A method to account for and estimate underreporting in crash frequency research

Dynamic short-term crash analysis and prediction at toll plazas for proactive safety management

Toll plazas are commonly recognized as bottlenecks on toll roads, where vehicles are prone to crashes. However, there has been a lack of research analyzing and predicting dynamic short-term crash risk specifically at toll plazas. This study utilizes traffic, geometric, and weather data to analyze and predict dynamic short-term collision occurrence probability at mainline toll plazas. A random-effects logit regression model is employed to identify crash precursors and assess their impacts on the probability of crash occurrence at toll plazas. Meanwhile, a Long Short-Term Memory Convolutional Neural Network (LSTM-CNN) network is applied for crash prediction. The results of random-effects logit regression model indicate that the flow standard deviation of downstream, upstream occupancy, speed difference and occupancy difference between upstream and downstream positively influence the probability of crash occurrence. Conversely, an increase in the proportion of ETC lanes negatively impacts the probability of crash occurrence. Additionally, there appears a higher likelihood of crashes occurring during summer at toll plaza area. Furthermore, to address the issue of data imbalance, Synthetic Minority Oversampling Techniques (SMOTE) and class weight methods were employed. Stacked Sparse AutoEncoder-Long Short-Term Memory (SSAE-LSTM) and CatBoost were developed and their performance was compared with the proposed model. The results demonstrated that the LSTM-CNN model outperformed the other models in terms of the Area Under the Curve (AUC) values and the true positive rate. The findings of this study can assist engineers in selecting suitable traffic control strategies to improve traffic safety in toll plaza areas. Moreover, the developed collision prediction model can be incorporated into a real-time safety management system to proactively prevent traffic crash.

Spatial and temporal prediction of secondary crashes combining stacked sparse auto-encoder and long short-term memory

Secondary crashes occur within the spatial and temporal impact area of primary crashes, resulting in traffic delays and safety problems. While most existing studies focus on the likelihood of secondary crashes, predicting the spatio-temporal location of secondary crashes could offer valuable insights for implementing prevention strategies. This includes guiding the deployment of emergency response measures and determining appropriate speed limits. The main objective of this study is to develop a prediction method for the spatial and temporal locations of secondary crashes. A hybrid deep learning model SSAE-LSTM is proposed by combining stacked sparse auto-encoder (SSAE) and long short-term memory network (LSTM). Traffic and crash data on the California I-880 highway covering the period of 2017-2021 are collected. The identification of secondary crashes is performed by the speed contour map method. The time and distance gaps between primary and secondary crashes are modeled using multiple 5-minute interval traffic variables as inputs. Multiple models are developed for benchmarking purposes, including PCA-LSTM, which incorporates principal component analysis (PCA) and LSTM, SSAE-SVM, which incorporates SSAE and support vector machine (SVM), and back propagation neural network (BPNN). The performance comparison indicates that the hybrid SSAE-LSTM model outperforms the other models in terms of both spatial and temporal prediction. In particular, SSAE4-LSTM1 (with 4 SSAE layers and 1 LSTM layer) demonstrates superior spatial prediction performance, while SSAE4-LSTM2 (with 4 SSAE layers and 2 LSTM layers) excels in temporal prediction. A joint spatio-temporal evaluation is also conducted to measure the overall accuracy of the optimal models over different permitted spatio-temporal ranges. Finally, practical suggestions are provided for secondary crash prevention.

A functional approach for characterizing safety risk of signalized intersections at the movement level: An exploratory analysis

Safety evaluation of signalized intersections is often conducted by developing statistical and data-driven methods based on data aggregated at certain temporal and spatial levels (e.g., yearly, hourly, or per signal cycle; intersection or approach leg). However, such aggregations are subject to a major simplification that masks the underlying spatio-temporal safety risk patterns within the data aggregation levels. Consequently, high-resolution analysis such as safety risk within signal cycles and at traffic movement level cannot be performed. This study contributes to the literature by proposing a new functional data analysis (FDA) approach for a novel characterization of safety risk patterns of signalized intersections. Functional data smoothing methods that can mitigate overfitting and account for the nonnegative characteristics of safety risk are proposed to model the time series of safety risk within signal cycles at the traffic movement level. Functional analysis of variance method (FANOVA) that can compare the group level differences of functional curves is used to test differences of safety risk functions among different traffic movements. A typical signalized intersection with representative signal types and channelizations is selected as the study location and approximately 1-hour traffic video data recorded by an unmanned aerial vehicle are used to extract traffic conflicts. New movement-level safety risk patterns are characterized based on the safety risk functions that can reveal the temporal distribution of risk within signal cycles. Most of the tested traffic movements have significantly distinct functional risk patterns according to the FANOVA results while risk patterns for most of the traffic movements cannot be differentiated based on the data aggregated at the cycle and approach levels. The proposed functional approach has the potential to be used for facilitating proactive safety management, calibrating microsimulation models for safety evaluation, and optimizing signal timing while considering traffic safety at more disaggregated levels.

Fusing crash data and surrogate safety measures for safety assessment: Development of a structural equation model with conditional autoregressive spatial effect and random parameters

Most existing efforts to assess safety performance require sufficient crash data, which generally takes a few years to collect and suffers from certain limitations (such as long data collection time, under-reporting issue and so on). Alternatively, the surrogate safety measure (SSMs) based approach that can assess traffic safety by capturing the more frequent "near-crash" situations have been developed, but it is criticized for the potential sampling and measurement errors. This study proposes a new safety performance measure-Risk Status (RS), by fusing crash data and SSMs. Real-world connected vehicle data collected in the Safety Pilot Model Deployment (SPMD) project in Ann Arbor, Michigan is used to extract SSMs. With RS treated as a latent variable, a structural equation model with conditional autoregressive spatial effect and corridor-level random parameters is developed to model the interrelationship among RS, crash frequency, risk identified by SSMs, and contributing factors. The modeling results confirm the proposed interrelationship and the necessity to account for both spatial autocorrelation and unobserved heterogeneity. RS can integrate both crash frequency and SSMs together while controlling for observed and unobserved factors. RS is found to be a more reliable criterion for safety assessment in an implementation case of hotspot identification.

Assessing the explanatory and predictive performance of a random parameters count model with heterogeneity in means and variances

Random parameters model has been demonstrated to be an effective method to account for unobserved heterogeneity that commonly exists in highway crash data. However, the predefined single distribution for each random parameter may limit how the unobserved heterogeneity is captured. A more flexible approach is to develop a random parameters model with heterogeneity in means and variances by allowing the mean and variance of potential each random parameter to be an estimable function of explanatory variables. This burgeoning technique for modelling unobserved heterogeneity has been increasingly applied to various safety evaluation scenarios recently. However, the predictive performance of this emerging method, which determines the practicability of the model for a specific circumstance, has never been investigated as far as our knowledge. In addition, the explanatory power by including heterogeneous means and variances of random parameters need to be further investigated to confirm the potential merits of this method in crash data analysis. In this paper, a random parameters negative binomial with heterogeneity in means and variances (RPNBHMV) model, a standard random parameters negative binomial (RPNB) model and a traditional fixed parameters negative binomial (NB) model were estimated using the same dataset. The explanatory and predictive performance of the three models were thoroughly evaluated and compared. Results showed that: 1) the RPNB model fitted the data significantly better than the NB model, and the RPNBHMV model further improved the statistical fit of the RPNB model but the improvement was slight; 2) more insights into interactions of safety factors were inferred from the RPNBHMV model, which demonstrates the explanatory benefit of this model; 3) the RPNBHMV and RPNB models had both advantages (e.g., produced overall better prediction accuracy) and disadvantages (e.g., provided reduced prediction accuracy across the range of explanatory variables) when applied to in-sample observations (i.e., observations used to estimate the model); 4) the RPNBHMV and RPNB models might be less precise than the NB model when applied to out-of-sample observations. These findings indicate that the RPNBHMV model offers more insights and may be used for explanatory safety analysis for sites where reliable data can be collected. However, the simple NB model is more reliable - at least with the dataset used in this study - than its random parameters model counterparts for other sites where the data are unavailable or unreliable, which is a common safety evaluation scenario in practice.

Bayesian spatial-temporal model for the main and interaction effects of roadway and weather characteristics on freeway crash incidence

This study attempts to examine the main and interaction effects of roadway and weather conditions on crash incidence, using the comprehensive crash, traffic and weather data from the Kaiyang Freeway in China in 2014. The dependent variable is monthly crash count on a roadway segment (with homogeneous horizontal and vertical profiles). A Bayesian spatio-temporal model is proposed to measure the association between crash frequency and possible risk factors including traffic composition, presence of curve and slope, weather conditions, and their interactions. The proposed model can also accommodate the unstructured random effect, and spatio-temporal correlation and interactions. Results of parameter estimation indicate that the interactions between wind speed and slope, between precipitation and curve, and between visibility and slope are significantly correlated to the increase in the freeway crash risk, while the interaction between precipitation and slope is significantly correlated to the reduction in the freeway crash risk, respectively. These findings are indicative of the design and implementation of real-time traffic management and control measures, e.g. variable message sign, that could mitigate the crash risk under the adverse weather conditions.

Commercial truck crash injury severity analysis using gradient boosting data mining model

INTRODUCTION

Truck crashes contribute to a large number of injuries and fatalities. This study seeks to identify the contributing factors affecting truck crash severity using 2010 to 2016 North Dakota and Colorado crash data provided by the Federal Motor Carrier Safety Administration.

METHOD

To fulfill a gap of previous studies, broad considerations of company and driver characteristics, such as company size and driver's license class, along with vehicle types and crash characteristics are researched. Gradient boosting, a data mining technique, is applied to comprehensively analyze the relationship between crash severities and a set of heterogeneous risk factors.

RESULTS

Twenty five variables were tested and 22 of them are identified as significant variables contributing to injury severities, however, top 11 variables account for more than 80% of injury forecasting. The relative variable importance analysis is conducted and furthermore marginal effects of all contributing factors are also illustrated in this research. Several factors such as trucking company attributes (e.g., company size), safety inspection values, trucking company commerce status (e.g., interstate or intrastate), time of day, driver's age, first harmful events, and registration condition are found to be significantly associated with crash injury severity. Even though most of the identified contributing factors are significant for all four levels of crash severity, their relative importance and marginal effect are all different.

CONCLUSIONS

For the first time, trucking company and driver characteristics are proved to have significant impact on truck crash injury severity. Some of the results in this study reinforce previous studies' conclusions.

PRACTICAL APPLICATIONS

Findings in this study can be helpful for transportation agencies to reduce injury severity, and develop efficient strategies to improve safety.