Aggregate nonparametric safety analysis of traffic zones

Exploring the significant variables related to specific types of crashes is vitally important in the planning stage of a transportation network. This paper aims to identify and examine important variables associated with total crashes and severe crashes per traffic analysis zone (TAZ) in four counties of the state of Florida by applying nonparametric statistical techniques such as data mining and random forest. The intention of investigating these factors in such aggregate level analysis is to incorporate proactive safety measures in transportation planning. Total and severe crashes per TAZ were modeled to provide predictive decision trees. The variables which carried higher weight of importance for total crashes per TAZ were – total number of intersections per TAZ, airport trip productions, light truck productions, and total roadway segment length with 35 mph posted speed limit. The other significant variables identified for total crashes were total roadway length with 15 mph posted speed limit, total roadway length with 65 mph posted speed limit, and non-home based work productions. For severe crashes, total number of intersections per TAZ, light truck productions, total roadway length with 35 mph posted speed limit, and total roadway length with 65 mph posted speed limit were among the significant variables. These variables were further verified and supported by the random forest results.     

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.     

Multilevel data and Bayesian analysis in traffic safety

Background

Traditional crash prediction models, such as generalized linear regression model, are incapable of taking into account multilevel data structure. Therefore they suffer from a common underlying limitation that each observation (e.g. a crash or a vehicle involvement) in the estimation procedure corresponds to an individual situation in which the residuals exhibit independence.

Problem

However, this “independence” assumption may often not hold true since multilevel data structures exist extensively because of the traffic data collection and clustering process. Disregarding the possible within-group correlations may lead to production of models with unreliable parameter estimates and statistical inferences.

Proposed theory

In this paper, a 5 × ST-level hierarchy is proposed to represent the general framework of multilevel data structures in traffic safety, i.e. [Geographic region level − Traffic site level − Traffic crash level − Driver-vehicle unit level − Occupant level] × Spatiotemporal level. The involvement and emphasis for different sub-groups of these levels depend on different research purposes and also rely on the heterogeneity examination on crash data employed. To properly accommodate the potential cross-group heterogeneity and spatiotemporal correlation due to the multilevel data structure, a Bayesian hierarchical approach that explicitly specifies multilevel structure and reliably yields parameter estimates is introduced and recommended.

Case studies

Using Bayesian hierarchical models, the results from several case studies are highlighted to show the improvements on model fitting and predictive performance over traditional models by appropriately accounting for the multilevel data structure.     

Utilizing support vector machine in real-time crash risk evaluation

Real-time crash risk evaluation models will likely play a key role in Active Traffic Management (ATM). Models have been developed to predict crash occurrence in order to proactively improve traffic safety. Previous real-time crash risk evaluation studies mainly employed logistic regression and neural network models which have a linear functional form and over-fitting drawbacks, respectively. Moreover, these studies mostly focused on estimating the models but barely investigated the models’ predictive abilities. In this study, support vector machine (SVM), a recently proposed statistical learning model was introduced to evaluate real-time crash risk. The data has been split into a training dataset (used for developing the models) and scoring datasets (meant for assessing the models’ predictive power). Classification and regression tree (CART) model has been developed to select the most important explanatory variables and based on the results, three candidates Bayesian logistic regression models have been estimated with accounting for different levels unobserved heterogeneity. Then SVM models with different kernel functions have been developed and compared to the Bayesian logistic regression model. Model comparisons based on areas under the ROC curve (AUC) demonstrated that the SVM model with Radial-basis kernel function outperformed the others. Moreover, several extension analyses have been conducted to evaluate the effect of sample size on SVM models’ predictive capability; the importance of variable selection before developing SVM models; and the effect of the explanatory variables in the SVM models. Results indicate that (1) smaller sample size would enhance the SVM model's classification accuracy, (2) variable selection procedure is needed prior to the SVM model estimation, and (3) explanatory variables have identical effects on crash occurrence for the SVM models and logistic regression models.     

Investigation of road network features and safety performance

The analysis of road network designs can provide useful information to transportation planners as they seek to improve the safety of road networks. The objectives of this study were to compare and define the effective road network indices and to analyze the relationship between road network structure and traffic safety at the level of the Traffic Analysis Zone (TAZ). One problem in comparing different road networks is establishing criteria that can be used to scale networks in terms of their structures. Based on data from Orange and Hillsborough Counties in Florida, road network structural properties within TAZs were scaled using 3 indices: Closeness Centrality, Betweenness Centrality, and Meshedness Coefficient. The Meshedness Coefficient performed best in capturing the structural features of the road network. Bayesian Conditional Autoregressive (CAR) models were developed to assess the safety of various network configurations as measured by total crashes, crashes on state roads, and crashes on local roads. The models’ results showed that crash frequencies on local roads were closely related to factors within the TAZs (e.g., zonal network structure, TAZ population), while crash frequencies on state roads were closely related to the road and traffic features of state roads. For the safety effects of different networks, the Grid type was associated with the highest frequency of crashes, followed by the Mixed type, the Loops & Lollipops type, and the Sparse type. This study shows that it is possible to develop a quantitative scale for structural properties of a road network, and to use that scale to calculate the relationships between network structural properties and safety.     

Corridor-level signalized intersection safety analysis in Shanghai,China using Bayesian hierarchical models

Most traffic crashes in Chinese cities occur at signalized intersections. Research on the intersection safety problem in China is still in its early stage. The recent development of an advanced traffic information system in Shanghai enables in-depth intersection safety analyses using road design, traffic operation, and crash data. In Shanghai, the road network density is relatively high and the distance between signalized intersections is small, averaging about 200 m. Adjacent signalized intersections located along the same corridor share similar traffic flows, and signals are usually coordinated. Therefore, when studying intersection safety in Shanghai, it is essential to account for intersection correlations within corridors. In this study, data for 195 signalized intersections along 22 corridors in the urban areas of Shanghai were collected. Mean speeds and speed variances of corridors were acquired from taxis equipped with Global Positioning Systems (GPS). Bayesian hierarchical models were applied to identify crash risk factors at both the intersection and the corridor levels. Results showed that intersections along corridors with lower mean speeds were associated with fewer crashes than those with higher speeds, and those intersections along two-way roads, under elevated roads, and in close proximity to each other, tended to have higher crash frequencies.     

Empirical Bayes before-after safety studies: lessons learned from two decades of experience and future directions

The empirical Bayes (EB) methodology has been applied for over 20 years now in conducting statistically defendable before-after studies of the safety effect of treatments applied to roadway sites. The appeal of the methodology is that it corrects for regression to the mean and traffic volume and other changes not due to the measure. There is, therefore, a natural tendency to put a stamp of approval on any study that uses this methodology, and to assume that the results can then be used in specifying crash modification factors for use in developing treatments for hazardous locations, or in designing new roads using tools such as the interactive highway safety design model (IHSDM). At the other extreme are skeptics who suggest that the increased sophistication and data needs of the EB methodology are not worth the effort since alternative, less complex methods can produce equally valid results. The primary objective of this paper is to capitalize on experience gained from two decades of conducting EB studies around the world to illustrate that the EB methodology, if properly undertaken, produces results that could be substantially different and less biased than those from more conventional types of studies. A secondary objective is to emphasize that caution is needed in assessing the validity of studies undertaken with the EB methodology and in using these results for providing crash modification factors. To this end, a number of issues that are critical to the proper conduct and interpretation of EB evaluations are raised and illustrated based on lessons learned from recent experience with these studies. These include: amalgamating the effects on different crash types; the specification of the reference/comparison groups; and accounting for traffic volume changes. Current and future directions, including the improvements offered by a full Bayes approach, are discussed.     

Sequential application of heterogeneous models for the safetyanalysis of a control system: a case study

This paper shows how heterogeneous stochastic modelling techniques of increasing modelling power can be applied to assess the safety of a digital control system. First, a Fault-Tree (FT) has been built to model the system, assuming two-state components and independent failures. Then, the FT is automatically converted into a Bayesian Network, allowing to include more modelling details and localized dependencies. Finally, in order to accommodate repair activities and perform an availability analysis, the FT is converted into a Stochastic Petri Net (SPN). Moving from a combinatorial model (the FT) to a state space based model (the SPN) increases the modelling flexibility, but incurs into the state space explosion problem. In order to alleviate the state space explosion problem, this paper resorts to the use of a particular type of high level (coloured) Petri nets called SWN. A digital control system is considered as a case study, and safety measures have been evaluated, referring to the emergent standard IEC 61508.     

Estimation of rear-end vehicle crash frequencies in urban road tunnels

According to The Handbook of Tunnel Fire Safety, over 90% (55 out of 61 cases) of fires in road tunnels are caused by vehicle crashes (especially rear-end crashes). It is thus important to develop a proper methodology that is able to estimate the rear-end vehicle crash frequency in road tunnels. In this paper, we first analyze the time to collision (TTC) data collected from two road tunnels of Singapore and conclude that Inverse Gaussian distribution is the best-fitted distribution to the TTC data. An Inverse Gaussian regression model is hence used to establish the relationship between the TTC and its contributing factors. We then proceed to introduce a new concept of exposure to traffic conflicts as the mean sojourn time in a given time period that vehicles are exposed to dangerous scenarios, namely, the TTC is lower than a predetermined threshold value. We further establish the relationship between the proposed exposure to traffic conflicts and crash count by using negative binomial regression models. Based on the limited data samples used in this study, the negative binomial regression models perform well although a further study using more data is needed.     

Bayesian prediction in threshold autoregressive models with exponential white noise

In this paper, we develop a Bayesian analysis of a threshold autoregressive model with exponential noise. An approximate Bayes methodology, which is introduced here, and the Gibbs sampler are used to compute marginal posterior densities for the parameters of the model, including the threshold parameter, and to compute one-step-ahead predictive density functions. The proposed methodogy is illustrated with a simulation study and a real example.     

Full Bayesian evaluation of the safety effects of reducing the posted speed limit in urban residential area

Full Bayesian (FB) before–after evaluation is a newer approach than the empirical Bayesian (EB) evaluation in traffic safety research. While a number of earlier studies have conducted univariate and multivariate FB before–after safety evaluations and compared the results with the EB method, often contradictory conclusions have been drawn. To this end, the objectives of the current study were to (i) perform a before–after safety evaluation using both the univariate and multivariate FB methods in order to enhance our understanding of these methodologies, (ii) perform the EB evaluation and compare the results with those of the FB methods and (iii) apply the FB and EB methods to evaluate the safety effects of reducing the urban residential posted speed limit (PSL) for policy recommendation. In addition to three years of crash data for both the before and after periods, traffic volume, road geometry and other relevant data for both the treated and reference sites were collected and used. According to the model goodness-of-fit criteria, the current study found that the multivariate FB model for crash severities outperformed the univariate FB models. Moreover, in terms of statistical significance of the safety effects, the EB and FB methods led to opposite conclusions when the safety effects were relatively small with high standard deviation. Therefore, caution should be taken in drawing conclusions from the EB method. Based on the FB method, the PSL reduction was found effective in reducing crashes of all severities and thus is recommended for improving safety on urban residential collector roads.     

Bayesian-network-based safety risk assessment for steel construction projects

There are four primary accident types at steel building construction (SC) projects: falls (tumbles), object falls, object collapse, and electrocution. Several systematic safety risk assessment approaches, such as fault tree analysis (FTA) and failure mode and effect criticality analysis (FMECA), have been used to evaluate safety risks at SC projects. However, these traditional methods ineffectively address dependencies among safety factors at various levels that fail to provide early warnings to prevent occupational accidents. To overcome the limitations of traditional approaches, this study addresses the development of a safety risk-assessment model for SC projects by establishing the Bayesian networks (BN) based on fault tree (FT) transformation. The BN-based safety risk-assessment model was validated against the safety inspection records of six SC building projects and nine projects in which site accidents occurred. The ranks of posterior probabilities from the BN model were highly consistent with the accidents that occurred at each project site. The model accurately provides site safety-management abilities by calculating the probabilities of safety risks and further analyzing the causes of accidents based on their relationships in BNs. In practice, based on the analysis of accident risks and significant safety factors, proper preventive safety management strategies can be established to reduce the occurrence of accidents on SC sites.     

Estimating the safety performance of urban road transportation networks

Transportation planning models are typically used to estimate future traffic patterns, peak period traffic, travel time, and various environmental or other related traffic flow characteristics. Unfortunately, traffic safety is seldom, if ever, explicitly considered proactively during the transportation planning process. This omission is attributed to various factors, including the lack of available tools needed to estimate the number of crashes during this process. To help fill this void, the research on which this paper is based aimed, as a primary objective, to develop a tool that would allow the estimation of crashes on digital or coded urban transportation networks during the planning process. The secondary objective of the research was to describe how the predictive models should be applied on these networks and explain the important issues and limitations surrounding their application. To accomplish these objectives, safety performance functions specifically created for this work were applied to two sample digital networks created with the help of EMME/2, a software package widely used in transportation planning. The results showed that it is possible to predict crashes on digital transportation networks, but confirmed the reality that the accuracy of the predictions is directly related to the precision of the traffic flow estimates. The crash predictions are also sensitive to how the digital network is coded, and it is shown how appropriate adjustments can be made.     

Safety modeling of suburban arterials in Shanghai,China

As urbanization accelerates in Shanghai, land continues to develop along suburban arterials which results in more access points along the roadways and more congested suburban arterials; all these changes have led to deterioration in traffic safety. In-depth safety analysis is needed to understand the relationship between roadway geometric design, access features, traffic characteristics, and safety. This study examined 161 road segments (each between two adjacent signalized intersections) of eight suburban arterials in Shanghai. Information on signal spacing, geometric design, access features, traffic characteristics, and surrounding area types were collected. The effect of these factors on total crash occurrence was investigated. To account for the hierarchical data structure, hierarchical Bayesian models were developed for total crashes. To identify diverse effects on different crash injury severity, the total crashes were separated into minor injury and severe injury crashes. Bivariate hierarchical Bayesian models were developed for minor injury and severe injury to account for the correlation among different severity levels. The modeling results show that the density of signal spacing along arterials has a significant influence on minor injury, severe injury, and total crash frequencies. The non-uniform signal spacing has a significant impact on the occurrence of minor injury crashes. At the segment-level, higher frequencies of minor injury, severe injury, and total crashes tend to occur for the segments with curves, those with a higher density of access points, those with a higher percentage of heavy vehicles, and those in inner suburban areas. This study is useful for applications such as related engineering safety improvements and making access management policy.     

Spatial analysis of fatal and injury crashes in Pennsylvania

Using injury and fatal crash data for Pennsylvania for 1996-2000, full Bayes (FB) hierarchical models (with spatial and temporal effects and space-time interactions) are compared to traditional negative binomial (NB) estimates of annual county-level crash frequency. Covariates include socio-demographics, weather conditions, transportation infrastructure and amount of travel. FB hierarchical models are generally consistent with the NB estimates. Counties with a higher percentage of the population under poverty level, higher percentage of their population in age groups 0-14, 15-24, and over 64 and those with increased road mileage and road density have significantly increased crash risk. Total precipitation is significant and positive in the NB models, but not significant with FB. Spatial correlation, time trend, and space-time interactions are significant in the FB injury crash models. County-level FB models reveal the existence of spatial correlation in crash data and provide a mechanism to quantify, and reduce the effect of, this correlation. Addressing spatial correlation is likely to be even more important in road segment and intersection-level crash models, where spatial correlation is likely to be even more pronounced.     

Validation of reliability computational models using Bayes networks

This paper proposes a methodology based on Bayesian statistics to assess the validity of reliability computational models when full-scale testing is not possible. Sub-module validation results are used to derive a validation measure for the overall reliability estimate. Bayes networks are used for the propagation and updating of validation information from the sub-modules to the overall model prediction. The methodology includes uncertainty in the experimental measurement, and the posterior and prior distributions of the model output are used to compute a validation metric based on Bayesian hypothesis testing. Validation of a reliability prediction model for an engine blade under high-cycle fatigue is illustrated using the proposed methodology.