Xuedong Yan

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.

Using hierarchical tree-based regression model to predict train-vehicle crashes at passive highway-rail grade crossings.

This paper applies a nonparametric statistical method, hierarchical tree-based regression (HTBR), to explore train-vehicle crash prediction and analysis at passive highway-rail grade crossings. Using the Federal Railroad Administration (FRA) database, the research focuses on 27 years of train-vehicle accident history in the United States from 1980 through 2006. A cross-sectional statistical analysis based on HTBR is conducted for public highway-rail grade crossings that were upgraded from crossbuck-only to stop signs without involvement of other traffic-control devices or automatic countermeasures. In this study, HTBR models are developed to predict train-vehicle crash frequencies for passive grade crossings controlled by crossbucks only and crossbucks combined with stop signs respectively, and assess how the crash frequencies change after the stop-sign treatment is applied at the crossbuck-only-controlled crossings. The study results indicate that stop-sign treatment is an effective engineering countermeasure to improve safety at the passive grade crossings. Decision makers and traffic engineers can use the HTBR models to examine train-vehicle crash frequency at passive crossings and assess the potential effectiveness of stop-sign treatment based on specific attributes of the given crossings.

Relationships between Heavy Metal Concentrations in Roadside Topsoil and Distance to Road Edge Based on Field Observations in the Qinghai-Tibet Plateau, China

This study investigated the spatial distribution of copper (Cu), zinc (Zn), cadmium (Cd), lead (Pb), chromium (Cr), cobalt (Co), nickel (Ni) and arsenic (As) in roadside topsoil in the Qinghai-Tibet Plateau and evaluated the potential environmental risks of these roadside heavy metals due to traffic emissions. A total of 120 topsoil samples were collected along five road segments in the Qinghai-Tibet Plateau. The nonlinear regression method was used to formulize the relationship between the metal concentrations in roadside soils and roadside distance. The Hakanson potential ecological risk index method was applied to assess the degrees of heavy metal contaminations. The regression results showed that both of the heavy metals’ concentrations and their ecological risk indices decreased exponentially with the increase of roadside distance. The large R square values of the regression models indicate that the exponential regression method can suitably describe the relationship between heavy metal accumulation and roadside distance. For the entire study region, there was a moderate level of potential ecological risk within a 10 m roadside distance. However, Cd was the only prominent heavy metal which posed potential hazard to the local soil ecosystem. Overall, the rank of risk contribution to the local environments among the eight heavy metals was Cd > As > Ni > Pb > Cu > Co > Zn > Cr. Considering that Cd is a more hazardous heavy metal than other elements for public health, the local government should pay special attention to this traffic-related environmental issue.

Analyses of Rear-End Crashes Based on Classification Tree Models

Objective. Signalized intersections are accident-prone areas especially for rear-end crashes due to the fact that the diversity of the braking behaviors of drivers increases during the signal change. The objective of this article is to improve knowledge of the relationship between rear-end crashes occurring at signalized intersections and a series of potential traffic risk factors classified by driver characteristics, environments, and vehicle types. Methods. Based on the 2001 Florida crash database, the classification tree method and Quasi-induced exposure concept were used to perform the statistical analysis. Two binary classification tree models were developed in this study. One was used for the crash comparison between rear-end and non-rear-end to identify those specific trends of the rear-end crashes. The other was constructed for the comparison between striking vehicles/drivers (at-fault) and struck vehicles/drivers (not-at-fault) to find more complex crash pattern associated with the traffic attributes of driver, vehicle, and environment. Results. The modeling results showed that the rear-end crashes are over-presented in the higher speed limits (45–55 mph); the rear-end crash propensity for daytime is apparently larger than nighttime; and the reduction of braking capacity due to wet and slippery road surface conditions would definitely contribute to rear-end crashes, especially at intersections with higher speed limits. The tree model segmented drivers into four homogeneous age groups: < 21 years, 21–31 years, 32–75 years, and > 75 years. The youngest driver group shows the largest crash propensity; in the 21–31 age group, the male drivers are over-involved in rear-end crashes under adverse weather conditions and the 32–75 years drivers driving large size vehicles have a larger crash propensity compared to those driving passenger vehicles. Conclusions. Combined with the quasi-induced exposure concept, the classification tree method is a proper statistical tool for traffic-safety analysis to investigate crash propensity. Compared to the logistic regression models, tree models have advantages for handling continuous independent variables and easily explaining the complex interaction effect with more than two independent variables. This research recommended that at signalized intersections with higher speed limits, reducing the speed limit to 40 mph efficiently contribute to a lower accident rate. Drivers involved in alcohol use may increase not only rear-end crash risk but also the driver injury severity. Education and enforcement countermeasures should focus on the driver group younger than 21 years. Further studies are suggested to compare crash risk distributions of the driver age for other main crash types to seek corresponding traffic countermeasures.

Relationship between Heavy Metal Concentrations in Soils and Grasses of Roadside Farmland in Nepal

Transportation activities can contribute to accumulation of heavy metals in roadside soil and grass, which could potentially compromise public health and the environment if the roadways cross farmland areas. Particularly, heavy metals may enter the food chain as a result of their uptake by roadside edible grasses. This research was conducted to investigate heavy metal (Cu, Zn, Cd, and Pb) concentrations in roadside farmland soils and corresponding grasses around Kathmandu, Nepal. Four factors were considered for the experimental design, including sample type, sampling location, roadside distance, and tree protection. A total of 60 grass samples and 60 topsoil samples were collected under dry weather conditions. The Multivariate Analysis of Variance (MANOVA) results indicate that the concentrations of Cu, Zn, and Pb in the soil samples are significantly higher than those in the grass samples; the concentrations of Cu and Pb in the suburban roadside farmland are higher than those in the rural mountainous roadside farmland; and the concentrations of Cu and Zn at the sampling locations with roadside trees are significantly lower than those without tree protection. The analysis of transfer factor, which is calculated as the ratio of heavy-metal concentrations in grass to those in the corresponding soil, indicates that the uptake capabilities of heavy metals from soil to grass is in the order of Zn > Cu > Pb. Additionally, it is found that as the soils’ heavy-metal concentrations increase, the capability of heavy-metal transfer to the grass decreases, and this relationship can be characterized by an exponential regression model.

Train–Vehicle Crash Risk Comparison Between Before and After Stop Signs Installed at Highway–Rail Grade Crossings

Objective: The safety benefit of stop sign treatment employed at passive highway–rail crossings has been a subject of research for many years. The objectives of this research is to investigate whether and to what degree the crash rate has changed at previously passive grade crossings after stop signs were implemented and examine whether and how the crash characteristics (associated with vehicle type, crossing surrounding, crossing design, crash severity, etc.) changed subsequently. Methods: Federal Railroad Administration grade crossing databases during the 26-year period (1980–2005) were applied in this study. Among the stop-controlled grade crossings, a total of 7394 “target” crossings were identified to be once crossbucks controlled and subsequently upgraded with the installation of stop signs without the implementation of other traffic control devices during the study period. Each target crossing was further divided into two time periods: when it was controlled by crossbucks only (before) and when it was controlled by stop signs (after). Both annual crash rate analysis and crash propensity analysis of before–after stop sign installation are conducted to quantify the safety benefit of stop sign treatment. Results: It was found that during the 26-year period (1980–2005), the annual crash rates when the crossings were controlled by crossbucks-only were consistently higher than the crash rates when the crossings were controlled by stop signs. The further crash propensity analysis indicated that the stop sign treatment was especially effective at crossings with higher annual average daily traffic (AADT), advanced warning signs, sight distance problem, adverse lighting conditions; the motorist-stopped-on-crossing, did-not-stop, and injury crash risks were also significantly reduced after stop signs were applied. Conclusions: The finding of this study suggested that the vehicle volume should be included into the guideline for stop sign use. Therefore, engineers and decision makers are encouraged to routinely check available sight distances at passive crossings controlled by crossbucks only and add stop signs to the crossings with insufficient sight distances. Additionally, it is suggested that advanced warning signs should be jointly used at stop-controlled crossings to maximize the safety effect. However, stop signs were less effective at crossings with higher train speeds or track classifications, where active warning devices may be a better safety solution for grade crossings.

Evaluation of Effectiveness of Stop-Sign Treatment at Highway–Railroad Grade Crossings

The safety benefit of stop-sign treatment employed at passive highway–rail crossings has been a subject of research for many years. The objective of this study is to assess the effectiveness of the stop-sign treatment on crossing safety. Using the Federal Railroad Administration database, the research focused on 26 years of vehicle–train accident history in the United States from 1980 through 2005. A before-and-after and cross-sectional statistical analysis was conducted for 7,394 public highway–railroad grade crossings that were upgraded from cross buck only to stop signs without involvement in other traffic-control devices (TCDs) or automatic countermeasures. The study found that accident rates based on annual accident frequency per 1,000 crossings were significantly higher during the period when crossings were controlled by cross bucks only than when they were controlled by stop signs. Further, this study developed negative binomial accident prediction models, respectively, for paved and unpaved highway–rail grade-crossings that include effect for stop-sign treatment. Based on specific attributes of the current crossbuck-only-controlled crossings, decision makers and traffic engineers can use the models to examine the accident risks at crossings and assess the potential effectiveness of stop-sign treatment.

Development of a subway operation incident delay model using accelerated failure time approaches.

This study aims to develop a subway operational incident delay model using the parametric accelerated time failure (AFT) approach. Six parametric AFT models including the log-logistic, lognormal and Weibull models, with fixed and random parameters are built based on the Hong Kong subway operation incident data from 2005 to 2012, respectively. In addition, the Weibull model with gamma heterogeneity is also considered to compare the model performance. The goodness-of-fit test results show that the log-logistic AFT model with random parameters is most suitable for estimating the subway incident delay. First, the results show that a longer subway operation incident delay is highly correlated with the following factors: power cable failure, signal cable failure, turnout communication disruption and crashes involving a casualty. Vehicle failure makes the least impact on the increment of subway operation incident delay. According to these results, several possible measures, such as the use of short-distance and wireless communication technology (e.g., Wifi and Zigbee) are suggested to shorten the delay caused by subway operation incidents. Finally, the temporal transferability test results show that the developed log-logistic AFT model with random parameters is stable over time.

Investigation of work zone crash casualty patterns using association rules

Investigation of the casualty crash characteristics and contributory factors is one of the high-priority issues in traffic safety analysis. In this paper, we propose a method based on association rules to analyze the characteristics and contributory factors of work zone crash casualties. A case study is conducted using the Michigan M-94/I-94/I-94BL/I-94BR work zone crash data from 2004 to 2008. The obtained association rules are divided into two parts including rules with high-lift, and rules with high-support for the further analysis. The results show that almost all the high-lift rules contain either environmental or occupant characteristics. The majority of association rules are centered on specific characteristics, such as drinking driving, the highway with more than 4 lanes, speed-limit over 40mph and not use of traffic control devices. It should be pointed out that some stronger associated rules were found in the high-support part. With the network visualization, the association rule method can provide more understandable results for investigating the patterns of work zone crash casualties.

Effects of fog, driver experience and gender on driving behavior on S-curved road segments

Driving on curved roads has been recognized as a significant safety issue for many years. However, driver behavior and the interactions among variables that affect driver performance on curves is complicated and not well understood. Previous studies have investigated various factors that influence driver performance on right- or left-turn curves, but have paid little attention to the effects of foggy weather, driver experience and gender on driver performance on complex curves. A driving simulator experiment was conducted in this study to evaluate the relationships between driving behavior on a continuous S-curve and foggy weather, driver experience and gender. The process of negotiating a curve was divided into three stages consisting of a straight segment, the transition from the straight segment to the S-curve and the S-curve. The experimental results indicated that drivers tended to drive more cautiously in heavy fog, but the driving risk was still increased, especially in the transition stage from the straight segment to the S-curve. The non-professional (NP) drivers were less sensitive to the impending change in the road geometry, and less skilled in both longitudinal and lateral vehicle control than the professional drivers. The NP female drivers in particular were found to be the most vulnerable group in S-curve driving.