This article delves deeply into the passenger satisfaction survey data of Guangzhou Metro Line 3 in 2024, and constructs an evaluation index system covering six dimensions, including Off-station perception. Drawing on the concept of Location Potential, a “Station Location Potential Index”(SLPI) was established to classify stations into three levels: “below-expectation”, “meet-expectation” and “exceed-expectation” stations. Structural Equation Modeling with Partial Least Squares(PLS-SEM) was applied to estimate the drivers of satisfaction across these different station levels. The key finding is the significant spatial heterogeneity in the drivers of passenger satisfaction across various station tiers. For the below-expectation stations, “operation-facility” and “entry-exit” are the primary determinants of evaluation, with “responsiveness”s is key to service recovery. For the meet-expectation stations, “off-station perception” becomes the primary influencing factor, indicating a significant increase in passenger demand for connection between rail transit and city life. For the exceed-expectation stations, “responsiveness”, “environment”, and “interchange” gain significant importance and are key to crafting a superior experience. The result provides an empirical basis for precision management and service enhancement.

Highlights What are the main findings? AD-YOLO improves small object detection in traffic-dense UAV images by integrating adaptive orientation-aware feature extraction, dual-path cross-scale feature fusion, and a reparametrized large-kernel fusion module. AD-YOLO outperforms baseline models in detection accuracy with acceptable computational costs, demonstrating its strong robustness and application potential under complex aerial perspectives. What are the implications of the main findings? Jointly modeling object orientations, multi-scale contexts, and bidirectional feature interactions contributes to enhancing the detection of densely distributed, scale-varying objects in UAV images. AD-YOLO offers a concise yet effective approach to boosting detection performance on dense and small objects in UAV imagery, without requiring extensive modifications to the original framework.Highlights What are the main findings? AD-YOLO improves small object detection in traffic-dense UAV images by integrating adaptive orientation-aware feature extraction, dual-path cross-scale feature fusion, and a reparametrized large-kernel fusion module. AD-YOLO outperforms baseline models in detection accuracy with acceptable computational costs, demonstrating its strong robustness and application potential under complex aerial perspectives. What are the implications of the main findings? Jointly modeling object orientations, multi-scale contexts, and bidirectional feature interactions contributes to enhancing the detection of densely distributed, scale-varying objects in UAV images. AD-YOLO offers a concise yet effective approach to boosting detection performance on dense and small objects in UAV imagery, without requiring extensive modifications to the original framework.Abstract The densely distributed, scale-varying objects in unmanned aerial vehicle (UAV) images, together with their dynamic, diverse, and unconstrained backgrounds, make conventional detection methods prone to missed detections, false alarms, and localization biases. To improve UAV vision tasks, we propose AD-YOLO, a unified method tailored for small object detection in traffic-dense settings. First, a module combining an adaptive rotation convolution unit and grouped directional attention with mixed-kernel features is introduced to enhance the model's orientation invariance and multi-scale discrimination. Then, a dual-path collaborative feature pyramid network is proposed to jointly refine the model's semantic and spatial details via a multi-directional context aggregation path and a hierarchical semantic progressive fusion path. Last, a hierarchically dense reparameterized large-kernel module is designed to produce broader receptive fields with reduced computational complexity. Extensive experiments on the VisDrone2019 and UAVDT datasets demonstrate that AD-YOLO outperforms state-of-the-art methods in detection accuracy while maintaining favorable computational efficiency.

With the continuous expansion of urban rail networks and increasing passenger volumes, attention to subway passenger satisfaction has grown. This paper uses passenger satisfaction survey data from Guangzhou Metro Line 2 in 2023 as a case study. After classifying stations by POI, a multiscale geographically weighted regression (MGWR) model and a generalized ordered logit model were established to analyze the impact of spatial attributes and individual factors on satisfaction. The results show that mixed land use positively affects satisfaction, while land use intensity, service ratios, housing prices, and transfer station numbers have negative effects. By quantifying travel demand levels at different stations using a modified potential formula, the paper analyzes individual factors affecting passengers at different demand levels. The findings indicate that passengers at both high and low demand stations report lower satisfaction, with occupation being the most influential factor at high-demand stations and education level at others. This study provides insights for targeted measures to improve passenger satisfaction.

Pandemic emergencies often result in a surge in demand for emergency supplies in affected regions. To expedite the recovery process, ensuring prompt and reliable emergency supplies is crucial, yet the demand for these supplies often exhibits stochastic and time-varying characteristics as the emergency evolves. This paper introduces a two-echelon multi-period locationrouting allocation problem, with the objective of concurrent optimization of facility location, transportation routing, and the allocation of various types of emergency supplies. We integrate a constraint relationship equation that links supplies and cure rate into the established SEIR model to illustrate how emergency supplies counter the spread of the epidemic. In terms of the uncertain demand for various emergency supplies across multiple emergency periods, we quantify the nominal supply demand utilizing an enhanced SEIR model. On this basis, a set of uncertain demands for emergency supplies that are subject to random disturbances in real demand is established. We devise a multi-objective robust optimization model employing a budget-of-uncertainty robust approach to minimize total transportation time, total transportation cost, and shortage of emergency supplies. Subsequently, we develop a customized multi-objective discrete gray wolf algorithm aimed at enhancing solution efficiency. Our model is applied to a real-world case study in Shanghai during the 2022 COVID-19 pandemic. Results have proved that the algorithm is effective and feasible. Managerial insights are also provided.

Pick/drop stand road (PDSR) is necessary for specific functions, but it very easily generates congestion and thus results in a less safe and comfortable ride for passengers. The Internet of Vehicles has offered new ideas and technical means to solve the problem. First, this study divides the PDSR into different functional zones and constructs the coordinated control environment of connected vehicles (CVs). A car-following coordinated control strategy (CF-CCS) and a lane-changing coordinated control strategy (LC-CCS) are proposed based on a mixed platoon of CVs and human-driven vehicles. For CF-CCS, this study offers a model for car-following coordinated platoon control based on terminal state prediction of the leading vehicle, with platoon safety and passengers' comfort as optimization objectives, and solves it using a modified particle swarm algorithm. For LC-CCS, a model for lane-changing coordinated control of platoon is proposed, and a numerical approach is utilized to solve it. It is based on finding the best area for lane-changing location while considering the safety of lane changing and the efficiency of vehicle traffic on the PDSR. Last, VISSIM and MATLAB simulations are conducted to prove that the proposed two strategies can effectively ease congestion, significantly increase the speed of traffic flow on PDSR, and provide passengers with safer and more comfortable rides.

Large-scale centralized emergency rescue staff scheduling under emergencies has always been a challenge. With a specific focus on nucleic acid testing amid the COVID-19 pandemic, this study introduces a model that considers dynamic changes in staff supply at rescue points, the varying demand at sampling points over time, and their impact on the spread of the epidemic. We quantified the emergency weight of sampling points by the demand and the regional risk degree and adopted robust optimization with base constraint. With the objectives of minimum comprehensive service distance, maximum weighted value of demand satisfaction rate, and minimum loss caused by unmet demand, a multi-stage, multi-medical rescue point medical staff scheduling model for nucleic acid sampling points with the influence caused by uncertain demand disturbance was built, and an improved NSGA-II-HC algorithm was designed to solve the optimization problem under the influence of demand disturbances. The testing results proved that the NSGA-II-HC algorithm can tackle the issue of insufficient uniformity and diversity in the solution set of NSGA-II while Multi-Objective Particle Swarm Optimization (MOPSO) failed to do so. Taking the epidemic data of Guangzhou city in May 2021 as a case study, our model and algorithm are verified to be feasible. We offer a scheme selection strategy according to the loss degree of two conflicting objectives, the comprehensive service distance and shortage loss. The results suggest that, compared with the demand deterministic model, the decision mechanism under the robust model can reduce the deviation from the optimization objective due to the demand disruption

The development of modular self-driving bus provides a new technology and way for the development of passenger remote access system in large airports in the future. In this paper, taking Guangzhou Baiyun International Airport as an example,based on the analysis of the current passenger travel characteristics and problems of Guangzhou Baiyun International Airport, an airport remote parking passenger connection system based on modular autopilot bus is designed, and the operation effect is simulated and evaluated. The results show that the use of the system can solve the problems of insufficient parking space and land-side traffic congestion in the airport terminal to a certain extent, better meet the needs of passengers than the traditional shuttle bus system, and can also effectively reduce the operating cost of the system, which is conducive to the diversification and convenience of airport traffic.

在四网融合背景下,联通城际铁路和地铁票务系统,实现付费区无缝换乘对于轨道交通票价一体化发展具有重要意义.通过对比国内外付费区换乘票制票价的实践经验,结合粤港澳大湾区城际铁路和地铁线网票价的发展基础,探讨一体化发展趋势下城际铁路和地铁的换乘模式与票制,分析计程票制下城际铁路和地铁计价、跨制式和跨区段计价等不同策略的适应性.结果表明,付费区换乘模式及同制式跨区段采取"不分段、不重新起步"和跨制式采取"分段、重新起步"的计价策略更符合粤港澳大湾区轨道交通票价的一体化发展.

港湾式公交中途站的合理配置是公交线路优化的基础约束条件和重要技术支撑.文中通过对公交车进出港湾式中途站全过程总耗时的分析,借助排队论和插队间隙理论,在HCM分析模型的基础上,研究通行能力、饱和度、不同线路到站概率等因素对中途站对应最优经停线路数量的影响,构建影响中途站停靠承载能力修正模型;经模型分析,港湾式中途站停靠线路的车辆到站情况不同,则到达概率不同,在降低对道路交通影响的情况下可经停的线路存在差异;不同道路的运行速度、饱和度和线路车辆到达概率直接影响车站对公交线路的承载能力.

The emergence of connected and autonomous vehicles (CAV) is of great significance to the development of transportation systems. This paper proposes a multiple-factors aware car-following (MACF) model for CAVs with the consideration of multiple factors including vehicle co-optimization velocity, velocity difference of multiple PVs, and space headway of multiple PVs. The Next Generation Simulation (NGSIM) dataset and the genetic algorithm are used to calibrate the parameters of the model. The stability of the MACF model is first theoretically proved and then empirically verified via numerical simulation experiments. In addition, the VISSIM software is partially redeveloped based on the MACF model to analyze mixed traffic flows consisting of human-driven vehicles and CAVs. Results show that the integration of CAVs based on the MACF model effectively improves the average velocity and throughput of the system.

Learning from the public transportation subsidy model, a certain subsidy to shared bicycles on campus is considered here, and supplemented by subsidy penalties linked to service quality. By considering the elastic demand affected by price, a multi-mode stochastic user equilibrium bi-level programming model with the goal of maximizing social welfare is established, and a genetic algorithm combined with the MAS algorithm is designed to solve the optimal price. Actual campus case data is used to verify the model and algorithm. The results show that there exists a certain subsidy rate that can make corporate profits and bicycle sharing rate reach the optimal rate, setting the ratio linked to service quality too high will limit the growth of social welfare, and the amount of shared bicycle must reach a certain scale in order to make society welfare advantage greater than the profit advantage.

将私人小汽车通勤者划分为Ⅰ类堵车焦虑型和Ⅱ类准点焦虑型两种类型,引入出行敏感系数(堵车焦虑系数、准点焦虑系数)和瓶颈忍耐系数来刻画两类通勤者的出行焦虑程度,在标准瓶颈模型的基础上构建新的出行成本函数,并将其应用于分析单一线路和混行线路下的出发率、高峰期起止时刻和通勤者构成等指标.研究结果表明:当道路上仅有某一类通勤者存在时,堵车焦虑系数、瓶颈忍耐系数以及准点焦虑系数的增加均会使单一线路通勤者的系统总感知出行成本降低;混行情形1中Ⅱ类通勤者的准点焦虑系数增大时,该类通勤者数量增加,系统总感知出行成本减小,高峰期前移,瓶颈忍耐系数增大时,系统总感知出行成本减少,高峰期后移;混行情形2中Ⅰ类通勤者的堵车焦虑系数与瓶颈忍耐系数对该类群体数量的影响具有两面性,不同瓶颈忍耐系数与堵车焦虑系数的组合下,该类群体的吸引力呈现单调减小、先减后增和单调增加三种不同的趋势.

Researchers usually conduct a questionnaire survey at bus stops to obtain data regarding the satisfaction of bus passengers with waiting times. The results are affected by many factors. Among them, the land use of the bus stops was proved to have an important impact on the survey results. The main contribution of this paper is the introduction of propensity score matching (PSM) into the evaluation of passenger waiting time satisfaction. By eliminating interference factors, this paper can quantify the impact of the various land use types of bus stops. On this basis, a method to modify the survey results of passenger waiting time satisfaction is proposed. This paper takes data pertaining to passenger satisfaction with bus service quality in Guangzhou City in 2018–2019 as an example, and the findings include that: different land use types of sites have different effects (positive or negative) on passenger waiting time scores. Also, matching propensity scores can balance the distribution of covariates between the treatment and control groups, effectively excluding the interference of other factors. After correcting the original rating results, this study found that 61.84% of the routes' waiting time ratings were overestimated. This finding indicates that data correction is necessary to accurately identify passenger waiting time ratings. The waiting time satisfaction of people using residential stations can best reflect the actual level. Therefore, it is suggested that stations with residential land use in each administrative area should be taken as representatives to conduct a waiting satisfaction survey.

Customized bus (CB) is an increasingly popular mode of transportation in many cities around the world. However, studies on CB network design have mostly overlooked three options that may further improve system performance: passenger-route assignment, passenger transfer, and modular vehicles. To bridge this gap, this paper proposes to design a transfer-based CB network with a modular fleet while simultaneously optimizing the passenger-route assignment. To solve the optimal network structure with this new design paradigm, we formulate the network design problem into a nonlinear mixed integer optimization model. A linearization approach and a particle swarm optimization (PSO) algorithm are proposed to solve the exact and near-optimal solution(s) to the model, respectively. Numerical experiments are conducted on the Sioux Falls network and a large-scale network in Chengdu, China. Results show that the customized PSO algorithm efficiently provides high quality near-optimal solutions compared with CPLEX, the genetic algorithm, and the simulated annealing algorithm. Results also show that incorporating passenger-route assignment optimization and the transfer operation produces a more costeffective CB operational network with less operational costs and higher service quality. The benefit increases as the passenger demand grows.

In the bus transit network, bus stop configurations influence the system's operational efficiency, but setting the proper number of lines for bus stops is a challenge. We define the line-based bus stop capacity as the maximum number of bus lines that can be accommodated by a bus stop. Based on queuing theory, a mathematical model is established to describe the quantitative relationship between the line-based bus stop capacity and associated factors including bus stop capacity and departure interval of bus lines. Then, an iterative method is proposed to calculate the line-based bus stop capacity. Sensitivity analysis concerning major influencing parameters is conducted. The results show that the Acceptable Inbound Failure Rate and the boarding and alighting time are sensitive to line-based bus stop capacity.

为对比"夜间慢速充电+白天快速充电"、"夜间慢速充电"两种不同的整车直接充电模式下,电动公交车的车辆发车计划与充电费用,考虑时刻表、电量限制、充电站距离、允许车辆等待发车时间阈值的影响,建立并运行模型.结果发现,两种模式需要的最小公交车辆数、单日需要的充电费用均不同,建议企业应考虑车辆数、充电费用等因素选取适合的充电模式.此外,还对公交车前往充电站的距离、公交车到达站点等待发车的时间阈值与充电费用的关系进行了敏感性分析,据此建议合理设置充电站位置,并根据具体条件设置合适的等待发车时间阈值,以减少充电费用.

出租车是城市客运交通中最为活跃的方式之一,制定与城市经济发展相适应的出租车运价,可直接作用于居民的交通出行成本,从而影响出行者的选择行为,达到平衡交通方式和出行时空的目的.基础费率作为出租车运价制定的基础和依据,确定一种合理的出租车基础费率模型具有重要的实际意义.以多方式城市交通路网为研究基础,建立双层规划模型来研究城市客运出租车基础费率问题,上层目标为交通系统的总成本最小,下层为多方式交通网络平衡模型.采用遗传算法与MSA迭代加权法的混合算法对模型进行求解.为验证模型的适用性,构建了简单路网进行算例分析.结果 表明:基于系统总成本最小,该模型可求得较为合理的出租车最优基础费率,且系统内各交通方式的分担率也较为合理.在政府指导出租车定价的过程中,可根据出租车基础费率与运价体系的关系,设计最优的出租车定价方案,从而达到降低社会总成本的目的.

为了研究网络预约出租汽车这一新兴服务行业的合理营运模式,在定性研究的基础上利用有效的决策模型对不同运营模式的实施效果展开量化分析,基于博弈论构建了一种涉及网络预约出租汽车营运四方博弈模型,即管理者(网络预约出租汽车营运政策制定者)、网络预约出租汽车平台、乘客和传统出租车企业的Stackelberg博弈模型.

This paper studies the problem of deploying electric bicycle (e-bike) sharing stations and determining their capacities, i.e. the number of shared e-bikes and charging piles, considering travelers’ responses to the charging demands and different deployment schemes. Given a one-way station-based setting, we propose an e-bike sharing network where the generalized trip cost is measured as the sum of the delay cost at stations and the travel time en-route. To estimate the trip costs, we modeled the pick-up and drop-off e-bikes at each sharing station as two different queues affected by e-bikes’ charging demands, and described the traffic flow of shared e-bike on each route based on Greenshield’s model. Further, the e-bike sharing station deployment problem was then formulated as a bi-level programming model, taking into account the government’s and individual travelers’ profits. The uniqueness of solution was proved. For the purpose of solution approach, this bi-level model was then reformulated into a single-level mixed-integer programming model, and a hybrid particle swarm optimization algorithm was proposed to solve the single-level model. Numerical experiments were presented to demonstrate the validity of the proposed model and solution technique. More importantly, through numerical experiments, further insights for designing an e-bike sharing system were examined and discussed: 1) sharing stations are bottlenecks in the e-bike sharing network, since the charging activities cause travelers large delay costs; 2) a well-designed quick-charging technology and reservation policy could be incorporated into e-bike sharing systems to reduce system costs; 3) the proposed hybrid particle swarm optimization algorithm shows good solution quality and convergence performance.