Search

Short-term traffic flow prediction is a core branch of intelligent traffic systems (ITS) and plays an important role in traffic management. The graph convolution network (GCN) is widely used in traffic prediction models to efficiently handle the graphical structural data of road networks. However, the influence weights among different road sections are usually distinct in real life and are difficult to analyze manually. The traditional GCN mechanism, which relies on a manually set adjacency matrix, is unable to dynamically learn such spatial patterns during training. To address this drawback, this study proposes a novel location graph convolutional network (location-GCN). The location-GCN solves this problem by adding a new learnable matrix to the GCN mechanism, using the absolute value of this matrix to represent the distinct influence levels among different nodes. Subsequently, long short-term memory (LSTM) is employed in the proposed traffic prediction model. Moreover, trigonometric function encoding was used in this study to enable the short-term input sequence to convey long-term periodic information. Finally, the proposed model was compared with the baseline models and evaluated on two real-world traffic flow datasets. The results show that our model is more accurate and robust than the other representative traffic prediction models. [ABSTRACT FROM AUTHOR]

Purpose – The purpose of this paper is to propose a bi-level programming optimization model to reduce traffic congestion of transportation network while evacuating people to safe shelters during disasters or special events. Design/methodology/approach – The previous optimization model for contra flow configuration only considered the character of the manager. However, the traffic condition is not only controlled by managers, but also depended on the root choice of travelers. A bi-level programming optimization model, which considered managers and evacuees' character, is proposed to optimize the contra flow of transportation network in evacuation during special events. The upper level model aims to minimize the total evacuation time, while the lower level based on user equilibrium assignment. A solution method based on discrete particle swarm optimization and Frank-Wolfe algorithm is employed to solve the bi-level programming problem. Findings – It is found that the bi-level programming based contra flow optimization model can improve evacuation efficiency and decrease evacuation time 30 per cent or more. With the increase of traffic demand, the evacuation time will decrease significantly by contra flow configuration. Research limitations/implications – In the optimization model, the background traffic is ignored for simplification and the contra flow is configured absolutely as 0 or 1, which ensures vehicles do not go back into the evacuation area. Practical implications – An efficient optimization model for traffic managers to reduce congestion and evacuation time of evacuation network. Originality/value – The new bi-level programming model not only considers managers' character, but also considers evacuees' reaction. The paper is aimed to optimize contra flow for transportation network. [ABSTRACT FROM AUTHOR]