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.

Travel route safety estimation based on conflict simulation.

With the aim of providing travelers information about the safety levels of selectable routes, it is necessary to develop a method that can properly estimate the safety of alternative travel routes. This paper proposes a conflict-based approach for travel route safety estimation (TRSE). It is developed on the basis of the classical safety evaluation model where both the amount of exposure to safety risk and the risk under unit exposure are measured to estimate the route safety. A combination of a set of dynamic and static factors related to traffic flow characteristics and roadway features are selected to estimate conflict exposure and potential conflict risk. A route-based method is employed where two parallel estimations of conflict are conducted for both the component segments links and intersection turning links. Three machine learning models (i.e., random forest, k-nearest neighbor, and support vector machine) are tested in conflict risk estimation. A fuzzy reasoning process based on the fuzzy logic algorithm is employed to conduct the route safety estimation. The proposed TRSE is tested on a four-horizontal and six-vertical network extracted from a real road network in China. Conflict simulation results were obtained by Vissim and SSAM tools. The results illustrate the practicability and effectiveness of the proposed TRSE approach.

Transferability of safety performance functions and hotspot identification for freeways of the United States and China.

Safety performance function (SPF) has been a vital tool in traffic safety evaluation including finding contributing factors to crashes, identifying hotspots, and assessing safety effects of countermeasures. In the United States (U.S.), the Highway Safety Manual provides a number of SPFs for a variety of road facilities. Due to the limited availability of traffic data in many regions, the transferability of SPFs has been an important topic in traffic safety analysis and has been evaluated by several studies. Nevertheless, the international transferability of freeway SPFs and the applicability of transferred SPFs on hotspot identification has been rarely investigated. Based on data from two Chinese cities, Shanghai and Suzhou, and three U.S. states, Texas, New York and Florida, this study analyzes the transferability of freeway SPFs between Chinese and U.S. regions. These SPFs are then transferred to the other country and their performance on hotspot identification is investigated. SPFs were developed in the frameworks of Poisson, Poisson-lognormal and negative binomial regressions for the five localities separately, and were calibrated using the calibration functions before being transferred. Without calibration, the poor model transferability was found between the two countries, while after calibration, the transferred SPFs between Shanghai/Suzhou and Texas/New York showed satisfactory performance on both model fitting and hotspot identification. However, the transferability of SPFs between Florida and the Chinese cities turned out to be unsatisfactory regardless of whether being calibrated or not, which was attributable to the considerable difference in traffic flow. The findings of this study are expected to be a good reference for researchers and practitioners who want to understand the transferability and applicability of SPFs in the international context.