significantly influence the injury severity score as well. The implications of the findings are discussed.

An analysis of highway work zone safety practices in Pakistan.

Highway Work Zones (HWZs) present a major hazard for road users, construction workers and equipment, and significantly contribute to occurrence of road crashes worldwide. The present study focuses on analysing the current state of safety measures at HWZs in Pakistan. A more direct approach is adopted by comparing safety measures at randomly selected HWZs in Pakistan with well-established safety procedures in Manual on Uniform Traffic Control Devices (MUTCD). HWZ safety measures such as traffic signs, markings, safety measures for pedestrians, workers and construction machinery, police enforcement, speed control measures, provision of advance warning area, buffer spaces, transition areas, and tapers for eight different HWZs were studied and compared with MUTCD standards. The results revealed that majority of the HWZs in Pakistan do not conform to any standard layout especially for safety and speed control measures. An enhanced need for special efforts towards improving safety at HWZs in Pakistan is highlighted.

A heterogeneity based case-control analysis of motorcyclist's injury crashes: Evidence from motorcycle crash causation study.

The main objective of this study is to quantify how different "policy-sensitive" factors are associated with risk of motorcycle injury crashes, while controlling for rider-specific, psycho-physiological, and other observed/unobserved factors. The analysis utilizes data from a matched case-control design collected through the FHWA's Motorcycle Crash Causation Study. In particular, 351 cases (motorcyclists involved in injury crashes) are analyzed vis-à-vis similarly-at-risk 702 matched controls (motorcyclists not involved in crashes). Unlike traditional conditional estimation of relative risks, the paper presents heterogeneity based statistical analysis that accounts for the possibility of both within and between matched case-control variations. Overall, the correlations between key risk factors and injury crash propensity exhibit significant observed and unobserved heterogeneity. The results of best-fit random parameters logit model with heterogeneity-in-means show that riders with partial helmet coverage (U.S. DOT compliant helmets with partial coverage, least intrusive covering only the top half of the cranium) have a significantly lower risk of injury crash involvement. Lack of motorcycle rider conspicuity captured by dark (red) upper body clothing is associated with significantly higher injury crash risk (odds ratio 3.87, 95% CI: 1.63, 9.61). Importantly, a rider's motorcycle-oriented lower clothing (e.g., cannot easily get stuck in the machinery) significantly lowers the odds of injury crash involvement. Regarding the effectiveness of training, formal motorcycle driving training in recent years was associated with lower injury crash propensity. Finally, riders with less sleep prior to crash/interview exhibited 1.97 times higher odds of crash involvement compared to riders who had more than 5 h of sleep. Methodologically, the conclusion is that the correlations of several rider, exposure, apparel, and riding history related factors with crash risk are not homogeneous and in fact vary in magnitude as well as direction. The study results indicate the need to develop appropriate countermeasures, such as refresher motorcycle training courses, prevention of sleep-deprived/fatigued riding, and riding under the influence of alcohol and drugs.

Contributory fault and level of personal injury to drivers involved in head-on collisions: Application of copula-based bivariate ordinal models.

The main objective of this study is to simultaneously investigate the degree of injury severity sustained by drivers involved in head-on collisions with respect to fault status designation. This is complicated to answer due to many issues, one of which is the potential presence of correlation between injury outcomes of drivers involved in the same head-on collision. To address this concern, we present seemingly unrelated bivariate ordered response models by analyzing the joint injury severity probability distribution of at-fault and not-at-fault drivers. Moreover, the assumption of bivariate normality of residuals and the linear form of stochastic dependence implied by such models may be unduly restrictive. To test this, Archimedean copula structures and normal mixture marginals are integrated into the joint estimation framework, which can characterize complex forms of stochastic dependencies and non-normality in residual terms. The models are estimated using 2013 Virginia police reported two-vehicle head-on collision data, where exactly one driver is at-fault. The results suggest that both at-fault and not-at-fault drivers sustained serious/fatal injuries in 8% of crashes, whereas, in 4% of the cases, the not-at-fault driver sustained a serious/fatal injury with no injury to the at-fault driver at all. Furthermore, if the at-fault driver is fatigued, apparently asleep, or has been drinking the not-at-fault driver is more likely to sustain a severe/fatal injury, controlling for other factors and potential correlations between the injury outcomes. While not-at-fault vehicle speed affects injury severity of at-fault driver, the effect is smaller than the effect of at-fault vehicle speed on at-fault injury outcome. Contrarily, and importantly, the effect of at-fault vehicle speed on injury severity of not-at-fault driver is almost equal to the effect of not-at-fault vehicle speed on injury outcome of not-at-fault driver. Compared to traditional ordered probability models, the study provides evidence that copula based bivariate models can provide more reliable estimates and richer insights. Practical implications of the results are discussed.

Do safety performance functions used for predicting crash frequency vary across space? Applying geographically weighted regressions to account for spatial heterogeneity.

Safety Performance Functions (SPFs) provide a basis for identifying locations where countermeasures can be effective. While SPFs in the Highway Safety Manual (HSM) were calibrated based on data from select states, calibration factors can be developed to localize SPFs to other states. Calibration factors typically provide a coarse adjustment-time and space stationarity of associations between crash frequencies and various factors is still assumed, implying that the SPF functional form is transferable. However, with increasing availability of statewide geo-referenced safety data, new spatial analysis methods, and increasing computational power, it is possible to relax the stationarity assumption. Specifically, to address spatial heterogeneity in SPFs, this study proposes relaxing SPFs (referring to them as Localized SPFs (L-SPFs)) that can be developed by using sophisticated geo-spatial modeling techniques that allow correlates of crash frequencies to vary in space. For demonstration, a 2013 geo-referenced freeway crash and traffic database from Virginia is used. As a potential methodological alternative, crash frequencies are predicted by estimating Geographically Weighted Negative Binomial Regressions. This model significantly outperforms the traditional negative binomial model in terms of model goodness-of-fit, providing a better and fuller understanding of spatial variations in modeled relationships. Our study results uncover significant spatial variations in parameter estimates for Annual Average Daily Traffic (AADT) and segment length. Ignoring such variations can result in prediction errors. The results indicate low transferability of a single statewide SPF highlighting the importance of developing L-SPFs. From a practical standpoint, L-SPFs can better predict crash frequencies and support prioritizing safety improvements in specific locations.