Global climate change has rendered transportation systems vulnerable to extreme weather conditions. A thorough assessment of network resilience is crucial to maintain the smooth operation of the road network. This study develops a three-dimensional resilience evaluation model that accounts for the impact of capacity degradation and sheds light on the resilience of road networks. We investigate the degradation mechanism of road network performance in a rainfall environment. Specific rainstorm weather scenarios were generated using the Monte Carlo simulation method. The road traffic network in the Chengdu metropolitan area was selected for validation and analysis. When meteorological conditions exceed the secondary intensity, the road network becomes more vulnerable to increased traffic flow. Betweenness and degree centrality are indicators of regional importance. Different recovery strategy priorities must be developed for different interference scenarios.

ObjectiveThe bus-style operation of intercity railways is an important measure to support and guide the development of urban agglomerations and metropolitan areas. Reasonably optimizing station stop schemes and service frequency is an effective way to meet travel demand within the region, enhance the efficiency of railway transportation enterprises, and promote the integration of urban agglomerations and metropolitan areas. Therefore, it is necessary to conduct an optimization study on the train service frequency of intercity railways under the bus-style operation mode. MethodBased on the characteristics of the bus-style operation of intercity railways, an alternative set of train stop schemes is proposed. Passenger travel and enterprise operation costs are analyzed, and a bi-objective optimization model is constructed with the two mentioned costs as objectives, train service frequency as the decision variable, and taking into account the constraints such as section capacity, passenger flow delivery volume, and station capacity. The NSGA-Ⅱ (Non-dominated Sorting Genetic Algorithm Ⅱ) algorithm is adopted to solve the proposed optimization model. Considering the interests of both passengers and enterprises, fuzzy logic compromise is applied to screen the solution set of the Pareto optimal frontier. Taking the bus-style operation of the Changsha-Zhuzhou-Xiangtan Intercity Railway as a case study, the obtained peak-hour intercity travel passenger flow is preprocessed, the optimization results and model-related parameters are discussed and analyzed. Result & Conclusion After optimization, six train station stop schemes (stop at every station, non-stop service and selective stop), and corresponding service frequencies (4 trains, 5 trains, or 6 trains) under Pareto non-dominated solutions are obtained. According to the screening results, four trains should be in operation during the peak hours, among which the service frequencies for the stop-at-every-station, non-stop service, and selective stop schemes are 1 train, 1 train, and 2 trains respectively. When the station stop cost fluctuates within the interval of 500-800 CNY, most of the compromise solutions are distributed along three optimal compromise distribution lines. Under such circumstances, within the acceptable range of different passenger travel costs, the uncertainty in train stop costs has the least impact on the enterprise operating costs.

Due to accelerating urbanization, there is a growing imbalance in passenger demand in terms of time and space distribution along the lines. To address this issue, urban rail transit lines have started implementing multi-routing and multi-composition operational plans to better manage complex passenger flows and reduce operating costs. Based on existing research, a space-time network diagram of multi-composition trains was designed using the physical network diagram as a reference. This optimizes transfer behavior within the network diagram and increases the space-time state of the transfer nodes. An optimization model for the multi-composition operational plan was constructed using composition and network constraints. The dual objective was to minimize enterprise costs and passenger travel time. Using the Beijing Changping Line as an example, the target cost reduction was 6.99% when the Lagrangian relaxation algorithm was employed. This suggests that operating multi-component trains can be more beneficial for enterprises and passengers. Finally, analysis of the network, composition types and routes showed that multi-component trains can reduce the infeasibility of passenger flow under line network conditions and increase the feasibility of passenger travel.

Coordinating timetabling and station stop plans is of significant importance. It enhances passenger travel and maximizes railway revenue. We analyze passenger flows and calculate the probability of a passenger selecting a particular train. This helps us distribute passenger flow effectively. Based on this, we propose a multi-objective optimization model for intercity railway timetables and station stop plans. The model aims to lower operational costs and reduce travel expenses. It includes constraints like departure times, stop durations, and stop frequencies. We use an Adaptive Large Neighborhood Search (ALNS) algorithm to solve it under bounded rationality. A case study on a specific intercity railway demonstrates the effectiveness of both the model and the algorithm. We also study key parameters to understand their effects on timetabling and station stops. The findings reveal that passenger allocation plays a significant role in coordinated optimization. When the minimum load factor threshold rises, there are fewer train stop plans. At the same time, passenger travel costs increase while enterprise costs decrease. This aligns with the operator's approach. Ultimately, this study offers valuable decision-making insights for railway operators and passengers alike.

[Objective]Re-coupling operation mode can ef-fectively improve the matching degree between the passenger flow and the transport capacity of urban rail transit,while it al-so poses challenges to the formulation of the train stock utiliza-tion plan.Therefore,it is necessary to conduct research on the connection between the train stock and different marshalling train trips under this mode.[Method]By analyzing the cha-racteristics of urban rail transit train stock utilization under the re-coupling operation mode and the connection situation of the train stock during the transition period from peak to off-peak passenger flow,aiming at minimizing the number of train stock in use and the connection time cost,and ensuring the balance of train stock utilization,relevant constraints such as the u-niqueness of train service connection,the consistency of train stock,maintenance plans,regular node connections,and con-nection of coupled train service nodes are comprehensively con-sidered.The MTSP(multiple traveling salesman problem)is introduced to construct an optimized train stock utilization mod-el during the transition period from peak to off-peak passenger flow in urban rail transit.Moreover,a MOFA(multi-objective firefly algorithm)based on MCS(multiply cooperative strate-gies)is designed.In the case study of a certain urban rail tran-sit line with significant imbalance of passenger flow in different periods,the feasibility and rationality of the above optimized model are verified.[Result & Conclusion]The optimized train stock utilization scheme of urban rail transit under the re-coupling operation mode can reasonably optimize the connec-tion between different marshalling train trips and reduce the op-eration costs for enterprises.

The cross-line operation of suburban railway (SR) and urban rail transit (URT) is a key way to improve transportation efficiency and promote regional transportation integration. Based on the cross-line operation of SR and URT, this study introduces the event activity network theory, constructs a mixed logit model to quantify the passenger route choice behavior, and proposes the supply and demand capacity matching index (SDCIM), which is used to measure the matching degree of line capacity and passenger demand. On this basis, a bi-objective nonlinear integer programming model is further constructed with the objectives of minimizing SDCMI and the total costs of cross-line operation. The research results show that after implementing cross-line operation, SDCMI is reduced by 37.1%, and the total costs are reduced by 10.3%. This study presents a theoretical framework and provides practical guidance for decision-making in multi-level rail transit cross-line operations, taking into account both efficiency and cost.

Short-term passenger flow prediction for urban rail transit is imperative for effective real-time scheduling, efficient resource allocation, and prompt emergency response. However, conventional models face challenges in accurately capturing passenger flow features, which are influenced by numerous sources. Consequently, to address the limitations of traditional models, this paper investigates the improved dual-channel convolutional neural network (DC-CNN) and long-short-term memory network (LSTM) to construct a combined model to deal with complex spatiotemporal data, capturing non-linear features and sequence dependencies. Secondly, the impact of data quality, station heterogeneity, and model parameter training strategy on prediction accuracy was considered. A multi-dimensional index system was constructed based on the ontological attribute characteristics of subway stations and location environment features. The research sample was classified as a typical station using the k-means clustering algorithm. Furthermore, the model training phase incorporates a hybrid tuning strategy that integrates manual parameter optimization and automated hyperparameter search algorithms. An early-stop mechanism is also integrated to balance the model performance and training efficiency. To conclude the research, an example analysis is carried out by combining Hangzhou Metro AFC data. Based on the station POI data and historical passenger flow data, the stations are grouped into four categories (Cluster 0, 1, 2, 3). The MAPE values of the passenger flow prediction results are 14.02%, 18.12%, 13.57%, and 13.82%, respectively. Through experimental validation, it is determined that the prediction accuracy of the DC-CNN-BiLSTM prediction model optimized by considering site heterogeneity and using an improved training strategy is enhanced in all types of sites. Furthermore, the average absolute percentage error can be reduced by up to 1.62% compared with that of the original data without site segmentation. Additionally, the sensitivity of different hyperparameters to the prediction accuracy of the model demonstrates significant heterogeneity. The present study has the potential to contribute to the theoretical framework for predicting short-time passenger flow in urban rail transit.

The train operation plan under skip-stop and overtaking mode is an effective measure to match passenger demand. In order to obtain a more reasonable operation plan, a concept of benefit loss that can reflect boarding passenger loading state and its duration is proposed. Based on the description of the train operation plan, an optimization model with strong compatibility is proposed by introducing some special station sets, such as no-skipping station set and overtaking station set. Some constraints about rolling stock scheduling are also considered to improve the comprehensiveness of the model. Finally, a local adaptive hybrid solution strategy is developed to handle the proposed mode by decomposing the problem into three sub-problems. The results show that: 1) When consider benefit loss, although total waiting time is increased by 3.28

The structural evolution and spreading dynamics of urban rail transit networks are key determinants of system performance and resilience. Recent work has improved path planning and the quantification of spreading processes. However, most studies still examine small or homogeneous networks and ignore node-scale growth, structural heterogeneity, and functional differentiation typical of large cities. This gap reduces model realism and weakens predictions of congestion and information diffusion that inform sustainable urban planning. To address this gap, we couple a path-based node-evolution framework with an SIR model under mean-field approximation and build a multidimensional representation of large-scale rail systems. Simulations show how node-scale changes reshape network evolution, reveal distinct spreading rates across line-level clusters, and map the timing of high-transmission-rate routes. The results clarify the hierarchical and quantitative flows of passengers and information in heterogeneous networks, improve model scalability, and offer guidance for strengthening infrastructure resilience and forecasting megacity dynamics.

To address the issue of imbalance between the supply and demand of rescue personnel at the early stage of emergencies,we construct an evolutionary game model for dispatching emergency rescue personnel.The approach integrates considerations of road damage to rescue networks by emergencies and secondary disasters,as well as the competitive psychology of disaster-stricken populations with bounded rationality towards rescue personnel dispatch.From the perspective of disaster victims,we analyze the game dynamics of dispatching rescuers across multiple disaster sites.Utilizing replicator dynamic equations,we simulate the dynamic evolution of strategies for rescue personnel dispatch at each disaster site,thereby constructing a multi-stage model for dispatching rescue personnel across multiple rescue and disaster sites.Case studies illustrate the dynamic evolution and optimal dispatching schemes for each disaster site.Results demonstrate the practicality and feasibility of solving rescue dispatch plans by comprehensively considering the multi-stage dynamic characteristics of emergency rescue processes and the bounded rational game theory among affected points.Additionally,sensitivity analysis of parameters provides dispatch plans suitable for three different rescue scenarios.The research provides a reference basis for decision-making in emergency rescue personnel dispatch.

The stable and efficient operation of rail transit networks (RTNs) is critical for the integrated development of metropolitan areas. However, numerous studies have indicated that RTNs are prone to large-scale cascading failures when subjected to disturbances. To address the limitations of traditional cascading failure models, this paper proposes an innovative cascading failure model for metropolitan areas RTNs, which incorporates nonlinear load fluctuations and the bounded rationality of passengers. This model aims to capture the cascading failure characteristics of RTNs with chaotic properties under 12 combination strategies. A single- and dual-parameter coupling analysis of chaotic evolution parameters and prospect theory parameters are conducted. Taking the RTN in the Chengdu metropolitan area as an example, both the static characteristics and cascading failure features of the network are analyzed. The findings reveal the following: (i) the RTN is a assortativity network and lacks small-world and scale-free properties. (ii) During network disturbances, a higher level of passenger familiarity with the network increases the likelihood of large-scale cascading failures. (iii) When passengers tend to avoid risks, stations with higher carrying capacity are more prone to failures. This study holds significant implications for ensuring the stable and reliable operation of rail transit systems within metropolitan areas.

Vehicle lane‐changing behaviour is often regarded as transient traffic behaviour while ignoring behavioural characteristics of the lane‐changing process. A combined prediction model based on wavelet transform (WT) and dual‐channel neural network (DCNN) is proposed to explore the selection behaviour of lane‐changing distance by taking lane‐changing behaviour in an urban inter‐tunnel weaving section. Firstly, the extracted lane‐changing data are analysed for correlation and noise reduction, and the main factors affecting lane‐changing distance are taken as input variables of the model. The trajectory data of the inter‐tunnel weaving section of the “Jiuhuashan‐Xi'anmen” tunnel in Nanjing, China, are used to improve the prediction of vehicle lane‐changing distance by training the model. The results show that the proposed WT‐DCNN model has high prediction performance when compared with existing artificial neural network (ANN), DCNN and wavelet neural network (WNN) models. The characterization and study of the typical lane‐changing behaviour in the weaving section can lay the theoretical foundation for the development of an urban inter‐tunnel weaving section management scheme.

During peak passenger flow periods, congestion propagation directly affects the operational safety and efficiency of multi-mode rail transit interconnections. By analyzing the key factors affecting congestion propagation, such as the train stop schedule, and considering parameters such as the basic reproduction number and propagation threshold, this study proposes a multi-mode rail transit susceptible-infected-recovered-susceptible (MRT-SIRS) epidemic model to analyze passenger flow congestion propagation. Simulation experiments and sensitivity analyses using data from the multi-mode rail transit in Beijing, China, were conducted to examine the influence mechanism of key factors on congestion propagation. The degree of influence of each factor was investigated using Gray correlation analysis. Each key factor, including the propagation and recovery rates, influences congestion propagation differently. The results of this study may provide theoretical support for the efficient operation and management of multi-mode rail transit systems.

Due to the complex operational characteristics of multi-level rail transit networks, such as cross-system and multi-level, passenger flow congestion must not only consider the steady state of homogeneous transportation networks but also reveal the deep-seated mechanism of congestion spreading between heterogeneous transportation networks. An analysis theory of travel paths based on Improved Prospect Theory (IPT) is proposed using generalized travel time and congestion degree as dual reference points. By organically integrating passenger travel modes and routes, a two-layer model of passenger travel mode selection based on Nested Logit-Improved Prospect Theory (NL-IPT) is constructed. On this basis, considering key influencing factors such as the stopping scheme, an improved Susceptible-Infected-Recovered (SIR) model of multi-level rail transit passenger flow congestion propagation under bounded rationality conditions is proposed. Taking the multi-level rail transit in Beijing, China, as an example, the propagation process of passenger flow congestion in multi-level rail transit is simulated and analyzed. Through the sensitivity analysis of critical factors such as gain and loss sensitivity coefficient, propagation rate, and recovery rate, the mechanism of the influence of key parameters on passenger flow congestion propagation is revealed. The results show that when the proportion of waiting passengers heading to subsequent stops of the arriving train is greater than or equal to 0.6, there will be slight fluctuations in the initial stage of congestion propagation. When this proportion decreases by 80%, the congestion propagation range decreases by 23.3%. The research provides a reference for the operation plans and management optimization of multi-level rail transit.

Short-term traffic volume prediction is crux for alleviating traffic gridlock. Considering the insufficiently extracted spatiotemporal and periodic characteristics of traffic stream in existing traffic volume forecast studies, this research presents a short-term traffic volume forecast model (WRNCLTCL) that considers spatiotemporal and periodic characteristics. Firstly, the wavelet threshold is used to denoise the initial traffic stream data. Secondly, the CNN-LSTM model is employed to capture the spatiotemporal features of the traffic stream. Considering the degradation problem that may be caused by the model with the increase of network depth, add residual neural units based on CNN. We employ the TCNLSTM model to acquire the periodic characteristics of the traffic stream. Finally, we combine the extracted spatiotemporal and periodic features, and utilize the fully connected layer to obtain the ultimate forecasted results. WRNCL-TCL is applied to real data sets in two different scenarios to validate the forecasting capability of the suggested model. Compared to the benchmark model and the ablation experiment, the consequences suggest that the proposed model exhibits favorable predictive capabilities and can serve as a theoretical foundation for traffic control.

In the initial stage of an emergency, rescue resources, such as rescuers, are often scarce, and victims at disaster sites under the condition of bounded rationality demonstrate competitive psychology in relation to the dispatch of rescue workers. Based on this, this study incorporates factors such as the bounded rationality of the disaster-stricken populations and their competitive psychology in terms of the number of rescuers into the analytical framework from the perspective of the disaster-stricken people and constructs an evolutionary game model with a multi-strategy set. The dynamic evolution process (EP) and the best rescuer dispatching scheme for each disaster site are shown through an example, and a disturbance analysis of the identified parameters is then carried out. The results reveal that the proposed rescuer dispatching model considers the multistage dynamic features of the emergency rescue process and the bounded rational game psychology among disaster sites, resulting in a more realistic dispatch scheme. For three different scenarios, such as the distance between the rescue point and the disaster site being relatively far and relatively close, and the subsequent rescue stage when the disaster situation is preliminarily alleviated, decision makers should make dispatch plans with timeliness, demand satisfaction, and comprehensive consideration of timeliness and demand satisfaction as the main decisions.

Traffic flow forecasting is critical in transportation research. However, the excessive nonlinearity and complexity of spatial and temporal correlations in traffic flow critically restrict the prediction accuracy. To cope with the challenge, a parallel–series combined deep learning prediction model is proposed. Firstly, the traffic flow data is decomposed into unique time spans consistent with positive rules (weekly, daily, and hourly cycles). Then, a deep learning model named Transformer-Graph Convolutional Attention Networks (TRGCAT) is further used to predict the multi-flow in the traffic network. TRGCAT firstly encodes and concatenates the current hourly, daily, and weekly periodic features of all traffic nodes in parallel, which decodes the middle output with spatial features, and ultimately predicts future single-step, short-term, and long-term multi-step traffic flow by using Graph Convolutional Attention Network (GCAT) in series. We conduct numerical experiments on open-source datasets PeMS, the results show that TRGCAT does better in spatial and temporal fusion and may acquire extra aggressive forecasting consequences than baseline methods.

Optimization of the travel paths is necessary to shorten travel time and reduce travel costs. This paper proposes the analysis theory of the travelers' low carbon route choice behavior based on improved prospect theory by analyzing the punctuality rate of travel time, the travelers' comfort level on the travelers' intermodal transport route optimization. Dual reference points are set according to the whether travelers have the prior experience as the judgment basis, constructing an optimization model of travelers' intermodal transport under bounded rationality. The results show that, in the travelers' intermodal transport route considering carbon emissions, the travel route selected by travelers will change with the road state. In this problem, the gain sensitivity coefficient alpha and the loss sensitivity coefficient beta have influence on the decision. When traffic crowding or bad weather occurs, travelers tend to be more sensitive to time. The setting of each parameter has an important impact on the comprehensive prospect values.

路网交通事故预测是实现道路管控、路线规划的最重要方式之一.考虑到路网中各路段特征与环境因素的影响,建立基于图卷积神经网络(GCN)和门控循环单元(GRU)的时空门控图卷积(STGRGCN)模型预测交通事故风险.通过GCN提取出道路间的空间关联性,通过GRU提取出环境因素中的时间关联性,再通过GCN与GRU的复合模块提取出时空关联性.选取美国全国交通事故数据集中洛杉矶市和休斯顿市相关数据对模型进行检验,STGRGCN模型的均方根误差、平均绝对误差以及召回率在两个城市分别为4.09、2.14、0.714和5.79、3.24、0.683,优于已有统计模型、机器学习模型以及复合模型.设计该除各模块的消融实验,证明该模型各模块皆有助于提升预测性能.

Traffic volume forecast the key to alleviating traffic congestion. However, the relationship between traffic data and outside factors makes the problem more complex. Existing traffic flow forecasting studies seldom consider the relationship between traffic data and outside factors. Therefore, we propose SGA-KGCN-LSTM to solve this problem. The model combines Savitzky-Golay (SG) filter, Knowledge Graph (KG), Graph Convolution Network (GCN), Long Short-Term Memory (LSTM), and Self-Attention Mechanism. Firstly, the SG filter can be employed to reduce the noise of traffic volume data. Then, aiming at the relationship between traffic data and outside factors, the KG theory is introduced, and the knowledge representation is applied to get the embedding of relevant knowledge. Secondly, new road features are obtained by integrating embedded information and traffic flow characteristics. GCN can be employed to acquire the spatial characteristic of traffic stream, and LSTM can be employed to extract the temporal characteristics of traffic stream. Finally, the input feature information is given enough weight by self-attention mechanism. The outcome is subsequently acquired by utilizing the fully connected layer to attain the ultimate consequence. The trajectory data of the Luohu taxi in Shenzhen are used in the experiment. The experimental consequence indicate that the SGA-KGCN-LSTM has high forecasting precision compared with the benchmark model and the ablation experiment.