Current Issue

25 October 2025, Volume 25 Issue 5 Previous Issue   

---

For Selected:

Download Citations EndNote Ris BibTeX

Toggle Thumbnails

Select

A Review of AI-driven Trajectory Prediction Methods for Autonomous Vehicles TIAN Daxin, XIAO Xiao, ZHOU Jianshan 2025, 25(5): 1-24.  DOI: 10.16097/j.cnki.1009-6744.2025.05.001 Abstract ( )   PDF (1838KB) ( )   In autonomous driving systems, trajectory prediction plays an important role in connecting vehicle's perception and decision-making, and enhancing driving safety and overall system robustness. In recent years, with the continuous advancement of artificial intelligence (AI), the AI-driven trajectory prediction methods have seen significant progress in terms of accuracy, adaptability, and the capability to model complex traffic environment. This paper provides a systematic review of mainstream trajectory prediction methods in autonomous driving, with a focus on predictive model frameworks. It first revisits traditional physics-based approaches, and then highlights current research trends, including modeling paradigms based on classical machine learning, deep neural networks, and reinforcement learning. Additionally, recent developments in explainable AI techniques aimed at improving model transparency and safety are discussed. Based on comparative analysis, the paper evaluates the strengths and limitations of various models in interaction modeling, multimodal uncertainty, and generalization capability. Furthermore, it organizes trajectory prediction evaluation metrics and publicly available trajectory prediction datasets according to their characteristics and application scenarios, and summarizes representative real-world deployments from both domestic and international sources. At last, considering the existing research bottlenecks and future development trends, the paper outlines potential directions for future studies, such as enhancing model interpretability, effectively integrating multimodal information, and designing unified frameworks for joint prediction and planning. The purpose of this review is to provide insights and references that can be used in the future research and applications. References | Related Articles | Metrics

Select

Review of Connected and Autonomous Vehicle Dedicated Lane Setup CHENG Guozhu, WANG Wenzhi, YANG Zihan, WANG Guopeng, CHEN Yongsheng, GU Shuang 2025, 25(5): 25-39.  DOI: 10.16097/j.cnki.1009-6744.2025.05.002 Abstract ( )   PDF (2429KB) ( )   With the advancement of information and communication technologies, autonomous vehicles (AV) and connected and autonomous vehicles have emerged as promising solutions to address traffic congestion, to enhance traffic safety, and to improve overall traffic efficiency. This paper provides a comprehensive review of the methods for setting up dedicated lanes for connected and autonomous vehicle (CAV). It begins by tracing the evolution of CAV dedicated lanes and elaborating on the background and significance of their implementation. Based on relevant literatures, it then delves into the methodologies for calculating road capacity, providing a foundation for predicting the impact of CAV dedicated lanes on traffic operations, evaluating strategies, and making necessary adjustments. Furthermore, the paper conducts an in-depth analysis on the strategies for setting up CAV dedicated lanes, including the conditions based on CAV penetration rates and traffic demand. It also explores the determination of the number and location of lanes, access methods, and lane separation approaches under various influencing factors. Finally, the paper proposes that future research should focus on understanding the changes in influencing factors post-implementation of CAV dedicated lanes and their alignment with real-world traffic conditions. It also emphasizes the need for establishing specific standards for setting up CAV dedicated lanes to ensure their function effectively across different traffic scenarios. References | Related Articles | Metrics

Select

Impact of Two-stage Warning on Takeover Performance of L3 Autonomous Vehicles WU Fuwei, YANG Hui, MA Yong, CUI Tongchuan, YAN Shengyu, ZHANG Zhi 2025, 25(5): 40-49.  DOI: 10.16097/j.cnki.1009-6744.2025.05.003 Abstract ( )   PDF (1826KB) ( )   To investigate the impact of two-stage warning method on takeover performance in Level 3 autonomous vehicles, the takeover experiments of autonomous driving were conducted on a highway takeover scenario by using a three-degree-of-freedom driving simulator. First, the data on takeover behavior was collected from 25 participants. Second, the impacts from different takeover methods (single-stage and two-stage) and warning time intervals (2, 3, 4 s) were analyzed on takeover time, lane change completion time, maximum steering wheel angle, maximum longitudinal deceleration, and time-to-crossing distance in this study. Finally, based on these metrics, a fuzzy comprehensive evaluation model using the entropy weight method was established to quantitatively assess the takeover performance of different takeover modes. The results indicate that compared to the single-stage takeover mode, the two-stage takeover warning method significantly reduces the time for a driver to take over his/her vehicle, the lane change completion time, and significantly increases the time-distance at the moment of lane crossing. On the premise that the takeover request time is 5 seconds, a 2-second warning time interval in the two-stage warning allows a driver to gain situational awareness and take over, while a 3-second warning time interval is already sufficient to effectively improve the lateral stability of vehicles. Compared to 2-second and 3-second warning intervals, the 4-second warning interval reduced lane change completion time by 6.7% and 9.6%, respectively, and decreased maximum longitudinal deceleration by 64.1% and 56.4%, respectively, with significant differences. In the fuzzy comprehensive evaluation, the two-stage takeover mode with a 4-second warning time interval scored the highest, demonstrating superior takeover performance. The research conclusions provide theoretical references for the application of two-stage warning in the takeover process of Level 3 autonomous vehicles. References | Related Articles | Metrics

Select

Optimization Method for Autonomous Bus Operation Considering Adaptive Control Points DOU Xueping, YANG Minghui, XIONG Jie 2025, 25(5): 50-58.  DOI: 10.16097/j.cnki.1009-6744.2025.05.004 Abstract ( )   PDF (1959KB) ( )   To address the issue of uncertain operation times for autonomous buses in time-varying traffic environments, this study proposes an optimization method for multi-objective operation that incorporates adaptive control points. The method overcomes the limitations of traditional line-level control schemes for human-driven buses, by constructing a stochastic mixed-integer nonlinear programming model, jointly optimizing the number and locations of time control points along with the scheduled arrival times, and minimizing the weighted sum of negative utilities from both arrival time deviations and travel time deviations. This approach leverages the synergistic advantages of adaptive control point strategies and autonomous driving technology to enhance bus service quality under uncertain conditions. Through the linearization techniques and Monte Carlo simulation, the proposed model is further transformed into a deterministic mixed-integer linear programming model, which is efficiently solved by using a branch-and-bound algorithm. The proposed method is validated through multiple comparative experiments on a bus route with 19 stops and 12 consecutive trips. The results demonstrate that, compared to fixed control point strategies, the proposed variable control point strategy and optimization method reduce mean arrival time deviation disutility by 11.35% and mean travel time deviation disutility by 58.87% during peak hours. During off-peak hours, the reductions reach 18.21% and 38.68%, respectively. Sensitivity analysis further reveals that the optimal operational performance is achieved when the number of regular stops between adjacent control points is set to 20% of the total stops on the bus route, under the experimental conditions specified in this study. References | Related Articles | Metrics

Select

Cooperative Optimization of Lane Allocation and Vehicle Trajectory at Intersections Under Connected-and-Automated-Vehicle Environment SONG Lang, HU Xiaowei, YU Shanchuan, AN Shi 2025, 25(5): 59-71.  DOI: 10.16097/j.cnki.1009-6744.2025.05.005 Abstract ( )   PDF (2673KB) ( )   In the collaborative optimization of intersection signal timing and Connected and Automated Vehicle (CAV) trajectory planning, the CAV exit, left turn, through, and right turn lanes can be assigned dynamically in the operation period. Based on the characteristics of CAV technology, this paper proposes a set of dynamic control rules for lane assignment under CAV, named as "flexible lane strategy". Compared to the existing fixed lane strategy, the proposed strategy can adjust the number of exit lanes and entrance lanes (including left turn, through, right turn) for different directions of traffic flow during operation. Lane assignment, signal timing and CAV trajectory planning are incorporated into a unified optimization framework to build a mixed integer linear programming optimization model. Meanwhile, feasible phase and sequence schemes can be automatically generated according to lane assignment in each direction, and the effectiveness of the model is verified through a case study. The results show that the optimization model can generate the optimal lane assignment scheme according to the traffic demand of each flow direction, especially when the lane assignment of the fixed lane strategy does not match the traffic composition of each flow direction, the flexible lane strategy helps to improve the intersection traffic efficiency. In low flow scenario, the flexible Lane strategy reduces average vehicle delay by 4.08%. In high-traffic scenarios, the fixed lane strategy at the intersection will be in a supersaturated state, while the flexible lane strategy can still meet the demand. References | Related Articles | Metrics

Select

Parking Choice Behavior Analysis of Autonomous Vehicle Users Considering Psychological Factors HAN Yan, YUAN Changyin, TANG Xintian, GUAN Hongzhi 2025, 25(5): 72-82.  DOI: 10.16097/j.cnki.1009-6744.2025.05.006 Abstract ( )   PDF (2669KB) ( )   Autonomous vehicles equipped with long-range autonomous valet parking function drop off the users at their destinations and then autonomously idle to the remote parking lot to park. It can present an innovative solution to the spatial and temporal misallocation of parking resources in urban centers, and at the same time bring a new problem of choice. Based on the theory of consumer purchase decision, the parking choice mechanism of Privately-owned Automated Vehicles (PAVs) is discussed by introducing two psychological latent variables: perceived risk and waiting attitude. An empirical survey on the parking choice behavior of PAVs users was designed. And an Integrated Choice and Latent Variable (ICLV) model and MNL model were developed for the parking choice behavior of AV users. The results indicate that psychological latent variables, along with personal characteristics, travel attributes, and parking program attributes, significantly influence the parking decisions of users. Moreover, the ICLV model, which incorporates psychological latent variables, demonstrates a notably superior fit over the traditional multinomial Logit model. Destination parking fee is elastic to changes in the probability of parking choice. When the destination parking fee increases from 11 yuan·h-1 to 15 yuan·h-1, the probability of parking choice will decrease from 59.9% to 34.4%, and the probability of proximal parking slots and remote parking slots will increase from 11.8% and 28.3% to 19.4% and 46.2%, respectively. The probability of choosing destination parking will increase from 12.0% and 24.4% to 53.3% and 52.0% when the risk and pick-up waiting aversions perceived by users increase from 1 to 5, respectively. The research findings can provide a theoretical basis for differential parking pricing in regional parking lots in the era of autonomous vehicles. References | Related Articles | Metrics

Select

Estimating Link Travel Time Under Sparse License Plate Recognition Device Coverage WANG Dianhai, WANG Yifei, HUANG Yulang, LIU Yong, ZENG Jiaqi 2025, 25(5): 83-90.  DOI: 10.16097/j.cnki.1009-6744.2025.05.007 Abstract ( )   PDF (2206KB) ( )   License Plate Recognition (LPR) devices are crucial traffic state detectors in urban road networks, but the high cost limits their deployment scale and density. This paper proposes a method for urban road network link travel time estimation under sparse devices deployment. Considering the interference of dwell trajectory travel time outliers, this study develops a mixed-integer optimization model for the link travel time measurement, and proposes an alternating solution method based on fixed-point iteration. First, path travel time distributions are calculated based on link travel time distributions to identify abnormal travel times. A travel time assignment method is then designed to allocate normal trajectory travel times to individual links. To ensure computational stability, Bayesian updating is applied to the travel time distribution parameters of high-traffic links, and the parameter proportionality is extended to links with insufficient traffic flow. Link travel times and abnormal trajectories are jointly estimated through iteration procession. Experiments on a real LPR dataset from Hangzhou demonstrate that the proposed method achieves a mean percentage error (MAPE) of 13.29% under a 70% device penetration rate. Compared to the gradient descent method, the MAPE is reduced by 7.69%, and the number of iterations is reduced by 99.4%. Furthermore, for urban scenarios with even sparser LPR device coverage, when the device coverage rate drops to 30%, the proposed method results in the MAPE of 18.51%. References | Related Articles | Metrics

Select

Sporadic Road Congestion Early Warning Based on Viscous Vehicle Group Identification HAN Baorui, JI Yuxuan, LI Gen, YANG Zheng, YAN Rongtian, XU Shengrui 2025, 25(5): 91-102.  DOI: 10.16097/j.cnki.1009-6744.2025.05.008 Abstract ( )   PDF (3538KB) ( )   To investigate an early-warning method for sporadic road congestion considering vehicle grouping, this paper proposes a framework to identify Viscous Vehicle Groups (VVG) based on dual car-following modes and multi-parameter fused hierarchical clustering. First, using two microscopic car-following patterns: Normal Following (NF) and Staggered Following (SF), the study defines the Normal Following Units (NFU) and Staggered Following Units (SFU) within a VVG. Then, with respect to the three dimensions: sporadic congestion characteristics, spatiotemporal similarity, and structural dynamic stability, five core parameters were selected for quantifying these features: vehicle speed, time gap, spatial gap, speed difference, and acceleration difference. Parameters are screened through the Kolmogorov-Smirnov tests and statistical analysis, and the bottom-up hierarchical clustering is applied to determine threshold ranges for NFU and SFU parameters and to identify VVGs. The case study is performed using real-world traffic flow data measured on roads in Nanjing city. The results show that: (1) NFU and SFU parameter distributions differ significantly, warranting separate treatment of the two unit types. (2) Among 3088 vehicle samples, 243 vehicles were successfully identified as participating in VVGs, validating the model's detection capability. (3) The VVGs predominantly emerge before sporadic congestion, serving as effective precursors, with an early warning success rate of 93.33%. This study provides theoretical and methodological support for the early warning of sporadic road congestion. References | Related Articles | Metrics

Select

A Data-driven Cellular Transmission Model for Self-learning on Expressways LIN Peiqun, HUANG Chaoshuo, ZHOU Chuhao, PANG Chonghao, DENG Kaiyu 2025, 25(5): 103-113.  DOI: 10.16097/j.cnki.1009-6744.2025.05.009 Abstract ( )   PDF (2270KB) ( )   An efficient traffic simulation model can offer a scientific basis on proactive traffic management and network optimization for transportation authorities by providing the changes of real-time and short-term traffic flow. However, in complex traffic scenarios, model parameters are susceptible to environmental influences, leading to the decrease in simulation accuracy. This paper proposes a data-driven Self-Learning Cellular Transmission Model (SL-CTM), which employs a data-driven approach, adaptively fitting the input features, internal states, and output flows of cells. It can autonomously learn the parameters that typically require manual calibration in the cellular transmission model, thereby effectively avoiding the complex parameter calibration process and enhancing simulation accuracy and operational efficiency. Validation results based on the empirical data from the South Second Ring Expressway and the Fokai Expressway in Guangdong Province indicate that, compared with the Random Forest model, the SL-CTM reduces the Weighted Mean Absolute Percentage Error (WMAPE) of traffic flow simulation by 17.55% and 15.83% on the two expressways, respectively. Compared with the Long Short-Term Memory (LSTM) model, the SL-CTM achieves reductions of 12.37% and 10.50% in WMAPE, respectively. These findings demonstrate that the SL-CTM is capable of achieving a stronger responsiveness to sudden traffic flow variations while requiring fewer initial features. Compared to the SUMO simulation software, SL-CTM achieves a 55.90% reduction in WMAPE and a 72.57% improvement in simulation speed, which exhibits superior performance in high-traffic scenarios. The study demonstrates that SL-CTM significantly improves the accuracy and computational efficiency of traffic simulation, providing more reliable technical support for dynamic traffic management in complex traffic environments. References | Related Articles | Metrics

Select

Multi-factor Equilibrium Allocation Model and Application of Container Freight Flow in Western Land-Sea New Corridor JIANG Jun, HUANG Haoran, LI Huomei, FU Xiaona, OUYANG Fan 2025, 25(5): 114-123.  DOI: 10.16097/j.cnki.1009-6744.2025.05.010 Abstract ( )   PDF (1821KB) ( )   The allocation study of container freight flows in the Western Land-Sea New Corridor can effectively enhance the corridor's transportation efficiency and achieve cost reduction and logistics efficiency improvement. In view of the current development of the container transport network along the corridor, this study identifies transportation cost, transit time, and carbon emissions as the core factors influencing freight flow allocation. Based on user equilibrium and system optimum theories, this paper develops a freight flow allocation model with the objective of minimizing multi-factor generalized transportation cost. An improved Frank-Wolfe algorithm is designed to solve the model, incorporating a dynamic impedance matrix to optimize the adjustment mechanism of path impedance. Key nodes along the corridor from Chongqing to ASEAN countries and connected regions are selected to construct a multi-node, multi-path container transport network. Freight volume is allocated based on actual transportation demand data from June 2024. The results show that the proposed model effectively balances transportation cost, transit time, and carbon emissions. Compared with the user equilibrium strategy and the actual distribution scheme, the system optimum allocation strategy reduces total generalized cost by 2.94% and 5.34%, respectively. Based on the findings, it is recommended that the corridor's operation platform appropriately increase the supply of rail-sea intermodal transport and adopt freight rate adjustment and other effective measures to guide container cargo shifting toward international rail transport modes. References | Related Articles | Metrics

Select

Path Optimization of Reefer Container Intermodal Transportation Under Transfer Time Uncertainty WANG Yan, WANG Zekai, SONG Meixia, FANG Lixuan 2025, 25(5): 124-134.  DOI: 10.16097/j.cnki.1009-6744.2025.05.011 Abstract ( )   PDF (1895KB) ( )   Accompanied by the rapid development of cold chain logistics market and the increase in the importance of intermodal transport of reefer containers, a multi-objective path optimization model is constructed with the objectives of minimizing the total transportation cost and maximizing the ratio of time efficiency, which aims at the uncertainty problem of transit time during the intermodal transport of reefer containers in cold chain logistics. A triangular fuzzy number is used to characterize the uncertainty of transit time, and then the transformation of a fuzzy model to a deterministic model is realized through the opportunity constrained planning. In addition, an improved NSGA-II algorithm with adaptive crossover and variance probabilities is designed to solve the model and further compared with the traditional NSGA-II algorithm. The results show that the constructed model can effectively reduce the transportation cost and improve the transportation timeliness. The improved NSGA-II algorithm shows significant advantages in terms of solution set size and convergence speed, which are 16.2% and 21.7% higher than those of the traditional algorithm, respectively. Further the sensitivity analysis of railroad tariffs shows that when the railroad tariffs are reduced to 40%, all transportation modes would switch to railroad transportation. The total cost of transportation would be reduced by 19.8% and the time efficiency ratio is improved by 49.0% at that time. This study provides scientific decision support for cold chain logistics enterprises to choose the path, which helps to meet the challenge of transit time uncertainty and optimizes the intermodal transport path of reefer containers. References | Related Articles | Metrics

Select

Optimal Subsidy Strategy for Electric Vehicle Charging Infrastructure Considering Power Disparities WANG Fulin, HUANG Haijun 2025, 25(5): 135-144.  DOI: 10.16097/j.cnki.1009-6744.2025.05.012 Abstract ( )   PDF (2068KB) ( )   In order to address the issue of localized supply-demand mismatches and limited incentives in the current subsidy policies for charging infrastructure, this study develops a tripartite dynamic game model involving the government, charging operators and consumers. It utilizes operational data selecting from typical Chinese cities to conduct multi-scenario simulations, with the objective of examining the effects of construction subsidies, operational subsidies, and their combinations on the adoption of electric vehicles (EVs) and the resulting socio-environmental outcomes. The findings indicate that a differentiated construction subsidy model based on charging power can more effectively guide the deployment of DC/AC charging piles in public areas with a high demand for fast charging. In residential areas, uniform per-unit subsidies have been shown to be more effective in encouraging consumer adoption and expanding the EV user base. Simulation analysis reveals that as the industry decarbonizes, preferences for fast-charging intensify and infrastructure costs decline, public-area subsidies are suggested to transition from construction-focused to hybrid approaches. When the annual environmental benefits per vehicle exceed 16000 yuan, the operation- oriented subsidies have been demonstrated to be the most effective strategy for promoting electric vehicle adoption. As the charging infrastructure industry enters a phase of high-quality development, the government should progressively reduce the scale of construction subsidies while enhancing operational subsidies. This strategy would optimize the effectiveness of policy and ensure the maximization of environmental benefits. References | Related Articles | Metrics

Select

Optimization of Multi-allocation Multimodal Express Network Considering Hub Failures TIAN Shuaihui, CHEN Anqi, SUN Jiazheng, WANG Kejie, ZHANG Hengbo 2025, 25(5): 145-157.  DOI: 10.16097/j.cnki.1009-6744.2025.05.013 Abstract ( )   PDF (2049KB) ( )   To mitigate the issue of rising operational costs in express delivery networks caused by hub failures, this study proposes an implementation based on the backup hub strategy to reduce the impact of hub failures on the network. Queuing theory is employed to quantify the cargo processing time at the hub nodes. In the traditional single-allocation hub-and-spoke express delivery networks, hub failures can lead to significantly increment in network transportation costs and congestion costs. Therefore, in order to alleviate hub congestion and reduce operational costs, this study considers a network structure where non-hub nodes can be allocated to multiple hub nodes, as well as the diversity of transportation modes between hubs, and then constructs an optimized multi-allocation multimodal express delivery network model that accounts for hub failures. To validate the effectiveness of the optimized model, this study employs an enhanced genetic algorithm to solve the model using Turkey network dataset as the experimental benchmark. Experimental results demonstrate that, compared to the conventional multi-allocation multimodal express delivery network model, the proposed optimized model reduces the total failure cost by 29.3% and decreases the post-failure transportation cost by 36.4% These findings confirm that the proposed model effectively mitigates the impact of network disruptions caused by hub failures, and provides valuable insights for the optimization of express delivery network layouts. References | Related Articles | Metrics

Select

Choice Behavior of Connecting Transportation Modes for Arriving Air Passengers Under Synergy Between Product Attributes and Group Preferences WU Xiaoyu, ZHANG Xiaoqiang, CHEN Yuling 2025, 25(5): 158-168.  DOI: 10.16097/j.cnki.1009-6744.2025.05.014 Abstract ( )   PDF (2510KB) ( )   Cross-platform passenger ticket data, after personal information desensitization, cannot directly reveal the decision-making outcomes of connecting transportation modes for arriving airport passengers. To uncover the behavioral mechanisms behind passengers mode choices during airport arrival and departure processes, this paper introduces a synergy mechanism integrating transport product attributes and passenger group preferences. Generalized cost characteristics of inbound flight products are utilized to identify heterogeneous choice preferences across passenger groups. Generalized cost functions for inbound flights and connecting modes are established, incorporating four evaluation metrics: efficiency, economy, reliability, and safety. Sensitivity analysis of influencing factors is performed using Sobol' indices. An improved K-Means clustering algorithm categorizes flight products, while principal component analysis computes feature weights, classifying passengers into three typical groups: economy-oriented, efficiency-oriented, and experience-oriented. Prospect utilities of connecting modes for each passenger group are calculated using generalized cost values. An indifference perception utility threshold is introduced to account for bounded rationality in decision-making, optimized via multinomial Logit model fitting. The final model determines choice probabilities of connecting modes. A case study of passengers arriving at Nanning Wuxi Airport traveling to Nanning Railway Station is conducted using 2024 air and high-speed rail ticket data (January to December). Results show an optimal indifference perception utility threshold of 0.54, with model performance metrics: Mean Absolute Error (MAE) is 12.32, Root Mean Squared Error (RMSE) is 15.35, and Mean Absolute Percentage Error (MAPE) is 8.9%. Daily mode share rates are 20.04% (high-speed rail), 23.15% (bus), 26.25% (taxi), and 30.56% (private vehicles). This study provides references for integrated transportation hubs planning and operation optimizations. References | Related Articles | Metrics

Select

Nonlinear Effects of Built Environment and Subjective Perception on Bike-sharing and Metro Transfer Trips YIN Chaoying, ZHOU Yue, XU Zhenyu, SHAO Chunfu, WANG Xiaoquan, QI Xin 2025, 25(5): 169-178.  DOI: 10.16097/j.cnki.1009-6744.2025.05.015 Abstract ( )   PDF (1722KB) ( )   This study uses bike-sharing data from Shanghai to investigate the nonlinear effects of built environment factors on bike-sharing and metro transfer trips. The dependent variable was extracted by establishing transfer identification buffers, while independent variables were selected from both objective built environment and subjective perception. An XGBoost model was developed to investigate the nonlinear relationships between built environment factors and transfer trips. The results indicate that population density, distance to city center, and perceived safety significantly influence bike-sharing and metro transfer behavior, with relative importance scores of 25.0%、13.4%, and 9.8%. Moreover, built environment factors exhibit nonlinear effects and threshold effects on both bike-sharing and metro transfer trips, with varying operational mechanisms across different variables. Population density and perceived safety demonstrate positive effects on transfer trips, whereas distance to the city center shows a negative effect. This study provides reference for optimizing urban transportation planning and enhancing the coordinated service between bike-sharing and metro systems. References | Related Articles | Metrics

Select

Traffic Conflict Risk Identification in Multi-vehicle Linked U-turn Areas HU Liwei, GONG Qi, ZHAO Xueting, ZHOU Zeyu, CHEN Jiale, PAN Jiangxiong, YANG Can, MA Siyue 2025, 25(5): 179-192.  DOI: 10.16097/j.cnki.1009-6744.2025.05.016 Abstract ( )   PDF (3658KB) ( )   In order to quantitatively identify the motor vehicle conflict risk points in urban multi-vehicle linked turnaround areas and reduce the accident rate, this paper proposes a conflict risk identification process for multi-vehicle linked turnaround areas: firstly, continuous and high-precision multi-vehicle trajectory video is acquired by UAV aerial photography, and the displacement, speed and other states of each vehicle are tracked and extracted at frame level with the help of Tracker software; then, the risk quantification idea based on the time-to-collision (TTC) risk quantification idea, improve the traditional TTC algorithm (ETTC) for the geometric characteristics of the U-turn area, and draw cumulative distribution curves based on the collected ETTC and TTC data, from which the key tertiles representing minor, general and serious conflicts are selected as the thresholds; finally, spatially map the conflict events to the twenty subsections in the study area, and combine the frequency and severity of conflicts in the various sections to determine the risk quantification of each vehicle. Finally, the conflict events were spatially mapped to the twenty subsectors of the study area, and the risk level of each subsector was graded by combining the frequency and severity of the conflict. It was found that the highest severity rate of 25.51% was found in zone 15 of horizontal conflicts, and the highest severity rate of 21.95% was found in zones 2 and 3 of vertical conflicts, and the high-risk areas were concentrated in the middle two lanes of the roadway and at the 4th parking space. The research results can provide a scientific basis for the traffic management department to optimize the traffic safety management of multi-vehicle linked U-turn area. References | Related Articles | Metrics

Select

Optimization of High-speed Rail Seat Allocation Considering Multi-dimensional Travel Demand of Passengers JING Yun, WANG Qi, WANG Siyuan, WU Mingze 2025, 25(5): 193-204.  DOI: 10.16097/j.cnki.1009-6744.2025.05.017 Abstract ( )   PDF (2287KB) ( )   Combined with the characteristics of the multi-dimensional travel demand from high-speed railway passengers in China, the key to the optimal allocation of transportation capacity and the improvement of high-speed rail operation efficiency and service quality is to study the scientific and reasonable seat allocation method. In order to solve the problem of seat allocation in the high-speed railway (HSR) network, based on the time-space network that integrates passenger preferences, this paper constructs a two-level planning seat allocation model to distinguish passenger categories under elastic demand. The model was re-represented into a single-layer mixed integer linear programming model (MILP) by using the model transformation, large M method and piecewise linear method. The GUROBI was used to solve the problem. The results of the Beijing-Shanghai high-speed railway case show that under the same potential demand, the flexible demand seat allocation scheme considering passenger classification increases the actual passenger demand by 1.97% and the seat revenue by 4.49% compared with the seat sharing scheme. Compared with the scheme of non-passenger classification, the demand of actual passenger increased by 1.16% and the seat revenue increased by 2.27%. It indicates that the seat allocation model considering passenger classification under flexible demand can provide a reference for high-speed rail enterprises to optimize the seat allocation structure, attract potential passengers, and obtain more seat revenue under the condition of limited resources. References | Related Articles | Metrics

Select

Equilibrium Analysis of Bi-mode Morning Commute Considering Modular Bus Supply Constraints LI Chuanyao, ZHANG Mengdan, TANG Tieqiao 2025, 25(5): 205-214.  DOI: 10.16097/j.cnki.1009-6744.2025.05.018 Abstract ( )   PDF (1953KB) ( )   The development of autonomous driving technology has potentials to alleviate urban traffic congestion. However, its induced surge in travel demand may exacerbate the bottleneck of per-vehicle transport efficiency. Modular bus breaks through per-vehicle capacity limitations via dynamic fleet reconfiguration, forming collaborative capacity synergy with autonomous private cars to address supply-demand imbalance during morning commute peak. This paper develops a bi-mode bottleneck model integrating modular buses and private cars in autonomous driving context, derives equilibrium solutions under bus supply constraints, and investigates supply strategies targeting both travel cost minimization and operator revenue maximization. Results indicate that: increasing initial supply or supply rate can partially alleviate demand-supply imbalances when modular buses are undersupplied, the total travel cost will still rise. There exist two thresholds of supply rate that nullify the marginal contributions of initial and final supply, and the key to revenue optimization lies in dynamically coordinating the combination of initial and final supply. Initial supply and supply rate directly influence queue lengths during mode-switching phases, thereby affecting queuing costs and home waiting time for private car commuters. Increasing initial supply under high supply rate effectively reduces commuting delays. This paper integrates modular bus supply constraints into bi-mode traffic equilibrium framework, proposing synergistic supply strategies that establish theoretical foundations for morning-peak congestion mitigation and dynamic fleet optimization in single-bottleneck scenarios. References | Related Articles | Metrics

Select

Vehicle Yielding Behavior Analysis for Unsignalized Midblock Pedestrian Crosswalks CHEN Wenqiang, XUE Panpan, WANG Tao, GU Yulei 2025, 25(5): 215-225.  DOI: 10.16097/j.cnki.1009-6744.2025.05.019 Abstract ( )   PDF (2337KB) ( )   Unsignalized midblock crosswalks are high-risk zones for traffic accidents, with vehicle yielding behavior serving as a core factor in balancing pedestrian safety and traffic efficiency. Considering low vehicle yielding rates and oversimplified violation criteria at such crosswalks, this paper proposes a heterogeneous vehicle yielding behavior model and a decision-space partitioning framework. The study collect 1511 vehicle-pedestrian interaction events at a midblock crosswalk on Yanta Road in Xi'an city and analyzed them using kinetic modeling and Logistic regression. The results show that: (1) Heterogeneous yielding behaviors exist across vehicle types, buses exhibit the highest yielding rates, taxis demonstrate the highest approach speeds driven by economic incentives, and private vehicles show the lowest yielding rates. (2) Velocity and distance are identified as core kinetic variables, a 1 m·s-1 increase in vehicle speed reduces yielding probability by 18.6%, while a 1 m extension in longitudinal distance increases yielding probability by 13.7%. Pedestrian speed shows a disproportionate effect, with a 1 m·s-1 rise significantly elevating yielding probability by 317%. (3) Social-environmental factors drive behavioral compliance, parallel vehicles trigger a "normative effect", increasing yielding probability by 146%. During low-traffic periods, violation risks escalate, necessitating warning facilities to compensate for regulatory gaps. This study provides theoretical support for differentiated law enforcement, vehicle-specific control strategies, and infrastructure optimization. References | Related Articles | Metrics

Select

Demand Forecasting and Allocation for High-speed Rail Express Network Development YANG Juhua, TANG Jia, TIAN Zhiqiang, SONG Qi 2025, 25(5): 226-235.  DOI: 10.16097/j.cnki.1009-6744.2025.05.020 Abstract ( )   PDF (2742KB) ( )   With the rapid development of China’s express delivery industry and the strategic integration of high-speed rail freight, accurately forecasting high-speed rail express transport demand and optimizing its spatial allocation have become critical. This study uses a comprehensive approach integrating network clustering, hybrid forecasting, and the gravity model for the demand forecast and allocation. Considering regional differences in economic and demographic indicators, this paper established a hierarchical high-speed rail express transport network through cluster analysis of 50 major cities. A self-adaptive weight allocation algorithm, evaluated using the root mean square error (RMSE) metric, is designed to integrate grey prediction, ARIMA, and multiple regression models for forecasting high-speed rail express transport volumes. To enhance prediction accuracy, spatial proximity is incorporated to adjust model weights. Furthermore, an improved Logit model is developed to quantify the market share of high-speed rail express across different transport distances. By comprehensively considering impedance factors such as Gross Domestic Product (GDP), total retail sales of consumer goods, and network distance, this paper develops a doubly constrained gravity model, incorporating a mixed-effects model to analyze the impact of regional economic heterogeneity on networked express volume. This approach enables the Origin-Destination (OD) express shipment matrix within the high-speed rail express transport network to better reflect real-world conditions, providing a reliable reference for high-speed rail express transport network planning. References | Related Articles | Metrics

Select

Anomaly Detection and Incremental Clustering Method for High-speed Train Delay Data SHEN Pengju, SONG Liying, LIN Chuanqian 2025, 25(5): 236-247.  DOI: 10.16097/j.cnki.1009-6744.2025.05.021 Abstract ( )   PDF (2709KB) ( )   To improve the real-time anomaly detection and the dynamic evolution of clustering structures in high-speed train operation data, this paper proposes a Posterior Classification-based Incremental Dirichlet Process Mixture Model (PC-IDPMM) for incremental clustering and anomaly detection. The method adopts a two-stage framework: in the offline phase, a clustering model is developed and anomalous samples are identified; in the online phase, new samples are rapidly classified using posterior probabilities, and density-based clustering is applied to extract potential new structures, enabling structural expansion and parameter updating. To validate model performance, experiments are conducted using real-world data from the Guangzhou-Shenzhen high-speed railway. Results show that PC-IDPMM maintains cluster consistency while achieving stable updates of statistical features, with an AUC (Area Under the Curve) of 90.55%, outperforming several offline methods. Compared to offline models, the training time and memory usage are reduced by approximately 85% and 80%, respectively. The PC-IDPMM enables real-time anomaly warnings based on upstream station data before a train completes its route, supporting early-stage dispatch interventions that reduce cumulative delay from 572 minutes to 320 minutes. These results demonstrate the model’s real-time capability and practical value in high-frequency data environments. References | Related Articles | Metrics

Select

Evaluating Public Welfare Level of Time-sharing Pricing Strategy for Urban Rail Transit MENG Ran, YANG Yang, SONG Zilong 2025, 25(5): 248-260.  DOI: 10.16097/j.cnki.1009-6744.2025.05.022 Abstract ( )   PDF (3205KB) ( )   This paper investigates the urban rail transit fare scheme and public welfare level of the time-sharing pricing strategy. Considering the interaction between urban rail transit fare and passenger flow, the study defines the fare rate of each time period as the decision variable and develops a two-level programming model with the goal of minimizing passengers total travel time of based on transportation mode and passenger travel choice behavior. A solution algorithm is designed based on a hybrid multi-population genetic algorithm with nested Method of Successive Averages. The field data was used to calibrate the key parameters in the model and analyze the fare rates and public welfare levels of different time-sharing pricing schemes. A time-sharing pricing scheme is then proposed based on public welfare. The results show that the schemes that can significantly improve the public welfare level of urban rail transit system include differential pricing (the lower and upper limits of peak and off-peak fares are the same) and price reduction strategy during the off-peak period. The off-peak and peak fares for the former are 0.1 and 0.2 yuan person-1⋅km-1, which reduces the total generalized cost and total travel time by 12.75% and 8.23% respectively compared with the traditional fare scheme. The scheme that significantly reduces the public welfare level is the strategy that increases price during peak hours. References | Related Articles | Metrics

Select

Multi-task Cooperative Prediction of Short-term Origin-Destination Flows in Urban Rail Transit Networks YANG Jing, HOU Yuqing, XIE Yuchen, YANG Anan 2025, 25(5): 261-270.  DOI: 10.16097/j.cnki.1009-6744.2025.05.023 Abstract ( )   PDF (2159KB) ( )   Aiming to address the insufficient modeling of the correlation between OD flows and station-level inbound and outbound flows, this paper proposes a spatio-temporal graph neural network model (MSTGN) incorporating multi-task learning (MTL) to overcome the limitation of representation capability while using the traditional GNNs in existing short-term OD prediction methods. MSTGN takes the prediction of OD flow as the main task and the prediction of station passenger flow as the auxiliary task, leveraging a message-passing graph neural network (MPNN) to build an edge-node collaborative spatio-temporal propagation module that models OD flows (edge features) and station flows (node features) separately, thus enabling efficient characterization of complex spatio-temporal coupling patterns. Moreover, a task consistency constraint and a dynamic weighting mechanism are introduced to enhance a multi-task collaboration. The comparative experiments and ablation studies conducted on empirical datasets from the subway systems in Beijing and Hangzhou demonstrate that MSTGN achieves an improvement of approximately 5.5%~6.0% in MAE, RMSE, and WMAPE compared with the suboptimal model HIAM, verifying the cross-city generalization ability of this model. The MPNN module shows the most significant contribution, with a performance degradation of 6.12% upon ablation. These results indicate that MSTGN offers notable advantages in improving the accuracy and system adaptability of short-term OD prediction, providing a strong technical support for intelligent scheduling and resource optimization in urban rail transit systems. References | Related Articles | Metrics

Select

Optimization of Age-Friendly Customized Bus Route Services TANG Chunyan, DAI Xinyi, ZHANG Jiyu 2025, 25(5): 271-279.  DOI: 10.16097/j.cnki.1009-6744.2025.05.024 Abstract ( )   PDF (1949KB) ( )   With the acceleration of population aging, the elderly are increasingly facing problems such as transportation facilities that do not meet their needs and difficulties in traveling. Customized buses, as a demand-driven public transportation service mode, provide a new solution for age-friendly travel. However, existing customized bus service modes are primarily designed for commuters and cannot adequately address the diverse travel needs of the elderly in terms of social equity. This paper proposes an age-friendly customized bus service mode tailored to the travel needs of the elderly. To enhance the elderly passengers' experience and improve the effectiveness of barrier-free services, the proposed bus service mode incorporates a differentiated boarding and alighting time strategy. Additionally, a dynamic heterogeneous seating configuration strategy is introduced to ensure that different types of passengers are allocated more appropriate seating resources and achieve greater comfort. An optimization model for age friendly customized bus routes is developed, with the goal of minimizing both the vehicle operating costs and the passengers' travel costs. A genetic algorithm is designed to solve this optimization problem and determine the routing plan and vehicle-passenger assignment scheme. The proposed method is validated and analyzed using a large-scale case study from Fengtai District, Beijing. The results demonstrate that, compared to traditional fixed heterogeneous seating configurations, the proposed service mode of dynamic heterogeneous seating configuration significantly reduces the total system cost while effectively shortening the average travel time for elderly passengers. This would help to improve the travel experience of elderly passengers, especially those who need time efficiency in their travel. References | Related Articles | Metrics

Select

Optimization Algorithm for Ride-hailing Dispatch Under Concentrated Supply and Sparse Demand WANG Jiangfeng, SONG Zhifan, LI Yunfei, QI Chongkai, YAN Xuedong, LI Qingshan 2025, 25(5): 280-290.  DOI: 10.16097/j.cnki.1009-6744.2025.05.025 Abstract ( )   PDF (2451KB) ( )   This study addresses the structural imbalance of “concentrated supply and sparse demand” in ride-hailing services across emerging urban areas by introducing incentive strategies to optimize dispatch decisions and alleviate spatial mismatches between supply and demand. A reinforcement learning-based dispatch optimization algorithm is proposed, integrating incentive mechanisms into four key modules: environment construction, feasible decision selection, incentive embedding, and policy optimization. Specifically, a Markov Decision Process-based environment is developed to model tasks such as order matching, idle vehicle repositioning, and charging management. A joint incentive strategy combining direct financial incentives (via reduced platform commission) and indirect regulatory adjustments (via prioritized dispatch guarantees) is designed, and an Actor-Critic algorithm is used to maximize platform revenue. Empirical analysis using data from Xiong’an New Area in Hebei Province demonstrates that the proposed method significantly improves the service rate of sparse-demand orders with sparse demand. Under fixed fleet size, the service rates of sparse-demand orders respectively increased by 24.70%, 2.53%, and 26.09% through commission reduction, prioritized dispatch, and joint incentives, while maintaining or increasing overall service rate and platform revenue. As the ride-hailing vehicle fleet expands, the sparse-demand and overall order fulfillment rates reach 61.20% and 81.55% respectively, when the fleet size reaches 120 vehicles. Sensitivity analysis further indicates that a 24% commission rate and a 4-period dispatch guarantee yield a favorable balance between platform revenue and global service performance. References | Related Articles | Metrics

Select

Charging Station Location and Capacity Determination Algorithm Based on Taxi Trajectory Data ZHANG Wenhui, QIAO Zifan, CHEN Deqi 2025, 25(5): 291-301.  DOI: 10.16097/j.cnki.1009-6744.2025.05.026 Abstract ( )   PDF (3151KB) ( )   To address the imbalance between the supply and demand of electric vehicle (EV) charging facilities, this study integrates both dynamic and static trajectory data of taxis, and employs an energy consumption estimation model to establish a state-of-charge (SOC) estimation method. By analyzing trajectory points containing SOC and dwell time information, the potential charging demand is identified. A multi-criteria decision-making function is constructed based on the Pareto optimality principle to recognize short-term dwell points with significant charging demand. The density-based spatial clustering algorithm (DBSCAN) is applied for unsupervised spatial analysis, using the resulting cluster centers as candidate locations for charging stations. A queuing model is used to simulate the EV charging process and determine the required number of charging piles. Subsequently, a multi-objective optimization model for charging station siting is formulated, and a hybrid solution strategy is designed combining the Ant Colony Optimization (ACO) and Genetic Algorithm (GA). Dual heuristic functions are introduced by incorporating Point of Interest (POI) weights and land calibration prices. Using taxi trajectory data from Shenzhen city as a case study, the optimal plan yields 40 charging stations with a total social cost of 35.394 million yuan. The research results provide a theoretical basis for optimizing the siting and capacity planning of urban EV charging stations. References | Related Articles | Metrics

Select

Aviation Multi-stage Operational Risk Evolution Under Heterogeneous Risk Propagation Mechanisms LIU Zhaoxuan, ZHOU Zhexu, FANG Jing 2025, 25(5): 302-311.  DOI: 10.16097/j.cnki.1009-6744.2025.05.027 Abstract ( )   PDF (3061KB) ( )   Aircraft operational risk analysis is essential for civil aviation safety guarantee. Current studies mainly focus on homogeneous risk propagation during one operational phase, while research on heterogeneous risk propagation across multiple phases is limited. This paper investigates both the cruise and approach/departure phase by constructing directed risk networks covering: personnel, equipment, environment, and management. Typical node influence metrics are introduced including node degree, closeness centrality, betweenness centrality, and PageRank. Besides, three models, i.e., probabilistic, robust and Susceptible Infected Recovered (SIR) are employed to characterize heterogeneous risk propagation. Results indicate that critical nodes such as abnormal flight states, runway excursion, near miss, and emergency descent are of paramount importance in the multi-phase operation risk network. Among the three propagation models, node degree-based identification exhibits the most significant risk diffusion effects under both the probabilistic and SIR models, whereas PageRank has the highest accuracy under the robust propagation model. The parameter sensitivity analysis reveals that nodes with both high transmission and diffusion parameters represent key vulnerabilities. Targeted redundancy measures, such as capacity enhancement, can significantly improve network safety. Notably, reinforcing intermediate nodes proves to be more effective than reinforcing initial nodes. References | Related Articles | Metrics

Select

Mixed-skilled Task Assignment in Aviation Considering Ground Staff Satisfaction ZENG Peibin, WU Weitiao, KE Weipeng, WANG Fa 2025, 25(5): 312-319.  DOI: 10.16097/j.cnki.1009-6744.2025.05.028 Abstract ( )   PDF (1843KB) ( )   The job satisfaction of airport ground staff is crucial for enhancing work efficiency. To addressing the issues of ground staff shortages and overlooked satisfaction, this paper proposes a multi-objective mixed integer programming model for mixed-skilled task assignment incorporating ground staff satisfaction. While ensuring task completion rates, this study establishes a multidimensional satisfaction index system that considers factors such as workload balance, adequate meal breaks, and frequency of simultaneous multi-aircraft operations. A multi-objective mixed-integer programming model is developed and solved using the solver Gurobi 12.0. Through case studies using typical data from Xiamen Gaoqi International Airport on peak workdays, the results show that compared to existing scheduling methods, the proposed approach significantly improves the work experience of ground staff. In terms of fairness, the deviation in average work hours per person is reduced from 68.5 minutes to 27.4 minutes, a 60% reduction. For comfortable meal times, the proportion of comfortable meal time increases from 71.2% to 98.5%. Through the optimization of the dual aircraft coordination mechanism, the average number of ground staff required per 100 audit tasks increased from 87 to 97, effectively easing the work pressure of ground staff. The proposed method can effectively alleviate the contradiction between operational efficiency and employee satisfaction, providing a decision framework for airline ground staff management considering both the completion rate of audit tasks and the sustainability of human resources. References | Related Articles | Metrics

Select

Route Planning Relief Delivery and Victim Evacuation Based on Rescue Utility in Flood Disaster Scenarios LIU Changshi, WAN Cheng, WANG Feng, CHEN Baoxi, YUE Junyu 2025, 25(5): 320-332.  DOI: 10.16097/j.cnki.1009-6744.2025.05.029 Abstract ( )   PDF (2095KB) ( )   The key components of flood rescue are the relief delivery and the evacuation of stranded residents in disaster-stricken areas. To scientifically assess the actual effectiveness of rescue efforts based on the arrival times of transport vehicles at demand points, this paper introduces the concept of rescue utility into the joint routing of relief delivery and victim evacuation in flood disaster scenarios. A collaborative routing model is developed with the dual objectives of minimizing the total rescue time and maximizing the total rescue utility of all affected locations. The model comprehensively considers factors such as the differing applicability of trucks and assault boats, the expected rescue times of disaster-stricken sites, as well as the speed, quantity, and capacity of the available transport vehicles. An improved Non-dominated Sorting Genetic Algorithm-II is designed to solve the proposed model. Numerical experiments are conducted based on the data from the 2024 flood disaster in Nanchang, Jiangxi Province of China. The results demonstrate that the proposed approach enables effective joint routing of relief delivery and victim evacuation. Compared to strategies that prioritize evacuating residents before delivering emergency supplies or delivering emergency supplies before evacuating residents, the proposed rescue strategy respectively reduces total rescue time by 138 minutes and 123 minutes and increases total rescue utility respectively by 108.2 units and 77.5 units. References | Related Articles | Metrics

Select

Online Optimization Methods for Dynamic Lane Configuration at Highway Toll Plazas MA Feihu, CHEN Xiaoyan, SUN Cuiyu, TIAN Xingtong 2025, 25(5): 333-342.  DOI: 10.16097/j.cnki.1009-6744.2025.05.030 Abstract ( )   PDF (2380KB) ( )   This study addresses the issue of lane configuration optimization at highway toll plazas by proposing a dynamic lane configuration strategy based on reinforcement learning. A simulation environment for highway toll plazas is constructed based on the traffic behavior of vehicles passing through a toll plaza. The complex problem of lane configuration is transformed into a clear quantitative objective function that considers the operational costs of toll plazas, user delay, and congestion penalties. The lane resource allocation strategy of toll plazas is optimized dynamically through the training of a reinforcement learning network. The model is capable of real-time learning and dynamic adjustment of lane configurations to adapt to the dynamic changes in traffic flow and patterns. Experiments comparing the reinforcement learning optimization method with traditional offline optimization methods show that the proximal policy optimization (PPO) method reduces the average queue length by 12.45% and narrows the fluctuation range of average travel time by 26.94%. The PPO algorithm demonstrates advantages in reducing queue length and decreasing the fluctuation in travel time, especially during peak hours. The dynamic lane configuration strategy exhibits higher adaptability and flexibility, enhancing the operational efficiency of toll plazas. References | Related Articles | Metrics

Select

Evaluation and Diagnostic Method for Effectiveness of Guidance Signs in Comprehensive Passenger Hubs Based on Wayfinding Experiments WENG Jiancheng, WANG Yihao, CHEN Xurui, GAO Runhong, HU Song 2025, 25(5): 343-352.  DOI: 10.16097/j.cnki.1009-6744.2025.05.031 Abstract ( )   PDF (2172KB) ( )   The guidance sign system in a comprehensive passenger hub plays a vital role in directing arriving passengers to connection points for various transfer modes. Given the wide range of information categories, significant variation in user demands, and the complexity of transfer routes, the effectiveness of guidance signs is important to support efficient and convenient passenger transfers and enhance the service quality of the hub. This study investigates the relationship between guidance signs and passengers’ routing decisions based on their wayfinding behaviors. The study proposes an index system to evaluate the effectiveness of guidance signs, and develops a three-dimensional virtual simulation platform for the comprehensive passenger hub based on the Unity engine and virtual reality (VR) technology. The case study is performed in Beijing South Railway Station with different wayfinding scenarios. Two key indicators: optimal path ratio and wayfinding speed are included. Combined with behavioral metrics such as the number of hesitation pauses, backtracking instances, and incorrect turns, the study conducts a quantitative evaluation and diagnostic analysis of the effectiveness of guidance signs. The results show that the optimal path ratio exceeds 85% in simple wayfinding tasks, and in complex tasks the optimal path ratio is below 50%. Passengers tend to exhibit more hesitation, backtracking, and errors in areas with dense information and at decision points, allowing for precise identification of locations within the hub where guidance signs require optimization. Although the current guidance sign system, designed in accordance with visual recognition standards, meets basic visibility requirements, its effectiveness varies significantly for different wayfinding tasks. This study provides a quantitative diagnostic approach and targeted recommendations for the optimization of guidance sign design in comprehensive passenger hubs. References | Related Articles | Metrics

Select

Assessment of Pollution and Carbon Reduction Potential of Medium-and Heavy-Duty Battery Electric Trucks JIANG Zhijuan, CHEN Meiling, WEI Xintian, HE Jingwen 2025, 25(5): 353-364.  DOI: 10.16097/j.cnki.1009-6744.2025.05.032 Abstract ( )   PDF (2983KB) ( )   The electrification of freight trucks is considered as an important measure for reducing emissions from road freight transport. However, the emission reduction potential of battery electric trucks (BETs) varies significantly depending on the electricity-generation method, weight class, and technical standard. A granular analysis of the electricity mix and vehicle parameters is essential for refining emission-reduction policies. This study evaluated the air-pollutant emissions of BETs and conventional diesel internal combustion engine trucks (diesel-ICETs) in China across their fuel-cycles to explore the potential of using BETs to reduce pollution and carbon emissions. Data on the crude oil mix, electricity mix, and truck technology of China were used in a life-cycle assessment (LCA) to analyze emission factors related to the energy production pathways and vehicle operations. The greenhouse gases, regulated emissions, and energy use in transportation (GREET) model was applied to assess the vehicle emission intensity (g·km-1) and freight emission intensity (g·t-1·km-1) for BETs and diesel-ICETs (China VI) of different weight classes. The results showed that replacing diesel-ICETs with BETs significantly reduces vehicle emission intensities for six pollutants: greenhouse gases (GHGs), volatile organic compounds (VOCs), CO, NOx , PM2.5 , and PM10 , as well as the freight emission intensities for four pollutants: GHG, CO, NOx , and VOCs. However, due to electricity mix of China and technical limitations of BETs, the adoption of BETs increases SOx emissions and exacerbates PM2.5 and PM10 emissions in road freight. References | Related Articles | Metrics

Select

Emission Reduction Potential and Multi-path Synergy of Sustainable Aviation Fuel in China's Civil Aviation TIAN Lijun, CHEN Xuegong, WANG Qi, XU Xinzhe, WANG Yating, QU Xixi 2025, 25(5): 365-374.  DOI: 10.16097/j.cnki.1009-6744.2025.05.033 Abstract ( )   PDF (2031KB) ( )   To address the contradiction between the rapid growth of China's civil aviation and deep decarbonization, the single path of sustainable aviation fuel (SAF) is confronted with the predicament of technical ceiling and offset by demand growth. This study constructs the FLEET model integrating mixed integer programming (MIP) and system dynamics (SD), establishes a two-way feedback mechanism of "micro-operation-macro-policy", and quantifies the synergistic emission reduction effect of multiple paths (SAF, new technology aircraft, and operation optimization). The core findings include: (1) Potential and limitations of SAF: When the SAF blending ratio is in the range of 0~30%, every 10% increase can reduce emissions by 12.7%; after exceeding 30%, the marginal benefit decreases significantly, which is restricted by raw material gap and PIL technical bottlenecks. (2) Necessity of multi-path synergy: under the mandatory blending scenario (SAF 50% + GDP annual growth rate 5.5%), carbon emissions in 2050 would still increase by 73% compared with 2019, and the increase is affected by the fluctuations in transport volume (±0.5%) and the actual blending efficiency of SAF (±5%). The optimal combination scenario (SAF 65% + new technology aircraft 40% + operation optimization 30% + market mechanism 15%) can achieve a 49.19% emission reduction in 2050. The emission reduction per unit policy incentive is 2.3 times of the single SAF path, forming technical complementarity, cost synergy, and emission reduction multiplier effect. (3) Carbon quota price threshold and heterogeneity: The carbon quota price exceeding 200 yuan per ton is the key threshold to trigger the technological leap of airlines; there are subject and regional heterogeneities in response (the emission reduction efficiency of eastern hubs is 1.35 tons per 10000 yuan larger than 0.87 tons per 10000 yuan of western branches). It is revealed that China's civil aviation emission reduction needs to break through the dependence on a single technology, and a three-stage "technology-raw material-policy" synergy roadmap is proposed (2025-2035 HEFA scale; 2035-2045 raw material diversification; 2045-2050 policy deepening), which provides a quantitative basis for phased policy design and engineering application. References | Related Articles | Metrics