Aemal J. Khattak

Motor vehicle drivers’ injuries in train–motor vehicle crashes

The objectives of this research were to: (1) identify a more suitable model for modeling injury severity of motor vehicle drivers involved in train-motor vehicle crashes at highway-rail grade crossings from among three commonly used injury severity models and (2) to investigate factors associated with injury severity levels of motor vehicle drivers involved in train-motor vehicle crashes at such crossings. The 2009-2013 highway-rail grade crossing crash data and the national highway-rail crossing inventory data were combined to produce the analysis dataset. Four-year (2009-2012) data were used for model estimation while 2013 data were used for model validation. The three injury severity levels-fatal, injury and no injury-were based on the reported intensity of motor-vehicle drivers injuries at highway-rail grade crossings. The three injury severity models evaluated were: ordered probit, multinomial logit and random parameter logit. A comparison of the three models based on different criteria showed that the random parameter logit model and multinomial logit model were more suitable for injury severity analysis of motor vehicle drivers involved in crashes at highway-rail grade crossings. Some of the factors that increased the likelihood of more severe crashes included higher train and vehicle speeds, freight trains, older drivers, and female drivers. Where feasible, reducing train and motor vehicle speeds and nighttime lighting may help reduce injury severities of motor vehicle drivers.

Weather impacts on single-vehicle truck crash injury severity.

INTRODUCTIONnThe focus of this paper is on illustrating the feasibility of aggregating data from disparate sources to investigate the relationship between single-vehicle truck crash injury severity and detailed weather conditions. Specifically, this paper presents: (a) a methodology that combines detailed 15-min weather station data with crash and roadway data, and (b) an empirical investigation of the effects of weather on crash-related injury severities of single-vehicle truck crashes.nnnMETHODnRandom parameters ordinal and multinomial regression models were used to investigate crash injury severity under different weather conditions, taking into account the individual unobserved heterogeneity. The adopted methodology allowed consideration of environmental, roadway, and climate-related variables in single-vehicle truck crash injury severity.nnnRESULTS AND CONCLUSIONSnResults showed that wind speed, rain, humidity, and air temperature were linked with single-vehicle truck crash injury severity. Greater recorded wind speed added to the severity of injuries in single-vehicle truck crashes in general. Rain and warmer air temperatures were linked to more severe crash injuries in single-vehicle truck crashes while higher levels of humidity were linked to less severe injuries. Random parameters ordered logit and multinomial logit, respectively, revealed some individual heterogeneity in the data and showed that integrating comprehensive weather data with crash data provided useful insights into factors associated with single-vehicle truck crash injury severity.nnnPRACTICAL APPLICATIONSnThe research provided a practical method that combined comprehensive 15-min weather station data with crash and roadway data, thereby providing useful insights into crash injury severity of single-vehicle trucks. Those insights are useful for future truck driver educational programs and for truck safety in different weather conditions.

Comparison of four statistical and machine learning methods for crash severity prediction

Crash severity prediction models enable different agencies to predict the severity of a reported crash with unknown severity or the severity of crashes that may be expected to occur sometime in the future. This paper had three main objectives: comparison of the performance of four statistical and machine learning methods including Multinomial Logit (MNL), Nearest Neighbor Classification (NNC), Support Vector Machines (SVM) and Random Forests (RF), in predicting traffic crash severity; developing a crash costs-based approach for comparison of crash severity prediction methods; and investigating the effects of data clustering methods comprising K-means Clustering (KC) and Latent Class Clustering (LCC), on the performance of crash severity prediction models. The 2012-2015 reported crash data from Nebraska, United States was obtained and two-vehicle crashes were extracted as the analysis data. The dataset was split into training/estimation (2012-2014) and validation (2015) subsets. The four prediction methods were trained/estimated using the training/estimation dataset and the correct prediction rates for each crash severity level, overall correct prediction rate and a proposed crash costs-based accuracy measure were obtained for the validation dataset. The correct prediction rates and the proposed approach showed NNC had the best prediction performance in overall and in more severe crashes. RF and SVM had the next two sufficient performances and MNL was the weakest method. Data clustering did not affect the prediction results of SVM, but KC improved the prediction performance of MNL, NNC and RF, while LCC caused improvement in MNL and RF but weakened the performance of NNC. Overall correct prediction rate had almost the exact opposite results compared to the proposed approach, showing that neglecting the crash costs can lead to misjudgment in choosing the right prediction method.

Distracted Motor Vehicle Driving at Highway–Rail Grade Crossings

Distracted driving of a motor vehicle increases a drivers susceptibility to crash involvement. Such driving in the vicinity of highway–rail grade crossings (HRGCs) is especially perilous because the distracted driver may not be aware of the presence of an approaching train. Train-involved motor vehicle crashes invariably cause more severe injuries and greater property damage than other motor vehicle crashes. Because distracted driving at HRGCs is somewhat underresearched, the objectives of this research were to (a) investigate the occurrence of distracted motor vehicle driving at HRGCs, (b) identify characteristics of those distracted motorists, and (c) ascertain empirically factors associated with distracted driving at HRGCs. Data were collected with video recordings at six HRGCs in Nebraska and were statistically analyzed. Results showed that about one-third of drivers observed at HRGCs were distracted. The presence of intersecting roads near the crossings increased distracted driving. Drivers of multiunit trucks at HRGCs were less often distracted than other drivers at HRGCs. The presence of front-seat passengers in vehicles significantly contributed to driver distraction. Findings from the research support the need for driver education about distracted driving at HRGCs. Enforcement of distracted driving laws at HRGCs as well as strengthening of existing laws may also reduce instances of distracted driving at HRGCs.

Injury severity in crashes reported in proximity of rail crossings: The role of driver inattention

ABSTRACT The objective of this article is to quantitatively investigate the impact of motor vehicle driver inattention on severity of drivers injuries sustained in crashes reported at or near highway-rail grade crossings (HRGCs). The Nebraska Department of Roads (USA) supplied 2002 to 2013 data on reported motor vehicle crashes at or near HRGCs. Besides information on driver behavior, the data set included information on seatbelt usage, presence of passengers, drivers age, gender, weather, train involvement, highway speed limit, road surface condition, and lighting condition. Random parameters binary logit regression models were used to investigate of the role of inattentive driving in severity of injuries sustained by motor vehicle drivers. The main findings were that driver inattention led to more severe injuries compared to attentive driving in vicinity of HRGCs. The probability of a driver suffering an injury in a single-vehicle crash increased by 9.7% when the driver was inattentive; in two-vehicle crashes the probability of at least one injured driver increased by 14.6% when any one of the involved drivers was inattentive to the driving task. The effects of inattentive driving on driver injury outcomes were not statistically different than the effects of driving under influence of alcohol/drugs or aggressive driving.

Factors associated with self-reported inattentive driving at highway-rail grade crossings

This research identified factors associated with inattentive driving at Highway-Rail Grade Crossings (HRGCs) by investigating drivers self-reported inattentive driving experiences and factors pertaining to their socioeconomic, personality, attitudinal, and other characteristics. A random selection of 2500 households in Nebraska received a survey questionnaire designed for licensed motor vehicle drivers; respondents returned 980 questionnaires. Factor analysis identified latent variables evaluating drivers patience and inclination to wait for trains, attitudes toward new technology, law enforcement or education regarding HRGC safety, and the propensity to commit serious traffic violations at HRGCs. The investigation utilized a structural equation model for analysis. This model indicated that drivers with a higher risk of inattentive driving at HRGCs were: female, younger in age, from households with higher incomes, with shorter tenure (in years) in their current city of residence, more frequently used HRGCs, received less information on safety at HRGCs, had less patience to wait for trains to pass and had less interest in safety improvement technology, law enforcement or safety education at HRGCs. These research findings provide useful information for future research and to policy makers for improving public safety. Additionally, the results are useful for safety educational program providers for targeted program delivery to drivers that are more vulnerable to distracted driving at HRGCs.

Investigation of Train Warning Times and Gate Violations

The objective of this research was to investigate the relationship between train warning times and instances of motorists driving around fully deployed crossing gates at a highway–rail grade crossing equipped with median barriers (lane separators). The purpose of the median barriers is to deter motorists from driving around fully deployed gates. The premise for this research was the expectation that longer train warning times were accompanied by more numerous occurrences of driving around fully deployed crossing gates, notwithstanding the presence of median barriers. Data on motorists’ opportunities to drive around fully deployed crossing gates, actual instances of driving around those gates, warning times experienced by motorists, and some additional variables were collected at a study site during a 2-month period. The study site was equipped with median barriers on both sides of the crossing to discourage driving around fully deployed gates. Data analysis showed that after violation opportunities and certain other variables were controlled for, longer train warning times were indeed associated with more frequent instances of motorists driving around gates. The finding emphasizes the provision of adequate but not excessively long warning times at highway–rail grade crossings equipped with dual-quadrant gates. On the basis of the research findings, the paper provides options for improving safety at highway–rail grade crossings.

Injury Severity of Truck Drivers in Crashes at Highway-Rail Grade Crossings in the United States

The physical and operational characteristics of large trucks distinguish them from other types of vehicles in terms of facility design needs and safety requirements. A critical node in the surface transportation network is the highway-rail grade crossing (HRGC) because it represents a conflict point between different modes of transportation. The focus of this research was to identify factors related to different injury severity levels of truck/truck-trailer drivers in crashes reported at HRGCs. This study utilized a mixed logit model to investigate injury severity of those drivers and relied on 2007–2014 Federal Railroad Administration (FRA) crash and inventory data involving trucks/truck-trailers. Results showed that truck/truck-trailer drivers’ injuries in crashes reported at HRGCs were positively associated with train speed, when train struck the road user (truck/truck-trailer), when the driver “went around crossing gates”, older drivers, crashes reported in rural areas, and crashes at crossings with a minimum crossing angle of 60–90 degrees. Presence of crossbucks, gates, track obstructions, and HRGCs located within 500 feet of a highway were associated with relatively less severe driver injuries. The paper provides recommendations for safety improvements at HRGCs and recommendations for future research.

A clustering approach to injury severity in pedestrian-train crashes at highway-rail grade crossings

ABSTRACT This research studied potential factors associated with pedestrian injury severity levels sustained in train-pedestrian crashes at highway-rail grade crossings (HRGCs) using the Federal Railroad Administrations ten-year data. The analysis focused on nonsuicide pedestrian crashes and took into consideration the unobserved heterogeneity. Latent class clustering (LCC) addressed unobserved heterogeneity by creating distinct subgroups with relatively homogeneous attributes within each subgroup. HRGC inventory variables were the basis for the LCC; the process split the dataset into five distinguished clusters. Binary logit models for each cluster and the complete data set were estimated. A generalized linear mixed model, based on the complete data set, allowed examination of the clustering and comparison of the modeling results. Findings provided justification for the use of LCC as the first step in accounting for unobserved heterogeneity. Different HRGC, pedestrian, and crash characteristics were associated with pedestrian injury severity across different clusters. Higher train speed was associated with more severe injury propensity, regardless of the conditions of the HRGCs. Other variables including freight train involvement, train hitting pedestrian, HRGCs with the absence of flashing lights, advance warnings, rural areas, lower visibility conditions, and older pedestrians increased pedestrian injury severity levels with varying effects in different clusters.

A Spatial–Temporal Multitask Collaborative Learning Model for Multistep Traffic Flow Prediction

Traffic flow prediction is a fundamental capability for successful deployment of intelligent transportation systems. Traditionally, multiple related prediction tasks are undertaken individually, without considering the relationships among the tasks. This paper presents a spatial–temporal multitask collaborative learning model for multistep traffic flow prediction. The novel approach learns multiple related prediction tasks collaboratively by extracting and utilizing appropriate shared information across tasks. First, each traffic flow prediction problem is formulated as a supervised machine-learning task. Next, the sparse features shared across multiple tasks are learned by solving a regularized optimization problem. To deal with the non-convex and non-smooth challenges, the optimization problem is then transformed into an equivalent convex problem. Finally, the global optimal solution of the convex problem is found by solving a variation of this problem using an alternating minimization algorithm. The proposed model incorporates both the spatial correlation between different observed stations and the intrinsic relationship between different traffic flow parameters, as well as the coarse-grain temporal correlation between different days in a week and the fine-grain temporal correlation between different prediction steps. Application of the proposed model to a real case study for SR180-E freeway in Fresno, California showed its effectiveness, robustness and advantages for multistep traffic flow prediction.