Modeling crash outcome probabilities at rural intersections: application of hierarchical binomial logistic models

It is important to examine the nature of the relationships between roadway, environmental, and traffic factors and motor vehicle crashes, with the aim to improve the collective understanding of causal mechanisms involved in crashes and to better predict their occurrence. Statistical models of motor vehicle crashes are one path of inquiry often used to gain these initial insights. Recent efforts have focused on the estimation of negative binomial and Poisson regression models (and related deviants) due to their relatively good fit to crash data. Of course analysts constantly seek methods that offer greater consistency with the data generating mechanism (motor vehicle crashes in this case), provide better statistical fit, and provide insight into data structure that was previously unavailable. One such opportunity exists with some types of crash data, in particular crash-level data that are collected across roadway segments, intersections, etc. It is argued in this paper that some crash data possess hierarchical structure that has not routinely been exploited. This paper describes the application of binomial multilevel models of crash types using 548 motor vehicle crashes collected from 91 two-lane rural intersections in the state of Georgia. Crash prediction models are estimated for angle, rear-end, and sideswipe (both same direction and opposite direction) crashes. The contributions of the paper are the realization of hierarchical data structure and the application of a theoretically appealing and suitable analysis approach for multilevel data, yielding insights into intersection-related crashes by crash type.     

Investigation of pedestrian crashes on two-way two-lane rural roads in Ethiopia

Understanding pedestrian crash causes and contributing factors in developing countries is critically important as they account for about 55% of all traffic crashes. Not surprisingly, considerable attention in the literature has been paid to road traffic crash prediction models and methodologies in developing countries of late. Despite this interest, there are significant challenges confronting safety managers in developing countries. For example, in spite of the prominence of pedestrian crashes occurring on two-way two-lane rural roads, it has proven difficult to develop pedestrian crash prediction models due to a lack of both traffic and pedestrian exposure data. This general lack of available data has further hampered identification of pedestrian crash causes and subsequent estimation of pedestrian safety performance functions. The challenges are similar across developing nations, where little is known about the relationship between pedestrian crashes, traffic flow, and road environment variables on rural two-way roads, and where unique predictor variables may be needed to capture the unique crash risk circumstances. This paper describes pedestrian crash safety performance functions for two-way two-lane rural roads in Ethiopia as a function of traffic flow, pedestrian flows, and road geometry characteristics. In particular, random parameter negative binomial model was used to investigate pedestrian crashes. The models and their interpretations make important contributions to road crash analysis and prevention in developing countries. They also assist in the identification of the contributing factors to pedestrian crashes, with the intent to identify potential design and operational improvements.     

Estimation of rear-end crash potential using vehicle trajectory data

Recent advancement in traffic surveillance systems has allowed for obtaining more detailed vehicular movement such as individual vehicle trajectory data. Understanding the characteristics of interactions between leading vehicle and following in the traffic flow stream is a backbone for designing and evaluating more sophisticated traffic and vehicle control strategies. This study proposes a methodology for estimating rear-end crash potential, as a probabilistic measure, in real time based on the analysis of vehicular movements. The methodology presented in this study consists of two components. The first estimates the probability that a vehicle's trajectory belonging to either ‘changing lane’ or ‘going straight’. A binary logistic regression (BLR) is used to model the lane-changing decision of the subject vehicle. The other component derives crash probability by an exponential decay function using time-to-collision (TTC) between the subject vehicle and the front vehicle. Also, an aggregated measure, crash risk index (CRI) is used in the analysis to accumulate rear-end crash potential for each subject vehicle. The result of this study can be used in developing traffic control and information systems, in particular, for crash prevention.     

Poisson, Poisson-gamma and zero-inflated regression models of motor vehicle crashes: balancing statistical fit and theory

There has been considerable research conducted over the last 20 years focused on predicting motor vehicle crashes on transportation facilities. The range of statistical models commonly applied includes binomial, Poisson, Poisson-gamma (or negative binomial), zero-inflated Poisson and negative binomial models (ZIP and ZINB), and multinomial probability models. Given the range of possible modeling approaches and the host of assumptions with each modeling approach, making an intelligent choice for modeling motor vehicle crash data is difficult. There is little discussion in the literature comparing different statistical modeling approaches, identifying which statistical models are most appropriate for modeling crash data, and providing a strong justification from basic crash principles. In the recent literature, it has been suggested that the motor vehicle crash process can successfully be modeled by assuming a dual-state data-generating process, which implies that entities (e.g., intersections, road segments, pedestrian crossings, etc.) exist in one of two states-perfectly safe and unsafe. As a result, the ZIP and ZINB are two models that have been applied to account for the preponderance of "excess" zeros frequently observed in crash count data. The objective of this study is to provide defensible guidance on how to appropriate model crash data. We first examine the motor vehicle crash process using theoretical principles and a basic understanding of the crash process. It is shown that the fundamental crash process follows a Bernoulli trial with unequal probability of independent events, also known as Poisson trials. We examine the evolution of statistical models as they apply to the motor vehicle crash process, and indicate how well they statistically approximate the crash process. We also present the theory behind dual-state process count models, and note why they have become popular for modeling crash data. A simulation experiment is then conducted to demonstrate how crash data give rise to "excess" zeros frequently observed in crash data. It is shown that the Poisson and other mixed probabilistic structures are approximations assumed for modeling the motor vehicle crash process. Furthermore, it is demonstrated that under certain (fairly common) circumstances excess zeros are observed-and that these circumstances arise from low exposure and/or inappropriate selection of time/space scales and not an underlying dual state process. In conclusion, carefully selecting the time/space scales for analysis, including an improved set of explanatory variables and/or unobserved heterogeneity effects in count regression models, or applying small-area statistical methods (observations with low exposure) represent the most defensible modeling approaches for datasets with a preponderance of zeros.     

Exploring the effects of roadway characteristics on the frequency and severity of head-on crashes: Case studies from Malaysian Federal Roads

Head-on crashes are among the most severe collision types and of great concern to road safety authorities. Therefore, it justifies more efforts to reduce both the frequency and severity of this collision type. To this end, it is necessary to first identify factors associating with the crash occurrence. This can be done by developing crash prediction models that relate crash outcomes to a set of contributing factors. This study intends to identify the factors affecting both the frequency and severity of head-on crashes that occurred on 448 segments of five federal roads in Malaysia. Data on road characteristics and crash history were collected on the study segments during a 4-year period between 2007 and 2010. The frequency of head-on crashes were fitted by developing and comparing seven count-data models including Poisson, standard negative binomial (NB), random-effect negative binomial, hurdle Poisson, hurdle negative binomial, zero-inflated Poisson, and zero-inflated negative binomial models. To model crash severity, a random-effect generalized ordered probit model (REGOPM) was used given a head-on crash had occurred. With respect to the crash frequency, the random-effect negative binomial (RENB) model was found to outperform the other models according to goodness of fit measures. Based on the results of the model, the variables horizontal curvature, terrain type, heavy-vehicle traffic, and access points were found to be positively related to the frequency of head-on crashes, while posted speed limit and shoulder width decreased the crash frequency. With regard to the crash severity, the results of REGOPM showed that horizontal curvature, paved shoulder width, terrain type, and side friction were associated with more severe crashes, whereas land use, access points, and presence of median reduced the probability of severe crashes. Based on the results of this study, some potential countermeasures were proposed to minimize the risk of head-on crashes.     

A crash-prediction model for road tunnels

Considerable research has been carried out into open roads to establish relationships between crashes and traffic flow, geometry of infrastructure and environmental factors, whereas crash-prediction models for road tunnels, have rarely been investigated. In addition different results have been sometimes obtained regarding the effects of traffic and geometry on crashes in road tunnels. However, most research has focused on tunnels where traffic and geometric conditions, as well as driving behaviour, differ from those in Italy. Thus, in this paper crash prediction-models that had not yet been proposed for Italian road tunnels have been developed. For the purpose, a 4-year monitoring period extending from 2006 to 2009 was considered. The tunnels investigated are single-tube ones with unidirectional traffic. The Bivariate Negative Binomial regression model, jointly applied to non-severe crashes (accidents involving material-damage only) and severe crashes (fatal and injury accidents only), was used to model the frequency of accident occurrence. The year effect on severe crashes was also analyzed by the Random Effects Binomial regression model and the Negative Multinomial regression model. Regression parameters were estimated by the Maximum Likelihood Method. The Cumulative Residual Method was used to test the adequacy of the regression model through the range of annual average daily traffic per lane. The candidate set of variables was: tunnel length (L), annual average daily traffic per lane (AADT_L), percentage of trucks (%Tr), number of lanes (N_L), and the presence of a sidewalk. Both for non-severe crashes and severe crashes, prediction-models showed that significant variables are: L, AADT_L, %Tr, and N_L. A significant year effect consisting in a systematic reduction of severe crashes over time was also detected. The analysis developed in this paper appears to be useful for many applications such as the estimation of accident reductions due to improvement in existing tunnels and/or to modifications of traffic control systems, as well as for the prediction of accidents when different tunnel design options are compared.     

In-depth analysis of drivers’ merging behavior and rear-end crash risks in work zone merging areas

This study investigates the drivers’ merging behavior and the rear-end crash risk in work zone merging areas during the entire merging implementation period from the time of starting a merging maneuver to that of completing the maneuver. With the merging traffic data from a work zone site in Singapore, a mixed probit model is developed to describe the merging behavior, and two surrogate safety measures including the time to collision (TTC) and deceleration rate to avoid the crash (DRAC) are adopted to compute the rear-end crash risk between the merging vehicle and its neighboring vehicles. Results show that the merging vehicle has a bigger probability of completing a merging maneuver quickly under one of the following situations: (i) the merging vehicle moves relatively fast; (ii) the merging lead vehicle is a heavy vehicle; and (iii) there is a sizable gap in the adjacent through lane. Results indicate that the rear-end crash risk does not monotonically increase as the merging vehicle speed increases. The merging vehicle's rear-end crash risk is also affected by the vehicle type. There is a biggest increment of rear-end crash risk if the merging lead vehicle belongs to a heavy vehicle. Although the reduced remaining distance to work zone could urge the merging vehicle to complete a merging maneuver quickly, it might lead to an increased rear-end crash risk. Interestingly, it is found that the rear-end crash risk could be generally increased over the elapsed time after the merging maneuver being triggered.     

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.     

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.     

A multinomial logit model-Bayesian network hybrid approach for driver injury severity analyses in rear-end crashes

Rear-end crash is one of the most common types of traffic crashes in the U.S. A good understanding of its characteristics and contributing factors is of practical importance. Previously, both multinomial Logit models and Bayesian network methods have been used in crash modeling and analysis, respectively, although each of them has its own application restrictions and limitations. In this study, a hybrid approach is developed to combine multinomial logit models and Bayesian network methods for comprehensively analyzing driver injury severities in rear-end crashes based on state-wide crash data collected in New Mexico from 2010 to 2011. A multinomial logit model is developed to investigate and identify significant contributing factors for rear-end crash driver injury severities classified into three categories: no injury, injury, and fatality. Then, the identified significant factors are utilized to establish a Bayesian network to explicitly formulate statistical associations between injury severity outcomes and explanatory attributes, including driver behavior, demographic features, vehicle factors, geometric and environmental characteristics, etc. The test results demonstrate that the proposed hybrid approach performs reasonably well. The Bayesian network reference analyses indicate that the factors including truck-involvement, inferior lighting conditions, windy weather conditions, the number of vehicles involved, etc. could significantly increase driver injury severities in rear-end crashes. The developed methodology and estimation results provide insights for developing effective countermeasures to reduce rear-end crash injury severities and improve traffic system safety performance.