This study is aimed at investigating the resilience degradation caused by traffic accidents and developing relevant resilience optimization strategies. A two-stage accident resilience triangle framework was proposed by comparing the differences between natural disasters and traffic accidents. To maximize system resilience, a network-wide traffic signal optimization model was presented. Spillback constraints and equilibrium constraints were established to enhance the capacity of urban-road networks to minimize congestion escalation, in addition to rapid recovery. A two-level algorithm based on greedy strategy and gradient descent was designed to solve the proposed non-linear programming model. In the experiment, a virtual road network was constructed based on the Simulation of Urban Mobility (SUMO) platform for validation and sensitivity analysis. The experimental results revealed that: (1) Compared to the traditional resilience framework, the proposed two-stage accident resilience framework can more reasonably describe the change mechanism of road network resilience under disturbance. (2) The proposed resilience-based traffic signal optimization model improved the system resilience under different conditions of traffic demand, accident severity, and rescue time in terms of the maximum performance degradation and recovery time. Precisely, the resilience loss is reduced by a maximum of 1.4%. Finally, the proposed model was further implemented with field data. The resilience improvement was significant during the evening rush hour. The results of this study contribute toward transportation resilience research and accident rescue strategies with respect to traffic management and control.

For urban extra-long underwater tunnels, the obstacle space formed by the tunnel walls on both sides has an impact on the driver's driving. The aim of this study is to investigate the shy away characteristics of drivers in urban extra-long underwater tunnels. Using trajectory offset and speed data obtained from real vehicle tests, the driving behaviour at different lanes of an urban extra-long underwater tunnel was investigated, and a theory of shy away effects and indicators of sidewall shy away deviation for quantitative analysis were proposed. The results show that the left-hand lane has the largest offset and driving speed from the sidewall compared to the other two lanes. In the centre lane there is a large fluctuation in the amount of deflection per 50 seconds of driving, increasing the risk of two-lane collisions. When the lateral clearances are increased from 0.5 m to 2.19 m on the left and 1.29 m on the right, the safety needs of drivers can be better met. The results of this study have implications for improving traffic safety in urban extra-long underwater tunnels and for the improvement of tunnel traffic safety facilities.

Traffic prediction is essential for advanced traffic planning, design, management, and network sustainability. Current prediction methods are mostly offline, which fail to capture the real-time variation of traffic flows. This paper establishes a sustainable online generative adversarial network (GAN) by combining bidirectional long short-term memory (BiLSTM) and a convolutional neural network (CNN) as the generative model and discriminative model, respectively, to keep learning with continuous feedback. BiLSTM constantly generates temporal candidate flows based on valuable memory units, and CNN screens out the best spatial prediction by returning the feedback gradient to BiLSTM. Multi-dimensional indicators are selected to map the multi-view fusion local trend for accurate prediction. To balance computing efficiency and accuracy, different batch sizes are pre-tested and allocated to different lanes. The models are trained with rectified adaptive moment estimation (RAdam) by dividing the dataset into the training and testing sets with a rolling time-domain scheme. In comparison with the autoregressive integrated moving average (ARIMA), BiLSTM, generating adversarial network for traffic flow (GAN-TF), and generating adversarial network for non-signal traffic (GAN-NST), the proposed improved generating adversarial network for traffic flow (IGAN-TF) successfully generates more accurate and stable flows and performs better.

About 90 of traffic crashes are caused by human factors, within which traffic violations are one of the most typical and common causes. In order to investigate the relationship between traffic violations and traffic crashes, this research targets signalized intersections in two Chinese cities: Yinchuan and Suqian. Thirty-one intersections are selected as the research sites, and additionally, the traffic volume, traffic violation, and traffic crash data of each intersection are collected for one year. A Whites test is conducted to test the homoscedasticity of the data and a multiple linear regression model is employed to investigate the relationship between traffic crashes and violations. The results show the following: (1) although the research sites are located in two different cities, the data is homoscedastic, which suggests that the above result may be statistically stable between different cities. (2) There is a significant multiple linear regression relationship (R²=0.782, adjusted R²=0.716) between the total number of traffic crashes and traffic violations. Among the chosen 7 independent variables, four are significantly related to the dependent variable, namely, driving commercial vehicle during internship, wrong-way entry, speeding, and traffic-light violation. (3) With the increase of annual average daily traffic (AADT), the number of total crashes goes up; however, the injury-or-fatality rate decreases, which means that intersections with smaller traffic volumes tend to have higher traffic crash severity. Based on the above conclusions, it is possible to conduct more targeted enforcement to improve the safety of intersections.