Rule extraction from an optimized neural network for traffic crash frequency modeling

Effects of geometric attributes of horizontal and sag vertical curve combinations on freeway crash frequency

The geometric design of the combinations of horizontal and sag vertical curves (sag combinations or sag combined curves) is vital to road safety. However, there is little research that investigates the safety effects of their geometric attributes based on the analysis of real-world crash data. To this end, the crash, traffic, geometric design, and roadway configuration data are collected from 157 sag combinations on six freeways in Washington State, during 2011-2017. Poisson, negative binomial (NB), hierarchical Poisson, and hierarchical NB models are developed for analyzing the crash frequency of sag combinations. The models are estimated and compared in the context of Bayesian inference. The results indicate that significant over-dispersion and cross-group heterogeneity exist in the crash data and that the hierarchical NB model yields the best overall performance. The parameter estimates show that: five geometric attributes, including horizontal curvature, vertical curvature, departure grade, the ratio of horizontal curvature to vertical curvature, and the layout of front dislocation, have significant effects on the crash frequency of sag combinations. Freeway section length, annual average daily traffic, and speed limits are also important predictors of crash frequency. The analysis results and the proposed model are useful for evaluating the safety performance of freeway sag combinations and optimizing their geometric design based on substantive safety evaluation.

Extraction of the association rules from artificial neural networks based on the multiobjective optimization

Artificial Neural Network (ANN) is one of the powerful techniques of machine learning. It has shown its effectiveness in both prediction and classification problems. However, in some fields there is still some reticence towards their use mainly the fact that they do not justify their answers. The lack of transparency on how ANN makes decisions motivated us to develop our rule extraction algorithm that extracts comprehensible rules with high accuracy and high fidelity. The aim is to generate a set of rules that mimic the decision of ANN and cover a larger set of patterns. The obtained rule sets should satisfy a well-balanced trade-off between the fidelity, the accuracy and the comprehensibility. The proposed algorithm consists of a three steps: ANN learning phase, rule extraction phase and rule simplification phase. The rule extraction phase is based on the extraction of the association rules while the rules simplification procedure is based on the laws of Boolean algebra. To evaluate the performance of our algorithm, the system has been studied using four datasets, and then compared with other rule extraction methods. The results show that our proposal offers a small set of rules having the highest accuracy and fidelity values.

On the interpretability of machine learning methods in crash frequency modeling and crash modification factor development

Machine learning (ML) model interpretability has attracted much attention recently given the promising performance of ML methods in crash frequency studies. Extracting accurate relationship between risk factors and crash frequency is important for understanding the causal effects of risk factors and developing safety countermeasures. However, there is no study that comprehensively summarizes ML model interpretation methods and provides guidance for safety researchers and practitioners. This research aims to fill this gap. Model-based and post-hoc ML interpretation methods are critically evaluated and compared to study their suitability in crash frequency modeling. These methods include classification and regression tree (CART), multivariate adaptive regression splines (MARS), Local Interpretable Model-agnostic Explanations (LIME), Local Sensitivity Analysis (LSA), Partial Dependence Plots (PDP), Global Sensitivity Analysis (GSA), and SHapley Additive exPlanations (SHAP). Model-based interpretation methods cannot reveal the detailed interaction relationships among risk factors. LIME can only be used to analyze the effects of a risk factor at the prediction level. LSA and PDP assume that different risk factors are independently distributed. Both GSA and SHAP can account for the potential correlation among risk factors. However, only SHAP can visualize the detailed relationships between crash outcomes and risk factors. This study also demonstrates the potential and benefits of using ML and SHAP to derive Crash Modification Factors (CMF). Finally, it is emphasized that statistical and ML models may not directly differentiate causation from correlation. Understanding the differences between them is critical for developing reliable safety countermeasures.

Parametric and Non-Parametric Analyses for Pedestrian Crash Severity Prediction in Great Britain

The study aims to investigate the factors that are associated with fatal and severe vehicle–pedestrian crashes in Great Britain by developing four parametric models and five non-parametric tools to predict the crash severity. Even though the models have already been applied to model the pedestrian injury severity, a comparative analysis to assess the predictive power of such modeling techniques is limited. Hence, this study contributes to the road safety literature by comparing the models by their capabilities of identifying the significant explanatory variables, and by their performances in terms of the F-measure, the G-mean, and the area under curve. The analyses were carried out using data that refer to the vehicle–pedestrian crashes that occurred in the period of 2016–2018. The parametric models confirm their advantages in offering easy-to-interpret outputs and understandable relations between the dependent and independent variables, whereas the non-parametric tools exhibited higher classification accuracies, identified more explanatory variables, and provided insights into the interdependencies among the factors. The study results suggest that the combined use of parametric and non-parametric methods may effectively overcome the limits of each group of methods, with satisfactory prediction accuracies and the interpretation of the factors contributing to fatal and serious crashes. In the conclusion, several engineering, social, and management pedestrian safety countermeasures are recommended.

Quantifying and comparing the effects of key risk factors on various types of roadway segment crashes with LightGBM and SHAP

Understanding and quantifying the effects of risk factors on crash frequency is of great importance for developing cost-effective safety countermeasures. In this paper, the effects of key crash contributing factors on total crashes and crashes of different collision types are analyzed separately and compared. A novel Machine Learning (ML) method, Light Gradient Boosting Machine (LightGBM), is introduced to model a Texas dataset consisting of vehicle crashes occurred from 2015 to 2017. Compared with other commonly used ML methods such as eXtreme Gradient Boosting (XGBoost), LightGBM performs significantly better in terms of mean absolute error (MAE) and root mean squared error (RMSE). In addition, the SHapley Additive explanation (SHAP) approach is employed to interpret the LightGBM outputs. Significant risk factors are identified, including speed limits, area type, number of lanes, roadway functional class, shoulder width and shoulder type. With the SHAP method, the importance, total effects, and main and interaction effects of risk factors are quantified. The results suggest that the importance of risk factors vary across collision types. Speed limit is a more important risk factor than right/left shoulder width, lane width, and median width for Rear-End (RE) crashes, while the opposite relationship is found for Run-Off-Road (ROR) crashes. Also, it is found that narrow lanes (8ft to 11ft) increase the risk for all types of crashes (i.e., Total, ROR, and RE) in this study. For road segments with 5 or 6 lanes in both directions combined, a lane width greater than or equal to 12ft may help reduce the risk of all types of crashes. These results have important implications for developing accurate crash modification factors and cost-effective safety countermeasures.

Use of Decision Trees for the Development of Decision Support Systems for the Control of Grinding Circuits

Grinding circuits can exhibit strong nonlinear behaviour, which may make automatic supervisory control difficult and, as a result, operators still play an important role in the control of many of these circuits. Since the experience among operators may be highly variable, control of grinding circuits may not be optimal and could benefit from automated decision support. This could be based on heuristics from process experts, but increasingly could also be derived from plant data. In this paper, the latter approach, based on the use of decision trees to develop rule-based decision support systems, is considered. The focus is on compact, easy to understand rules that are well supported by the data. The approach is demonstrated by means of an industrial case study. In the case study, the decision trees were not only able to capture operational heuristics in a compact intelligible format, but were also able to identify the most influential variables as reliably as more sophisticated models, such as random forests.

Application of machine learning and genetic optimization algorithms for modeling and optimizing soybean yield using its component traits

Improving genetic yield potential in major food grade crops such as soybean (Glycine max L.) is the most sustainable way to address the growing global food demand and its security concerns. Yield is a complex trait and reliant on various related variables called yield components. In this study, the five most important yield component traits in soybean were measured using a panel of 250 genotypes grown in four environments. These traits were the number of nodes per plant (NP), number of non-reproductive nodes per plant (NRNP), number of reproductive nodes per plant (RNP), number of pods per plant (PP), and the ratio of number of pods to number of nodes per plant (P/N). These data were used for predicting the total soybean seed yield using the Multilayer Perceptron (MLP), Radial Basis Function (RBF), and Random Forest (RF), machine learning (ML) algorithms, individually and collectively through an ensemble method based on bagging strategy (E-B). The RBF algorithm with highest Coefficient of Determination (R2) value of 0.81 and the lowest Mean Absolute Errors (MAE) and Root Mean Square Error (RMSE) values of 148.61 kg.ha-1, and 185.31 kg.ha-1, respectively, was the most accurate algorithm and, therefore, selected as the metaClassifier for the E-B algorithm. Using the E-B algorithm, we were able to increase the prediction accuracy by improving the values of R2, MAE, and RMSE by 0.1, 0.24 kg.ha-1, and 0.96 kg.ha-1, respectively. Furthermore, for the first time in this study, we allied the E-B with the genetic algorithm (GA) to model the optimum values of yield components in an ideotype genotype in which the yield is maximized. The results revealed a better understanding of the relationships between soybean yield and its components, which can be used for selecting parental lines and designing promising crosses for developing cultivars with improved genetic yield potential.

Safety critical event prediction through unified analysis of driver and vehicle volatilities: Application of deep learning methods

Transportation safety is highly correlated with driving behavior, especially human error playing a key role in a large portion of crashes. Modern instrumentation and computational resources allow for the monitorization of driver, vehicle, and roadway/environment to extract leading indicators of crashes from multi-dimensional data streams. To quantify variations that are beyond normal in driver behavior and vehicle kinematics, the concept of volatility is applied. The study measures driver-vehicle volatilities using the naturalistic driving data. By integrating and fusing multiple real-time streams of data, i.e., driver distraction, vehicular movements and kinematics, and instability in driving, this study aims to predict occurrence of safety critical events and generate appropriate feedback to drivers and surrounding vehicles. The naturalistic driving data is used which contains 7566 normal driving events, and 1315 severe events (i.e., crash and near-crash), vehicle kinematics, and driver behavior collected from more than 3500 drivers. In order to capture the local dependency and volatility in time-series data 1D-Convolutional Neural Network (1D-CNN), Long Short-Term Memory (LSTM), and 1DCNN-LSTM are applied. Vehicle kinematics, driving volatility, and impaired driving (in terms of distraction) are used as the input parameters. The results reveal that the 1DCNN-LSTM model provides the best performance, with 95.45% accuracy and prediction of 73.4% of crashes with a precision of 95.67%. Additional features are extracted with the CNN layers and temporal dependency between observations is addressed, which helps the network learn driving patterns and volatile behavior. The model can be used to monitor driving behavior in real-time and provide warnings and alerts to drivers in low-level automated vehicles, reducing their crash risk.

Applying machine learning and geolocation techniques to social media data (Twitter) to develop a resource for urban planning

With all the recent attention focused on big data, it is easy to overlook that basic vital statistics remain difficult to obtain in most of the world. What makes this frustrating is that private companies hold potentially useful data, but it is not accessible by the people who can use it to track poverty, reduce disease, or build urban infrastructure. This project set out to test whether we can transform an openly available dataset (Twitter) into a resource for urban planning and development. We test our hypothesis by creating road traffic crash location data, which is scarce in most resource-poor environments but essential for addressing the number one cause of mortality for children over five and young adults. The research project scraped 874,588 traffic related tweets in Nairobi, Kenya, applied a machine learning model to capture the occurrence of a crash, and developed an improved geoparsing algorithm to identify its location. We geolocate 32,991 crash reports in Twitter for 2012–2020 and cluster them into 22,872 unique crashes during this period. For a subset of crashes reported on Twitter, a motorcycle delivery service was dispatched in real-time to verify the crash and its location; the results show 92% accuracy. To our knowledge this is the first geolocated dataset of crashes for the city and allowed us to produce the first crash map for Nairobi. Using a spatial clustering algorithm, we are able to locate portions of the road network (<1%) where 50% of the crashes identified occurred. Even with limitations in the representativeness of the data, the results can provide urban planners with useful information that can be used to target road safety improvements where resources are limited. The work shows how twitter data might be used to create other types of essential data for urban planning in resource poor environments.

The application of artificial neural networks in modeling and predicting the effects of melatonin on morphological responses of citrus to drought stress

Drought stress as one of the most devastating abiotic stresses affects agricultural and horticultural productivity in many parts of the world. The application of melatonin can be considered as a promising approach for alleviating the negative impact of drought stress. Modeling of morphological responses to drought stress can be helpful to predict the optimal condition for improving plant productivity. The objective of the current study is modeling and predicting morphological responses (leaf length, number of leaves/plants, crown diameter, plant height, and internode length) of citrus to drought stress, based on four input variables including melatonin concentrations, days after applying treatments, citrus species, and level of drought stress, using different Artificial Neural Networks (ANNs) including Generalized Regression Neural Network (GRNN), Radial basis function (RBF), and Multilayer Perceptron (MLP). The results indicated a higher accuracy of GRNN as compared to RBF and MLP. The great accordance between the experimental and predicted data of morphological responses for both training and testing processes support the excellent efficiency of developed GRNN models. Also, GRNN was connected to Non-dominated Sorting Genetic Algorithm-II (NSGA-II) to optimize input variables for obtaining the best morphological responses. Generally, the validation experiment showed that ANN-NSGA-II can be considered as a promising and reliable computational tool for studying and predicting plant morphological and physiological responses to drought stress.

Early warning model for passenger disturbance due to flight delays

Disruptive behavior by passengers delayed at airport terminals not only affects personal safety but also reduces civil aviation efficiency and passenger satisfaction. This study investigated the causal mechanisms of disruptive behavior by delayed passengers in three aspects: environmental, managerial, and personal. Data on flight delays at Shenzhen Airport in 2018 were collected and analyzed. The main factors leading to disruptive behavior by delayed passengers were identified, and an early warning model for disturbances was developed using multiple logistic regression and a back-propagation(BP) neural network. The results indicated that the proposed model and method were feasible. Compared to the logistic regression model, the BP neural network model had advantages in predicting disturbances by delayed passengers, showing higher prediction accuracy. The BP network weight analysis method was used to obtain the influence weight of each factor on behavior change of delayed passengers. The influence weight of different factors was obtained, providing an assistant decision-making method to address disruption from flight-delayed passengers.

Fifty Years of Accident Analysis & Prevention: A Bibliometric and Scientometric Overview

Accident Analysis & Prevention (AA&P) is a leading academic journal established in 1969 that serves as an important scientific communication platform for road safety studies. To celebrate its 50^th anniversary of publishing outstanding and insightful studies, a multi-dimensional statistical and visualized analysis of the AA&P publications between 1969 and 2018 was performed using the Web of Science (WoS) Core Collection database, bibliometrics and mapping-knowledge-domain (MKD) analytical methods, and scientometric tools. It was shown that the annual number of AA&P's publications has grown exponentially and that over the course of its development, AA&P has been a leader in the field of road safety, both in terms of innovation and dissemination. By determining its key source countries and organizations, core authors, highly co-cited published documents, and high burst-strength publications, we showed that AA&P's areas of focus include the "effects of hazard and risk perception on driving behavior", "crash frequency modeling analysis", "intentional driving violations and aberrant driving behavior", "epidemiology, assessment and prevention of road traffic injuries", and "crash-injury severity modeling analysis". Furthermore, the key burst papers that have played an important role in advancing research and guiding AA&P in new directions - particularly those in the fields of crash frequency and crash-injury severity modeling analyses were identified. Finally, a modified Haddon matrix in the era of intelligent, connected and autonomous transportation systems is proposed to provide new insights into the emerging generation of road safety studies.

A new spatial count data model with Bayesian additive regression trees for accident hot spot identification

The identification of accident hot spots is a central task of road safety management. Bayesian count data models have emerged as the workhorse method for producing probabilistic rankings of hazardous sites in road networks. Typically, these methods assume simple linear link function specifications, which, however, limit the predictive power of a model. Furthermore, extensive specification searches are precluded by complex model structures arising from the need to account for unobserved heterogeneity and spatial correlations. Modern machine learning (ML) methods offer ways to automate the specification of the link function. However, these methods do not capture estimation uncertainty, and it is also difficult to incorporate spatial correlations. In light of these gaps in the literature, this paper proposes a new spatial negative binomial model which uses Bayesian additive regression trees to endogenously select the specification of the link function. Posterior inference in the proposed model is made feasible with the help of the Pólya-Gamma data augmentation technique. We test the performance of this new model on a crash count data set from a metropolitan highway network. The empirical results show that the proposed model performs at least as well as a baseline spatial count data model with random parameters in terms of goodness of fit and site ranking ability.

A comparative study of two methods to predict the incidence of hepatitis B in Guangxi, China

In recent years, the incidence of hepatitis B (HB) in Guangxi is higher than that of the national level; it has been increasing, so it is urgent to do a good predictive research of HB incidence, which can help analyze the early warning of hepatitis B in Guangxi, China. In the study, the feasibility of predicting HB incidence in Guangxi by autoregressive integrated moving average (ARIMA) model method and Elman neural network (ElmanNN) method was discussed respectively, and the prediction accuracy of the two models was compared. Finally, we established the ARIMA (0, 1, 1) model and ElmanNN with 8 neurons. Both ARIMA (0, 1, 1) model and ElmanNN model had good performance, and their prediction accuracy were high. The fitting and prediction root-mean-square error (RMSE) and mean absolute error (MAE) of ElmanNN were smaller than those of ARIMA (0, 1, 1) model, which indicated that ElmanNN was superior to ARIMA (0, 1, 1) model in predicting the incidence of hepatitis B in Guangxi. Based on the ElmanNN, the HB incidence from September 2019 to December 2020 in Guangxi was predicted, the predicted results showed that the incidence of HB in 2020 was slightly higher than that in 2019 and the change trend was similar to that in 2019, for 2021 and beyond, the ElmanNN model could be used to continue the predictive analysis.

An optimal network screening method of hotspot identification for highway crashes with dynamic site length

We propose a novel network screening method for hotspot (i.e., sites that suffer from high collision concentration and have high potential for safety improvement) identification based on the optimization framework to maximize the total summation of a selected safety measure for all hotspots considering a resource constraint for conducting detailed engineering studies (DES). The proposed method allows the length of each hotspot to be determined dynamically based on constraints the users impose. The calculation of the Dynamic Site Length (DSL) method is based on Dynamic Programming, and it is shown to be effective to find the close-to-optimal solution with computationally feasible complexity. The screening method has been demonstrated using historical crash data from extended freeway routes in San Francisco, California. Using the Empirical Bayesian (EB) estimate as a safety measure, we compare the performance of the proposed DSL method with other conventional screening methods, Sliding Window (SW) and Continuous Risk Profile (CRP), in terms of their optimal objective value (i.e., performance of detecting sites under the highest risk). Moreover, their spatio-temporal consistency is compared through the site and method consistency tests. Findings show that DSL can outperform SW and CRP in investigating more hotspots under the same amount of resources allocated to DES by pinpointing hotspot locations with greater accuracy and showing improved spatio-temporal consistency.

Effect of Music Listening on Physiological Condition, Mental Workload, and Driving Performance with Consideration of Driver Temperament

This paper presents the study on the association between in-vehicle music listening, physiological and psychological response, and driving performance, using the driving simulator approach, with which personality (temperament) was considered. The performance indicators considered were the standard deviation of speed, lane crossing frequency, perceived mental workload, and mean and variability of heart rate. Additionally, effects of the presence of music and music genre (light music versus rock music) were considered. Twenty participants of different personalities (in particular five, four, seven, and four being choleric, sanguine, phlegmatic, and melancholic, respectively) completed a total of 60 driving simulator tests. Results of mixed analysis of variance (M-ANOVA) indicated that the effects of music genre and driver character on driving performance were significant. The arousal level perceived mental workload, standard deviation of speed, and frequency of lane crossing were higher when driving under the influence of rock music than that when driving under the influence of light music or an absence of music. Additionally, phlegmatic drivers generally had lower arousal levels and choleric drivers had a greater mental workload and were more likely distracted by music listening. Such findings should imply the development of cost-effective driver education, training, and management measures that could mitigate driver distraction. Therefore, the safety awareness and safety performance of drivers could be enhanced.

Roadway traffic crash prediction using a state-space model based support vector regression approach

Conventional traffic crash analyzing methods focus on identifying the relationship between traffic crash outcomes and impact risk factors and explaining the effects of risk factors, which ignore the changes of roadway systems and can lead to inaccurate results in traffic crash predictions. To address this issue, an innovative two-step method is proposed and a support vector regression (SVR) model is formulated into state-space model (SSM) framework for traffic crash prediction. The SSM was developed in the first step to identify the dynamic evolution process of the roadway systems that are caused by the changes of traffic flow and predict the changes of impact factors in roadway systems. Using the predicted impact factors, the SVR model was incorporated in the second step to perform the traffic crash prediction. A five-year dataset that obtained from 1152 roadway segments in Tennessee was employed to validate the model effectiveness. The proposed models result in an average prediction MAPE of 7.59%, a MAE of 0.11, and a RMSD of 0.32. For the performance comparison, a SVR model and a multivariate negative binomial (MVNB) model were developed to do the same task. The results show that the proposed model has superior performances in terms of prediction accuracy compared to the SVR and MVNB models. Compared to the SVR and MVNB models, the benefit of incorporating a state-space model to identify the changes of roadway systems is significant evident in the proposed models for all crash types, and the prediction accuracy that measured by MAPE can be improved by 4.360% and 6.445% on average, respectively. Apart from accuracy improvement, the proposed models are more robust and the predictions can retain a smoother pattern. Furthermore, the results show that the proposed model has a more precise and synchronized response behavior to the high variations of the observed data, especially for the phenomenon of extra zeros.

Identification of Vehicle-Pedestrian Collision Hotspots at the Micro-Level Using Network Kernel Density Estimation and Random Forests: A Case Study in Shanghai, China

The improvement of pedestrian safety plays a crucial role in developing a safe and friendly walking environments, which can contribute to urban sustainability. A preliminary step in improving pedestrian safety is to identify hazardous road locations for pedestrians. This study proposes a framework for the identification of vehicle-pedestrian collision hot spots by integrating the information about both the likelihood of the occurrence of vehicle-pedestrian collisions and the potential for the reduction in vehicle-pedestrian crashes. First, a vehicle-pedestrian collision density surface was produced via network kernel density estimation. By assigning a threshold value, possible vehicle-pedestrian hot spots were identified. To obtain the potential for vehicle-pedestrian collision reduction, random forests was employed to model the density with a set of variables describing vehicle and pedestrian flows. The potential for crash reduction was then measured as the difference between the observed vehicle-pedestrian crash density and the prediction produced by the random forests models. The final hotspots were determined by excluding those with a crash reduction value of no more than zero. The method was applied to the identification of hazardous road locations for pedestrians in a district in Shanghai, China. The result indicates that the method is useful for decision-making support.

The effect of gender, occupation and experience on behavior while driving on a freeway deceleration lane based on field operational test data

Deceleration lanes improve traffic flow by reducing interference, increasing capacity and enhancing safety. However, accident rates are higher on these interchange segments than on other freeway segments. It is important to attempt to reduce traffic accidents on these interchange segments by further exploring the behavior of different types of drivers on a highway deceleration lane. In this study, with field operational test (FOT) data from 89 driving instances (derived from 46 participants driving the test road twice) on a typical freeway deceleration lane, section speed profiles, vehicle trajectories, lane position and other key parameters were obtained. The lane-change characteristics and speed profiles of drivers with different genders, occupations and experiences were analyzed. The significant disparities between them reflects the risk associated with different groups of drivers. The study shows that male drivers changed to the outside lane earlier; professional drivers and experienced drivers made the last lane change as early as possible to enter the deceleration lane; and the speed of the vehicles entering the exit ramp was significantly higher than the speed limit. This research work provides ground truth data for deceleration lane design, driver ability training and off-ramp traffic safety management.

Examining the influence of link function misspecification in conventional regression models for developing crash modification factors

This study further examined the use of regression models for developing crash modification factors (CMFs), specifically focusing on the misspecification in the link function. The primary objectives were to validate the accuracy of CMFs derived from the commonly used regression models (i.e., generalized linear models or GLMs with additive linear link functions) when some of the variables have nonlinear relationships and quantify the amount of bias as a function of the nonlinearity. Using the concept of artificial realistic data, various linear and nonlinear crash modification functions (CM-Functions) were assumed for three variables. Crash counts were randomly generated based on these CM-Functions. CMFs were then derived from regression models for three different scenarios. The results were compared with the assumed true values. The main findings are summarized as follows: (1) when some variables have nonlinear relationships with crash risk, the CMFs for these variables derived from the commonly used GLMs are all biased, especially around areas away from the baseline conditions (e.g., boundary areas); (2) with the increase in nonlinearity (i.e., nonlinear relationship becomes stronger), the bias becomes more significant; (3) the quality of CMFs for other variables having linear relationships can be influenced when mixed with those having nonlinear relationships, but the accuracy may still be acceptable; and (4) the misuse of the link function for one or more variables can also lead to biased estimates for other parameters. This study raised the importance of the link function when using regression models for developing CMFs.

A Bayesian spatial random parameters Tobit model for analyzing crash rates on roadway segments

This study develops a Bayesian spatial random parameters Tobit model to analyze crash rates on road segments, in which both spatial correlation between adjacent sites and unobserved heterogeneity across observations are accounted for. The crash-rate data for a three-year period on road segments within a road network in Florida, are collected to compare the performance of the proposed model with that of a (fixed parameters) Tobit model and a spatial (fixed parameters) Tobit model in the Bayesian context. Significant spatial effect is found in both spatial models and the results of Deviance Information Criteria (DIC) show that the inclusion of spatial correlation in the Tobit regression considerably improves model fit, which indicates the reasonableness of considering cross-segment spatial correlation. The spatial random parameters Tobit regression has lower DIC value than does the spatial Tobit regression, suggesting that accommodating the unobserved heterogeneity is able to further improve model fit when the spatial correlation has been considered. Moreover, the random parameters Tobit model provides a more comprehensive understanding of the effect of speed limit on crash rates than does its fixed parameters counterpart, which suggests that it could be considered as a good alternative for crash rate analysis.

The interactive effect on injury severity of driver-vehicle units in two-vehicle crashes

INTRODUCTION

This study sets out to investigate the interactive effect on injury severity of driver-vehicle units in two-vehicle crashes.

METHOD

A Bayesian hierarchical ordered logit model is proposed to relate the variation and correlation of injury severity of drivers involved in two-vehicle crashes to the factors of both driver-vehicle units and the crash configurations. A total of 6417 crash records with 12,834 vehicles involved in Florida are used for model calibration.

RESULTS

The results show that older, female and not-at-fault drivers and those without use of safety equipment are more likely to be injured but less likely to injure the drivers in the other vehicles. New vehicles and lower speed ratios are associated with lower injury degree of both drivers involved. Compared with automobiles, vans, pick-ups, light trucks, median trucks, and heavy trucks possess better self-protection and stronger aggressivity. The points of impact closer to the driver's seat in general indicate a higher risk to the own drivers while engine cover and vehicle rear are the least hazardous to other drivers. Head-on crashes are significantly more severe than angle and rear-end crashes. We found that more severe crashes occurred on roadways than on shoulders or safety zones.

CONCLUSIONS

Based on these results, some suggestions for traffic safety education, enforcement and engineering are made. Moreover, significant within-crash correlation is found in the crash data, which demonstrates the applicability of the proposed model.