<rss version="2.0" xmlns:atom="https://www.w3.org/2005/Atom">
  <channel>
    <title>Transport Research International Documentation (TRID)</title>
    <link>https://trid.trb.org/</link>
    <atom:link href="https://trid.trb.org/Record/RSS?s=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" rel="self" type="application/rss+xml" />
    <description></description>
    <language>en-us</language>
    <copyright>Copyright © 2026. National Academy of Sciences. All rights reserved.</copyright>
    <docs>http://blogs.law.harvard.edu/tech/rss</docs>
    <managingEditor>tris-trb@nas.edu (Bill McLeod)</managingEditor>
    <webMaster>tris-trb@nas.edu (Bill McLeod)</webMaster>
    <image>
      <title>Transport Research International Documentation (TRID)</title>
      <url>https://trid.trb.org/Images/PageHeader-wTitle.jpg</url>
      <link>https://trid.trb.org/</link>
    </image>
    <item>
      <title>Energy-saving optimization of train speed Profile based on online Policy deep reinforcement learning algorithm under emergency conditions</title>
      <link>https://trid.trb.org/View/2687105</link>
      <description><![CDATA[In the domain of energy-efficient train-speed-profile optimization, off-policy reinforcement-learning (RL) algorithms and their variants have long dominated the academic landscape. Such approaches typically rely on an experience-replay buffer to reuse historical samples, yet they suffer from low training efficiency, poor numerical stability, and insufficient real-time responsiveness—limitations that become critical in complex metro systems or under emergency conditions. To overcome these drawbacks, we propose an on-policy RL framework termed the Energy-Saving Maximization Advantage Distribution (ES-MAD) algorithm. ES-MAD abandons the replay buffer and instead employs dual policy networks (new vs. old): data collected by the newly explored policy are immediately leveraged to update the old policy, thereby markedly improving sample efficiency. In addition, a normalized-and-scaled advantage function is introduced to reduce variance, which significantly enhances model stability in highly dynamic or emergency scenarios. Finally, a multi-objective, self-adaptive reward mechanism is designed that seamlessly integrates energy efficiency, punctuality, ride comfort, and stopping accuracy; this reward structure requires no re-tuning when the system switches between normal and emergency operating modes. Experimental validation on the Yizhuang metro line demonstrates that ES-MAD exhibits strong robustness and generalization capability under regular service, sudden schedule changes, and multi-interval comprehensive tests. Compared with state-of-the-art off-policy deep RL baselines, ES-MAD achieves 24 % and 26 % additional energy-saving under normal and emergency conditions, respectively.]]></description>
      <pubDate>Fri, 31 Jul 2026 09:23:49 GMT</pubDate>
      <guid>https://trid.trb.org/View/2687105</guid>
    </item>
    <item>
      <title>Dynamic Evolution of Metro Network Vulnerability: Shifting Mechanisms Under Cascading Failures</title>
      <link>https://trid.trb.org/View/2692956</link>
      <description><![CDATA[The rapid expansion of metro networks introduces evolving systemic vulnerabilities that are poorly captured by static, snapshot-based analyses. This study proposes an integrated framework to dynamically assess vulnerability in the Beijing Metro over a 21-year period. We first employ unsupervised learning to objectively partition the network’s evolution into Formation and Densification stages. A multi-scale critical station identification model is then applied within each stage, integrating micro, meso, and macro indicators for comprehensive ranking. Cascading failure propagation is simulated via a heterogeneous agent-based model that captures passenger behavioral diversity. Our results reveal a fundamental shift in failure mechanisms: from topology-driven “structural fragmentation” in the early Formation Stage to passenger flow-dominated “overload cascades” in the later Densification Stage. This transition significantly amplifies systemic risk, with the comprehensive vulnerability integral leaping from 0.467 to 0.769 and the cascade amplification factor intensifying from 0.861 to 1.214. These findings underscore a critical “efficiency-vulnerability trade-off” in metro network development. Our work provides a mechanistic understanding of vulnerability evolution and offers practical tools for enhancing the resilience and safety of urban rail transit systems.]]></description>
      <pubDate>Fri, 31 Jul 2026 09:23:48 GMT</pubDate>
      <guid>https://trid.trb.org/View/2692956</guid>
    </item>
    <item>
      <title>Human-like hierarchical decision-making for real-time train regulation strategies in urban rail transit lines</title>
      <link>https://trid.trb.org/View/2692657</link>
      <description><![CDATA[Efficient real-time train regulation strategies are vital for ensuring reliable service quality and resilience in urban rail transit (URT) systems. Existing automated train regulation methods often face limitations in dealing flexibly with complex, unpredictable operational disturbances, whereas purely manual dispatching approaches struggle with scalability and consistency. This paper proposes a novel human-like, multi-level train regulation strategy that integrates the cognitive adaptability of human dispatchers with model-based rolling-horizon optimization. A hierarchical dispatching framework, inspired by human decision-making processes, dynamically adjusts regulation strategies based on real-time train delays and passenger congestion conditions. Furthermore, to meet the real-time requirements of train rescheduling in the operational phase, a tightening McCormick relaxation algorithm is developed to efficiently solve the associated train regulation problems within the rolling horizon framework. Numerical experiments based on actual operational data from the Beijing Yizhuang metro line illustrate the proposed strategy’s superior adaptability and robustness compared with traditional rule-based dispatching and fully automated single-level optimization strategies, especially under mixed disturbance scenarios. These findings highlight the potential of human-like collaborative frameworks in enhancing operational robustness and decision-making efficiency within intelligent transportation systems.]]></description>
      <pubDate>Tue, 21 Jul 2026 09:50:45 GMT</pubDate>
      <guid>https://trid.trb.org/View/2692657</guid>
    </item>
    <item>
      <title>Joint Optimization of Passenger Flow Control and Train Skip-Stopping for Overcrowded Metro Lines: A Multi-Agent Reinforcement Learning Approach</title>
      <link>https://trid.trb.org/View/2663055</link>
      <description><![CDATA[During rush hours, the capacity of metro in megacities is insufficient to meet the travel demand, resulting in oversaturation and high risk on platform in stations, especially transfer stations. This paper addresses this problem through the joint optimization of some operational interventions, aiming to alleviate passenger overloads while maintaining travel efficiency. To make the model more realistic, the stochastic characteristics of passengers are considered, including the probability distribution of passenger arrival time, inbound and transfer walking times. To provide a high-quality solution for the complex constraint model, three cooperative agents—governing passenger inflow, transfer flows, and train skip-stopping mode—are architected within improved Double Deep Q learning Network (IDDQN) to form a multi-agent reinforcement learning solution. Empirical validation on Beijing Metro Line 13 and Changping Line demonstrates that the multi-agent framework proposed in this paper can eliminate 100% of passenger over-limit flow while reducing the average waiting time of passengers. It also has a significant improvement in reducing stochastic characteristic impact and accelerating convergence.]]></description>
      <pubDate>Thu, 09 Jul 2026 13:32:10 GMT</pubDate>
      <guid>https://trid.trb.org/View/2663055</guid>
    </item>
    <item>
      <title>Responsible AI-Driven Timetable Optimization: A Circular Economy Framework for Energy-Regenerative Rail Transit</title>
      <link>https://trid.trb.org/View/2713420</link>
      <description><![CDATA[Under the rapid expansion of urban rail transit (URT) as the backbone of sustainable urban mobility, this study advances circular-economy practices through the responsible deployment of artificial intelligence. We present the Train Timetable Energy-Saving Deep Reinforcement Learning (TES-DRL) framework—a two-stage, progressive AI architecture that deliberately minimizes computational and environmental overheads. In Stage 1, a heuristic scheduler generates a baseline timetable, drastically shrinking the subsequent decision space. Stage 2 reframes timetable optimization as a Markov Decision Process in which a centralized-training, centralized-execution deep reinforcement learning agent pursues dual sustainability objectives: (i) a reward function that explicitly minimizes net energy consumption and (ii) an energy-recovery-maximization routine that transforms regenerative braking into reusable system resources. Robust punctuality constraints are embedded to preserve service reliability, foregrounding the ethical dimension of technology deployment. Empirical validation on real-world operational data from Beijing’s Yizhuang Line demonstrates that TES-DRL reduces overall energy use by 10 %, translating directly into a measurable contraction of the rail system’s carbon footprint. The resulting AI-generated energy-saving strategies constitute reusable digital assets that accelerate the transition toward circular transportation systems. The centralized-training design guarantees scalability, offering a transferable blueprint for emission-reduction across global supply chains. Our findings substantiate that responsible technological innovation—exemplified by AI-driven, precision timing control—can systematically steer sustainable manufacturing transformations while establishing transport ecosystems with minimized environmental footprints.]]></description>
      <pubDate>Thu, 18 Jun 2026 09:09:29 GMT</pubDate>
      <guid>https://trid.trb.org/View/2713420</guid>
    </item>
    <item>
      <title>Multi-objective parallel optimization of train energy-saving curves considering different passenger loads</title>
      <link>https://trid.trb.org/View/2669702</link>
      <description><![CDATA[This study investigates the optimization of train trajectories between stations under variable passenger loads to enhance the energy efficiency of inter-station travel. First, a bi-objective optimization model is formulated to simultaneously minimize inter-station running time and energy consumption. The model incorporates passenger load variations and is constructed using a time-step discretization approach integrated within a working regime sequence. Second, a solution framework is developed by integrating a multi-objective optimization algorithm with a parallel computing architecture to significantly improve computational efficiency. Finally, a case study is conducted using operational data from the Beijing Yizhuang Line. Simulation results demonstrate significant improvements: the optimized trajectories achieve substantial energy savings ranging from 25.0% to 37.8% compared to actual operational data, highlighting the model’s practical effectiveness in reducing operational costs and environmental impact. Furthermore, the parallel computing architecture achieves an 18-fold average reduction in computational time, demonstrating its critical role in making computationally intensive multi-scenario optimization tasks feasible for practical implementation. Additionally, the optimization process yields a three-dimensional Pareto surface that elucidates the trade-offs among the passenger numbers, inter-station running time, and energy consumption.]]></description>
      <pubDate>Tue, 26 May 2026 09:41:00 GMT</pubDate>
      <guid>https://trid.trb.org/View/2669702</guid>
    </item>
    <item>
      <title>Risk Control of the Automatic Train Supervision System in Rail Transit Systems: Under a Van der Pol Equation-Based Framework</title>
      <link>https://trid.trb.org/View/2697864</link>
      <description><![CDATA[Automatic train supervision (ATS) systems are a core safety component in metro operations. Its redundant design results in extremely scarce failure data, rendering traditional data-driven risk analysis ineffective. Consequently, existing studies often substitute reliability analysis for risk analysis. To overcome the limitations of static and vague reliability methods, this study employs the van der Pol equation to dynamically quantify inherent risk oscillations in ATS systems, providing managers with actionable control measures. Our paper begins by analyzing ATS risk characteristics and examining the feasibility of using the van der Pol equation to model risk state changes. Then, we establish a risk state equation derived from this framework and analyze the system’s risk dynamics. Finally, to control risk, we integrate a risk control function into the equation. A case study of Beijing Metro Line 2 demonstrates the method’s applicability. The proposed methodology enables accurate risk state judgment, potential risk prediction, and precise control implementation. By applying differential equation theory, it reduces reliance on historical data or expert knowledge while addressing inaccuracies from missing critical data. This work establishes a novel framework for system risk control and offers practical guidance for operators.]]></description>
      <pubDate>Sat, 02 May 2026 15:47:06 GMT</pubDate>
      <guid>https://trid.trb.org/View/2697864</guid>
    </item>
    <item>
      <title>Safety Evaluation of Operational Metro Shield Tunnels Using Improved Game Theory and Dynamic Variable Weight Theory</title>
      <link>https://trid.trb.org/View/2660562</link>
      <description><![CDATA[The safety assessment of operational shield tunnels involves complexity, randomness, and uncertainty. Traditional constant weight methods fail to account for the dynamic changes in structural state caused by the interaction among different defects. Therefore, this study proposes a novel safety evaluation framework for operational tunnels by integrating game theory and variable weight theory. This evaluation model is applied to four tunnel sections of Beijing Metro Line 8 and is compared with three other evaluation models (including conventional weighting methods, fuzzy comprehensive evaluation, etc.). Sensitivity analysis identified U31, U32, U33, and U52 as the key indicators affecting the tunnel’s structural safety. On-site investigation results demonstrate that the proposed model provides more accurate evaluations, thereby verifying its feasibility. Furthermore, this study has also established an evaluation framework that is applicable to the comprehensive assessment of the entire tunnel section. Evaluating an entire tunnel section as a single unit may conceal local high-risk areas, leading to inaccurate assessment results. It facilitated managers taking safeguard measures in a timely manner based on the evaluation results and ensuring the safety and reliability of the tunnel structure.]]></description>
      <pubDate>Wed, 22 Apr 2026 14:04:04 GMT</pubDate>
      <guid>https://trid.trb.org/View/2660562</guid>
    </item>
    <item>
      <title>Train timetable optimization for urban railway systems under the virtual formation mode combined with the rolling stock utilization strategy</title>
      <link>https://trid.trb.org/View/2647870</link>
      <description><![CDATA[The distribution of passenger demands on certain urban railway lines exhibits obvious spatiotemporal imbalances, posing challenges for the traditional fixed formation mode. This paper presents the optimization of the virtual formation train timetable and rolling stock utilization strategy, which aims to maximize the quantity of connections and minimize the number of detained passengers. A mixed-integer nonlinear programming model (MINLP) is formulated to characterize this problem, in which the coupling/decoupling operations between different types of rolling stock are considered. By applying linearization techniques, the aforementioned MINLP model can be transformed into a mixed-integer linear programming (MILP) model. To effectively address the model, a two-stage (TS) optimization approach is designed to decompose the original problem into two sequential steps for the solution. In the first stage, a reduced-scale optimization problem is solved, focusing solely on a subset of services; then, the partial binary variables obtained from the first stage are incorporated into the original problem for further resolution in the second stage. Furthermore, we design an accelerated technique of bound contraction based on logical inference to enhance the solving efficiency of the second stage. Five sets of numerical experiments based on the Beijing metro Yizhuang line are conducted to verify the effectiveness and practicability of the model and algorithm. The experimental results illustrate that the virtual formation mode can effectively address the spatiotemporal imbalances of passenger demands on the line. The proposed TS approach is also proven to exhibit greater efficiency than traditional heuristic algorithms, such as genetic algorithm (GA), for large-scale problems.]]></description>
      <pubDate>Mon, 02 Feb 2026 09:32:47 GMT</pubDate>
      <guid>https://trid.trb.org/View/2647870</guid>
    </item>
    <item>
      <title>Metro train rescheduling approach during the transitional period</title>
      <link>https://trid.trb.org/View/2613687</link>
      <description><![CDATA[Metro systems will inevitably be affected by train delays. Train delays during the peak to off-peak transitional period potentially lead to passengers being stranded. To address this problem, this study develops a train rescheduling model that comprehensively considers train delays, service cancellations, and the passenger service quality. The rescheduling strategy of adding temporary train services is particularly considered based on operational characteristics during the transitional period. Next, the model is transformed into a mixed-integer linear programming (MILP) model using linearization methods. To improve computational efficiency, a two-stage approach is proposed. Finally, numerical examples based on practical data from the Beijing Metro Yizhuang line are conducted to verify the effectiveness and efficiency of the proposed approach. The computational results show that the model proposed in this study can reduce the number of stranded passengers during the transitional period, and the two-stage approach can obtain high-quality feasible solutions in a short time.]]></description>
      <pubDate>Mon, 26 Jan 2026 14:44:19 GMT</pubDate>
      <guid>https://trid.trb.org/View/2613687</guid>
    </item>
    <item>
      <title>Digital twin-based resilience analysis and emergency maintenance with generative AI of smart urban metro systems</title>
      <link>https://trid.trb.org/View/2587731</link>
      <description><![CDATA[With the rapid expansion and growing complexity of urban metro systems, operational disturbances increasingly threaten system stability and service continuity. However, existing recovery strategies remain insufficient in terms of emergency response efficiency and intelligent decision-making. To address these challenges, this study proposes a digital twin metro system architecture with generative AI enabling real-time system status monitoring, and gives a new resilience model to optimize the emergency maintenance decision-making.. Furthermore, the study investigates performance and cost variations across failure and recovery phases, and introduces a cost importance-based recovery strategy that enhances system resilience by optimizing the station recovery sequence. Finally, using the Beijing metro as a case study, the impact of different recovery strategies is evaluated under both single-line section and multi-line section attack scenarios. The results indicate that the cost importance-based strategy (CIBRS) outperforms traditional approaches (BBRS, DBRS, ECBRS, RGBRS). In the single-line section attack, it achieves a resilience value of 0.747, with improvements of 74.94%, 35.34%, 95.55%, and 1.91%, respectively. Under the multi-line section attack, the resilience value reaches 0.777, with corresponding improvements of 79.45%, 46.33%, 85.44%, and 5.00%. The result confirms the superior adaptability and robustness of the proposed method in complex scenarios. This study offers valuable insights for intelligent metro system management and emergency maintenance.]]></description>
      <pubDate>Fri, 26 Sep 2025 13:39:50 GMT</pubDate>
      <guid>https://trid.trb.org/View/2587731</guid>
    </item>
    <item>
      <title>Enhanced intelligent train operation algorithms for metro train based on expert system and deep reinforcement learning</title>
      <link>https://trid.trb.org/View/2563852</link>
      <description><![CDATA[In recent decades, automatic train operation (ATO) systems have been gradually adopted by many metro systems, primarily due to their cost-effectiveness and practicality. However, a critical examination reveals computational constraints, adaptability to unforeseen conditions and multi-objective balancing that the authors' research aims to address. In this paper, expert knowledge is combined with deep reinforcement learning algorithm (Proximal Policy Optimization, PPO) and two enhanced intelligent train operation algorithms (EITO) are proposed. The first algorithm, EITOₑ, is based on an expert system containing expert rules and a heuristic expert inference method. On the basis of EITOₑ, the authors propose EITOₚ algorithm using the PPO algorithm to optimize multiple objectives by designing reinforcement learning strategies, rewards, and value functions. The authors also develop the double minimal-time distribution (DMTD) calculation method in the EITO implementation to achieve longer coasting distances and further optimize the energy consumption. Compared with previous works, EITO enables the control of continuous train operation without reference to offline speed profiles and optimizes several key performance indicators online. Finally, the authors conducted comparative tests of the manual driving, intelligent driving algorithm (ITOR, STON), and the algorithms proposed in this paper, EITO, using real line data from the Yizhuang Line of Beijing Metro (YLBS). The test results show that the EITO outperform the current intelligent driving algorithms and manual driving in terms of energy consumption and passengers' comfort. In addition, the authors further validated the robustness of EITO by selecting some complex lines with speed limits, gradients and different running times for testing on the YLBS. Overall, the EITOₚ algorithm has the best performance.]]></description>
      <pubDate>Fri, 12 Sep 2025 13:38:37 GMT</pubDate>
      <guid>https://trid.trb.org/View/2563852</guid>
    </item>
    <item>
      <title>Mechanism of Asset Devaluation in the Planning Stage of Urban Rail Transit PPP Projects in China: A System Dynamics Approach</title>
      <link>https://trid.trb.org/View/2550924</link>
      <description><![CDATA[Urban rail transit public–private partnership (PPP) projects are characterized by their significant public attributes and strong externalities. They often exhibit low profitability and require substantial capital for operation. Coupled with the absence of a robust payment guarantee system, these factors can easily result in the asset value of PPP projects in the market failing to meet anticipated targets. This situation frequently leads to a proliferation of inefficient and ineffective investments related to assets each year, posing a risk of asset devaluation. Based on a retrospective analysis and categorization of previous research topics, the performance of asset devaluation can be divided into two distinct types: cost-driven asset devaluation and functional asset devaluation. Cost-driven asset devaluation is defined as cost overruns in a PPP project, while functional asset devaluation refers to functional redundancy resulting from agreed-upon functions in a project that either fails to meet public demand or exceeds it. The asset planning phase represents the most vital and pivotal link within the entire asset lifecycle management. The rationality of goal setting directly influences the government’s subsequent project expenditures. Consequently, this paper adopts an array of methodologies to thoroughly examine the influence mechanism of asset devaluation in the asset planning stage. First, risk factors (R) contributing to asset devaluation were identified by integrating literature reviews, case studies, and expert (E) interviews. Second, system dynamics modeling was utilized to articulate the relationships among these risk factors. Finally, this study utilized the simulation modeling of the Beijing Metro Line 4 project to identify the critical risk factors affecting cost-driven asset devaluation and functional asset devaluation, respectively: the reasonableness of return mechanism measurement and the reasonableness of market forecast. The critical factors impacting both cost-driven and functional asset devaluation were identified as: professionalism in feasibility study preparation, and clarity of risk-sharing framework, among others. Findings enrich the research on asset management in PPP projects and provide theoretical guidance for the accurate diagnosis and precise prevention of asset devaluation.]]></description>
      <pubDate>Thu, 21 Aug 2025 09:19:22 GMT</pubDate>
      <guid>https://trid.trb.org/View/2550924</guid>
    </item>
    <item>
      <title>A Continuous Autonomous Train Positioning Method Using Stereo Vision and Object Tracking</title>
      <link>https://trid.trb.org/View/2553169</link>
      <description><![CDATA[Accurate and reliable train positioning is essential for safe railway operations. However, traditional infrastructure-based methods lack the ability for autonomy and continuous positioning. To solve the problem, the authors propose a novel modular visual processing framework for continuous train positioning. Environmental information captured by stereo vision sensors is processed using a feature extraction algorithm to detect landmarks along the railway. The detection results are input into a dual-stream parallel processing architecture, where an anchor-based stereo matching module calculates the depth values of detected landmarks, providing essential spatial information. Simultaneously, a dynamic region of interest-based multiobject tracking module assigns unique identification codes to each landmark, facilitating the integration of landmark detection results with their corresponding depth values. This correlation allows for the tracking of changes in landmark positions and the calculation of displacement caused by train motion, leading to a more accurate estimation of the train’s location. Subsequently, the train mileage is matched against a database track map to obtain the final precise position. Field tests on Beijing Metro Line 9 and the Capital Airport Line demonstrate an average mean relative error of 0.19% and positioning accuracy within 1 m over a range of 100 m.]]></description>
      <pubDate>Thu, 26 Jun 2025 11:42:14 GMT</pubDate>
      <guid>https://trid.trb.org/View/2553169</guid>
    </item>
    <item>
      <title>Optimization study of dynamic emergency feeder bus paths with the sudden interruption of urban railway traffic</title>
      <link>https://trid.trb.org/View/2557113</link>
      <description><![CDATA[The rapid expansion of urban rail transit networks has increased their vulnerability to disruptions. When a metro line experiences sudden interruptions, it can severely reduce passenger mobility and degrade the overall transportation system performance. Existing bus feeder programs are often inadequate in responding effectively to dynamic and real-time fluctuations in passenger flow during such disruptions, particularly when combined with complex road traffic conditions. The authors propose a novel hybrid metaheuristic algorithm for the emergency feeder bus routes with time-window constraints to address this issue. The algorithm combines the Max-Min Ant System (MMAS) and Simulated Annealing (SA) to enhance search performance. A Back Propagation (BP) neural network estimates the emergency demand at each affected station, using historical and structural factors. These estimates are integrated into the hybrid optimization process, improving routing efficiency. The model aims to minimize the total operational time of emergency buses, ensuring timely evacuation. A case study using Beijing Metro validates the model’s effectiveness. Results indicate a 1.7-hour reduction in total passenger travel time and an 84.7% decrease in computation time compared to the Gurobi exact algorithm. These improvements facilitate the identification of optimal feeder paths under varying traffic conditions. The study provides practical guidance for enhancing emergency response strategies in metro systems.]]></description>
      <pubDate>Thu, 05 Jun 2025 14:01:10 GMT</pubDate>
      <guid>https://trid.trb.org/View/2557113</guid>
    </item>
  </channel>
</rss>