To address the limitations of traditional manual inspection of rail transit infrastructure, such as low efficiency, safety risks, and the dependence of existing rail-mounted detection equipment on maintenance time gaps, which leads to blind spots and limited coverage, this study develops an integrated “End-Edge-Cloud-Surveillance” framework for autonomous unmanned aerial vehicle (UAV)-based intelligent inspection in rail transit. At the “End” layer, multi-source perception combining visible light, infrared, and LiDAR, together with visual-inertial state estimation, enables autonomous perception and task-level navigation. At the “Edge” layer, beyond-visual-line-of-sight (BVLOS) communication and secure, efficient data transmission mechanisms are established, alongside lightweight onboard inference for real-time defect and risk detection. At the “Cloud” and “Surveillance” layers, cross-scenario and multi-target inspection applications are conducted with global data analytics, while a low-altitude surveillance system integrating cooperative and non-cooperative surveillance is established to ensure regulatory compliance and operational safety throughout the entire process. The results demonstrate that this work systematically identifies the unique challenges and characteristics of the rail transit domain and, for the first time, unifies UAV-based rail transit inspection within a full-chain “End-Edge-Cloud-Surveillance” framework. This provides a generalizable reference framework for the future deployment of autonomous UAVs in rail infrastructure inspection.

To address the issue of excessive ground settlement caused by improper design of tunneling parameters in slurry shield machines, this study proposes an intelligent optimization method for shield tunneling parameters that integrates machine learning algorithms. First, a five-step optimization framework is established, comprising geological information encoding, engineering data processing, shield response parameter prediction, ground settlement prediction, and control parameter optimization. Then, a data preprocessing pipeline tailored to the characteristics of shield tunneling data is developed to construct a sample database. Next, solution algorithms are respectively formulated for shield response prediction, ground settlement prediction, and control parameter optimization by applying Long Short-Term Memory (LSTM) neural networks and the Particle Swarm Optimization (PSO) algorithm. Finally, the proposed method is validated using a case study of the large-diameter slurry shield tunnel section between the Jingha Expressway and Luyuan North Street in the Beijing East Sixth Ring Road underground renovation project. The results indicate that geological encoding is an effective means of incorporating unstructured geological information into machine learning models. The shield response prediction model and ground settlement prediction model, both incorporating geological encoding as input, achieve an R² of 0.92 on the test data, demonstrating strong predictive performance. Under the settlement control standard of -10 to 5 mm, the excavation parameters generated by the intelligent optimization method reduce the average ground settlement by 29% compared to the measured data, offering valuable guidance for practical engineering applications.

This study addresses lane-change obstacle avoidance for autonomous vehicles under sudden road hazards and proposes SafeLC-DelayDDPG, a vehicle control algorithm based on Deep Reinforcement Learning (DRL). The task is formulated as a Markov Decision Process (MDP), and a structured hybrid state space is constructed by integrating local observations, lane-level semantic information, and the ego vehicle’s global states to enhance environmental perception and risk sensitivity. The action space consists of continuous front-wheel steering angle and longitudinal acceleration. The reward function is centered on a two-dimensional time-to-collision (2D-TTC) metric, balancing safety, efficiency, comfort, and traffic-rule compliance, and employs a TTC-conditioned dynamic weighting mechanism that prioritizes safety under high risk and efficiency under low risk. Furthermore, delayed policy updates and target policy smoothing are introduced, and the Critic network loss is refined to mitigate the training instability and Q-value overestimation issues inherent in Deep Deterministic Policy Gradient (DDPG). The proposed method is validated through traffic simulations across diverse scenarios. Experimental results show that, compared with multiple baseline algorithms, SafeLC-DelayDDPG achieves superior safety and efficiency: during training, the first-attempt and consecutive obstacle-avoidance success rates improve by up to 17.9% and 60.5%, respectively; the safety metric by up to 7.6%; and the average speed by up to 2.1%. In cross-scenario tests, the first-attempt and consecutive success rates improve by up to 13.3% and 44.1%, the safety metric by up to 9.8%, and the average speed by up to 0.6%.

For the problem of energy-efficient operation in urban rail transit, existing studies can be grouped into two main strands: Classical methods and technology-driven approaches. The classical strand reviews optimization of train speed profiles, train timetables, and their joint optimization, showing that these methods have developed into mature modeling and solution frameworks capable of reducing traction energy consumption and peak power while enhancing the utilization of regenerative braking energy under safety and service constraints. The second strand summarizes technology-driven progress and discusses the enabling effects and challenges of emerging technologies across three domains: Power supply, control, and operations. On the power side, efforts focus on advancing “source-grid-storage-train” integration, coordinating reversible converters, energy storage, and renewable sources to achieve load shifting and peak suppression. On the control side, autonomous operation and virtual coupling shift the problem toward a dynamic perspective under multi-train coordination. On the operations side, cross-line through services, multi-route patterns, and flexible train formations upgrade capacity supply and reduce inefficient traction. The findings suggest that classical methods already provide well-established frameworks for safe and effective energy saving, while new technology-driven approaches move beyond train-level offline optimization and progressively drive research toward system-level coordination that spans power supply, control, and operations. Future research should place greater emphasis on developing unified benchmark and evaluation systems, while also advancing multi-level coordination and integrated optimization across diverse domains.

Due to constantly interacting targets, rapidly shifting environments, and the inherent heterogeneity of multi-sensor data, dynamic traffic scenarios impose stringent demands on the perceptual robustness and decision reliability of intelligent systems. Multi-modal learning emerges as a critical solution to overcome bottlenecks in dynamic scenario understanding by fusing heterogeneous modalities. This paper offers a systematic review on multimodal robust learning for dynamic traffic scenarios. First, we clarify the definition of multimodal dynamic traffic scenarios, and analyzes the types and dynamic characteristics of multi-source modalities (e.g., optical, radio frequency, and acoustic). Next, we lay out the fundamental principles of multi-modal learning, paying particular attention to key techniques that enhance robustness across data-level processing, model architectures, and training strategies. Furthermore, we delve into core challenges currently confronting the field and future research directions. The review highlights current challenging of data imperfections, model limitations, and the absence of evaluation benchmarks, and chart future directions toward three aspects: technological innovation, technology integration, and collaborative industry efforts. Our aim is to provide a systematic reference for both theoretical research and practical deployment of multimodal robust learning in dynamic traffic scenarios, facilitate the evolution of intelligent transportation systems from being “available in limited scenarios” to achieving “reliability across all conditions,” and provide critical technical support for the widespread adoption of intelligent transportation solutions.

To address the challenges of precise governance arising from the spatiotemporal heterogeneity and diverse patterns of pollutant and carbon emissions from road mobile sources in China, this study develops a machine learning-based analytical framework for emission characteristics, driving factors, and mitigation pathways. First, a transport-carbon-environment dataset is constructed by integrating multi-source data to analyze the spatiotemporal distribution characteristics of emissions. Second, a knowledge-constrained non-negative matrix factorization model is developed to identify distinct emission patterns and their underlying drivers. Finally, a multi-pattern integrated multiple linear regression model is applied to evaluate effective pathways for pollution and carbon reduction. The results indicate that in recent years, particulate matter emissions from road mobile sources have shown a decreasing trend, while carbon dioxide emissions continue to increase, with emissions of NO _x, CO, and VOCs remaining substantial. Three primary emission patterns are identified across China: a “CO₂-PM Reduction Pattern,” distributed across border and coastal regions and driven mainly by road passenger and freight transport activities and vehicle mileage (37.2%); a “Pollutant Reduction Pattern,” found in municipalities like Beijing and Tianjin and provinces such as Guangdong, mainly influenced by the Gross Transport Product (30.1%); and a “CO₂-NO _x Pollution Pattern,” located in the central and western regions, affected by multiple factors including railway development and its electrification, as well as the use of light-duty transport vehicles (23.6%). Pathway analysis indicates that accelerating economic development in the transport sector has the most significant impact on reduction, potentially reducing CO₂ and NO _x by up to 6.7% and 4.5%, respectively. Other significant measures include restructuring the transport energy system, reducing road freight and vehicle mileage, and promoting the shift from road to rail alongside railway electrification. The proposed framework provides a quantitative basis and strategic guidance for coordinated pollution and carbon reduction from road mobile sources in the context of China’s dual-carbon goals of carbon peaking and carbon neutrality.

The transformation and upgrading of the railway transportation production system represent a core issue and critical challenge currently faced by railway transportation enterprises in their transition toward modern logistics enterprises. To support the construction of a modern logistics-oriented railway transportation production system, the study first analyzes key aspects including top-level design, theoretical architecture, critical technologies, and implementation evaluation indicators. Subsequently, the key scientific and technological bottlenecks that need to be overcome in building the new system are discussed, proposing a key technological architecture oriented to modern logistics. Finally, based on this architecture, a simulation prototype system for railway production organization capable of optimization, evaluation, and simulation is designed to validate the feasibility and effectiveness of the proposed theoretical framework and technical pathways. The research demonstrates that this study offers certain theoretical reference value and practical significance for enhancing the operational efficiency and service quality of railway logistics, as well as promoting the transformation and development of transportation production. Additionally, the developed prototype system can provide a reusable development paradigm and decision-support tool for subsequent system optimization and engineering applications.

Visual intelligence, as a core branch of AI, seeks to endow machines with human-like capabilities for visual understanding and interaction. Since the breakthrough of deep learning in computer vision in 2012, the field has undergone four progressive stages of evolution. The first stage, represented by AlexNet, VGGNet and ResNet, leveraged large annotated datasets such as ImageNet to achieve remarkable success in closed-domain tasks (e.g., image classification and object detection), but its dependence on labeled data highlighted inherent limitations. The second stage witnessed the rise of self-supervised learning, with models such as MoCo, DION and MAE learning powerful visual representations from massive unlabeled data through contrastive, distillation, and masked reconstruction methods. The third stage marked a shift toward multimodal intelligence, where models like CLIP and GPT-4V integrated vision and language, enabling open-vocabulary understanding and advancing toward fine-grained, intent-driven reasoning. The current frontier is world models, exemplified by Sora, which aim not only to perceive and describe but also to simulate and predict the physical world, paving the way for embodied intelligence capable of interacting with reality. The current frontier is world models, exemplified by Sora, which aim not only to perceive and describe but also to simulate and predict the physical world, paving the way for embodied intelligence capable of interacting with reality. This fundamental transformation from discriminative understanding to generative simulation of the world marks a new consensus: generative modeling is the new deep learning. This survey follows this developmental trajectory, analyzing the core ideas, representative models, and methodological paradigms at each stage, while highlighting ongoing challenges in robustness, reasoning, and generalization. This survey follows this developmental trajectory, analyzing the core ideas, representative models, and methodological paradigms at each stage, while highlighting ongoing challenges in robustness, reasoning, and generalization.

With the continuous evolution of intelligent railway systems, novel railway applications based on spatiotemporal information from the BeiDou Navigation Satellite System (BDS) have garnered significant attention. However, the lack of dedicated testing systems tailored to specific railway applications makes it difficult to meet the full lifecycle requirements of BDS-based railway systems. To address this problem, this paper focuses on exploring the requirements and development potential for specialized testing and evaluation of BDS-based railway applications. It investigates how to leverage effective mechanisms to build dedicated testing instruments and system environments that address the unique characteristics of the railway industry, thereby meeting the full lifecycle needs of BDS-based railway systems. First, existing relevant work is surveyed and summarized, outlining the development needs and vision for dedicated testing of BDS-based positioning in railway. Second, a comprehensive overall architecture for a dedicated railway BDS positioning testing system is established, centered on the concept of "zero-on-site testing". This architecture aims to adopt a modular approach and inter-connected collaborative mechanisms to forge a new pathway that balances conventional satellite positioning testing paradigms with the application characteristics of the railway industry. Third, this paper uses the train positioning unit of a specific type of train control system as a test subject example, detailing the setup of the dedicated test system and the implementation of testing. Finally, this paper analyzes and summarizes the research challenges that may arise in providing comprehensive and reliable testing and evaluation for the progressively developing applications of BDS in railway, offering corresponding approaches. The research results indicate that constructing a dedicated railway BDS positioning testing system can effectively enhance testing coverage, flexibility, adaptability, and compatibility, thereby providing strong support for the deep application of BDS in the railway industry.

Most existing quality evaluation methods for high-speed railway ballastless track construction are established based on traditional construction processes. As ballastless track construction progressively shifts towards intelligent construction, the existing evaluation methods can no longer adequately meet the requirements for quality evaluation of intelligent ballastless track construction.Taking the construction of the CRTS Ⅲ slab ballastless track as an example, this paper analyzes the key quality factors of intelligent construction processes, constructs an evaluation index system, and establishes a comprehensive quality evaluation model for intelligent ballastless track construction based on the Analytic Hierarchy Process (AHP) and the matter-element extension evaluation theory. Field tests are carried out on an intelligent construction test section of a high-speed railway for verification.The results show that the model established in this paper can objectively evaluate the quality of intelligent ballastless track construction. Compared with the traditional process, the guided intelligent construction process achieves a 32% improvement in the excellent rate. For the intelligent construction of the CRTS Ⅲ slab ballastless track, the construction quality of the track slab is the most important, followed by the quality of the self-compacting concrete construction, while the construction quality of the base slab has the least impact.The research results can provide a basis for evaluating the quality of intelligent construction of CRTS Ⅲ slab ballastless tracks for high-speed railways and offer a reference for the quality management of ballastless track construction in China’s high-speed railways.

To meet the stringent requirements for low latency, high reliability, and intelligence posed by emerging applications such as autonomous driving and the Industrial Internet, this study examines the limitations of traditional network architectures in heterogeneous resource integration, service adaptation, and intelligent decision-making. A smart computing integration network architecture based on a “three-layer, three-domain” framework is proposed, and its development prospects are discussed in relation to the deep integration of heterogeneous networks, computing resource scheduling, and network-native intelligence. The research results demonstrate that the proposed smart computing integration network architecture enables a paradigm shift from “passive connection” to “active service” through key technologies such as unified resource representation, computing-network demand analysis, and agile resource scheduling, thereby achieving comprehensive coordination between computing and networking. The research findings provide theoretical references and technical pathways for the construction of intelligent, efficient, and reliable emerging network infrastructures.

In response to the challenge of extracting information from railway signal system station yard diagrams, this study proposes a primitive detection model, YOLO11-AT, based on an improved YOLO11. By constructing a detection model that integrates object detection and keypoint detection, it achieves automatic extraction of primitives and keypoints. First, an Attentional Scale Sequence Fusion (ASF) module is incorporated into the neck network to fuse multi-scale features, thereby enhancing detection performance for small targets. Second, a Task-Aligned Dynamic Detection Head (TADDH) is implemented in the head network, which improves feature interaction between classification and localization tasks through task alignment, reduces feature conflicts, and increases detection accuracy for densely distributed targets. Finally, Slicing Aided Hyper Inference (SAHI) is applied to improve detection accuracy on high-resolution station yard images. Experiments are conducted on a constructed dataset containing multi-style station yard diagrams to validate the proposed method. The results show that, compared with YOLO11s-pose, the proposed YOLO11-AT improves precision, recall, mAP0.5, and mAP0.5-kp by 9%, 2.2%, 4.2%, and 3.2%, respectively, while reducing the number of parameters by 4.3%. Compared with existing mainstream detection models, YOLO11-AT achieves a better balance between detection accuracy and efficiency. The results indicate that the proposed method is adaptable to various styles of station yard diagrams and provide a feasible solution for automated information extraction from station yard drawings.

Current research on optimizing and controlling traction energy consumption in urban rail transit has largely focused on localized improvements within single systems, making it difficult to coordinate inter-system coupling and achieve global energy efficiency. To address this problem, this study proposes a multi-system collaborative optimization method that integrates power supply, vehicles, and operations. By analyzing the circulation paths of traction energy consumption and the inter-dependencies among decision variables, a four-layer global optimization framework is developed. The first layer optimizes single-train driving curves according to specified interval running times, identifying energy-minimal trajectories under constraints such as punctuality, comfort, and speed limits. The second layer optimizes inter-station running times by adjusting train travel durations between adjacent stations, thereby minimizing overall line energy consumption while ensuring that the total turnaround time equals the given value. The third layer focuses on optimizing multi-train energy-efficient timetables. By adjusting departure intervals and dwell times, this layer reduces total traction energy consumption while balancing passengers’ average waiting times through multi-train coordination. The fourth layer addresses threshold optimization of energy storage and inverter devices. The results of the first three layers of the train scheduling and control optimization process are used to inform the adoption of a grid-voltage-based control strategy. Simulation studies are undertaken in actual line operations. Simulation results demonstrate that the proposed global optimization reduces system traction energy consumption by 13.24%, while increasing passengers’ average waiting time by only 12 seconds. These results verify the effectiveness of the proposed coupled energy-saving optimization method and provide a theoretical foundation for the comprehensive optimization and control of traction energy consumption in urban rail transit.

To investigate the decision-making mechanisms of stranded passengers’ travel mode choice during unexpected metro service disruptions and to clarify the influence of key situational factors, this study constructs a baseline Multinomial Logit model incorporating mode-specific attributes, based on scenario simulations and Stated Preference (SP) surveys. A stepwise utility-correction approach is applied to quantify both the impact intensity and patterns of influence exerted by factors such as expected recovery time, weather conditions, travel time and purpose, remaining trip distance, crowd behavior, and the availability of alternative routes. Model improvements are evaluated using the Likelihood Ratio Test (LRT), Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and prediction error rates. The research results indicate that passenger decision-making demonstrates strong scenario dependence and a pronounced risk-averse tendency. The effects of situational factors vary considerably, with expected recovery time having the greatest impact (LR=183.98, prediction error reduced by 1.94%), followed by weather conditions (LR=102.63, error reduced by 1.16%). Trip purpose and remaining journey distance also show notable influence, while crowd behavior and alternative-route availability do not exhibit significant effects. Preferences shifts across public transport modes display similar patterns of change. however, waiting on-site, though frequently chosen as a passive option, maintains a considerable selection rate but declines significantly when systemic and situational risks overlap. Sensitivity to scenario factors differs markedly across modes, with those offering higher certainty and lower risk proving more attractive in disruption scenarios.

To address the issues of low detection accuracy, slow detection speed, and frequent missed or false detections in railway track foreign object intrusion detection, this study proposes a lightweight railway track foreign object intrusion detection algorithm based on shallow feature fusion (YOLO-LSF). First, building on the YOLOv8n feature extraction network, the C2f module is improved based on GhostConv to construct the C2f_Ghost module, thereby reducing both the parameter count and computational cost of the model. Second, the MLCA attention mechanism is introduced at the end of the backbone network to enhance the feature representation of the target area and optimize the feature extraction efficiency of the model. Third, deformable convolution DCNv2 is employed to replace some ordinary convolutions in the C2f module of YOLOv8n, constructing the C2f_DCNv2 module and further strengthening the model’s feature extraction capacity. Finally, shallow feature information from the backbone network is integrated into the neck network, effectivelt mitigating detail loss caused by multiple convolution operations and enhancing the model's ability to detect distant foreign objects (small targets). Experimental results show that on a self- constructed railway track foreign object intrusion detection dataset, compared with the original YOLOv8n algorithm, the YOLO-LSF algorithm achieves an improvement of 5.2% in average precision, 3.37% in FPS, a reduction of 20.1% in the number of parameters, and a decrease of 22.2% in computational complexity. These results verify that the proposed algorithm significantly enhances detection accuracy and speed in complex environments while reducing the likelihood of missed and false detections.

Amid the increasing frequency of extreme climate events worldwide, urban transportation systems face dual pressures from structural vulnerabilities and climate-related risks, highlighting the urgent need to enhance climate adaptability and comprehensive resilience. Firstly, this paper systematically reviews the primary impacts of climate change on urban transport network structures and residents’ travel behaviors, revealing issues such as reduced network accessibility and temporal-spatial mismatches in travel demand caused by meteorological hazards. Then, from the perspective of physical and social resilience, the study analyzes the evolutionary characteristics of infrastructure disturbance resistance and recovery capabilities, traveler risk-response behaviors, and governance system adaptation mechanisms. Finally, the paper summarizes sustainable mobility policies guided by low-carbon objectives and transport resilience policy frameworks centered on disturbance absorption and recovery, explores synergistic governance pathways formed through their integration, reviews recent research progress on urban transport networks in infrastructure optimization, coordinated allocation of transport capacity resources, and transport-energy system coupling, and proposes an integrated optimization approach for constructing a multi-modal synergistic, low-carbon resilience-integrated transport system. The findings indicate that enhancing climate adaptability in future urban transportation systems should focus on multi-modal network optimization and technology empowerment, strengthen social equity and policy coordination, and promote a governance shift from infrastructure resilience to system resilience and from single-objective to multi-objective coordination. It is necessary to deeply integrate the coupled development of transport and energy systems, establish a full-cycle resilience assessment framework covering resistance, absorption, recovery, and adaptation to fill the current evaluation gap that focuses on post-climate-disaster impacts while neglecting long-term evolutionary processes. Additionally, integrated modeling of physical and social resilience, combining network topology, travel behavior, and governance mechanism analysis, should be refined to provide customized infrastructure optimization solutions and policy support for cities with diverse climate characteristics.

To address the complex and dynamic spatiotemporal features in traffic flow prediction, this paper proposes an Attention-based Spatiotemporal UNet with Graph Convolutional Network (AST-UNet-GCN) model for long-term traffic flow forecasting. First, for temporal modeling, a U-Net-based feature pyramid architecture is introduced to capture multi-scale temporal features. The encoder-decoder structure enables effective extraction of features across different time scales. To enhance adaptability to sudden events, a short-term temporal feature extraction module based on attention mechanisms is designed. Furthermore, a global temporal feature extraction module utilizing deformable attention mechanisms is constructed to capture long-term temporal dependencies. Then, a Squeeze-and-Excitation (SE) feature fusion module is developed to improve the expressiveness of the convolutional neural network, enabling dynamic weighting of features at different time scales and enhancing the fusion of multi-scale temporal information. This approach effectively highlights key features while suppressing redundant information. Finally, for spatial modeling, a Graph Convolutional Network (GCN) is employed. By constructing the topological graph structure of the traffic network, the model captures spatial dependencies. A sigmoid-based feature fusion mechanism is further designed to explore the intricate dynamic relationships between spatial and temporal features, enabling comprehensive spatiotemporal modeling. Experimental results demonstrate that compared to other mainstream models, the AST-UNet-GCN model achieved reductions of 8.5% and 9.4% in MAE and RMSE metrics for short-term prediction, while reductions of 10.4% and 6.5% were observed for long-term prediction, respectively. These results demonstrate the model’s strong performance in traffic flow forecasting, particularly in immediate prediction accuracy and the stability of long-term trend forecasting.

To address issues arising from operational interference between mainline and cross-line trains, specifically the “long-distance but short-flow” problem for cross-line trains and the “passenger flow without sufficient train supply” issue for mainline trains, this study investigates the coordinated optimization of stop patterns and passenger flow allocation for both train types. The goal is to improve the matching between train capacity and passenger demand under existing operational conditions. First, the physical high-speed rail network is simplified to focus on a corridor where mainline and cross-line trains operate concurrently. The impact of different train operation schemes on passenger flow allocation along this corridor is analyzed. Subsequently, a service network is constructed based on train operation segments, and a mixed-integer programming model is developed to jointly optimize the operation schemes of mainline and cross-line trains, minimizing both railway operational costs and passengers’ travel time loss. To enhance computational efficiency for large-scale problems, a heuristic algorithm based on Lagrangian relaxation is designed, which decomposes the original problem into simpler subproblems by relaxing coupling constraints. Finally, the Wuhan-Guangzhou section of the Beijing-Guangzhou high-speed railway is used as a case study. High-quality feasible solutions are obtained within a short computation time, validating the proposed model’s and algorithm’s effectiveness. The research results indicate that compared to the sequential optimization approach, the coordinated optimization scheme reduces the number of train services by one. Analysis of the proportion and seat occupancy rate of cross-line passengers further confirms the improved passenger flow allocation under coordinated optimization. The division of roles between mainline and cross-line trains becomes more distinct, enhancing the match between train capacity and passenger flow under current operating conditions. These findings provide valuable insights for railway operators.

To enhance the efficiency of bus-metro connectivity, this study conducts a coordinated optimization of the spatial-temporal distribution of passenger flows and transfer efficiency for feeder bus routes serving metro station clusters. A multi-objective optimization model is proposed, aiming to minimize the total system cost while maximizing network transfer demand. A penalty mechanism for both transfer time and number of transfers is incorporated to constrain cases involving two or more transfers, encouraging the system to reduce unnecessary transfers during route design. To solve the model, a hybrid genetic particle swarm optimization algorithm is developed, integrating an adaptive elite retention strategy and a dynamic inertia weight adjustment mechanism. Results indicate that, compared with the original bus network, the optimized system increases bus carrying capacity by approximately 23%, reduces average passenger travel cost by about 9%, and improves algorithm efficiency with a 15.4% faster runtime compared to conventional genetic algorithms. The proposed model demonstrates superior performance across multiple key indicators, including transfer appeal and travel cost, thereby validating its effectiveness in improving the operational efficiency and service quality of feeder bus networks. This work offers valuable insights for the refined management and intelligent upgrading of urban public transportation systems.

The configuration of hot standby EMUs in high-speed railway networks directly affects emergency response efficiency and operational costs. To address issues such as coarse management and inadequate risk coverage in existing configuration approaches, this study proposes an optimization method based on comprehensive accident risk coverage across the network. First, considering both the timeliness and cost-effectiveness of emergency response, a multi-objective planning model is constructed with dual objectives to minimize emergency response time and configuration costs for hot standby EMUs. The model integrates point coverage and arc segment coverage methods to accurately represent the requirements for sequential tasks and inter-section rescue scenarios. It incorporates constraints related to emergency rescue point demands, response time limits, and full coverage of risk points, enabling effective adaptation to complex railway networks and the uncertainty of sudden incidents. Second, the ε-constraint method is used to solve the model, overcoming limitations associated with traditional weighted coefficient methods such as dimension normalization and parameter setting. This approach generates multiple sets of Pareto-optimal solutions, providing flexible options for decision-making. Finally, a case study is conducted on the high-speed railway network managed by a specific railway bureau. The research results indicate that the model produces reasonable configuration schemes aligned with various optimization objectives. As the number of hot standby EMUs increases, the maximum reduction in emergency response time reaches 29.3%, while configuration costs remain effectively controlled. Sensitivity analysis confirm the model’s strong adaptability to varying response time constraints and risk point distribution characteristics. This method provides a theoretical basis and practical guidance for the scientific configuration of hot standby EMUs and serves as a reference for enhancing the emergency management system of high-speed railways.

To address challenges such as passenger crowding, excessive waiting times, and inefficient use of transportation resources in metro-to-high-speed rail transfer scenarios, this study proposes an optimization method for metro-to-high-speed rail passenger flow dispersion based on Multi-Agent Reinforcement Learning (MARL). The method dynamically adjusts metro timetables to enhance passenger dispersion efficiency, alleviate crowding, and improve the utilization of transportation resources. First, the metro-to-high-speed rail passenger flow dispersion optimization problem is formulated as a Markov game by integrating the spatiotemporal information of metro operations and the spatiotemporal characteristics of passenger transfers, with general state features, action space, and a reward function specifically designed. Second, a multi-agent decision-making model is then developed using the Actor-Critic (AC) framework, and an asynchronous action coordination mechanism is introduced within a centralized training and distributed execution architecture to enhance training efficiency. Finally, an optimization study is conducted using the Tianjin West railway station as a case study. Results indicate that the proposed method significantly reduces passenger waiting times and improves metro operational efficiency. The average passenger waiting time decreases by 26.80%, while the average metro operational efficiency increases by 14.11%.

To fully utilize the transport capacity of high-speed railways, improve operational efficiency, and enhance economic benefits, this study addresses the vehicle-cargo matching optimization problem in high-speed rail express freight services. The research comprehensively considers three transport modes, passenger-freight mixed operation, reserved carriages, and dedicated EMU trains, and four categories of goods: fresh produce, urgent documents, electronic products, and valuables. A two-stage stochastic programming model is proposed. In the first stage, train operation plans are determined, including the selection of transport modes for each train and decisions on whether to dispatch dedicated EMU trains. In the second stage, based on the train plans from the first stage, optimal vehicle-cargo matching schemes are determined under various demand scenarios. To solve the model, a Mixed-Integer Programming-based Genetic Algorithm (MIP-GA) is designed. Finally, case studies are conducted using data from the Beijing-Shanghai high-speed railway, including four types of goods with stochastic demand, 5 500 containers, and 24 trains comprising a total of 254 carriages. The model is validated through analysis of the value of stochastic solutions and further tested across 12 case sizes to evaluate algorithm performance. The results show that the stochastic programming model effectively manages demand fluctuations, reducing costs by 3.8% and increasing the on-time delivery rate by 13% compared to the average-demand model. The MIP-GA significantly accelerates computation, saving an average of 75.88% in solving time, with an average optimality gap of only 0.20% compared to Gurobi, thereby enhancing computational efficiency without compromising solution quality.

With the continuous expansion and technological upgrading of China’s high-speed railway systems, the electromagnetic complexity of the operating environment has significantly increased. In particular, the frequent appearance of strong external electromagnetic interference sources has posed unprecedented challenges to the stable operation of the system. The traditional electromagnetic compatibility (EMC) framework, which focuses primarily on improving device immunity, shows evident limitations when dealing with random and strong external electromagnetic interference, such as delayed response and poor adaptability. These shortcomings make it difficult to meet the current high demands for system functional stability, mission continuity, and rapid recovery. To address these challenges, this paper systematically reviews the current research progress in the field of EMC for high-speed railways in China, with a particular focus on key technical aspects such as interference source modeling, coupling path identification, disturbance immunity evaluation of critical equipment, and system-level protection strategies. On this basis, a new research framework for electromagnetic safety is proposed, with train control capability retention as the core objective. This framework aims to shift from "passive immunity" to "intelligent defense and autonomous recovery," and it includes essential components such as intelligent interference identification and perception, multi-physical-field coupled network modeling, electromagnetic interference risk propagation mechanisms, active protection technologies, and experimental validation methods. Research findings indicate that electromagnetic safety, as an extension of the traditional EMC system, represents a key pathway to ensuring the high safety and high reliability of high-speed railway systems under complex electromagnetic environments. The outcomes of this study can provide theoretical foundations and technical references for the future design and engineering implementation of electromagnetic protection in high-speed railway systems.

To systematically summarize the main themes, methods, and findings in current research on the resilience of comprehensive transportation systems, identify existing gaps, and provide insights for future research, this study conducts a detailed review of relevant literature on the resilience of comprehensive transportation systems. More than one hundred articles are retrieved from major Chinese and international academic databases. The review synthesizes existing work from several perspectives, including research objectives and scopes, resilience evaluation indicators, integrated transportation network modeling approaches, disruption and recovery scenario design, and strategies for enhancing system resilience. Current research can be broadly categorized into two spatial scales, intercity and intra-urban, with primary objectives focusing on resilience assessment, identification of critical components, and resilience enhancement. Regarding resilience evaluation indicators, existing studies predominantly adopt two categories: network-analysis-based indicators and traveler-oriented indicators, alongside some integrated indicators and metrics tailored to multimodal transportation. For modeling integrated transportation networks, the literature is reviewed from the perspectives of multilayer network structures, disruption and recovery scenario configurations, and network dynamics. In terms of resilience enhancement, current strategies typically fall into three groups: pre-disaster prevention, post-disaster response, and optimized recovery planning. The results demonstrate that future research should place greater emphasis on demand-side factors and incorporate multimodal transportation characteristics more thoroughly. Disruption scenarios and passenger behavioral responses require more accurate modeling, with fuller consideration of slow modes of travel. Dynamic analyses should pay closer attention to realism and the influence of travel information guidance. Additionally, the complexity of integrated transportation systems and the heterogeneity across different transport modes warrant further investigation.

To address the limitations of existing road defect detection algorithms, such as low accuracy in complex backgrounds, limited generalization capability, and frequent missed detections of small objects, this study proposes an improved YOLOv8-based detection algorithm. First, a Coordinate Attention (CA) mechanism is integrated into the backbone network layer to introduce positional information, enabling the model to better capture spatial dependencies and enhancing its feature discrimination ability under complex background conditions. Second, the Path Aggregation Network (PANet) in the neck network layer is replaced with a weighted Bi-directional Feature Pyramid Network (BiFPN). By incorporating bidirectional connections and learnable weights, the network facilitates bidirectional information flow across different resolution levels, leading to more effective fusion of low-level positional features with high-level semantic features and improving multi-scale feature representation. Finally, small-object feature maps are introduced to more accurately capture the small-object characteristics and reduce missed detections, thereby improving detection precision. Experimental results show that on the RDD2022 road defect dataset, the improved algorithm increases mean Average Precision (mAP) by 3.1% compared to the original version, while reducing model parameters by 2.3%, achieving more accurate and rapid road defect detection.

To address the challenges of complex backgrounds, deep placement and difficulty in recognizing and measuring cracks in plate rubber bearings of beam bridges, this study proposes an automatic crack detection and parameter calculation method based on dual-stage semantic segmentation using the YOLOv8-ESF framework. First, the EfficientViT backbone is integrated into the backbone network. Based on this, the Bottleneck structure within the C2f module is reconstructed to form the C2f-Faster-EMA module, which replaces part of the original C2f modules in the YOLOv8n backbone and incorporates decoupled heads, thereby enhancing the model’s ability to capture multi-scale detail features of bearing cracks. Second, the improved YOLOv8n model is employed to segment and extract the entire bearing region. Subsequently, the same model is used to further segment crack regions within the extracted bearing image. Crack parameters are then obtained by extracting the skeleton centerline and searching for the maximum outer rectangle method. Finally, the model is validated and evaluated from three aspects: network architecture, crack dataset, and segmentation accuracy. Experimental results show that YOLOv8-ESF model achieves over 85% accuracy in terms of mPA, DSC, and IoU for both bearing region and crack recognition. In field tests on real bridges, the maximum deviation between the crack parameters calculated via the dual-stage semantic segmentation method and manual measurements is less than 0.1 mm, meeting practical engineering requirements.

To address the challenge of accurately calculating the braking force of eddy current brakes (ECBs) in rail transit, this study proposes a three-dimensional electromagnetic model and analyzes the effects of various parameters on braking performance. First, the expression for static air-gap magnetic flux density is derived using the equivalent magnetic circuit method and Maxwell’s equations. The influence of the transverse edge effect on braking force is examined, and a correction coefficient for the conductivity of the conductor plate is derived. An equivalent eddy current density model is then established, leading to analytical expressions for both air-gap flux density and eddy current density. Subsequently, the impact of the longitudinal end effect on braking force is analyzed, and an expression for the additional air-gap flux density is derived. Based on the relationship between eddy current density and air-gap flux density, an analytical ex-pression for braking force is obtained. Finally, a three-dimensional finite element model is constructed using the design parameters of the Shanghai TR-08 high-speed maglev train’s ECB to verify the accuracy of the proposed model. The three-dimensional electromagnetic model is also used to analyze the influence of different parameters on braking performance. The results indicate that the maximum error between the three-dimensional electromagnetic model and the three-dimensional finite element model is 5.18%, with an average error of 2.35%, validating the accuracy of the three-dimensional electromagnetic model. The braking force increases initially and then decreases with rising speed, exhibiting four distinct characteristics in its variation curve: linearity, criticality, attenuation, and stabilization. The additional braking force increases with speed. Properly widening the conductor plate can effectively mitigates the influence of the transverse edge effect. The braking force increases with higher ampere-turns of the excitation winding and a wider primary core, while it first increases and then decreases with increasing conductivity and permeability of the conductor plate and with greater primary pole distance. The braking force decreases as the air-gap thickness increases, and the speed corresponding to peak braking force also increases. A thinner conductor plate slows the growth of braking force. Both increasing the air-gap thickness and reducing the conductor plate thickness contribute to a more stable braking performance.

To explore the nonlinear effects of the built environment on urban rail transit station ridership, this study takes Beijing’s rail transit stations as its focus. Leveraging multi-source data, including Point of Interest (POI) data, mobile phone signaling data, and road network data, the built environment is finely characterized from four perspectives: socio-economic and demographic attributes, land use, multimodal connectivity, and station characteristics. A Gradient Boosting Decision Tree (GBDT) model is employed to reveal the relative importance, nonlinear influences, and threshold effects of these factors on station ridership across different time periods and travel directions. Results indicate that the GBDT model outperforms the Ordinary Least Squares (OLS) model, Adaptive Boosting (AdaBoost), and Random Forest (RF) models in terms of fitting performance. Socio-economic and demographic attributes exert the greatest influence on peak-period ridership, contributing over 40% to both morning inbound and evening outbound peak-period ridership. Land use attributes have the strongest impact during off-peak periods, accounting for 33.06% and 49.10% of inbound and outbound ridership, respectively. The most influential variables exhibit pronounced nonlinear and threshold effects on station ridership. Notably, stations located 15 to 22 km from the city center show significantly higher morning inbound and evening outbound ridership. Additionally, the proportion of passengers accessing rail transit via private cars is considerably higher during peak hours than during off-peak periods. These findings provide valuable theoretical support for the planning of urban rail transit networks and the spatial layout of urban land use.

The stable operation of switch machines is a key guarantee for the safety of High-Speed Railways (HSRs). With the increasing demand for intelligent railway systems, higher requirements are imposed on the precise perception and autonomous diagnosis of switch machine working conditions. To overcome the limitations of traditional methods in terms of diagnostic accuracy, computational real-time performance, and anti-interference capability, this paper constructs an intelligent diagnosis model that integrates cross-modal feature attention mechanism and Transformer-XL recursive memory mechanism. The proposed model enhances fault recognition and environmental adaptability under complex operating conditions. By introducing a cross-modal attention mechanism, it enables dynamic interaction between power curve signals and switch rail vibration signals, mitigating the judgment bias caused by missing single-modal information. The Transformer-XL with recursive memory mechanism dynamically adjusts the model’s perception of historical information, allowing it to extract state information across time windows. Additionally, 1D-CNN is incorporated for short-term dynamic feature extraction, optimizing global sequential modeling while improving noise robustness and reducing computational complexity. Experimental results demonstrate that the proposed model exhibits significant advantages in cross-modal feature representation, long-term dependency modeling, noise robustness, and computational efficiency. This study provides an intelligent diagnostic solution for HSR maintenance, featuring high efficiency, low resource consumption, and strong environmental adaptability. It facilitates the transition from reactive maintenance to predictive maintenance, thereby improving operational safety margins and robustness.

To address shortcomings of insufficient existing slow traffic systems and the lack of targeted optimization strategies, this study explores the impact of various indicators on slow traffic systems. First, based on an in-depth analysis of residents’ slow travel behavior, this study designs a questionnaire and selects the Honggutan New District of Nanchang as the case study area to perform the satisfaction survey on slow traffic system. Subsequently, hypotheses are developed and continuously refined using Structural Equation Modeling (SEM). An evaluation system comprising 20 third-level indicators is constructed across four dimensions: safety, convenience, comfort, and empathy. Standardized path coefficients are used to determine the weights of the indicators. The weights are then integrated into the Grey Clustering Model (GCM) for a comprehensive evaluation, leading to the development of a SEM-GCM-based model for assessing residents’ satisfaction with slow traffic travel. Empirical analysis is subsequently conducted to validate the model. Finally, an “Importance-Satisfaction” matrix is constructed based on the analysis of evaluation results. Optimization strategies are proposed for indicators identified as having “high importance but low satisfaction”.Results indicate that the overall satisfaction level of residents with slow traffic travel is Level 4 (Satisfied). The most influential third-level indicators are transfer convenience (0.058 8), bus stop coverage (0.058 4), and convenience of non-motorized vehicle parking (0.056 9), highlighting that convenience is the most critical factor, followed by safety, comfort, and empathy. The research findings can provide a valuable reference for enhancing satisfaction levels with slow traffic systems.

To investigate the variation patterns of the temperature field and thermal stress in seamless CRTS Ⅲ slab tracks on simply supported beam bridges under high-temperature conditions, an indirect heat-stress coupling analysis method is employed. A finite element model of the coupled structural system of the heat-ballastless track is established to analyze the distribution and evolution of the temperature field, as well as the longitudinal force, stress, and displacement distribution between track structure layers under high temperatures. The results indicate that the temperature of each structural layer of the ballastless track exhibits a wave-like variation with ambient temperature. Over the course of a day, the maximum and minimum temperatures, 53.1 ℃ and 26.4 ℃, respectively, occur at the top of the track slab. The amplitude of the temperature time-history curves decreases with increasing vertical depth of the track structure, and the peak temperature shows a time lag. In summer high-temperature conditions, the temperature gradient of the track slab approaches zero around 11:00 and 21:00; the positive gradient peaks at 85.5 ℃/m at 15:00, while the negative gradient peaks at -43.8 ℃/m at 03:00. As the depth of the track structure increases, the temperature gradient gradually decreases. Under single-day high-temperature conditions, the rail longitudinal force and track structure displacement reach their maximum around 18:00 every day, representing the most unfavorable state, while the longitudinal stress in the track slab and self-compacting concrete layer peaks between 14:00 and 16:00. These findings provide a t theoretical reference for monitoring and maintenance of track structures in high temperature regions during summer.

This study addresses the integrated optimization problem of order allocation, processing sequence, and shelf access sequence in a multi-picking-station scenario of an Autonomous Mobile Robot (AMR)-based parts-to-picker picking system. The Order Allocation and Sequencing Problem (OASP) in a multi-picking-station scenario is proposed, which jointly optimizes how orders are assigned to picking stations, the processing sequence of orders at each station, and the shelf access sequence. A mixed-integer programming model is formulated with the objective of minimizing total order picking time. A Variable Neighborhood Search Algorithm (VNSA) is developed, which batches orders based on order similarity to generate a greedy initial solution. The algorithm incorporates four types of local search neighborhoods, including shelf replacement and order reallocation jitter operators, order exchange/insertion, and shelf sequence adjustments, combined with a dynamic switching mechanism to iteratively improve the solution. The performance of VNSA is compared with that of the CPLEX solver. Results demonstrate that VNSA outperforms CPLEX in solution speed and accuracy on small-scale instances, and shows significant improvements in initial solution quality on large-scale instances, verifying the effectiveness of joint optimization of order allocation and sequencing. Moreover, order picking time exhibits a negative correlation with the number and capacity of picking stations, and a positive correlation with the load balancing coefficient.

To address the challenge of accurately predicting container transportation time, this study proposes a hybrid model, CNN-GRU-Attention (CGA), which integrates Convolutional Neural Networks (CNN), Gated Recurrent Units (GRU), and an Attention Mechanism (Attention). Key factors influencing railway container transportation time, such as transport distance and whether the shipment crosses bureau boundaries, are selected as input features. A sliding window approach is employed to segment the data before feeding it into the model. The CNN-GRU framework is used as the main framework to extract data features and capture long-term dependencies, while the attention module enhances the model’s ability to focus on critical information. The model’s performance is evaluated using Mean Square Error (MSE), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Coefficient of Determination (R²), and Mean Absolute Percentage Error (MAPE). Representative machine learning and deep learning models are used as benchmarks for comparison. Results indicate that the CGA model achieves an MSE of 77.84, RMSE of 8.82, MAE of 2.72, R ² of 0.958, and MAPE of 4.47%. Compared with other models, the CGA model has demonstrates superior prediction accuracy for railway container transportation time and delivers better overall forecasting performance.

To address the vulnerability of urban rail transit networks during operational disruptions, this study proposes a systematic evaluation method that integrates network structure, passenger flow dynamics, and passenger behavioral responses. First, a CPT-NL model of travel choice behavior is developed based on Cumulative Prospect Theory (CPT) and the Nested Logit (NL) model. This model captures passengers’ bounded rationality and risk aversion under emergency conditions. Second, a vulnerability analysis framework is designed for urban rail networks within multimodal transportation systems, establishing quantitative vulnerability metrics that reflect the combined impacts of operational disruptions and passenger flow dynamics. Finally, using Shenzhen’s rail transit network as a case study, peak-hour passenger flow data are employed to systematically simulate disruptions at sections and transfer stations, quantifying network performance losses and passenger travel impacts. The results indicate that failures at critical sections or transfer stations reduce network connectivity and trigger large-scale passenger congestion and significant travel delays. During section disruptions, the proportion of affected passengers reaches up to 14.78%, with network vulnerability peaking at 12.02%. During transfer station disruptions, the proportion of affected passengers peaks at 11.73%, and network vulnerability reaches 5.53%. The number of delayed passengers significantly exceeds the number who abandon rail travel due to disruptions, indicating that most passengers opt for detours within the rail network or multimodal travel alternatives. Operational management should strengthen maintenance and emergency preparedness for high-vulnerability nodes, while optimizing multimodal transportation scheduling to improve the resilience and operational stability of urban rail transit systems.

To ensure the long-term safe and stable operation of railways, this study employs a self-designed permeameter to investigate the permeability of ballast aggregates under various service conditions, aiming to reveal the evolution patterns of ballast bed seepage performance. First, clean ballast aggregates meeting the premium crushed stone ballast standard are selected to analyze the effects of gradation non-uniformity coefficient and mean particle size on ballast aggregates permeability. Next, fly ash and fine sand are introduced as fouling agents and added to the upper, middle, and lower layers of clean ballast aggregates to simulate fouling distributions in existing ballast bed. The impacts of fouling degree, composition, and distribution on ballast aggregates permeability are systematically examined. Finally, by varying initial densities and applying vertical loads, the effects of initial density and applied vertical load on ballast aggregates permeability are studied. Results indicate that, in the absence of contamination within the ballast aggregates, a smaller non-uniformity coefficient and larger average particle size correspond to higher permeability and hydraulic conductivity. Conversely, the presence of contamination within the ballast aggregates, with higher contamination levels, smaller contaminant particle sizes, and more concentrated distribution, results in a smaller hydraulic conductivity and weaker permeability. Moreover, clean ballast aggregates with lower initial density result in a greater hydraulic conductivity and gradually enhanced permeability, while contaminated aggregates with lower initial density show a reduced hydraulic conductivity and diminishing permeability. Additionally, the permeability of contaminated ballast aggregates decreases with higher vertical loads due to plastic deformation during loading, preventing permeability from fully recovering after unloading.

The proportion and scale of connected commercial vehicles significantly influence traffic flow speed, which in turn affects overall vehicular emissions across the road network. To address the lack of quantitative evaluation methods for the impact of connected commercial vehicle platooning on overall network emission reductions, this study develops a large-scale traffic simulation model based on SUMO to quantify emission reductions under future platooning scenarios. First, a large-scale highway simulation environment is established in SUMO by calibrating against real-world highway data. Next, within a joint SUMO-Python simulation framework, car-following and fuel consumption models for various vehicle types are configured, and emissions are calculated using the MOVES software. Using formation size and penetration rate as variables, platoons consisting of 2 to 8 commercial vehicles are simulated under free-flow conditions to analyze fuel consumption and emissions at different penetration levels. Finally, the study proposes optimal platoon sizes for the target road network at varying penetration rates. The research results indicate that average instantaneous fuel consumption of the platoon decreases as platoon size increases, reaching minimum values at platoon sizes of 4 and 8 vehicles, but rises with increasing penetration rates. Compared to a baseline scenario, higher platoon penetration rates and larger platoon sizes yield maximum reductions of 22.25% in fuel consumption, 21.09% in CO₂, 16.91% in HC, 19.36% in NOₓ, 15.23% in PM2.5, 17.39% in PM10, and 26.06% in CO emissions. Both emissions and fuel consumption decline as penetration rates rise; under fixed penetration, they further decrease with larger platoon sizes. Overall, increasing both penetration rate and platoon size leads to significant reductions in total fuel consumption and pollutant emissions from road traffic flow.

To address the insufficient adaptability of fixed signal timing in dynamic traffic flow environments, this study proposes an improved Gradient-Based Meta-Learning Deep Q-Network (GBML-DQN) algorithm integrating a meta-learning mechanism for adaptive signal control at isolated intersections. First, the state space is constructed based on lane density, the signal phase action space is defined, and a multi-objective weighted reward function is designed. Second, using DQN as the base architecture, the discount factor γ is dynamically adjusted via a meta-gradient strategy. Third, a Dueling structure is used to decouple state values from action advantages, and NoisyNet is employed to replace the traditional ε-greedy strategy. Consequently, two improved algorithms are constructed: Improved GBML-DQN-Dueling (I-GD-D) and Improved GBML-DQN-Noisy (I-GD-N). Finally, experimental validation is conducted on the SUMO simulation platform across high, medium, and low traffic volume scenarios. The experimental results indicate that I-GD-N exhibits superior robustness and adaptability across different traffic scenarios. Specifically, under medium traffic conditions using the Stochastic Gradient Descent (SGD) optimizer, the average delay is reduced to 20.55 s, representing an improvement of approximately 20% compared to DQN. While DQN exhibits higher stability than GBML-DQN in medium and low traffic scenarios due to its simpler structure, it suffers from significant policy degradation in high traffic scenarios or when using the Root Mean Square Propagation (RMSprop) optimizer, performing worse than even fixed signal timing. The dynamic γ adjustment mechanism adaptively optimizes based on traffic intensity; in high traffic scenarios, it significantly outperforms fixed γ strategies by reducing γ to rapidly respond to congestion. The research results provide a valuable reference for urban intersection signal control.

To address the issue of train delays caused by uncontrollable factors during operation, this study investigates the impact of various adjustment strategies under a variable train formation operation mode on both passengers and subsequent train services. First, a train delay adjustment model is developed based on the variable formation operation framework, incorporating constraints such as train formation, arrival and departure times, rolling stock circulation, headway intervals, and passenger flow calculations. Then, the model optimizes four key decision variables, train formation status, dwell time, section running time, and headway, to achieve two main objectives: minimizing the number of stranded passengers and minimizing the deviation between the actual and scheduled timetables. Finally, using Shanghai Metro Line 16 as a case study, three distinct delay scenarios are simulated and solved using the CPLEX solver. The research results indicate that in different scenarios, for shorter delays, there is no need to implement coupling or decoupling strategies. However, for longer delays, these strategies are necessary to mitigate operational impacts. Within the same scenario, allowing coupling and decoupling operations significantly reduces both the impact on subsequent trains and the total number of stranded passengers compared to scenarios where such adjustments are not permitted.

Addressing the unclear key factors and mechanisms influencing regional freight structure optimization, this study proposes a systematic research approach. Firstly, leveraging the Latent Dirichlet Allocation (LDA) topic model, eight pivotal factors are systematically extracted and identified: socioeconomic development level, industrial structure, natural geographical conditions, energy structure, environmental awareness, techno-economic attributes, scientific and technological progress, and transportation industry policies. Secondly, based on research hypotheses, a Structural Equation Model (SEM) is formulated to analyze the effects and interrelationships of these factors on regional freight structure. Finally, elasticity analysis is introduced to quantify the contribution of each factor to freight structure adjustment. The results indicate that all eight factors significantly promote regional freight structure adjustment, with varying degrees of impact and elasticity across regions. The eastern region should focus on optimizing industrial structure, leveraging geographical advantages, enhancing environmental awareness, and promoting green freight technologies and policy guidance to build an efficient and green freight system. The central region needs to optimize geographical conditions, advance transportation infrastructure, improve freight network connectivity and efficiency, and strengthen investment guidance to enhance techno-economic characteristics and facilitate freight structure optimization. The western region should accelerate the integration of ecological protection and technological development, emphasize techno-economic improvements, promote freight technology innovation, and optimize costs to drive green and innovative freight structure development. At the overall level, efforts should strengthen the dynamic alignment of industrial structure and techno-economic characteristics, improve policy precision and coordination, and promote innovation and upgrading of transportation modes and systems.

To address the complex and diverse characteristics of track fastener defects, as well as the low efficiency and high missed-detection rates of traditional detection methods, this study proposes a lightweight detection model, FPSI-YOLOv8s, based on the YOLOv8s framework. First, to reduce model complexity, FasterNet, featuring higher processing speed and fewer parameters, is adopted to replace the CSPDarkNet53 backbone in YOLOv8s for defect feature extraction. Second, the C2f module in the YOLOv8s neck is redesigned using Position-aware Recurrent Convolution(ParConv) to form a new FasterBlock module, enabling multi-scale feature fusion and further model lightweighting. Third, a Spatial Group-wise Enhance (SGE) attention mechanism is integrated after the SPPF layer to enhance the model’s sensitivity to defect features and mitigate accuracy degradation. Finally, the Inner-IoU loss function replaces CIoU to improve detection performance for objects of varying scales and shapes, while refined quality evaluation and gradient-gain strategies further enhance model robustness. Experimental results show that the improved model reduces model size by 29.78%, and decreases computational cost and parameter count by 29.93% and 30.46%, respectively, with only a 0.7% decrease in detection accuracy. These results demonstrate that the proposed model achieves significant lightweighting and improved operational efficiency while maintaining high accuracy, indicating strong application potential for rapid inspection of track fasteners.