<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>Deep reinforcement learning for day-to-day dynamic tolling in tradable credit schemes</title>
      <link>https://trid.trb.org/View/2643286</link>
      <description><![CDATA[Tradable credit schemes (TCS) are an increasingly studied alternative to congestion pricing, given their revenue neutrality and ability to address issues of equity through the initial credit allocation. Modeling TCS to aid future design and implementation is associated with challenges involving user and market behavior, demand-supply dynamics, and control mechanisms. In this paper, we focus on the latter and address the day-to-day dynamic tolling problem in TCS, which is formulated as a discrete-time Markov Decision Process and solved using reinforcement learning (RL) algorithms. Our results indicate that RL algorithms achieve travel times and social welfare comparable to a Bayesian optimization benchmark, with the potential to generalize across varying capacities and demand levels. We further assess the robustness of RL across different hyperparameters and apply regularization techniques to mitigate action oscillation, which generates practical tolling strategies that are transferable under day-to-day demand and supply variability. Finally, we discuss potential challenges – such as scaling to large networks – and show how transfer learning can be leveraged to improve computational efficiency and facilitate the practical deployment of RL-based TCS solutions (Code available at https://github.com/XiaoyiWu21/RL4TCS.git.).]]></description>
      <pubDate>Wed, 22 Apr 2026 16:15:32 GMT</pubDate>
      <guid>https://trid.trb.org/View/2643286</guid>
    </item>
    <item>
      <title>Distributed Massive MIMO-Aided Task Offloading in Satellite-Terrestrial Integrated Multi-Tier VEC Networks</title>
      <link>https://trid.trb.org/View/2674331</link>
      <description><![CDATA[This paper proposes a distributed massive multiple-input multiple-output (DM-MIMO) aided multi-tier vehicular edge computing (VEC) system. In particular, each vehicle terminal (VT) offloads its computational task to the roadside unit (RSU) by orthogonal frequency division multiple access (OFDMA), which can be computed locally at the RSU and offloaded to the central processing unit (CPU) via massive satellite access points (SAPs) for remote computation. By considering the partial task offloading model, we consider the joint optimization of the task offloading, subchannel allocation and precoding optimization to minimize the total cost in terms of total delay and energy consumption. To solve this non-convex problem, we transform the original problem into three sub-problems and use the alternate optimization algorithm to solve it. First, we transform the subcarrier allocation problem of discrete variables into the convex optimization problem of continuous variables. First, we transform the subcarrier allocation problem of discrete variables into the convex optimization problem of continuous variables. Then, we use multiple quadratic transformations and the Lagrange multiplier method to transform the non-convex subproblem of optimizing precoding vectors into a convex problem, while the task offloading subproblem is a convex problem. Given the subcarrier and the task allocation and precoding result, we finally find the joint optimized results by the iterative optimization algorithm. Simulation results show that our proposed algorithm is superior to other benchmarks.]]></description>
      <pubDate>Fri, 27 Feb 2026 11:29:05 GMT</pubDate>
      <guid>https://trid.trb.org/View/2674331</guid>
    </item>
    <item>
      <title>Real-time multi-objective optimization and simulation of intelligent compaction for railway subgrade construction</title>
      <link>https://trid.trb.org/View/2640854</link>
      <description><![CDATA[Efficient and quality-controlled compaction is significant in railway subgrade construction, influencing subsequent construction stages and long-term railway operational safety. This study proposes an innovative ensemble Coupled Optimization–Evaluation Algorithm (COEA), integrating intelligent (IC) compaction, surrogate modeling, multi-objective optimization, and discrete event simulation (DES) into a unified, real-time decision-making framework. Initially, a particle swarm optimized back-propagation neural network surrogate model predicts compaction quality, which then inputs to the Non-dominated Sorting Genetic Algorithm III to generate optimized construction schemes. These optimized construction schemes are then evaluated by DES, where the service entity time is predicted by the ARIMA-LSTM hybrid model that captures both linear trends and nonlinear fluctuations in construction times. Field test demonstrates that the COEA significantly enhances compaction efficiency, reducing compaction passes from 8 to 6 while maintaining quality, leading to an 11.4 % reduction in overall construction duration. The simulation predictions closely matched actual construction timelines with only a 1.05 % MAPE. This study provides a comprehensive methodology for improving railway subgrade compaction, precise controlling over compaction quality, time efficiency, and resource allocation, thus offering significant practical and economic benefits.]]></description>
      <pubDate>Tue, 17 Feb 2026 13:12:50 GMT</pubDate>
      <guid>https://trid.trb.org/View/2640854</guid>
    </item>
    <item>
      <title>From black-box to white-box: Interpretable deep reinforcement learning with Kolmogorov-Arnold networks for autonomous driving</title>
      <link>https://trid.trb.org/View/2614765</link>
      <description><![CDATA[While Deep Reinforcement Learning (DRL) has made significant progress in decision-making within complex traffic environments, the policies learned by traditional Artificial Neural Networks (ANNs) often lack interpretability. In this paper, we propose Kolmogorov-Arnold Network-based Interpretable DRL (KAN-IDRL), a novel framework that replaces black-box ANNs with inherently interpretable Kolmogorov-Arnold Networks (KANs) to develop autonomous driving decision-making policies. To ensure high decision performance while further improving interpretability, we introduce a dynamic pruning method that gradually removes unimportant neurons from the KAN during the training phase. To enhance the understanding of local decision-making in KAN-based models, we design a layer-wise relevance propagation rule tailored to KANs, which reveals the contribution of each input feature to the model’s decision. These feature contributions are then used to guide counterfactual searches, generating representative counterfactual explanations. Additionally, as a complement to attribution-based explanations, we employ gradient-based saliency methods to analyze the sensitivity of input features. The proposed framework is evaluated in both discrete decision-making and continuous control scenarios, and compared against traditional DRL architectures. The results demonstrate that KAN-IDRL achieves the highest or near-highest driving efficiency and safety while utilizing significantly smaller computation graphs. Finally, the framework’s ability to provide transparent decision-making is verified through a comprehensive analysis across three key dimensions: attribution, sensitivity, and counterfactual explanation.]]></description>
      <pubDate>Mon, 02 Feb 2026 09:33:48 GMT</pubDate>
      <guid>https://trid.trb.org/View/2614765</guid>
    </item>
    <item>
      <title>Mobile Edge Deployment and Resource Management for Maritime Wireless Networks</title>
      <link>https://trid.trb.org/View/2606287</link>
      <description><![CDATA[Mobile Edge Computing (MEC) has been envisioned as one of the key technologies for supplying computation and storage resources in Internet of Vessels (IoV) networks. Due to its flexible deployment, low cost and agile maneuverability, Unmanned Surface Vehicle (USV) has emerged as a promising solution, to provide communication and computation services for maritime users. In this paper, the authors study mobile edge deployment and resource management for MEC-assisted maritime wireless networks where USVs with diverse computation resources are deployed to provide edge computing services that complement the cloud-based services. To this end, the authors formulate an optimization problem to minimize the expected response time by jointly optimizing the deployment of mobile USVs and computation offloading decisions. To solve the mixed-integer nonlinear program problem, the authors propose a Dual-Layer Reinforcement Learning (DLRL) framework to attain a near-optimal solution. Specifically, a Deep Deterministic Policy Gradient (DDPG) algorithm is designed to obtain the best USV deployment in the outer layer learning, and a Q-learning algorithm is designed to determine the best computation offloading decisions in the inner layer learning. Numerical results demonstrate that the proposed solution outperforms some literature algorithms by effectively handling both continuous and discrete variables.]]></description>
      <pubDate>Mon, 13 Oct 2025 08:49:07 GMT</pubDate>
      <guid>https://trid.trb.org/View/2606287</guid>
    </item>
    <item>
      <title>Cuckoo Search-Enabled Task Scheduling and Cache Updating in Vehicular Edge-Fog Computing</title>
      <link>https://trid.trb.org/View/2606525</link>
      <description><![CDATA[With the emergence of computation-intensive vehicle applications, vehicle edge-fog computing (VEFC) is playing an increasingly important role in intelligent transportation systems. In this article, the authors study the task scheduling and cache updating (TSCU) problem in VEFC, where vehicles can choose three modes to process their tasks: local computing, vehicle-to-vehicle (V2V) offloading, and vehicle-to-infrastructure (V2I) offloading. Service providers can dynamically update their caches to improve the system offloading efficiency. To maximize the offloading efficiency, the authors formulate the TSCU problem as a mixed-integer non-linear programming (MINLP) problem and develop an improved discrete cuckoo search algorithm to solve the optimization problem. The authors' proposed algorithm explores the global environment via Levy flight. On the other hand, the algorithm convergence rate is accelerated by generating the initial population based on a greedy algorithm and its variants and updating the discard probability. Simulation results show that the authors' proposal has achieved an up to 12% improvement in offloading efficiency compared to the benchmark schemes.]]></description>
      <pubDate>Fri, 03 Oct 2025 16:16:10 GMT</pubDate>
      <guid>https://trid.trb.org/View/2606525</guid>
    </item>
    <item>
      <title>Rural Transportation System Design for Omnichannel Healthcare</title>
      <link>https://trid.trb.org/View/2576317</link>
      <description><![CDATA[This report investigates the deployment of Telehealth Kiosks (TKBs) as part of an omnichannel healthcare strategy to improve access and equity in rural areas. Using case studies from rural Missouri, the research integrates empirical data, discrete choice experiments, and optimization models to address critical barriers in healthcare accessibility. Travel time decay functions were calibrated to capture patient preferences for healthcare options, and these functions informed both continuous approximation and discrete optimization models to design optimal TKB networks. Key findings reveal the trade-offs between equity and efficiency, highlighting the importance of adaptive equity thresholds and strategic resource allocation. Practical implications for policymakers emphasize data-driven deployment strategies, scalable solutions, and leveraging emerging technologies. This study contributes a robust framework for designing equitable healthcare systems, offering insights to bridge access gaps in underserved rural regions.]]></description>
      <pubDate>Fri, 08 Aug 2025 08:50:11 GMT</pubDate>
      <guid>https://trid.trb.org/View/2576317</guid>
    </item>
    <item>
      <title>Causal reinforcement learning for train scheduling on single-track railway networks</title>
      <link>https://trid.trb.org/View/2567342</link>
      <description><![CDATA[Recent advancements in deep reinforcement learning have shown promise for large-scale railway scheduling. By decomposing complex railway networks into infrastructure-defined control units (e.g., sections and stations) managed by intelligent agents, this approach reduces scheduling complexity while enhancing scalability. Crucially, scheduling decisions across these units exhibit complex interdependencies posing a major challenge for existing reinforcement learning methods, particularly in single-track railway networks with spatio-temporal constraints. Achieving optimal scheduling requires understanding and leveraging interaction mechanisms between unit-level behaviors. However, agents managing individual units struggle to capture unobservable interactions, and even determining whether these interactions are discrete, continuous, or hybrid remains a major challenge, posing difficulties in modeling through deterministic variables. To address this, the authors introduce a latent variable into each agent’s probabilistic decision-making model to capture unobserved interactions, establishing a structural causal model for multi-agent decision-making for the complicated train scheduling task. By inferring latent variables from observed data, the authors disentangle interdependent decision processes. Each agent integrates a variational autoencoder with an end-to-end causal reinforcement learning framework to enhance collaborative scheduling in single-track networks. Experiments demonstrate state-of-the-art performance, marking the first application of causal modeling in railway scheduling and suggesting new research directions.]]></description>
      <pubDate>Fri, 11 Jul 2025 14:28:22 GMT</pubDate>
      <guid>https://trid.trb.org/View/2567342</guid>
    </item>
    <item>
      <title>Solving discrete network design problem using disjunctive constraints</title>
      <link>https://trid.trb.org/View/2548824</link>
      <description><![CDATA[This paper introduces a deterministic algorithm to solve the discrete network design problem (DNDP) efficiently. This non-convex bilevel optimization problem is well-known as a non deterministic polynomial (NP)-hard problem in strategic transportation planning. The proposed algorithm optimizes budget allocation for large-scale network improvements deterministically and with computational efficiency. It integrates disjunctive programming with an improved partial linearized subgradient method to enhance performance without significantly affecting solution quality. The authors evaluated the algorithm on the mid-scale Sioux Falls and large-scale Chicago networks. The authors assess the proposed algorithm's accuracy by examining the objective function's value, specifically the total travel time within the network. When tested on the mid-scale Sioux Falls network, the algorithm achieved an average 46% improvement in computational efficiency, compared to the best-performing method discussed in this paper, albeit with a 4.17% higher total travel time than the most accurate one, as the value of the objective function. In the application to the large-scale Chicago network, the efficiency improved by an average of 99.48% while the total travel time experienced a 4.34% increase. These findings indicate that the deterministic algorithm proposed in this research improves the computational speed while presenting a limited trade-off with solution precision. This deterministic approach offers a structured, predictable, and repeatable method for solving DNDP, which can advance transportation planning, particularly for large-scale network applications where computational efficiency is paramount.]]></description>
      <pubDate>Tue, 27 May 2025 09:30:44 GMT</pubDate>
      <guid>https://trid.trb.org/View/2548824</guid>
    </item>
    <item>
      <title>Multi-task deep reinforcement learning for dynamic scheduling of large-scale fleets in earthmoving operations</title>
      <link>https://trid.trb.org/View/2540148</link>
      <description><![CDATA[Large-scale earthwork transportation encounters queuing congestion and dynamic uncertainties, while existing methods ignore complex traffic behaviors and exhibit limited responsiveness and generalization. This paper proposes a multi-task Deep Reinforcement Learning (DRL) framework for the dynamic scheduling of large fleets across supply sites and traffic networks. In the framework, multiple agents interact in complex environments modeled by discrete-event simulation, utilizing long short-term memory networks that consider queuing behaviors and dynamic trends of transportation systems to allocate rational materials, supply sites, and routes collaboratively, with an invariant update strategy to balance generalization and task-specific optimization during training. Experiments demonstrate that the model generates dynamic schedules within 7 min, reducing transportation time by 24 %. The trained agent can adapt to the changing transportation demand in complex construction environments and enhance transportation efficiency. This paper demonstrates the potential of DRL in scheduling more complex construction projects and promoting real-time lean control of modern logistics.]]></description>
      <pubDate>Wed, 30 Apr 2025 16:57:08 GMT</pubDate>
      <guid>https://trid.trb.org/View/2540148</guid>
    </item>
    <item>
      <title>On the spatio-temporal optimization for the charging scheduling of battery electric buses</title>
      <link>https://trid.trb.org/View/2526829</link>
      <description><![CDATA[Battery electric buses (BEBs) play a crucial role in advancing energy efficiency, reducing emissions, and fostering sustainable public transport. Opportunity charging technology has proven effective in alleviating BEB range anxiety. As BEB networks expand, designing an optimized charging schedule that accounts for spatio-temporal complexities becomes essential. This paper proposes an innovative optimization scheme that leverages the spatio-temporal characteristics of BEB networks. By employing variable charging power, the scheme balances TOU with grid loads. The problem is formulated as a nonlinear, non-convex program with continuous time variables and transformed into a solvable mixed-integer linear programming model via discrete-event-based linear reconstruction. Applied to a real BEB network in Shanghai, the results demonstrate a 7.83% reduction in charging costs by improving grid peak power and a 2.42% cost reduction by increasing charging piles across terminals.]]></description>
      <pubDate>Thu, 27 Mar 2025 11:35:00 GMT</pubDate>
      <guid>https://trid.trb.org/View/2526829</guid>
    </item>
    <item>
      <title>Simulation and Numerical Methods for Solution of the Dynamic Disequilibrium Network Design Problem</title>
      <link>https://trid.trb.org/View/2263754</link>
      <description><![CDATA[This paper provides a succinct statement of the deterministic dynamic disequilibrium network design problem. The formulation developed here is currently the subject of of an effort to provide numerical tools for optimal dynamic infrastructure capital budgeting. The authors also present a numerical solution of a small example model using a discrete multigrid optimization approach.]]></description>
      <pubDate>Mon, 24 Feb 2025 14:46:51 GMT</pubDate>
      <guid>https://trid.trb.org/View/2263754</guid>
    </item>
    <item>
      <title>Planning and Dynamic Management of Autonomous Modular Mobility Services [supporting dataset]</title>
      <link>https://trid.trb.org/View/2479855</link>
      <description><![CDATA[This project aimed to address the challenges faced by on-demand mobility operators in understanding and addressing spatiotemporal random bipartite matching problems (ST-RBMPs). At the planning level, we developed analytical models to estimate the expected system performance in a static RBMP. At the operational level, we designed solution algorithms to improve the overall service efficiency in ST-RBMPs with different types of supply arrivals. Although our main focus was on the application of on-demand mobility services, these models can also be applied to other contexts such as resource allocation, target detection, etc. This project aimed to address the following research objectives:  1. Propose an analytical model with closed-form formulas (without statistical curve fitting) that estimate the expectation of the optimal matching distance for static RBMP, where the bipartite vertices are distributed randomly over a discrete network. These formulas can be incorporated into queuing and optimization models to identify the best operational strategies in on-demand mobility systems with closed- or open-loop resource arrivals. It helps determine the optimal decision timing for whether newly arriving customers should be matched instantly or pooled into a batch for matching.  2. For ST-RBMPs with closed-loop resources, where arriving customers shall be matched instantly, the objective is to propose a Pareto-improving strategy that allows matched vertices to be swapped among candidates with improved matching distances as the system evolves. This strategy could enhance system efficiency by reducing the overall expected matching distance and mitigating the so-called Wild Goose Chasing (WGC) phenomenon. Approximate analytic formulas can be derived from a series of differential equations and spatial probability models to estimate the expected system performance in the steady state.]]></description>
      <pubDate>Tue, 28 Jan 2025 14:52:40 GMT</pubDate>
      <guid>https://trid.trb.org/View/2479855</guid>
    </item>
    <item>
      <title>Railcar itinerary optimization in railway marshalling yards: A graph neural network based deep reinforcement learning method</title>
      <link>https://trid.trb.org/View/2480486</link>
      <description><![CDATA[The goal of Railcar Itinerary Optimization in Marshalling Yards (RIO-MY) is to achieve an effective integrated operation plan for both train shunting operations and train makeup, with the aim of minimizing the railcar dwell time in the railway marshalling yard. Due to complex interdependent decisions in the disassembly and assembly process of trains, conventional optimization methods for the problem face challenges in addressing the dynamic nature of traffic in the marshalling yard and offering highly efficient solutions. This paper introduces a novel approach to the RIO-MY problem using a graph neural network based deep reinforcement learning method. First, the authors model the solving process of RIO-MY as a Markov decision process, utilizing a tripartite graph to represent the operational state of a marshalling yard. Then the authors design a novel tripartite graph isomorphism network (TGIN) to learn informative embeddings on the graph, which are exploited to reason out the joint action to simultaneously decide on hump sequencing and classification track assignment. The TGIN based policy network is trained by the proximal policy optimization algorithm, with a reward tailored to well estimate railcar dwell time for each state. Moreover, the authors develop a discrete-event simulation of operations in the railway marshalling yard, which serves as the reinforcement learning environment and integrates typical heuristic rules of outbound train assembly and shunting locomotive scheduling. Extensive experiments on two real-world railway marshalling yards demonstrate that the proposed method outperforms conventional heuristic algorithms. Moreover, it achieves competitive performance to the mixed integer nonlinear programming model with significantly less computational time. In addition, the trained policy networks can favourably generalize to scenarios that are unseen during training and effectively handle disturbances in the train disassembly process.]]></description>
      <pubDate>Mon, 27 Jan 2025 15:39:52 GMT</pubDate>
      <guid>https://trid.trb.org/View/2480486</guid>
    </item>
    <item>
      <title>Average Distance of Random Bipartite Matching in Discrete Networks</title>
      <link>https://trid.trb.org/View/2479856</link>
      <description><![CDATA[The bipartite matching problem is widely applied in the field of transportation; e.g., to find optimal matches between supply and demand over time and space. Recent efforts have been made on developing analytical formulas to estimate the expected matching distance in bipartite matching with randomly distributed vertices in two- or higher-dimensional spaces, but no accurate formulas currently exist for one-dimensional problems. This paper presents a set of closed-form formulas, without curve-fitting, that can provide accurate average distance estimates for one-dimensional random bipartite matching problems (RBMP). The authors first focus on one-dimensional space and propose a new method that relates the corresponding matching distance to the area size between a random walk path and the x-axis. This result directly leads to a straightforward closed-form formula for balanced RBMPs. For unbalanced RBMPs, the authors first analyze the properties of an unbalanced random walk that can be related to balanced RBPMs after optimally removing a subset of unmatched points, and then derive a set of approximate formulas. Additionally, the authors build upon an optimal point removal strategy to derive a set of recursive formulas that can provide more accurate estimates. Then, the authors shift their focus to regular discrete networks, and use the one-dimensional results as building blocks to derive RBMP formulas. To verify the accuracy of the proposed formulas, a set of Monte-Carlo simulations are generated for a variety of matching problems settings. Results indicate that the proposed formulas provide quite accurate distance estimations for one-dimensional line segments and discrete networks under a variety of conditions.]]></description>
      <pubDate>Mon, 30 Dec 2024 09:58:10 GMT</pubDate>
      <guid>https://trid.trb.org/View/2479856</guid>
    </item>
  </channel>
</rss>