Chunjiao Dong

Multivariate random-parameters zero-inflated negative binomial regression model: An application to estimate crash frequencies at intersections

Crash data are collected through police reports and integrated with road inventory data for further analysis. Integrated police reports and inventory data yield correlated multivariate data for roadway entities (e.g., segments or intersections). Analysis of such data reveals important relationships that can help focus on high-risk situations and coming up with safety countermeasures. To understand relationships between crash frequencies and associated variables, while taking full advantage of the available data, multivariate random-parameters models are appropriate since they can simultaneously consider the correlation among the specific crash types and account for unobserved heterogeneity. However, a key issue that arises with correlated multivariate data is the number of crash-free samples increases, as crash counts have many categories. In this paper, we describe a multivariate random-parameters zero-inflated negative binomial (MRZINB) regression model for jointly modeling crash counts. The full Bayesian method is employed to estimate the model parameters. Crash frequencies at urban signalized intersections in Tennessee are analyzed. The paper investigates the performance of MZINB and MRZINB regression models in establishing the relationship between crash frequencies, pavement conditions, traffic factors, and geometric design features of roadway intersections. Compared to the MZINB model, the MRZINB model identifies additional statistically significant factors and provides better goodness of fit in developing the relationships. The empirical results show that MRZINB model possesses most of the desirable statistical properties in terms of its ability to accommodate unobserved heterogeneity and excess zero counts in correlated data. Notably, in the random-parameters MZINB model, the estimated parameters vary significantly across intersections for different crash types.

Differences in passenger car and large truck involved crash frequencies at urban signalized intersections: An exploratory analysis

The influence of intersection features on safety has been examined extensively because intersections experience a relatively large proportion of motor vehicle conflicts and crashes. Although there are distinct differences between passenger cars and large trucks-size, operating characteristics, dimensions, and weight-modeling crash counts across vehicle types is rarely addressed. This paper develops and presents a multivariate regression model of crash frequencies by collision vehicle type using crash data for urban signalized intersections in Tennessee. In addition, the performance of univariate Poisson-lognormal (UVPLN), multivariate Poisson (MVP), and multivariate Poisson-lognormal (MVPLN) regression models in establishing the relationship between crashes, traffic factors, and geometric design of roadway intersections is investigated. Bayesian methods are used to estimate the unknown parameters of these models. The evaluation results suggest that the MVPLN model possesses most of the desirable statistical properties in developing the relationships. Compared to the UVPLN and MVP models, the MVPLN model better identifies significant factors and predicts crash frequencies. The findings suggest that traffic volume, truck percentage, lighting condition, and intersection angle significantly affect intersection safety. Important differences in car, car-truck, and truck crash frequencies with respect to various risk factors were found to exist between models. The paper provides some new or more comprehensive observations that have not been covered in previous studies.

Identifying the factors contributing to the severity of truck-involved crashes

To address the dilemma between the need for truck transportation and the costs related to truck-involved crashes, the key is to identify the risk factors that significantly affect truck-involved crashes. The objective of this research is to estimate the effects of the characteristics of traffic, driver, geometry, and environment on severity of truck-involved crashes. Based on four crash severity categories (fatal/incapacitating, non-incapacitating, possible injury, and no injury/property damage only), a multinomial logit model is conducted to identify the risk factors. The investigation of risk ratios indicates that lower traffic volume with higher truck percentage is associated with more serious traffic crash with fatal/incapacitating injury while a non-standard geometric design is the main cause of non-incapacitating crashes. The influences of weather are significant for the possible-injury crashes while driver condition is the principal cause of no-injury/property-damage-only crashes. In addition, the statistical results demonstrate that the influence of the truck percentage is significant. One-unit change in the truck percentage will cause more than one times probability of being in an injury.

Predicting expressway crash frequency using a random effect negative binomial model: A case study in China

To investigate the relationship between crash frequency and potential influence factors, the accident data for events occurring on a 50km long expressway in China, including 567 crash records (2006-2008), were collected and analyzed. Both the fixed-length and the homogeneous longitudinal grade methods were applied to divide the study expressway section into segments. A negative binomial (NB) model and a random effect negative binomial (RENB) model were developed to predict crash frequency. The parameters of both models were determined using the maximum likelihood (ML) method, and the mixed stepwise procedure was applied to examine the significance of explanatory variables. Three explanatory variables, including longitudinal grade, road width, and ratio of longitudinal grade and curve radius (RGR), were found as significantly affecting crash frequency. The marginal effects of significant explanatory variables to the crash frequency were analyzed. The model performance was determined by the relative prediction error and the cumulative standardized residual. The results show that the RENB model outperforms the NB model. It was also found that the model performance with the fixed-length segment method is superior to that with the homogeneous longitudinal grade segment method.

Analyzing the effectiveness of implemented highway safety laws for traffic safety across U.S. states

ABSTRACT Since highway safety laws vary greatly from state to state in the U.S., there is a need to analyze the effectiveness and performances of the implemented highway safety laws. The random-parameter zero-truncated negative binomial (RZTNB) models are proposed to analyze the effects of highway safety laws on fatal crashes at state levels. The results show that the proposed models are useful in describing the relationships between the fatal crashes and the explanatory variables with better goodness of fit. By accounting for the heterogeneities, the RZTNB model outperforms the negative binomial model and reveals new insights. The findings indicate that (1) compared to the secondary ban, the primary handheld cellphone ban is more effective; (2) establishing reasonable and acceptable speed limits can enhance the traffic safety; and (3) the implemented speed camera system and ignition interlock device have weaknesses and alternative methods should be considered when upgrading laws and regulations.

Analyzing injury crashes using random-parameter bivariate regression models

ABSTRACT This paper proposes a random-parameter bivariate zero-inflated negative binomial (RBZINB) regression model for analyzing the effects of investigated variables on crash frequencies. A Bayesian approach is employed as the estimation method, which has the strength of accounting for the uncertainties related to models and parameter values. The modeling framework has been applied to the bivariate injury crash counts obtained from 1000 intersections in Tennessee over a five-year period. The results reveal that the proposed RBZINB model outperforms other investigated models and provides a superior fit. The proposed RBZINB model is useful in gaining new insights into how crash occurrences are influenced by the risk factors. In addition, the empirical studies show that the proposed RBZINB model has a smaller prediction bias and variance, as well as more accurate coverage probability in estimating model parameters and crash-free probabilities.

Combining the Statistical Model and Heuristic Model to Predict Flow Rate

Statistical and heuristic models have been proposed as applications that are well suited to short-term traffic flow prediction. However, traffic flow data often contain both linear and nonlinear patterns. Therefore, neither statistical nor heuristic models are adequate to model and predict traffic flow data. This paper discusses the relative merits of statistical and heuristic models for traffic flow prediction and summarizes the findings from a comparative study for their performances. Based on that, a hybrid support vector machine for regression (SVR) methodology that combines both statistical and heuristic models is proposed to take advantage of their unique strength in linear and nonlinear modeling. In addition, the dynamics of spatial-temporal patterns in traffic flow are considered in this study, and they are treated as part of the input data. The experiment results based on the real field data of a test region in Beijing suggest that the proposed method is able to provide accurate and reliable flow rate predictions under both low- and high-flow traffic conditions. The benefit from combining statistical and heuristic models as opposed to not combining [autoregressive integrated moving average (ARIMA) model or Elman neural network (NN)] is much more evident in all cases, and the accuracy can be improved by 9.04% on average. Regarding the incorporation of a combination of temporal and spatial characteristics, the use of the hybrid model is found helpful in a one-step-ahead flow rate prediction under high-flow traffic conditions, with a maximum 9.52% improvement on accuracy.

A study of factors affecting intersection crash frequencies using random-parameter multivariate zero-inflated models

Recent research demonstrates the appropriateness of multivariate regression models in crash count modelling when one specific type of crash counts needs to be analysed, since they can better handle the correlated issues in multiple crash counts. In this paper, a random-parameter multivariate zero-inflated Poisson (RMZIP) regression model is proposed as an alternative multivariate methodology for jointly modelling crash counts simultaneously. Using this RMZIP model, we are able to account for the heterogeneity due to the unobserved roadway geometric design features and traffic characteristics. Our formulation also has the merit of handling excess zeros in correlated crash counts, a phenomenon that is commonly found in practice. The Bayesian method is employed to estimate the model parameters. We use the proposed modelling framework to predict crash frequencies at urban signalized intersections in Tennessee. To investigate the model performances, three models – a fixed-parameter MZIP model, a random-parameter multivariate negative binomial (RMNB) model, and a random-parameter multivariate zero-inflated negative binomial (RMZINB) model – have been employed as the comparison methods. The comparison results show that the proposed RMZIP models provide a satisfied statistical fit with more variables producing statistically significant parameters. In other word, the RMZIP models have the potential to provide a fuller understanding of how the factors affect crash frequencies on specific roadway intersections. A variety of variables are found to significantly influence the crash frequencies by varying magnitudes. These variables result in random parameters and thereby their effects on crash frequencies are found to vary significantly across the sampled intersections.

Exploring the effects of state highway safety laws and sociocultural characteristics on fatal crashes.

ABSTRACT Objective: Distinguished from the traditional perspectives in crash analyses, which examined the effects of geometric design features, traffic factors, and other relevant attributes on the crash frequencies of roadway entities, our study focuses on exploring the effects of highway safety laws, as well as sociocultural characteristics, on fatal crashes across states. Methods: Law and regulation related data were collected from the Insurance Institute for Highway Safety, State Highway Safety Offices, and Governors Highway Safety Association. A variety of sociodemographic characteristics were obtained from the U.S. Census Bureau. In addition, cultural factors and other attributes from a variety of resources are considered and incorporated in the modeling process. These data and fatal crash counts were collected for the 50 U.S. states and the District of Columbia and were analyzed using zero-truncated negative binomial (ZTNB) regression models. Results: The results show that, in law and regulation–related factors, the use of speed cameras, no handheld cell phone ban, limited handheld cell phone ban, and no text messaging ban are found to have significant effects on fatal crashes. Regarding sociocultural characteristics, married couples with both husband and wife in the labor force are found to be associated with lower crash frequencies, the ratios of workers traveling to work by carpool, those driving alone, workers working outside the county of residence, language other than English and limited English fluency, and the number of licensed drivers are found to be associated with higher crash frequencies. Conclusions: Through reviewing and modeling existing state highway safety laws and sociocultural characteristics, the results reveal new insights that could influence policy making. In addition, the results would benefit amending existing laws and regulations and provide testimony about highway safety issues before lawmakers consider new legislation.

Effects of Roadway Geometric Design Features on Frequency of Truck-Related Crashes

Because truck-related crashes are a socioeconomic concern that can result in tremendous loss of life and property, unbiased, relatively accurate estimations of crash frequency are essential. A data set from Tennessee was used to examine the effects of roadway geometric design features and other relevant attributes on the frequency of truck-related crashes. Negative binomial (NB) and zero-inflated NB (ZINB) models were proposed to identify the risk factors that had significant effects on the frequency of crashes that involved large trucks. Differences in truck-related crashes were investigated across collision vehicle types, and three crash count models—total truck related, car–truck, and truck only—were developed under the ZINB and NB frameworks. Elasticities were estimated for these crash count models to identify the most critical variables contributing to crashes. Findings suggest that the ZINB models have most of the desirable statistical properties (i.e., better goodness of fit and more significant variables identified). Model results revealed seven factors significant for the frequency of truck-related crashes regardless of crash type: annual average daily traffic, percentage of trucks, segment length, degree of horizontal curvature, terrain type, median type, and posted speed limit. Results of elasticity estimation reveal that the percentage of trucks is the most critical variable of all explanatory variables for the frequency of truck-related crashes.