Based on relevant complex network theory, this paper analyzes the characterization parameters of traffic network complexity, such as the causes of traffic congestion and evacuation models. From the perspective of the traffic capacity of traffic network nodes, combined with the transit time of each road, model is proposed for selecting the weights of congestion evacuation nodes. According to the evacuation target, combined with characteristic parameters, such as node degree, node strength, clustering coefficient and closeness, the grey system evaluation method and the analytic hierarchy process (AHP)are combined. Based on the grey relational analysis model, a model is established for determining the priority connectivity evaluation value of each node. The complex characteristics of the actual traffic network in the Chaoyang District of Changchun City are analyzed, then the selection weights of each node of the traffic network are obtained. Having defined the distance between complex network nodes, a congestion evacuation path selection model is proposed, and an evacuation path scheme is given for specified start and end points.

With the advent of the data-driven era, deep learning approaches have been gradually introduced to short-term traffic flow prediction, which plays a vital role in the Intelligent Transportation System (ITS). A hybrid predicting model based on deep learning is proposed in this paper, including three steps. Firstly, an improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) method is applied to decompose the nonlinear time series of highway traffic flow to obtain the intrinsic mode function (IMF). The fuzzy entropy (FE) is then calculated to recombine subsequences, highlighting traffic flow dynamics in different frequencies and improving prediction efficiency. Finally, the Temporal Convolutional Network (TCN) is adopted to predict the recombined subsequences, and the final prediction result is reconstructed. Two sensors of US101-S on the main road and on-ramp were selected to measure the prediction effect. The results show that the prediction error of the proposed model on two sensors is notably decreased on single-step and multistep prediction, compared with the original TCN model. Furthermore, the proposed improved CEEMDAN-FE-X framework can be combined with prevailing prediction methods to increase the prediction accuracy, among which the improved CEEMDAN-FE-TCN model has the best performance and strong robustness.

An efficient freight transportation system is a core part of the modern urban logistics. The transportation system depends on an efficient urban road network. In urban road networks, the failure of some components leads to others failing in succession, triggering cascading failures. This may cause the collapse of the freight transportation system. A cascading failure model is proposed for analyzing the propagation of failures on urban road networks, which can provide guidance for the planning of freight transportation routes in case of emergency. When analyzing the cascading failure process, failed components are commonly deleted permanently. Travelers will avoid failed components, resulting in no increase in traffic flow on the failed components. At the same time, vehicles on these failed components will evacuate slowly to alleviate congestion. Considering this point, this paper proposes a dynamic cascading failure model. This model is established based on a hybrid routing strategy in which traffic flow is distributed according to global and local information. Specifically, the new input traffic demand is assigned along efficient paths in an urban road network (using global information). Efficient paths are determined according to the weights of all edges in urban road networks. Meanwhile, failed edges transfer some traffic flow to their normally working neighbors by self-organization (using local information). The proportion of transferred traffic flow is determined by the residual capacity of the failed edges’ normally working neighbors. The feasibility of the proposed model is verified in a partial road network in Changchun, China.

This paper develops a distributed cooperative control strategy (DCCS) to plan and optimize the collision‐free trajectory for the connected automated vehicle (CAV) at an isolated roundabout. The representation of the trajectories of CAVs is defined within a roundabout. The central angle and lane radius are used to indicate the location of vehicles, and the corresponding angular velocity is used to describe the vehicle's speed through the roundabout. This trajectory definition method simplifies the proposed problem into two parts: the trajectory optimization of each vehicle in a specific lane and the lane change coordination of CAVs in the roundabout. To solve the trajectory optimization problem, a mixed‐integer linear programming (MILP) model is formulated at the vehicle level to minimize the travel time and make the trajectory as smooth as possible. A model predictive control (MPC) framework is designed to coordinate lane‐change conflicts and push CAVs’ trajectories toward global optimization. The proposed framework integrates the vehicles’ trajectory optimization results so that all CAVs reach a consensus. Numerical experiments test the effect of the strategy in terms of traffic efficiency and mobility performance under different demand pattern scenarios. The results show that the proposed RA‐DCCS (Roundabout‐DCCS) can make all CAVs smooth and conflict‐free trajectories and enable vehicles to execute reasonable lane change decisions.

As a green and sustainable trip mode, shared bicycles play an essential role in completing short-distance trips in cities. This paper proposes a method to analyze the impact of the urban built environment on the distribution of shared bicycles in a small-scale space. First, the Fishnet function of ArcGIS is utilized to divide the study area into grids of 500 m × 500 m. Then, three indicators are proposed to describe the characteristics of the urban built environment, including point of information (POI) comprehensive index, the intensity of public transportation coverage, spatial accessibility, providing them the ways to be assigned to the grids. Finally, the multivariable linear regression model and support vector regression (SVR) models are applied to reveal the impacts of built environment factors on the spatial distribution of shared bicycles. Results show that SVR models based on linear kernel function, Gaussian radial basis kernel function, and Polynomial kernel function can achieve better analysis results. The SVR model based on the Gaussian radial basis function shows higher explanatory power (adjusted ${\mathrm{R}}^2$ = 0.978) than the multivariable linear regression model (adjusted ${\mathrm{R}}^2$ = 0.847). This paper can aid in understanding the demand and supply of shared bicycles and help operators or governments to improve service quality.

CAVs (Connected Autonomous Vehicles) can be effective in improving the efficiency of transportation, but heterogeneous multi-modal traffic flows may hinder this efficiency. This paper addresses the issue of heterogeneous traffic flows affecting the efficiency of transportation when CAVs enter the market and proposes a joint dedicated lane for CAVs and buses. In the bi-level program model for the joint dedicated lane, the lower-level is aimed at the multi-modal traffic assignment problem, while the upper-level is aimed at system optimality. For the lower-level, the paper examines the characteristics of various traffic flows in a mixed traffic flow, investigates the impact of CAV mixing on the road link’s capacity, calculates the travel time of various traffic modes accordingly, and generates a generalized travel cost function for each mode, which is solved using the diagonalized weighted successive averaging method (MSWA) algorithm. The upper-level issue considers the continuity of dedicated and non-dedicated road segments, and the goal is to reduce the overall cost for all travelers by utilizing the dedicated road deployment scheme as the decision variable, which is addressed using a genetic algorithm. Finally, numerical examples and sensitivity analyses are designed accordingly. The numerical example demonstrates that the joint dedicated lane not only lowers the overall cost of the system, but also enhances the efficiency of CAV and bus travel, optimizing the road network and promoting bus and CAV travel modes. The sensitivity analysis shows that in order to set up a joint dedicated lane, the frequency of bus departures and the penetration of CAVs are conditions that must be considered, and that the benefits of a joint dedicated lane can only be fully realized if the frequency of bus departures and the penetration of CAVs are appropriate.