Modeling work zone crash frequency by quantifying measurement errors in work zone length

Toward safer highway work zones: An empirical analysis of crash risks using improved safety potential field and machine learning techniques

Due to complex traffic conditions, transition areas in highway work zones are associated with a higher crash risk than other highway areas. Understanding risk-contributing features in transition areas is essential for ensuring traffic safety on highways. However, conventional surrogate safety measures (SSMs) are quite limited in identifying the crash risk in transition areas due to the complex traffic environment. To this end, this study proposes an improved safety potential field, named the Work-Zone Crash Risk Field (WCRF). The WCRF force can be used to measure the crash risk of individual vehicles that enter a work zone considering the influence of multiple features, upon which the overall crash risk of the road segment in a specific time window can be estimated. With the overall crash risk used as a label, the time-window-based traffic data are used to train and validate an eXtreme Gradient Boosting (XGBoost) classifier, and the Shapley Additive Explanations (SHAP) method is integrated with the XGBoost classifier to identify the key risk-contributing traffic features. To assess the proposed approach, a case study is conducted using real-time vehicle trajectory data collected in two work zones along a highway in China. The results demonstrate that the WCRF-based SSM outperforms conventional SSMs in identifying crash risks in work zone transition areas on highways. In addition, we perform lane-based analysis regarding the impact of setting up work zones on highway safety and investigate the heterogeneity in risk-contributing features across different work zones. Several interesting findings from the analysis are reported in this paper. Compared to existing SSMs, the WCRF-based SSM offers a more practical and comprehensive way to describe the crash risk in work zones. The approach using the developed WCRF technique offers improved capabilities in identifying key risk-contributing features, which is expected to facilitate the development of safety management strategies for work zones.

How effective is reducing traffic speed for safer work zones? Methodology and a case study in Pennsylvania

Transportation agencies post and enforce reduced speed limits in work zones to ensure work zone safety, since traffic speed is found to be associated with work zone crash risks. However, prior findings on the relationship between speed and crash rate in work zones are inconsistent. This may be attributed to the methods of statistical associations between traffic speed and crash risks that do not necessarily discover true causal relations. In fact, work zone presence could lead to the reduction of actual traffic speed that influences crash risks, where it may also directly impose effects on crash risks as a result of work zone configurations. The actual traffic speed (not posted speed limit) is also known as a "mediator" where work zones can indirectly impact the crash risks. It is challenging to rigorously separate the causal effect of traffic speed on work zone crash risk from that directly caused by work zones. The underlying causal relation could help to determine what reduced post speed limit (with enforcement) is necessary to ensure work zone safety under the most desired "actual traffic speed". This study proposes to use the sequential g-estimation and the regression discontinuity design to estimate the controlled direct effect of traffic speed on work zone crashes. Two research gaps are identified and filled: inaccurate inferences of the effect of reduced speed limit in work zones as a result of ignoring (1) potential post-treatment bias since traffic speed is a mediator; and (2) potential confounding bias caused by unobservable roadway characteristics. The proposed methodology was applied to 4008 work zones in Pennsylvania from 2015 to 2017, and the results were validated through a series of robustness tests. The results indicate that the direct causal effect of the presence of work zones on crash risk is significantly positive when the traffic speed is relatively low (i.e., lower than 55 mph in this case study), while traffic speed has a positive causal effect on crash occurrences when the actual traffic speed is high (i.e., greater or equal to 55 mph). It suggests that strictly enforcing reduced posted speed limits in work zones is particularly effective when the actual traffic speed is greater than 55 mph. This is particularly true on roadways with high traffic volume (i.e., AADT > 20,000 vehicles per day), long, and daytime work zones (i.e., > 3000 m). On the other hand, the effect of enforcing reduced speed on work zone safety is unclear when the actual speed is already low. In this case, improving work zone configurations and driving behaviors may be more effective in reducing crash risks.

Inferring heterogeneous treatment effects of work zones on crashes

The increasing number of work zone crashes has been a significant concern for road users, transportation agencies, and researchers. Crashes can be caused by work zones, and this effect changes across different work zone configurations, traffic volumes, roadway functional classifications, and weather conditions. This is typically represented by Crash Modification Functions (CMFunctions). However, current methods for developing work zone CMFunctions have two major limitations: (1) They focus on analyzing statistical associations and fail to mitigate the confounding bias due to possible unobservable roadway characteristics; and (2) They cannot address CMFunctions of multiple variables simultaneously, such as weather and traffic conditions, since they are represented using mixed data types (continuous and categorical) that could potentially affect the causal effect of work zones on crashes. In this study, we develop a method that utilizes causal forest with fixed-effect modeling to mitigate the confounding bias while identifying CMFunctions conditioning on various environmental characteristics, including work zone configurations, traffic volume, roadway functional classification, and weather conditions. The developed method was applied to 3378 work zones that occurred in Pennsylvania between 2015 and 2017. The results were validated via a series of robustness tests. The validations demonstrate that this method can mitigate the confounding bias and identify CMFunctions of multiple variables. The results also show that the causal effect of a work zone on crash occurrence is significantly positive (p<0.05) on roadways with high traffic volumes (e.g., > 20,000 vehicles per day) and on medium length (e.g., 2000 to 5000 m) work zones. It appears that having medium-long (e.g., between 6000 and 8000 m) work zones or long duration (e.g., longer than 4 h) work zones do not necessarily lead to extra crashes.

Identifying factors associated with roadside work zone collisions using machine learning techniques

Identifying factors that are associated with the probability of roadside work zone collisions enables decision makers to better assess and control the risk of scheduling a particular maintenance or construction activity by modifying the characteristics of the operation. This can be achieved by studying the effect of work zone properties on the risk of roadside work zone collisions. Much of the existing work in this area is based on data in the police traffic collision reports, which do not include data on the characteristics of the work zone itself. This paper develops a comprehensive data set of 42 features describing time, location, work zone characteristics, traffic volume, and road properties. Using recent machine learning techniques such as extreme gradient boosting classifiers on this extensive set of features allows for more accurate analysis to identify factors that affect the risk of work zone collisions or indicate higher than baseline chances of a roadside crash. Our statistical analysis reveals 10 important features and shows that four of these features are significantly associated with higher probabilities of roadside work zone collisions.

Cyclist-vehicle crash modeling with measurement error in traffic exposure

Exposure measures are always among the explanatory variables of any crash model. Regardless of the technique used to model crash, the mean crash frequency will increase with an increase in exposure since more crashes are likely to occur at higher exposure. For cyclist-vehicle crash models, bike and vehicle exposure measures are essential for an accurate and reliable estimate of the cyclist crash risk. However, traffic exposure measures are an example of variables that are measured with error. Generally, measurement error in regression estimates has three effects: 1) produce bias in parameter estimation for statistical models, 2) lead to a loss of explanation power, 3) mask important features of the data. This study proposes a full Bayesian Poisson Lognormal crash models that account for measurement error in traffic exposure measures (i.e., Vehicle Kilometers Travelled and Bike Kilometers Travelled). The underlying approach is to adjust the traffic exposure measures for measurement error to improve the accuracy of the crash model and crash model estimates. The full Bayesian models are developed using data for 134 traffic analysis zones (TAZs) in the city of Vancouver, Canada. The results show that Poisson Lognormal models that account for measurement error have a better fit for the modeled cyclist-vehicle crash data compared to traditional Poisson Lognormal models. The estimates of the Poisson Lognormal model that accounts for measurement error are consistent, with traditional Poisson Lognormal models' estimates except for the BKT and VKT estimates. Estimates of the BKT and VKT increased after introducing measurement error, which indicates an underestimation (downward bias) to BKT and VKT estimates in case of overlooking measurement error.

Modeling crash risk of horizontal curves using large-scale auto-extracted roadway geometry data

Highway horizontal curves (H-curves) provide a smooth transition between two tangent sections of roadways. They allow vehicles to adjust their travel directions gradually. However, the geometry changes of the highway sections with H-curves often raise safety concerns. In order to deploy effective safety countermeasures, a critical task is to understand the risk factors associated with H-curves. Existing studies have made efforts to probe the safety issues associated with H-curves, whereas they were limited to relatively small-scale examinations because of the challenges in identifying H-curves from large road networks. In addition, due to the lack of well-archived traffic and roadway information, gathering other data associated with the H-curves is also difficult. Regarding to these gaps, this study aims to leverage open-source data to analyze the crash risk of highway sections with H-curves. In particular, the present study highlights itself from two main aspects: (i) a H-curve extraction tool was developed to facilitate large-scale curve data collection through the analytics of different open source data; and (ii) a crash modeling framework was developed to quantify H-curve crash risk. A case study based on a statewide road network was performed to test the developed crash risk models with the collected curve data. The results show the opportunities of using the developed tool for large-scale data collection and analyze the safety impacts of H-curve geometric properties, elevation change, traffic exposure, among others. Findings of this study provide insights into the improvement of H-curve data collection and safety evaluation.

Does time of day matter at highway work zone crashes?

INTRODUCTION

It is widely agreed that highway work zones pose significant threats to road users because driving conditions in work zones are quite different from the normal ones, particularly when traffic volumes approach a highway capacity. Therefore, work zone safety is a critical aspect for state agencies and traffic engineers.

METHOD

In the current study, a total of 10,218 crashes that occurred in highway work zones in the state of Washington for the period between 2007 and 2013 were used. Time of day is disaggregated into four subgroups: (1) Morning from 6:00 to 11:00 a.m. (2) Midday from 12:00 to 5:00 p.m. (3) Night from 6:00 to 11:00 p.m., and (4) Late night from 12:00 to 5:00 a.m. Then, four mixed logit models were estimated to account and correct for heterogeneity in the crash data by considering three injury severity levels: severe injury, minor injury, and no injury.

RESULTS

The estimation results reveal that most contributing factors are uniquely significant in a specific time of day period, whereas three factors affect injury severity regardless of time of day such as the indicators of not deployed airbag, one passenger vehicle involved in the crash, and rear-end collision. Further, some factors were found to affect injury severity into two or three time periods, such as female drivers that found to decrease the probability of no injury in morning and night time periods, while increasing severe injury outcome in midday time.

CONCLUSIONS

The effect of time of day on injury severity of work-zone related crashes should be modeled separately rather than using a holistic model. Practical applications: As a starting point, findings of the current study can be used by transportation officials to reduce fatalities and injuries of work zone crashes by identifying factors that uniquely contribute to each time of day period.

Spatial analysis of traffic accidents near and between road intersections in a directed linear network

Although most of the literature on traffic safety analysis has been developed over areal zones, there is a growing interest in using the specific road structure of the region under investigation, which is known as a linear network in the field of spatial statistics. The use of linear networks entails several technical complications, ranging from the accurate location of traffic accidents to the definition of covariates at a spatial micro-level. Therefore, the primary goal of this study was to display a detailed analysis of a dataset of traffic accidents recorded in Valencia (Spain), which were located into a linear network representing more than 30 km of urban road structure corresponding to one district of the city. A set of traffic-related covariates was constructed at the road segment level for performing the analysis. Several issues and methodological approaches that are inherent to linear networks have been shown and discussed. In particular, the network was defined in a way that allowed the explicit investigation of traffic accidents around road intersections and the consideration of traffic flow directionality. Zero-inflated negative binomial count models accounting for spatial heterogeneity were used. Traffic safety at road intersections was specifically taken into account in the analysis by considering the higher variability and number of zeros that can be observed at these road entities and the differential contribution of the covariates depending on the proximity of a road intersection. To complement the results obtained from the count models fitted, coldspots and hotspots along the network were also detected, with explanatory objectives. The models confirmed that spatial heterogeneity, overdispersion and the close presence of road intersections explain the accident counts observed in the road network analyzed. Hotspot detection revealed that several covariates whose contribution was unclear in the modelling approaches may also be affecting accident counts at the road segment level.

A New Methodology for Before-After Safety Assessment Using Survival Analysis and Longitudinal Data

The widely used empirical Bayes (EB) and full Bayes (FB) methods for before-after safety assessment are sometimes limited because of the extensive data needs from additional reference sites. To address this issue, this study proposes a novel before-after safety evaluation methodology based on survival analysis and longitudinal data as an alternative to the EB/FB method. A Bayesian survival analysis (SARE) model with a random effect term to address the unobserved heterogeneity across sites is developed. The proposed survival analysis method is validated through a simulation study before its application. Subsequently, the SARE model is developed in a case study to evaluate the safety effectiveness of a recent red-light-running photo enforcement program in New Jersey. As demonstrated in the simulation and the case study, the survival analysis can provide valid estimates using only data from treated sites, and thus its results will not be affected by the selection of defective or insufficient reference sites. In addition, the proposed approach can take into account the censored data generated due to the transition from the before period to the after period, which has not been previously explored in the literature. Using individual crashes as units of analysis, survival analysis can incorporate longitudinal covariates such as the traffic volume and weather variation, and thus can explicitly account for the potential temporal heterogeneity.

Exploring the impacts of speed variances on safety performance of urban elevated expressways using GPS data

Speed variation on urban expressways has been frequently noted as a key factor associated with high crash risk. However, it was often difficult to capture the safety impact of speed variance with spaced sensor measurements. As an alternative, this paper aims to leverage the use of the floating car data (FCD) to capture the speed variance in a morning rush hour on urban elevated expressways and examine its effect on safety. A semi-automatic filtering process was introduced to distinguish taxi GPS data points on the elevated expressways from the ones on the surface roads under the expressways. Subsequently, the standard deviation of the cross-sectional speed mean (SDCSM) and the cross-section speed standard deviation (MCSSD) were derived to capture the spatial and temporal speed variances, respectively. Together with other explanatory variables, both hierarchical and non-hierarchical Poisson-gamma measurement error models were developed to model the crash frequencies of the expressways. The modeling results showed that the hierarchical model performed better and both SDCSM and MCSSD were found to be positively related to the crash occurrence. This secures the need for addressing the impact of speed variation when modeling crashes occurred on the elevated expressways.

Meta-analysis of the effect of road work zones on crash occurrence

There is strong evidence that work zones pose increased risk of crashes and injuries. The two most common risk factors associated with increased crash frequencies are work zone duration and length. However, relevant research on the topic is relatively limited. For that reason, this paper presents formal meta-analyses of studies that have estimated the relationship between the number of crashes and work zone duration and length, in order to provide overall estimates of those effects on crash frequencies. All studies presented in this paper are crash prediction models with similar specifications. According to the meta-analyses and after correcting for publication bias when it was considered appropriate, the summary estimates of regression coefficients were found to be 0.1703 for duration and 0.862 for length. These effects were significant for length but not for duration. However, the overall estimate of duration was significant before correcting for publication bias. Separate meta-analyses on the studies examining both duration and length was also carried out in order to have rough estimates of the combined effects. The estimate of duration was found to be 0.953, while for length was 0.847. Similar to previous meta-analyses the effect of duration after correcting for publication bias is not significant, while the effect of length was significant at a 95% level. Meta-regression findings indicate that the main factors influencing the overall estimates of the beta coefficients are study year and region for duration and study year and model specification for length.

A full Bayesian approach to appraise the safety effects of pedestrian countdown signals to drivers

Although they are meant for pedestrians, pedestrian countdown signals (PCSs) give cues to drivers about the length of the remaining green phase, hence affecting drivers' behavior at intersections. This study focuses on the evaluation of the safety effectiveness of PCSs to drivers, in the cities of Jacksonville and Gainesville, Florida, using crash modification factors (CMFs) and crash modification functions (CMFunctions). A full Bayes (FB) before-and-after with comparison group method was used to quantify the safety impacts of PCSs to drivers. The CMFs were established for distinctive categories of crashes based on crash type (rear-end and angle collisions) and severity level (total, fatal and injury (FI), and property damage only (PDO) collisions). The CMFs findings indicated that installing PCSs result in a significant improvement of drivers' safety, at a 95% Bayesian credible interval (BCI), for total, PDO, and rear-end collisions. The results of FI and angle crashes were not significant. The CMFunctions indicate that the treatment effectiveness varies considerably with post-treatment time and traffic volume. Nevertheless, the CMFs on rear-end crashes are observed to decline with post-treatment time. In summary, the results suggest the usefulness of PCSs for drivers. The findings of this study may prompt a need for a broader research to investigate the need to design PCSs that will serve the purpose not only of pedestrians, but drivers as well.

Work zone safety analysis and modeling: a state-of-the-art review

OBJECTIVE

Work zone safety is one of the top priorities for transportation agencies. In recent years, a considerable volume of research has sought to determine work zone crash characteristics and causal factors. Unlike other non-work zone-related safety studies (on both crash frequency and severity), there has not yet been a comprehensive review and assessment of methodological approaches for work zone safety. To address this deficit, this article aims to provide a comprehensive review of the existing extensive research efforts focused on work zone crash-related analysis and modeling, in the hopes of providing researchers and practitioners with a complete overview.

METHODS

Relevant literature published in the last 5 decades was retrieved from the National Work Zone Crash Information Clearinghouse and the Transport Research International Documentation database and other public digital libraries and search engines. Both peer-reviewed publications and research reports were obtained. Each study was carefully reviewed, and those that focused on either work zone crash data analysis or work zone safety modeling were identified. The most relevant studies are specifically examined and discussed in the article.

RESULTS

The identified studies were carefully synthesized to understand the state of knowledge on work zone safety. Agreement and inconsistency regarding the characteristics of the work zone crashes discussed in the descriptive studies were summarized. Progress and issues about the current practices on work zone crash frequency and severity modeling are also explored and discussed. The challenges facing work zone safety research are then presented.

CONCLUSIONS

The synthesis of the literature suggests that the presence of a work zone is likely to increase the crash rate. Crashes are not uniformly distributed within work zones and rear-end crashes are the most prevalent type of crashes in work zones. There was no across-the-board agreement among numerous papers reviewed on the relationship between work zone crashes and other factors such as time, weather, victim severity, traffic control devices, and facility types. Moreover, both work zone crash frequency and severity models still rely on relatively simple modeling techniques and approaches. In addition, work zone data limitations have caused a number of challenges in analyzing and modeling work zone safety. Additional efforts on data collection, developing a systematic data analysis framework, and using more advanced modeling approaches are suggested as future research tasks.