In recent years, with the expansion of rail transit networks and increasing complexity, along with the growing passenger demand, the operational load of the system has risen, while operational resilience has decreased. The negative impacts of emergencies, such as train delays and operational disruptions, have become more severe. There is an urgent need to study timetable rescheduling optimization methods to assist in decision-making, improve the quality of rescheduling schemes, and reduce the negative impacts of emergencies. This paper applies the bibliometric method to analyze the research hotspots related to timetable rescheduling. It categorizes studies on single-line rescheduling under disturbances or disruptions, focusing on scenarios involving minor disturbances and severe disruptions. Related research is further categorized based on train operation or passenger behavior, track equipment or train failures, static or dynamic input parameters, and micro or macro modelling perspectives. Regarding rescheduling problems under multi-line conditions, the paper first reviews studies that reschdule the timetable of only the affected lines. It then discusses multi-line collaborative rescheduling for failure and non-failure lines, based on both split-line operation and cross-line operation modes. Future research directions are suggested in five areas: enhancing the typicality of emergency scenarios, integrating more deeply with actual operational needs, improving the robustness of rescheduling solutions, flexible combining and applying multiple strategies, and enhancing the model-solving efficiency. The results indicate that the research on timetable rescheduling began with train delay propagation theory and has evolved from single-line rescheduling to multi-line collaborative rescheduling. Among them, the optimization methods for single-line rescheduling are widely applied, However, most existing multi-line collaborative rescheduling methods are focused on intercity rail systems, and are challenging to implement in urban rail transit systems due to their unique characteristics.

Existing research on regional railway network characteristics often overlooks the degree of connectivity between stations and tends to adopt a singular analytical perspective. To address these limitations, this study examines the railway passenger transport network of the Chengdu-Chongqing urban agglomeration from the perspective of community division, exploring the complex structural and functional characteristics of the railway network. First, considering both railway infrastructure and train operations, the Space L and Space P methods are employed to construct models of the railway passenger physical network and service network, respectively. Next, the overall network characteristics are analyzed through topological statistical indicators such as degree distribution and average path length. Finally, community structure theory is introduced to investigate the internal network characteristics. A community division method based on an improved particle swarm optimization algorithm is proposed to further analyze the community composition, geographical distribution, and internal connectivity of the Chengdu-Chongqing railway passenger transport network. The results indicate that the railway passenger physical network of the Chengdu-Chongqing urban agglomeration exhibits scale-free characteristics, while the service network demonstrates small-world characteristics. The physical network is divided into 12 communities, whose spatial distribution exhibits obvious geographical patterns closely aligned with the main railway lines. The service network is divided into 8 communities, with a spatial distribution that transcends geographical constraint, where stations within the same community display long-distance interactions. Additionally, each community network retains small-world characteristics. The study’s conclusions provide valuable insights for promoting the integrated development of the Chengdu-Chongqing railway passenger transport network and enhancing its overall reliability.

To fully utilize the interconnectivity advantages of the inner loop and connecting lines in high-speed railway hubs, aimed at serving both transfer passengers and urban travelers, this study explores an optimization method for adding extra train paths in mass transit operations with a high-speed railway hub loop. First, the organizational model for Electric Multiple Unit (EMU) trains on the hub loop is analyzed in terms of train scheduling, EMU sourcing, and routing schemes. Second, with the objective of increasing high-speed train operation frequency on the loop line, a comprehensive optimization model is developed for scheduling mass transit-type train paths, factoring in station routes, track utilization, and EMU routing schemes. Finally, a high-speed railway hub is selected as a case study. The research results demonstrate that, without altering original train assignments and using the same number of EMUs as the initial plan, the service frequency between hub loop stations can increase by 120%, and the minimum average departure interval between stations can be reduced to 14 minutes. The established model allows flexible application of various hub loop train scheduling plans, maximizing transportation capacity and advancing mass transit operations. These findings provide an organizational strategy for implementing mass transit-type operations of EMUs in high-speed railway hubs.

To address the issue of high cargo handling costs arising from improper container space allocation in the passenger-like transportation organization of freight, this study introduces the concept of “cargo loading and unloading offset”. By comprehensively considering container space selection strategies and station operational capacities, an optimization model for container passenger-like train operation plans and space allocation is developed, aiming to minimize the weighted train operation time and total station economic costs. The model is solved using the Gurobi optimizer, exploring space allocation decisions at a micro level. Its feasibility is validated through small-scale case studies, followed by an analysis of a specific railway line. The results indicate that the train operation plans and container space allocation strategies obtained through model solving can comprehensively consider the operational capabilities of various stations, minimizing system-wide handling costs. Compared with the centralized container space utilization strategy, when the station’s arrival and departure line capacity is constrained, the decentralized freight space utilization strategy can reduce the number of train stops at a smaller cost. This strategy effectively alleviate the pressure on the station’s arrival and departure lines, thereby enhancing the competitiveness of railway transportation.

The video surveillance system is a crucial euqipment in ensuring the safety of railway passenger stations. Existing studies on optimizing video surveillance layouts at railway passenger stations often overlooks the specific monitoring requirements of these stations, leading to potential inefficiencies in the collected image data. To address this issue, this study first analyzes the video surveillance coverage requirements of different areas within the stations, focusing on coverage clarity and orientation requirements. Subsequently, the impact of these coverage needs on the effective range of surveillance cameras is examined. Based on this, a 0-1 integer programming model is developed to optimize video surveillance layout, aiming to minimize the total deployment cost while ensuring effective coverage of target areas. This model is solved using GUROBI software and validated through a case study at Suzhou Railway Station. The results of the study indicate that, after a thorough analysis of the surveillance requirements in railway passenger stations, the proposed layout achieves a 129.3% improvement in overall effective coverage compared to layouts that do not consider these requirements. Specifically, areas requiring target discovery and identification see coverage improvements of 86.5% and 148.1%, respectively, with full coverage achieved in areas necessitating target confirmation. These results provide valuable insights for planning and formulating video surveillance layouts in railway passenger stations.

In addressing the impact of various uncertain factors, including transport time and freight volume, on multimodal transport path selection, a fuzzy robust regret model is proposed. This model integrates fuzzy opportunity constraint programming and robust optimization to comprehensively manage uncertain variables. First, triangular fuzzy numbers are used to represent the uncertainty in transport time parameters, while the robust optimization scenario method is applied to handle fluctuations in freight volume. Hybrid time window constraints are introduced, and fuzzy time parameters are clarified using fuzzy opportunity constraint programming. The model optimizes total cost, total carbon emissions, and total time as its objectives. Subsequently, to improve the convergence quality of the algorithm and maintain population diversity, an improved adaptive Non-dominated Sorting Genetic Algorithm Ⅱ (NSGA-Ⅱ) is designed with local optimization strategies. The algorithm’s performance is compared across four practical multimodal transport scenarios with different node sizes. Finally, a complex virtual example is used to validate the model and analyze the robustness of the solution in response to uncertainty. The results demonstrate that, in comparison with the conventional NSGA-Ⅱ, the improved daptive NSGA-Ⅱ exhibits superior performance in the two objectives of total cost and total carbon emission as node size increases. When the node network reaches 30, the adaptability of the two objectives experiences a decline by 25.22% and 26.39%, respectively. The robust solution obtained from the model can effectively adapt to an uncertain transportation environment and changes in the decision maker’s preferences. Multimodal transportation decision makers need to comprehensively consider the impact of uncertain factors, select appropriate regret values and confidence levels for fuzzy parameters, and achieve transportation solutions that align with their preferences.

To address the optimization of green multimodal transport routes for container transportation in the Yangtze River basin under uncertain conditions, this study incorporates trapezoidal fuzzy numbers and fuzzy chance-constrained programming theory to clarify uncertainty based on fuzzy demand and fuzzy time windows. A multimodal transport route optimization model is developed, factoring in turnover cost, time cost, and carbon emission cost. A genetic algorithm with an elite retention strategy is employed to optimize the multimodal transport routes for 13 cities within the Yangtze River basin. The results show that due to the economic shortcomings of road transportation, the constraints of arrival time windows, and the timeliness limitations of waterway transportation, railway transportation emerges as the dominant mode in the region’s multimodal transport system. Railway-waterway intermodal transport is identified as the primary form of intermodal transport in the Yangtze River basin. Fluctuations in carbon trading prices influence the proportion of railway and waterway transportation in multimodal transport schemes, with carbon trading policies effectively promoting the green transition of transportation modes. A positive correlation is observed between the fuzzy demand preference value and total cost; when the fuzzy demand preference value ranges between 0.5 and 0.7, the growth rate of total cost peaks. The fuzzy demand preference value of 0.5 is considered ideal for balancing total cost and customer demand. These findings provide valuable insights for enhancing the efficiency of multimodal container transportation in the Yangtze River basin and advancing the transportation industry’s green, low-carbon, and sustainable development.

To address the requirements of the “Road to Rail” policy and account for the choice behavior of shippers, a bi-level decision model for railway freight subsidies is constructed. The upper-level model incorporates both economic and environmental benefits and establishes a multi-objective nonlinear subsidy decision framework, with the single OD pair railway subsidy amount as the decision variable. The lower-level model considers the game behavior among shippers and develops a freight flow distribution model aimed at minimizing generalized transportation costs. Considering the characteristics of the model, the affiliation function is used to transform the model into a single-objective problem, and a simulated annealing algorithm alongside an iterative weighting method is designed for its solution. The model and algorithm are validated through an empirical analysis of the gravel aggregate freight network in Beijing. The results show that the OD pair subsidy amounts range between 0 and 0.12 yuan/(t·km), exhibiting significant spatial and individual differences. Under the subsidy scheme, 24% of freight shift from road to rail and 17% carbon emission reduction can be achieved with an economic expenditure of 2.6 million yuan. Furthermore, under the differentiated subsidy scheme, the economic expenditure required to achieve the same scale of freight shift and carbon emission reduction is only 52% of that under a uniform subsidy strategy. The weighting coefficients for economic and environmental benefits have a significant impact on subsidy decisions, suggesting that local governments should determine these weighting coefficients based on local characteristics. Increasing the upper limit of the total amount of railway freight subsidies can enhance the effectiveness of modal shift promotion and environmental benefits, although the marginal growth rate of environmental benefits continues to decline over time. A higher railway freight turnover ratio substantially increases financial subsidy pressures, and the adoption of complementary incentives beyond subsidies is recommended. Additionally, there is a clear negative correlation between total railway freight subsidies and highway freight rates, highlighting the need for the government to establish a flexible subsidy adjustment mechanism to accommodate freight market dynamics.

This study proposes a Metro Freight Network Layout Mathematical Optimization Model (MFNLMOM) to address the issue of urban freight transportation by utilizing the idle capacity of metro and suburban railway lines. It integrates metro, suburban railways, and road networks. First, a multi-objective mathematical optimization model for metro freight network layout is developed, comprehensively considering factors such as transportation costs, node reconstruction and expansion costs, investment expenses, transfer frequencies, and carbon emission costs. The model aims to minimize overall freight costs and transportation time. Next, passenger flow, freight flow, and metro operation logic modules are designed and simulated using Anylogic software to obtain transfer arc segment data within the freight network. Finally, the Gurobi mathematical optimization solver is employed to allocate freight network traffic flow. This is followed by a sensitivity analysis of carbon emission costs, transfer costs, and trans-portation costs. The research findings demonstrate that, compared to traditional road transportation schemes, the metro freight network demonstrates significant advantages, reducing total costs by 55.08% and transportation time by 44.24%. Additionally, compared to the full-station transfer model, the MFNLMOM partial-station transfer model achieves higher computational efficiency.

To enhance traffic operation efficiency and improve driver and passenger comfort in highway merging areas while ensuring safety, this study proposes an optimization method for highway merging order and trajectory planning for Connected and Autonomous Vehicles (CAVs) in a heterogeneous traffic flow environment where Human Driven Vehicles (HDVs) and CAVs coexist. First, vehicle travel time and delay are used as performance indicators to characterize traffic operation efficiency in the merging area, and a merging order optimization function is established. The Monte Carlo Tree Search (MCTS) algorithm is used and adjusted to determine the optimal merging order. Then, based on the optimized merging order, a CAV merging trajectory planning function, referred to as Minimize Acceleration and Jerk Trajectory Planning(MAJTP), is established. By applying optimal control theory, the analytical solution for the longitudinal optimal trajectory of CAVs is derived, forming a cooperative control strategy for highway merging. Finally, traffic simulations are conducted using the SUMO software and PYTHON libraries to validate the proposed method. Simulation results demonstrate that, at CAV penetration rates of 0.2, 0.4, 0.6 and 0.8, the MCTS-based merging order optimization method reduces cumulative delay by 5.75%, 8.84%, 12.24%, and 11.06%, respectively, compared to the First In First Out (FIFO) algorithm. Additionally, compared to the Minimize Acceleration Trajectory Planning (MATP) method, the MAJTP approach results in an average jerk value closer to zero, thereby enhancing ride comfort and verifying the effectiveness of the method. These findings provide theoretical support for traffic management and control strategies in highway merging areas.

Addressing the identification and optimization of critical urban bus lines, this study examines Xi’an’s bus system using a higher-order network model to identify and optimize its key bus lines. First, acknowledging the path dependency inherent to urban bus systems, a higher-order bus network is constructed based on the higher-order network model. Next, an improved weighted k-core decomposition method is proposed, incorporating four location attribute indicators: the road hierarchy of bus stops, rail transit connections at bus stops, Points of Interest (POIs) within the service areas of bus stops, and population density in the areas surrounding bus stops. This approach categorizes the higher-order bus network into core, bridge and peripheral layers. Finally, an empirical analysis is conducted using Xi’an as an example. Important bus lines are identified based on the average number of buses operating in each layer, and the most critical sections of the bus lines are identified according to their frequency of occurrence in significant network edges. Optimization suggestions are put forward for existing problems. The research results indicate that there are 234 critical bus line sections and 55 bus lines traversing the 6 most important sections of the bus lines in Xi’an. Additionally, challenges exist in the connectivity between the city’s new and suburban areas and the central urban areas. Specifically, 524 bus stops in the bridge layer lack direct routes to any stops in the core layer. Through the optimization of 13 non-direct bus lines, the direct connection rate of bus stops has increased by 4.72%, adding direct route options for 247 stops on 13 lines to core-layer stops, thereby enhancing the travel convenience for urban residents.

This study investigates the spatiotemporal heterogeneity in the impact of the built environment on taxi travel demand using taxi order data from Nanjing. By integrating multiple data sources, the built environment is characterized across five dimensions: density, diversity, design, destination accessibility, and public transport proximity. Considering both weekdays and weekends, a Multi-scale Geographically Weighted Regression (MGWR) model is developed to account for the scale differences among feature variables and to explore the interactive relationship between the built environment and taxi travel demand. The research data is visualized using ArcGIS to provide an intuitive representation of model results and spatial distribution characteristics. The findings show that, compared to the Ordinary Least Squares (OLS) and the Geographically Weighted Regression (GWR) models, the MGWR model exhibits superior performance, with the adjusted R${}_{\mathrm{a}\mathrm{d}\mathrm{j}}^{\mathrm{2}}$ improving by 52.42% and 4.62% on weekdays, and by 51.98% and 0.97% on weekends. Additionally, the AICc values decrease by 19.66% and 10.93% on weekdays, and by 21.37% and 8.35% on weekends, confirming the MGWR model’s better fitting performance. The regression coefficients reveal spatiotemporal heterogeneity in the impact of different built environment characteristics. Based on the average regression coefficients of significant spatial unit variables, population density, road density, slow road density, metro accessibility, and bus stop density positively influence taxi travel demand on both weekdays and weekends, while bus accessibility exerts a negative impact. On weekends, land use mix and distance to the city center negatively affect taxi travel demand, with a greater proportion of significant spatial units observed. The research findings provide valuable insights for optimizing urban functional structures and improving taxi operation policies.

To address the issues of existing methods for measuring car dependence, including their single-dimensional focus and lack of clarity regarding influencing mechanisms, this study conducts a comprehensive analysis of car dependence and its influencing factors, using travel survey data collected from car users in Kunming. Incorporating three dimensions, which are objective dependence, subjective dependence, and car usage conditions, the study employs a second-order structural equation model to comprehensively measure car dependence and analyze group-level differences. Subsequently, the Gradient Boosting Decision Tree (GBDT) model and the Shapley Additive exPlanations (SHAP) model are applied to explore the influence mechanisms of socio-economic attributes and built environment on car dependence. The results indicate that the loading coefficients for objective dependence, subjective dependence, and car usage conditions are 0.725, 0.242, and 0.852, respectively. Car dependence is constrained by car usage conditions, with objective dependence proving stronger than subjective factors, and varying significantly across demographic groups. The objective built environment accounts for 62.48% of the variance in car dependence, higher than 29.87% for socio-economic attributes and just 7.46% for the perceived built environment with a relatively low influence. The influence of the built environment on car dependence also exhibits nonlinear relationships and threshold effects. These findings present a novel framework for measuring car dependence and provide valuable insights for developing targeted strategies to reduce car dependence.

To address the issue of insufficient utilization of space-time resources in conventional tandem intersections, which leads to a loss of traffic efficiency at intersections, this study proposes a novel integrated model for tandem intersections that accounts for pedestrian crossings. First, the design form, control strategy, and applicable conditions of the new tandem intersection are discussed. Next, based on the operational patterns of traffic flow and pedestrian crossings, separate delay calculation models for vehicles and pedestrians are proposed. Then, considering pedestrian crossing safety and the operational efficiency of both vehicles and pedestrians, an optimal time-space resource allocation model is constructed based on the control strategy. Finally, the case is validated using VISSIM simulation software. The results show that the optimized design of the new tandem intersection can transform an oversaturated intersection to an unsaturated state, reducing flow saturation in all directions to below 0.8. Compared with the current situation, the capacity of key traffic directions is increased by over 40%, the average delay at the intersection is reduced by approximately over 50%, the average number of stops is decreased by approximately 30%, and the optimization effect outperforms that of the traditional Webster timing method. Compared with traditional tandem intersections, the proposed model introduces a pedestrian crossing phase at the pre-signal and controls right-turn traffic, further optimizing the utilization of space-time resources within the intersection. This enhances intersection capacity while providing new options for pedestrian crossings, offering a fresh perspective for the study of tandem intersections.

To promote the coordinated development of metropolitan multi-airport systems and enhance passenger travel experiences, this study investigates the problem of flight coordinative planning based on passenger travel demand. First, considering factors including the geographical distribution of passengers, their preferred travel time, and airport capacity at different periods, an optimization model is developed to determine flight routes and frequencies across the airport system. The objective is to minimize passengers’ ground transportation costs, the deviation from their preferred departure time, and airfare differences. Second, a mathematical programming model is developed, and a variable neighborhood search algorithm incorporating nine problem-specific neighborhood structures is designed. Finally, the proposed method is validated using flight planning cases involving Beijing’s dual-airport system (PEK and PKX) and routes connecting Chongqing Jiangbei, Yuncheng Yanhu, and Foshan Shadi airports. Results indicate that when average ticket prices are identical at both Beijing airports, PKX is allocated significantly fewer flights than PEK. However, when PKX’s average ticket price is reduced by 6% compared to PEK, the arrival and departure flight volumes between the two airports become balanced. Under the fare consistency strategy, 71% of departing passengers and 62% of arriving passengers can choose their preferred airport, while 58.77% of passengers experience travel time deviations within two hours. Although the 6% fare reduction strategy slightly lowers the satisfaction rate of passenger preferences, it attracts more travelers to non-preferred time slots and airports, further balancing passenger flow and flight density. This approach aligns with the dual-hub positioning of Beijing’s two major international airports and contributes to the efficient operation of the overall aviation network.

The use of electric Vertical Take-off and Landing (eVTOL) aircraft for low-altitude urban air transportation presents an effective solution to alleviate ground traffic congestion. This study addresses the path planning problem in future eVTOL ridesharing operations. First, a static eVTOL path planning mathematical model is developed, aiming to maximize the benefits for both eVTOL operators and passengers. The model considers constraints such as eVTOL vertiport capacity, aircraft load, and battery limitations. Ridesharing fairness is also taken into account, including passenger charging fairness and the equitable distribution of shared trip revenue between passengers and operators. Next, based on the problem’s characteristics, an Improved Adaptive Genetic Algorithm (IAGA) is proposed, incorporating customized crossover and mutation operations with adaptive probability adjustments. Finally, numerical experiments are conducted to solve the model, followed by a sensitivity analysis. The results show that, compared to non-ridesharing scenarios, ridesharing reduces total eVTOL flight distance by 16.52%, increases operator revenue by 9%, and decreases passenger expenditure by 9.01%. These findings confirm that the proposed ridesharing model ensures a fair distribution of benefits between passengers and operators. Both the enumeration algorithm and the IAGA achieve optimal solutions for small-scale cases. Compared to the Improved Genetic Algorithm (IGA), the IAGA reduces computation time by 51.47%. This proves that the IAGA can effectively solve ridesharing problems of various scales. The research results provide valuable insights for future eVTOL ridesharing pricing strategies.

Under the “Dual-Carbon” policy framework, there is a notable lack of multidimensional performance evaluations for ridesharing behaviors under various incentive measures and carbon reduction policies. To address this gap, this study proposes a Logit-based multi-class stochastic user equilibrium ridesharing assignment model. The model integrates both carbon costs and additional costs associated with specific policies, such as congestion pricing, into user travel costs. An evaluation system based on Environmental, Social, and Governance (ESG) criteria is developed. Lastly, numerical experiments are conducted on the model using the Sioux-Falls road network. The results demonstrate that congestion pricing policy generally benefit the environmental dimension. It also exhibits positive effects on social and governance dimensions at lower rates, but these benefits diminish with higher rates, potentially adversely affecting platform governance. Traffic restriction policy enhances ESG composite indicators when the network accommodates a high number of vehicles, though stricter restrictions on private car use reduce social and governance benefits. High-Occupancy Vehicle (HOV) lane policy improves the environmental dimension but has limited social impacts and can negatively affect governance indictors. Carbon cost reduction policy significantly promotes environmental benefits but also negatively affects governance. Among the four policies examined, traffic restriction policy outperforms in environmental dimensions, congestion pricing policy is optimal in social dimensions, and under lenient carbon emission constraints, congestion pricing policy also performs well in governance. Under stricter carbon emission constraints, however, HOV lane policy is more appropriate.

To explore the relationship between community identification and the carbon reductionbenefits of shared bicycles, this study examines the identification of carbon reduction benefit communities and their influencing factors in Xi’an, China. First, based on the Hello Bike order data in Xi’an in 2020, the spatiotemporal distribution characteristics of shared bicycle trips are analyzed. Second, the carbon reduction benefits of shared cycling are quantified, and their temporal variation characteristics are analyzed. Then, the Louvain algorithm is employed to identify communities in the central urban area of Xi’an based on carbon reduction benefits, followed by classification using the K-means clustering algorithm. Finally, the Gradient Boosting Decision Tree (GBDT) model is applied to explore the impact of the built environment on carbon reduction benefits. The results indicate that shared bicycle usage exhibits distinct morning and evening peak periods, with hotspots concentrated along subway lines and subway transfer stations. Shared bicycles significantly contribute to carbon reduction, with evident peak-hour effects. A total of 16 communities are identified based on carbon reduction benefits, with minimal overlap between these active communities and administrative divisions. The identified communities exhibit a “low-coupling, high-cohesion” structural pattern, where the city center contains a greater number of smaller communities, while peripheral areas have fewer but larger communities. Central communities demonstrate more significant carbon reduction benefits. Based on the weighted average degree, graph density, and average clustering coefficient of the community, the 16 communities are categorized into three categories: low, medium, and high carbon reduction areas. All built environmental factors positively influence carbon reduction benefits, though to varying extents. The findings provide valuable insights for the management and policy formulation of shared bicycle carbon reduction initiatives in Xi’an.