With the increasing number of new energy vehicles globally, the density and spatial distribution of urban public charging infrastructure lag behind demand. Moreover, supply-demand imbalance has become an increasingly prominent issue, posing a key bottleneck in the development of green mobility. To address this challenge, this study considers Shunyi District, Beijing, as a case study to propose a comprehensive evaluation and location optimization method for public charging stations using multisource spatiotemporal data. By combining multisource spatiotemporal data such as vehicle trajectories and points of interest, we constructed a spatiotemporal distribution model of urban charging demand to accurately characterize the dynamic charging loads in different functional zones. Furthermore, through traffic accessibility analysis and charging behavior simulation, the effectiveness of the layout of the existing stations is quantitatively assessed and service blind spots are identified. The results reveal that the service capacity in some high-demand areas of Shunyi District is insufficient, with considerable coverage gaps. To overcome this issue, we used the K-means clustering algorithm to identify the cores of unmet demand and proposed a priority-based construction plan for new stations. This study provides a theoretical basis and a practical approach for mitigating regional supply-demand imbalances and enhancing the scientific layout and systemic adaptability of urban public charging facilities.

To address the challenges of significant scale variations, numerous environmental interferences, high real-time requirements, and the difficulty in achieving a good balance between detection accuracy and computational cost faced by existing bicycle-type vehicle detection models, this study proposes YOLO-DBG, a lightweight and efficient bicycle-type vehicle detection model based on computer vision. First, a novel dual-branch pooling & depthwise separable convolution bottleneck module is designed, which synchronously extracts global contour and local detail features of bicycle-type vehicles through a differentiated feature aggregation strategy, thereby enhancing multiscale feature extraction capabilities and reducing model computational costs by integrating depthwise separable convolution. Second, a weighted bidirectional feature pyramid network architecture is introduced in the neck network, which enhances the fusion of key vehicle features through bidirectional cross-scale connections and a dynamic weighting mechanism, and effectively reduces model computational costs through node pruning. In addition, ghost convolution is used as a downsampling operator, which considerably compresses the model while maintaining the feature expression ability. These three modules work together to construct an effective lightweight network architecture. Experiments demonstrate that the proposed model achieves a 0.2% increase in mean average precision while reducing parameters, giga floating point operations, and model size by 55.8%, 37.0%, and 53.1%, respectively. The proposed method achieves ideal lightweighting without compromising detection accuracy, offering a novel solution for real-time detection of bicycle-type vehicles.

Selecting between fixed-route and flexible-route bus systems is a critical strategy for improving the operational efficiency of urban public transit. While most existing studies focus primarily on passenger demand, this study adopts a more comprehensive perspective by incorporating demand intensity, service area size, and block size. Based on the cost structures of both fixed- and flexible-route systems, mathematical models are developed to determine the optimal vehicle capacities and departure intervals that minimize total system costs. Closed-form solutions for these optimal values are derived for both transit modes. In addition, the study identifies the conditions under which each system is preferable, and derives threshold formulas for demand intensity, service area size, and block size. Research demonstrates that the operating costs of flexible bus systems (particularly pickup/drop-off costs) and passenger walking costs in fixed-route systems are pivotal determinants in transit mode selection. These cost factors exhibit threshold-dependent effects: only when their quantitative relationship meets specific conditions do demand intensity, service area scale, and block size significantly influence mode choice. Numerical simulations validate the proposed mode selection methodology between flexible and fixed-route bus systems. The findings provide actionable decision support for real-world transit planning applications.

The integrated dispatch of metro firefighting engines (MFEs) and trains during fire emergencies presents unique challenges due to confined tunnel spaces that restrict conventional fire truck access and the often uncertain fire location. This necessitates cross-line deployment of MFEs within the metro network. We decompose the problem into two subproblems: MFE rescue operations and evacuation of trapped trains. For each, a mixed-integer programming model is formulated. To enhance computational efficiency, we propose a line-based decomposition algorithm combined with a constraint filtering technique. The models are validated using the Suzhou metro network as a case study. Results demonstrate that the proposed approach efficiently generates solutions within practical timeframes, enabling effective cross-line MFE dispatch and rapid train evacuation. This research offers both theoretical insights and a practical decision-support tool for emergency management in metro fire scenarios, contributing significant value to metro operation safety and resilience.

To address the issue of low load factors associaated with conventional demand-responsive buses with fixed capacity, this study introduces a modular vehicle system and proposes a cross-regional demand-responsive bus dynamic scheduling optimization method based on modular vehicles. An objective function is established to minimize travel and operational costs for passengers and enterprises, respectively. The concept of coupling stations is introduced, and a dynamic fleet formation model considering coupling stations is designed, which allows for fleet reorganization and passenger transfer across two routes. Using the Big-M method, the model is linearized into a mixed-integer linear programming (MILP) problem for solutions, and the model is validated through a case study of two commuting routes from Jiukeshu to Wangfujing in Beijing. The experimental results show that compared to the introduction of conventional demand-responsive buses with fixed capacity, the introduction of modular buses effectively increases vehicle load factors, reduces operational costs for enterprises, and slightly reduces passenger travel costs. This indicates that the introduction of modular bus systems in urban commuting scenarios can provide a more flexible and efficient operational model for passenger travel and enterprise operations.

To explore the factors influencing the injury severity of cyclists in urban bicycle accidents and mitigate the impact of data heterogeneity and imbalance on the quantification of these factors, this study proposes a method integrating resampling, latent class analysis (LCA), and Bayesian networks (BNs) based on 3 895 bicycle accidents from the CRSS database. First, LCA was used to reclassify accident data into several sub-accident clusters with intra-cluster homogeneity and inter-cluster heterogeneity to reduce the impact of data heterogeneity. Second, random over-sampling (ROS), synthetic minority oversampling technique, and adaptive synthetic sampling approach were used to resample each accident cluster to reduce the impact of data imbalance. Finally, based on various resampled accident clusters, two BN structure learning algorithms and one parameter learning algorithm were applied and the optimal BN model for each accident cluster was selected based on AUC values to enable quantitative and heterogeneity analyses of factors influencing the injury severity of cyclists. Results show that when the overall accident data were divided into three homogeneous sub-clusters, the LCA model achieved an increased entropy value of 0.943. For the C1, C2, C3, and OD accident clusters, 10, 13, 9, and 12 key factors influencing the injury severity of cyclists were identified, respectively. The introduction of LCA and resampling into the BN considerably improved the BN model’s G-mean value, AUC value, and risk factor identification capabilities. Factors such as time period, cyclist’s gender, cyclist’s age, and weather conditions showed substantial heterogeneity across different accident clusters.

To address the efficiency loss caused by traffic congestion and frequent vehicle stops at signalized intersections, in this study we propose an intersection passing strategy based on multiple control modes. The interactions between vehicles and traffic signals as well as vehicle-to-vehicle interactions are analyzed, and six control modes are developed based on variations in traffic flow, signal cycles, and vehicle behavior. The strategy is dynamically adjusted to real-time traffic conditions. To minimize the impact of abrupt acceleration changes, the concept of vehicle jerk is introduced, and a straight-line trajectory model is developed accordingly. In addition, considering the collision-avoidance constraints of preceding vehicles, a polynomial method is used to construct an optimized lane-changing trajectory model to enhance the lane-changing efficiency. A comparative analysis of four control strategies demonstrates that the proposed multi-mode control strategy reduces delay times by 18.93% and 25.79% under low- and high-traffic flow conditions, respectively, compared to traditional strategies. Furthermore, by analyzing the displacement, speed, and acceleration curves of vehicles during travel, vehicle passing time is reduced by 2.60 to 23.52 s under different control modes, confirming the effectiveness of the proposed strategy in improving intersection traffic efficiency.

This study analyzed actual passenger travel demand based on city bus and subway card swipe data to provide data support for airport bus station selection and route optimization. First, travel demand centers were identified by analyzing card swipe data, and the K-medoids clustering algorithm was used to cluster city-wide travel demands. The resulting travel demand centers, along with major transportation hubs such as railway stations, were selected as shuttle bus stations to provide passengers with convenient transfer services. Second, optimal routes between stations were generated using network analysis and route optimization algorithms to improve the service coverage and operational efficiency of the airport shuttle. Finally, recommended routes were refined through manual adjustments of the optimal routes. A comparative analysis of route evaluation results revealed that the optimal airport shuttle routes significantly reduced the total walking distance of passengers. This study optimized airport bus station layout and route planning based on actual travel demand, providing scientific support for improving airport transfer services.

To alleviate queue lengths at bottom nodes in two-layer networks, reduce data packet travel time, and improve overall network traffic efficiency, this paper examines information packet transfer mechanisms between network layers. The study investigates how layer judgment frequency and network coupling strength influence transmission capability and stability in dynamic and static coupled networks with limited cache capacity. We compare the operational efficiency of two layer-switching approaches by analyzing packet loss rates, traffic density, and average travel times. Network stability and reliability are measured using loss threshold values. Results indicate that in two-layer dynamic networks with limited buffer space, both the layer-switching methodology and switching probability significantly affect transmission efficiency. By jointly optimizing the layer-switching approach and inter-layer edge distribution, the network's loss threshold can be increased, enhancing stable information transmission capacity and alleviating network congestion.

To address the limited accuracy of single-model forecasting and challenges faced by combined models in handling abnormal data fluctuations，this study proposes a novel forecasting method integrating mutation point correction into a pruned exact linear time （PELT）-grey prediction model（GM）-seasonal autoregressive integrated moving average （SARIMA） combined model. This method initially employs the PELT algorithm to detect fluctuations in freight turnover data and identify change points. The Grey GM（1，1） model is then used to correct anomalies at these change points，enabling the dataset to better meet the stationarity and randomness requirements for the SARIMA model. Finally，based on the optimized dataset，the SARIMA model is used to perform predictions on the refined data. Using freight turnover data from Beijing as a case study，comparative analysis of different hybrid models reveals that the proposed model exhibits superior performance than other combined models，with significant reductions in mean squared error and mean absolute error and a coefficient of determination close to 1. The PELT-GM-SARIMA model is structurally simple and can better adapt to time-series data with missing values or frequent anomalies，resulting in more accurate predictions. This study presents a more effective approach for traffic predictions in highway transportation planning and investment decision making.

Freeway traffic accidents seriously affect road safety and accessibility. Accurately predicting the durations of accidents is key to improving emergency response efficiency，alleviating traffic congestion，and reducing the risk of secondary accidents. This paper proposes an adaptive parameter-optimization model based on a deep belief network （DBN） and genetic algorithm （GA） for predicting traffic accident durations. Traffic accident data from freeways in Shandong province were collected from 2020 to 2022，including 16 variables such as road，temporal attributes，and environmental attributes. The Spearman correlation coefficient and box plots were used to analyze the correlation between each variable and the accident duration，ensuring the validity and significance of the selected variables. Based on this analysis，we developed an adaptive parameter optimization-based prediction model，GADBN，using numerous traffic accident data. This model integrates the global search and optimization capabilities of the GA to notably improve the predictive accuracy of the DBN. To validate the model effectiveness，experimental comparisons were conducted with other algorithms such as support vector regression，radial basis functions，XGBoost，and DBNs，with mean absolute percentage error (δ_MAPE) and root-mean-square error (δ_RMSE) being used as evaluation metrics. The experimental results showed that the GADBN model achieved δ_MAPE and δ_RMSE values of 16.49% and 9.12，respectively，outperforming the other comparison models，thereby demonstrating its effectiveness and practicality.

With the extensive development of intelligent transportation and eco-friendly travel, a low-energy task-offloading method based on edge computing in the internet of vehicles (IoV) is proposed to address the dual challenges of low-latency service demands and energy conservation in the IoV. In the context of multivehicle single-cell scenarios on public roads, this study explores the task-offloading requirements of vehicles in motion and systematically investigates the allocation of computational resources. To fully utilize computing resources, this study not only considers the computing power of vehicles but also introduces a new approach for offloading tasks to vehicle servers traveling in the same direction or parked along the roadside as well as to edge servers in roadside units. This enables the effective integration and efficient sharing of computing resources, thereby remarkably enhancing the processing capabilities of the IoV. Furthermore, this study employs an improved particle swarm optimization algorithm to optimize offloading power and task allocation ratios. Extensive simulation tests revealed that the proposed method significantly reduced the energy consumption of vehicle tasks and improved the service quality and energy efficiency of the IoV.It helps to promote green transportation and sustainable development, and lays a solid foundation for energy optimization and efficiency improvement of future intelligent transportation systems.

Energy-saving metro train control is closely related to the vertical track alignment (VTA) design, and both have a significant impact on operating costs. To further reduce operating costs based on optimized train control, this study proposed a collaborative optimization model for the VTA design phase. This model optimizes the bidirectional train control strategy and VTA of a metro section with the goal of minimizing energy consumption and maintenance costs simultaneously, while adhering to the constraints of scheduled train control and the requirements of the "Metro Design Code." Given the numerous factors affecting the maintenance costs of wheels and rails, a train-track dynamic simulation model was developed to calculate these costs. Based on this, an algorithm combining the pseudospectral method and brute force search was designed to solve the collaborative optimization model. The effectiveness of this optimization method was validated using three sections of the Guangzhou metro line. The results indicate that, compared to the method of optimizing scheduled train control alone on the actual VTA, the collaborative optimization model is more effective in saving operating costs, reducing the average operating costs by 21% across the studied sections. This study can provide novel approaches and theoretical support to further reduce metro operating costs, which contributes to promoting sustainable development of metro.

The prediction of highway tolls is affected by complex factors such as holidays and unexpected events. Traditional prediction methods often fail to fully account for intricate interactions between these multiple factors, resulting in less-than-ideal prediction accuracy. By leveraging the self-attention mechanism, large language models can better fit complex spatiotemporal data and have enhanced feature learning capabilities, making them highly effective for precise highway toll prediction. Therefore, this study proposes a highway toll prediction model based on iTransformer. This model embeds temporal information as an independent dimension into the input sequence and reverses the roles of the self-attention mechanism and feed-forward network, thereby allowing the model to more accurately capture the dynamic features of time series and correlations between multiple variables. Case studies show that the proposed model improves the average prediction accuracy by 23.47% and 17.84% compared with the SARIMA and LSTM models, respectively, in regular scenarios. In irregular scenarios, the model demonstrates even better predictive performance, improving the accuracy by 70.92% and 45.64%, respectively. A sensitivity analysis of the proposed model indicates that it is highly sensitive to the number of feed-forward network layers and stacked encoder layers but is less sensitive to changes in the number of attention heads. Thus, this study provides a new methodological approach for addressing the challenges associated with toll prediction in complex traffic environments and has significant implications in terms of improving the accuracy of highway toll predictions.

Lane detection is a remarkable practical application of computer vision technology in the field of transportation. However, existing semantic segmentation network models still face certain challenges such as insufficient accuracy and blurred edges in road semantic segmentation tasks. To address these issues, an improved lane segmentation network architecture based on the UNet model is proposed. First, a dual attention module (DAM) is introduced in the skip connections of the UNet model, which prioritizes the importance of lane lines and effectively reduces noise interference. Additionally, dynamic snake convolution (DSConv) is employed to replace traditional convolution methods, enhancing the network’s lane detection ability. To enhance the comprehensiveness and accuracy of lane detection in underexposed or dark backgrounds, an improved adaptive Gamma correction method is introduced in the image preprocessing stage. Furthermore, atrous spatial pyramid pooling (ASPP) technology is introduced at the end of the encoder to enhance network performance. Experimental results show that this model achieves an accuracy of 98.93% on the TuSimple dataset while meeting real-time requirements. Compared to five other semantic segmentation-based lane detection algorithms, the proposed algorithm demonstrates superior recognition performance, thus validating its effectiveness.

The joint optimization of train timetable and short-turn routing under the flexible composition mode are restricted by various factors such as train timetables, passenger dynamic equations, and train composition adaptability. The coupling of constraints increases the complexity of the problem, making it difficult to solve using traditional optimization methods.This paper introduces the quantum computing method to address the problem. We built a mixed-integer nonlinear programming model to minimize the number of gathered passengers across all stations along the transit line. Furthermore, we used the real coherent Ising machine(CIM) to solve this problem. The numerical results show that the real coherent Ising machine has obvious advantages in computing efficiency and optimization performance compared with other classical algorithms.

To address the challenges posed by complex traffic scenarios, particularly congested roads where traffic objects are densely packed and often occlude each other and small-scale objects are detected inaccurately, a new object detection model called YOLO-T (You Only Look Once-Transformer) is proposed. First, the CTNet backbone network is introduced, which has a deeper network structure and multiscale feature extraction module compared with CSPDarknet53. Not only can it better learn the multilevel features of dense objects but can also improve the model’s ability to handle complex traffic scenarios. Moreover, it directs the model’s focus toward the feature information of small objects, thereby improving the detection performance for small-scale objects. Second, Vit-Block is incorporated, which integrates more features by parallelly combining convolution and Transformer. This approach balances the relevance of local and contextual information, thereby enhancing detection accuracy. Finally, the Reasonable module is added after the Neck network, introducing attention mechanisms to further improve the robustness of the object detection algorithm against complex scenarios and occluded objects. Experimental results indicate that compared with baseline algorithms, YOLO-T achieves a 1.92% and 12.78% increase in detection accuracy on the KITTI and BDD100K datasets, respectively. This enhancement effectively boosts detection performance in complex traffic scenarios and can assist drivers to better predict the behaviors of other vehicles, thus reducing the occurrence of traffic accidents.

Currently, in surveillance video groups, traditional methods for searching camera videos involve traversing and searching through all cameras or performing repetitive searches in a network topology. These approaches result in low efficiency and poor accuracy in tracking individuals. To address this issue, we propose an efficient method for selecting surveillance camera videos based on the principles of the queuing and vertex-weighted directed graph theories. In this method, we treat cameras as vertices and construct a weighted directed graph. By calculating weights, we can determine the optimal monitoring paths considering the connections and weights between cameras. The key advantage of this method is its efficient selection of surveillance camera videos. Additionally, by combining the optimal movement paths of target passengers in urban rail transit nodes with individual tracking, we use the concept of vertex-weighted directed graphs to enhance the accuracy and efficiency of person recognition. The research results show the great significance of this method in improving the performance of surveillance systems and individual tracking capabilities. By applying the queuing and vertex-weighted directed graph theories for individual tracking, we offer an innovative approach to address practical problems and enhance system performance. This method holds great importance in enhancing surveillance system performance and individual tracking capabilities.

To improve the competitiveness of the railway-passenger transportation market and increase its operational revenue, this study investigates the multiobjective system optimization issue of multimodal railway-passenger transportation fares. A mathematical model was used to describe the equilibrium relationship among the demands of different railway-passenger transportation products. Sensitivity analysis was performed to provide a calculation method for the demand elasticity of multimodal railway-passenger transportation products, and a market demand function for multimodal railway-passenger transportation was formulated. Considering multiple optimization objectives such as market demand, passenger transportation revenue, and profit of railway-passenger transportation enterprises along with passenger transportation costs, we proposed a multiobjective bi-level planning model for describing the system optimization issue of multimodal railway-passenger transportation fares. Finally, we used real passenger transportation data of the railway line between Beijing and Tianjin to validate the proposed model. The results show that the proposed method can effectively balance multiple objectives such as passenger transportation demand, passenger transportation revenue, and profit, providing reference and support for railway-passenger transportation departments to develop scientifically reasonable fare systems in different market competition stages.

To study the invulnerability of urban public transport systems, the complex network theory is used to map and analyze the metro and bus systems. First, an improved double-layer complex network construction method is proposed based on the Space-L model. This method constructs connecting edges based on actual transfer distances and uses the peak-hour passenger capacity of lines as edge weights to develop a metro-bus double-layer network. Second, the characteristics of this network and its sub-network are analyzed using indicators such as degree, intensity, and betweenness. Finally, the random attack and intentional attack models are utilized to analyze the invulnerability of the metro-bus double-layer network and its sub-networks, respectively. The results show that the developed network exhibits a scale-free property and is vulnerabe to intentioanl attacks, exhibiting different sensitivities to various intentioanl attack indicators. Thus, the results of this study provide valuable guidelines to public transport systems for responding to emergencies and improving their robustness.

This study first constructs a topological structure model of the rail transit network based on the Space L network topology method. Herein, upon analyzing network characteristics using UCINET as a basis, the maximum connected subgraph ratio and overall network efficiency are selected as indicators to analyze the resilience of the rail transit network. Then, the article comprehensively analyzes the attribute values of nodes and employs the TOPSIS method to rank the importance of nodes using the coefficient of variation for weighting. Then network resilience is analyzed by destroying individual nodes with high importance, and network recovery is achieved through indicator-based ranking restoration strategies, ultimately yielding the average resilience value of the rail transit network. Furthermore, the resilience and robustness of the rail transit network of Nanjing in 2022 is analyzed based on the established model. Results show that nodes with high degree values often have a greater impact on resilience than other indicators. Prioritizing the repair of nodes with the highest degree values leads to the greatest increase in network efficiency, whereas repairing nodes with the highest closeness to the center has less impact on network efficiency.

The highly dynamic nature of subway airport line passenger flows and their susceptibility to the influence of airport flight schedules present challenges for accurate short-term forecasting of passenger flow. This study integrates airport flight information and historical passenger flow data from airport lines to construct a short-term passenger flow forecasting model based on a stacking ensemble model. The model incorporates random forest (RF), LightGBM (light gradient boosting machine), gradient boosting decision tree (GBDT), and logistic regression algorithms to act as ensemble learners. The proposed model is validated using data from the Beijing Subway Daxing Airport Line and is compared against two baseline models, namely informer and long short-term memory (LSTM) networks. The results indicate that the dual-channel prediction, which considers flight information and historical passenger flows, outperforms the single-channel prediction solely based on historical passenger flows. The results also indicate that the stacking model demonstrates superior performance across all metrics. Particularly, the best prediction performance is achieved at a 96 step (24 h) forecast horizon, with mean absolute error of 7.66 and 4.67 for inbound and outbound passenger flow predictions, respectively. Analysis of the impact of flight information characteristics on the prediction model reveals that departure flight information is of relatively lower importance than that of arrival flights, which is attributed to large differences in advance arrival times for departing passengers.

To make comprehensive use of the passenger transportation resources in a high-speed railway hub, this study explores the division of labor among passenger stations within a high-speed railway hub. Herein, a multiobjective programming model was developed with the objectives of minimizing the total train operation time and coordinating the capacities of the stations in the hub. The augmented ε-constraint algorithm was used to solve the approximate nondominated frontier of the model. Using the Zhengzhou high-speed railway hub as a case study, the differences between the existing plan and the optimized plan and their adaptabilities were qualitatively and quantitatively compared to validate the feasibility and effectiveness of the proposed model and algorithm. The results show that the augmented ε-constraint algorithm can identify high-quality representative nondominated solutions for the proposed model. Moreover, the optimized plan can reduce the train operation time within the hub and enhance the coordination of station capacity utilization. Thus, this study provides reliable references and reasonable suggestions for decision-making regarding capacity expansion, transformation, and optimization of high-speed railway hub areas.

This study investigates vehicle scheduling and path planning problems on field roads after large equipment transportation vehicles enter construction sites. Due to road width limitations and varying task priorities, vehicles have difficulty traveling in opposite directions on the same road. Furthermore, the large equipment transportation vehicles have different priorities depending on their loads and urgency of the transportation. To address these challenges, this study constructs an integer programming model based on spatiotemporal network technology that minimizes the total travel time of all vehicles on the site by considering road restrictions and vehicle priorities. Furthermore, vehicle flow balance and meeting avoidance constraints are incorporated into the model. Moreover, a heuristic algorithm is designed to efficiently solve the model and obtain the spatiotemporal path of each vehicle, thereby providing guidance for vehicle path planning and passing each other. The effectiveness of the proposed model and algorithm is demonstrated through multiple cases based on an actual wind farm road network. The computational results show that the algorithm can quickly solve the vehicle path planning problem at various scales. Additionally, it can guarantee short waiting time to avoid vehicle meeting while eliminate spatiotemporal conflicts. Moreover, the proposed approach showed high transportation efficiency.

Based on the standard requirements of ISO 26262 Road Vehicles-Functional Safety, this study analyzes the multiaxle electro-hydraulic steering system of special vehicles to enhance the system's safety and reliability. In this study, the Simulation X software was used to establish a detailed simulation model for the multiaxle special vehicle, and simulation experiments were conducted via fault injection. The simulation results and data were analyzed to assess the severity, exposure and controllability of the faults, thereby determining the corresponding automotive safety integrity level. Thus, based on fault injection simulation, the automotive functional safety analysis method can serve as a crucial means to assess architectural safety in the early stages of system design.

In this study, we build a model for the rotors and propellers of 60 kg combined wing aircraft based on the strip and momentum theory, and circularly calculated the increment of the upcoming flow as an intermediate variable to precisely determine the propellers’ performance. By comparing the obtained results with the experimental data, we corrected the model and calculated the mechanical performance of the propeller. Result showed that the model could evaluate the thrust and shaft power with a bias of less than 5% and less than 10%, respectively. Using this method, we drew the MAP curves representing the mechanical performance as the essential parameters in the power model and built a bridge between mechanical performance and controlling model. The results can support the study of mechanical modelling of combined wing aircraft.

The issue of the pricing of airport advertising revenue, a crucial component of non-aeronautical income in airports, holds significant importance in the operational management of airports. Currently, most airports in China commonly adopt a pricing mechanism based on historical price inertia, while also making adjustments to advertising prices by appropriately considering the total passenger flow for the current year. This pricing mechanism struggles to effectively reflect the true value of advertisements in different locations. This paper proposes a pricing mechanism based on passenger traffic to assess the relative value of advertising spaces within airport terminals. Utilizing a mathematical model combined with the physical layout, and flight and passenger data of the airport, we calculate the distribution of passenger traffic and subsequently evaluate the value of advertising spaces based on this information. Additionally, we apply this approach using sample data from the Capital International Airport. The findings demonstrate that the application of this model can reveal variations in the value of airport advertising spaces with the same media format across different spatial and temporal contexts. This lays the theoretical groundwork for airport advertising management entities to further implement differential dynamic pricing strategies and flexible advertising placement policies.

The safety potential field is utilized to characterize the distribution of safety risks around a vehicle during the driving process. However, when analyzing the safety potential field formed by moving vehicles, the existing models only focus on the vehicle motion but ignore the traffic environment information perceived by drivers. This study focuses on the construction of an improved safety potential field model and its application to the car-following model. Herein, the relative state influence factor is introduced to strengthen the influence of relative speed among vehicles, and the traffic state influence factor is introduced to reflect its influence on driving safety. Moreover, the vehicle type coefficient is introduced to adjust the distance to reflect its influence on driving safety in mixed vehicle type traffic. The car-following model is developed by using the preceptive safety potential field to establish the relationship between the motion state of the front vehicle and the behavior of the following vehicle. Furthermore, the genetic algorithm is employed to calibrate the proposed model, the intelligent driver model, and the car-following model based on the safety potential field. The results show that the root mean square errors of these three models mentioned before are 6.124, 8.515 and 7.248 respectively, which proves that the model proposed in this paper can describe car-following behavior more accurately. Therefore, this study can provide theoretical support for driving risk evaluation and vehicle control under a complex environment.

To enhance the navigation efficiency of ships in inland waterway navigation facilities and increase their operational capacity, a continuous gear shifting model and algorithm for ship lock chambers are proposed. This model comprises two scenarios: considering and not considering the sequence of ships entering the lock. First, a two-dimensional packing problem model was employed to establish a continuous gear shifting model for ship lock chambers. Then, an algorithm for solving the aforementioned continuous gear shifting model based on a greedy strategy was proposed. Finally, simulated ship data for vessels arriving at the lock was generated based on the Baise Junction Project. The proposed algorithm was used to calculate the lock chamber gear arrangement. Results indicate that, in the case of randomly generated data for 90 ships, 47 lock cycles were required for the gear arrangement considering the ship arrival sequence, with an average occupancy rate of 76.424%. However, only 45 locks were needed for the gear arrangement when the ship arrival sequence was not considered, with an average occupancy rate of 76.821%. The proposed model and algorithm can effectively shift gears continuously in the ship lock chamber under various conditions.

With the development of science and technology, the systematization, networking and intelligentization of the social technology system gradually deepen, forming the complexity of the system. The failures of these complex systems, such as traffic jams, rumor spreading, and financial collapse, can be regarded as a kind of "1+1<2" negative emergence of system capability, which is difficult to understand directly through the reduction analysis of system components. It challenges the classical reliability theory. Research on the complex systems reliability mainly focuses on failures laws, which includes two perspectives. One is the study of system vulnerability considering failure propagation. The other is the study of system adaptability considering failure recovery. System vulnerability studies focus on exploring the internal mechanism of system collapse, namely the failure propagation mechanism. System adaptability studies focus on the capacity to adapt and recover, including the system failure recovery mechanism. Based on this, the article introduces relevant research on reliability method.

With the development of network science and the emergence of complex systems theory, scholars have embarked on in-depth research on the structural and dynamic properties of complex networks. Among the dynamic properties of complex networks, cascading failures, as one of the most important research areas, describe a situation where a fault or error in a system or process leads to the failures of other related components or links. Various models and recovery strategies have been proposed for cascading failures in complex networks. This study analyzes the mechanisms of cascading failures, provides a comprehensive summary on the development of domestic and international cascading failure in complex networks, outlines the recovery strategies for addressing cascading failures, and highlights the existing issues and shortcomings in current research, providing valuable insights for future studies.

A functional relationship was constructed between the probability of inhaling viruses and social distance to characterize the viral transmission of subway passengers at the microscopic level. Formulas for calculating the increase and decrease of viral load were constructed based on establishing the viral load evolution equation. Normalized parameters were used within this equation to describe the effect of pandemic prevention measures. The viral load of each passenger was programmed through the Anylogic software’s secondary development interface to characterize the viral load change at the pre- and post-infection phases. In the initial simulation settings, 10% of the passengers were infected with the virus, including ordinary carriers and supercarriers. The evolution of the virus under different passenger number conditions within subway carriages was simulated, which was categorized into with-control and without-control scenarios. The simulation results showed the following: as the number of passengers increases, the passenger density increases, the virus transmission increases, and the individual viral load increases rapidly. Isolating passengers with a viral load greater than a threshold of 1 000 and prohibiting them from taking the subway can reduce the viral load of all passengers by an order of magnitude.

Frequent delays of flights at large international airports can affect their smooth operation, hence, the airport apron allocation problem needs to be robustly optimized. In this study, we proposed two integer linear-programing models for solving this problem and used two algorithms for performance comparison: the hill-climbing and large-neighborhood search (LNS) metaheuristic algorithms. In addition, we used the Monte Carlo method to evaluate the effectiveness of different objective functions in dealing with flight conflicts. The final test results show that the LNS algorithm not only improves the robustness of the gate allocation scheme for large airports but also excels in speed and quality, especially, when the square of idle time is used as the objective function.

An accurate prediction of short-term subway passenger flowscan effectively alleviate traffic congestion and improve the quality of travel services for urban residents. Herein, we propose a multitask learning-based model for the prediction of short-term subway passenger flows, which uses a residual convolutional neural network (NN) and a nested long short-term memory NN to extract the spatio-temporal correlation of traffic patterns, and introduces an attention mechanism to enhance the feature extraction performance of the NNs. Considering the characteristics of subway operations, the model selects train operation features, bus stops around subway stations, and point of interest data as external features to improve the accuracy of the prediction. Based on the historical data of the Beijing Subway, experiments were conducted in multiple time granularity scenarios, such as 10, 30, and 60 min. The results showed that the methodsuccessfully modeled and analyzed the inflow-outflow interaction through multitask learning, improved the prediction performance and generalization ability of the model, and providednovel approaches for the prediction of short-term subway passenger flows.

To address the challenges related to distance measurement of an approaching vehicle in fog,we developed an experimental platform to rapid image processing and real-time distance measurement.Firstly,we down-sampled the images through the dark channel algorithm to estimate atmospheric light values. Then, we introduced a tolerance mechanism to deal with the bright regions that do not satisfy the dark channel prior. This tolerance mechanism corrected the estimate with incorrect refractive index of such regions and effectively mitigated the issues of color distortion and low contrast. Secondly, we detected the vertical edges of an approaching vehicle using the edge detection and the improved Hough transform algorithms. Finally, we measured the safe distance from the approaching vehicle using the model. The results shows that the platform developed in this study can effectively measure the distance of the approaching vehiclein fog with a visibility <100 m, and can alert drivers in a timely and effective manner.

To study the traffic flow characteristics, the traffic data is analyzed using a complex network method. A box plot-clustering algorithm model is proposed to identify and fill in missing values and outliers in the initial data. The one-dimensional data is reconstructed into network nodes using the phase space reconstruction method. Additionally, the connection threshold is selected to determine the connection relationship of network nodes to convert the traffic sequence data as a complex network and analyze the structure and quantitative indicators of the network. The result shows that the structure of the complex network of traffic data can reflect the traffic flow state of the road section to a certain extent. The research optimizes the data preprocessing method and extends the application of complex networks into traffic data research.

To meet the needs of passengers for connection and evacuation at high-speed rail stations and enhance the role of high-speed rail stations as urban comprehensive transportation hubs, a dynamic route planning model of a high-speed rail feeder bus is established based on mixed demand that includes reservation and real-time demands. Based on the reservation demand, considering the operation cost of a bus company as well as the travel time cost, the route planning model is established before the start of operation. An improved genetic algorithm was designed using niche methods to solve the problem. After the start of operation, real-time demand can be inserted into the established vehicle route with temporary stations. To realize a dynamic route planning scheme, an integer planning model is established to minimize the variable cost of the system. Using the proposed method,30 demand groups were randomly generated and solved in the Beitaipingzhuang street area, Beijing. Results show that the model can generate an optimal dynamic route planning scheme for a high-speed rail feeder bus in two periods to satisfy the mixed demand. Compared with traditional genetic algorithm, niche genetic algorithm can effectively avoid premature and obtain better results, thus confirming the feasibility of the proposed model and the niche genetic algorithm.

In this paper, an approximate model and algorithm for the throughput rate are established by studying a docked bike-sharing system (DBSS) using stochastic user demands, routing matrix, and cycling times. A DBSS with a fixed number of bikes can be considered a closed queuing network with a buffered M/M/1 queue at each station, thus establishing an approximate model and algorithm for the throughput rate of DBSS. This algorithm can calculate the average number of bikes on roads and at stations. Moreover, it can estimate the average cycling time on roads and bike dwell time at stations and further determine the optimal number of bikes achieving the maximum throughput rate in the DBSS. Additionally, this paper proposes a method to determine whether a station is a bike surplus station or a bike deficient station under given user demands, routing matrix, cycling time matrix, and dock allocation. Finally, the approximate algorithm is verified in a real-world DBSS. The results show that the throughput rate of the DBSS increases in a step-wise manner with the increasing bike input under an superior limit. When the number of bike inputs exceeds the optimal quantity, there will be idle bikes, and the spatial distribution of bike surplus stations and bike deficient stations will remain unchanged.

Unmanned aerial vehicles (UAVs) have considerable application potential in urban logistics delivery. However, there are many uncertainties in urban low-altitude airspace operation scenarios. Therefore, it is essential to build a safe and orderly logistics UAV delivery network using scientific methods. From the perspectives of delivery economy, operational safety, and features of logistics UAVs, an integer programming model of multilevel hub-and-spoke network was constructed based on the original ground logistics delivery network. A network construction method was proposed, which combines partitioning around medoids(PAM) clustering with distance restrictions and integer programming. Three evaluation indicators were selected, i.e., delivery timeliness, network security, and network structure characteristics, to compare the constructed logistics UAV delivery network with the original ground delivery network. A logistics UAV delivery network was constructed in Jiangning District of Nanjing city to verify the feasibility of the proposed network construction method. The experimental results show that the UAV delivery network constructed using this method has good delivery timeliness while taking delivery costs and safety into account.

For larger enclosed communities, it is necessary to open the existing entrances or add some entrances to allow external vehicles or pedestrians to pass through for smooth urban traffic microcirculation and alleviating traffic congestion and the mutual interference between pedestrians and motor vehicles. Considering the actual situation of a community and the traffic distribution, with the goal of minimizing the total travel time and the cost of construction to open the community as the upper level model, the existing and alternative entrances are open to external vehicles or pedestrians as decision variables, and the combined (walking and car travel) mode choice and route choice with user equilibrium model as the lower level model, a bi-level programming model of decision-making optimization for opening closed communities was established. The genetic algorithm is applied for the upper level model and Frank-Wolfe algorithm is applied for the lower level model, and a solution algorithm of the bi-level programming model was proposed. Finally, the model and algorithm were verified by a sample, discovering the setting of traffic micro circulation and optimizing the plan, the total time spent has been reduced by about 26%. This proves that the model and algorithm proposed in this article have practical engineering application value, and can effectively reduce traffic congestion and improve traffic efficiency.