Cunlai Pu

Nowadays, high-speed rail (HSR) has become an increasingly popular choice for travel. However, delays occasionally occur during HSR trips, and these delays can propagate throughout the network, causing widespread disruptions. Therefore, controlling delay propagation is a critical task in HSR network maintenance. Given the limited resources available for delay control in real-world applications (such as budgets, personnel, and facilities), it is crucial to identify the optimal set of station nodes, of a given size, that can most effectively resist the spread of delays when controlled. In this paper, we formulate an optimization problem for immune node selection to control delay propagation. To solve this problem, we propose a node propagation centrality metric that integrates both the susceptibility and infectivity of a node to quantify the importance of nodes in delay propagation. Furthermore, we develop a genetic algorithm to find an approximately optimal set of immune nodes, using centrality-based initialization and a one-hop neighbor-based variable neighborhood search strategy. Comparative experiments on real-world HSR networks demonstrate the effectiveness of our algorithm. Our work helps identify the most influential nodes in HSR networks for limiting delay propagation, assisting railway personnel in making informed decisions.

Link weight prediction extends classical link prediction by estimating the strength of interactions rather than merely their existence, and it underpins a wide range of applications such as traffic engineering, social recommendation, and scientific collaboration analysis. However, the robustness of link weight prediction against adversarial perturbations remains largely unexplored.In this paper, we formalize the link weight prediction attack problem as an optimization task that aims to maximize the prediction error on a set of target links by adversarially manipulating the weight values of a limited number of links. Based on this formulation, we propose an iterative gradient-based attack framework for link weight prediction, termed IGA-LWP. By employing a self-attention-enhanced graph autoencoder as a surrogate predictor, IGA-LWP leverages backpropagated gradients to iteratively identify and perturb a small subset of links. Extensive experiments on four real-world weighted networks demonstrate that IGA-LWP significantly degrades prediction accuracy on target links compared with baseline methods. Moreover, the adversarial networks generated by IGA-LWP exhibit strong transferability across several representative link weight prediction models. These findings expose a fundamental vulnerability in weighted network inference and highlight the need for developing robust link weight prediction methods.

Train delays are a common issue in high-speed rail (HSR) networks. If not addressed promptly, they can propagate across multiple stations, reducing operational efficiency, negatively impacting passenger experience and causing economic losses. In this paper, we investigate strategies to mitigate the widespread propagation of delays in HSR networks from the perspective of resource allocation. First, we model delay propagation as a Normal-Delayed-Normal (NDN) spreading process, incorporating both self-induced delays at a station and delays caused by neighboring stations. We then formulate an optimization problem that seeks to allocate limited resources to stations to effectively control delays. To solve this problem, we propose heuristic resource allocation strategies based on network centrality measures, such as node degree and node betweenness. Furthermore, we develop a simulated annealing algorithm to find the optimal resource allocation solution. Experimental results based on data from a real-world HSR network show that degree-based strategies outperform other heuristic methods, while simulated annealing achieves the best overall performance, significantly enhancing the resilience of the HSR system against delay propagation.

The propagation of delays is a prevalent issue in high-speed railway (HSR) networks. However, the underlying properties of delay propagation have not been fully explored, hindering the effective management of this problem. In this paper, considering the impact of traffic flow, we model the delay propagation as a traffic-driven susceptible–infected–susceptible (SIS) epidemic spreading process. We introduce a traffic frequency matrix to estimate the traffic rate and further derive the propagation dynamic equations using a continuous-time Markov chain method. We then derive the epidemic threshold of the model, which is proportional to the recovery rate and inversely proportional to the average path length. Finally, through simulation, we study the impacts of various factors including the infection rate, recovery rate, and the degree and number of initial delay stations. Our work provides some insights for the prediction and control of delay propagation in HSR networks.

Purpose: To optimize train operations, dispatchers currently rely on experience for quick adjustments when delays occur. However, delay predictions often involve imprecise shifts based on known delay times. Real-time and accurate train delay predictions, facilitated by data-driven neural network models, can significantly reduce dispatcher stress and improve adjustment plans. Leveraging current train operation data, these models enable swift and precise predictions, addressing challenges posed by train delays in high-speed rail networks during unforeseen events.Design/methodology/approach: This paper proposes CBLA-net, a neural network architecture for predicting late arrival times. It combines CNN, Bi-LSTM, and attention mechanisms to extract features, handle time series data, and enhance information utilization. Trained on operational data from the Beijing-Tianjin line, it predicts the late arrival time of a target train at the next station using multidimensional input data from the target and preceding trains.Findings: This study evaluates our model's predictive performance using two data approaches: one considering full data and another focusing only on late arrivals. Results show precise and rapid predictions. Training with full data achieves a MAE of approximately 0.54 minutes and a RMSE of 0.65 minutes, surpassing the model trained solely on delay data (MAE: is about 1.02 min, RMSE: is about 1.52 min). Despite superior overall performance with full data, the model excels at predicting delays exceeding 15 minutes when trained exclusively on late arrivals. For enhanced adaptability to real-world train operations, training with full data is recommended.Originality/value: This paper introduces a novel neural network model, CBLA-net, for predicting train delay times. It innovatively compares and analyzes the model's performance using both full data and delay data formats. Additionally, the evaluation of the network's predictive capabilities considers different scenarios, providing a comprehensive demonstration of the model's predictive performance.

The network infrastructures, such as the power grids and Internet, are expanding in size due to the increasing needs of our society. This brings about the problem of expanding networks with a guarantee of robustness against network disturbances that may cause catastrophic consequences. In this paper, we study the optimal network expansion in terms of network robustness against cascading failures. Specifically, we consider the network expansion as a Markovian decision process and further propose a reinforcement learning based network expansion method. Simulation results in model networks and real-world networks demonstrate that our expansion method can greatly improve network robustness. Our work provides some insights for the optimal expansion of network infrastructures.

Transport processes are universal in real-world complex networks, such as communication and transportation networks. As the increase of transport demands in these complex networks, the problems of traffic congestion and transport delay become more and more serious, which call for a systematic network transport optimization. However, it is pretty challenging to improve transport capacity and efficiency simultaneously, since they are often contradictory in that improving one degenerates the other. In this paper, we formulate a multi-objective optimization problem including two objectives: maximizing the transport capacity and minimizing the average number of hops. In this problem, we explore the optimal edge weight assignments and the associated routing paths, corresponding to the optimal trade-off between the two objectives. To solve this problem, we provide a multi-objective evolutionary algorithm, namely network centrality guided multi-objective particle swarm optimization (NC-MOPSO). Specifically, within the framework of MOPSO, we propose a hybrid population initialization mechanism and a local search strategy by employing the network centrality theory to enhance the quality of initial solutions and strengthen the exploration of the search space, respectively. Simulation experiments performed on network models and real networks show that our algorithm has better performance than five state-of-the-art alternatives on several most-used metrics.

Realistic epidemic spreading is usually driven by traffic flow in networks, which is not captured in classic diffusion models. Moreover, the progress of a node’s infection from mild to severe phase has not been particularly addressed in previous epidemic modeling. To address these issues, we propose a novel traffic-driven epidemic spreading model by introducing a new epidemic state, that is, the severe state, which characterizes the serious infection of a node different from the initial mild infection. We derive the dynamic equations of our model with the tools of individual-based mean-field approximation and continuous-time Markov chain. We find that, besides infection and recovery rates, the epidemic threshold of our model is determined by the largest real eigenvalue of a communication frequency matrix we construct. Finally, we study how the epidemic spreading is influenced by representative distributions of infection control resources. In particular, we observe that the uniform and Weibull distributions of control resources, which have very close performance, are much better than the Pareto distribution in suppressing the epidemic spreading.