Transportation Research Part C-Emerging Technologies最新文献

{"title":"Collaborative optimization of train timetabling with maintenance window setting and maintenance scheduling considering fluctuating daily train volumes and dynamic maintenance demands","authors":"Xuze Ye ,&nbsp;Yijia Du ,&nbsp;Haonan Yang ,&nbsp;Jinshan Pan ,&nbsp;Shaoquan Ni ,&nbsp;Dingjun Chen","doi":"10.1016/j.trc.2026.105552","DOIUrl":"10.1016/j.trc.2026.105552","url":null,"abstract":"<div><div>Coordinating transport and maintenance in railway operations is difficult. The complexity induced by fluctuating daily train volumes on railway corridors has motivated us to propose a dynamic maintenance scheduling mode that increases the routine maintenance flexibility to scheduling more maintenance activities on days with lower train volumes to reduce the burden on track resources. Some activities of railway assets with relatively mature condition-monitoring technology are scheduled based on maintenance-condition thresholds. Within a planning horizon of several days, dynamic maintenance demands are satisfied by scheduling maintenance activities based on their flexible maintenance date or maintenance-condition threshold, which are coordinated with fluctuating daily train volumes. To reduce maintenance-induced interference on train operations and schedule the maximum possible maintenance activities within the planning horizon, an integer linear programming model is proposed to collaboratively optimize daily train timetabling with maintenance window setting and maintenance scheduling, where the duration and maintenance content of each maintenance window on each day are determined based on dynamic maintenance demands and daily train volumes. To solve large-scale instances, a Lagrangian relaxation-based heuristic algorithm is developed in which the primal problem is decomposed into multiple day- and section-specific subproblems. The effectiveness of the collaborative optimization method and performance of the algorithm are verified for a real-world case. Our method can effectively satisfy dynamic maintenance demands under fluctuating daily train volumes. Moreover, we develop an application of our method to the entire life cycle of railways using the rolling horizon framework.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"185 ","pages":"Article 105552"},"PeriodicalIF":7.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146152662","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Joint optimization of multimodal transit frequency and shared autonomous vehicle fleet size with hybrid metaheuristic and nonlinear programming","authors":"Max T.M. Ng ,&nbsp;Hani S. Mahmassani ,&nbsp;Draco Tong ,&nbsp;Ömer Verbas ,&nbsp;Taner Cokyasar","doi":"10.1016/j.trc.2026.105568","DOIUrl":"10.1016/j.trc.2026.105568","url":null,"abstract":"<div><div>Shared autonomous vehicles (SAVs) bring competition to traditional transit services but redesigning multimodal transit network can utilize SAVs as feeders to enhance service efficiency and coverage. This paper presents an optimization framework for the joint multimodal transit frequency and SAV fleet size problem, a variant of the transit network frequency setting problem. The objective is to maximize total transit ridership (including SAV-fed trips and subtracting boarding rejections) across multiple time periods under budget constraints, considering endogenous mode choice (transit, point-to-point SAVs, driving) and route selection, while allowing for strategic route removal by setting frequencies to zero. Due to the problem’s non-linear, non-convex nature and the computational challenges of large-scale networks, we develop a hybrid solution approach that combines a metaheuristic approach (particle swarm optimization) with nonlinear programming for local solution refinement. To ensure computational tractability, the framework integrates analytical approximation models for SAV waiting times based on fleet utilization, multimodal network assignment for route choice, and multinomial logit mode choice behavior, bypassing the need for computationally intensive simulations within the main optimization loop. Applied to the Chicago metropolitan area’s multimodal network, our method illustrates a 33.3% increase in transit ridership through optimized transit route frequencies and SAV integration, particularly enhancing off-peak service accessibility and strategically reallocating resources.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"185 ","pages":"Article 105568"},"PeriodicalIF":7.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146190816","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-agent reinforcement learning with a hybrid sequential reward feedback strategy for dynamic multi-modal traffic assignment","authors":"Dongyue Cun ,&nbsp;Muqing Du ,&nbsp;Lin Cheng ,&nbsp;Anthony Chen","doi":"10.1016/j.trc.2026.105545","DOIUrl":"10.1016/j.trc.2026.105545","url":null,"abstract":"<div><div>Urbanization and the expansion of transportation modes have exacerbated the challenges of understanding travelers’ decision-making processes regarding route choice across various transportation modes. This paper proposes a novel macroscopic hybrid sequential game method using multi-agent reinforcement learning (MARL) to address issues of computational efficiency and behavioral complexity in multi-modal transportation network simulations. Specifically, agents’ perception behaviors are modeled as a sequential decision-making process considering road capacity constraints, which helps estimate travel time under congestion effects in the multi-modal traffic assignment. In addition, a hybrid reward framework is proposed, providing system-level reward to guide the multi-agent system towards different Nash equilibria, thereby reducing policy fluctuations. To simulate interactions between agents of different transportation modes, a multi-edge representation and reward structures designed for car, bus, priority bus, and metro modes are adopted to handle the mixed traffic flow through the same road. Furthermore, our approach uses a mean-field multi-agent deep Q-learning method to consider both mode and route choice, simplifying agent interactions through mean-field theory and clustering agents with the same origin–destination (OD) demands. Experimental results demonstrate that the hybrid sequential feedback strategy outperforms the simultaneous feedback strategy regarding convergence speed, agent reward distribution, and network flow distribution. Furthermore, the proposed method is tested on the Sioux-Falls network to verify its computational efficiency in three network change scenarios (disruption, road reconstruction, and new road construction). These findings highlight the potential of the proposed MARL method for large-scale multi-modal transportation network analysis, particularly under various incident scenarios, providing an effective tool for urban transportation planning and project evaluation.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"185 ","pages":"Article 105545"},"PeriodicalIF":7.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146102562","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improving the fate of a Bike: A Usage-Continuity-Driven predictive framework for Bike-Sharing rebalancing under sparse demand","authors":"Ziyi Shi ,&nbsp;Chenlu Wang ,&nbsp;Linghao Wang ,&nbsp;Zheng Zhu ,&nbsp;Hai Yang","doi":"10.1016/j.trc.2026.105572","DOIUrl":"10.1016/j.trc.2026.105572","url":null,"abstract":"<div><div>The sustainable operation of shared micro-mobility service (like bike/e-bike sharing) generally requires demand prediction and bike rebalance to avoid the mismatch between bike inflow and outflow. Current rebalancing strategies typically rely on forecasting one-time site demand, and try to fulfill predicted values. However, related studies overlook the lagged effects of rebalance (e.g., where rebalanced bikes will go), resulting in operational inefficiency. Rebalanced bikes may be used once only to be stranded in a cold-demand zone, necessitating another costly rebalance. Therefore, this study introduces a usage-continuity-driven predictive framework for rebalance, where bike usage continuity is defined as the average of future usage frequency of bikes at one specific site. For sparse-demand systems, it is challenging to directly predict usage continuity due to sparsity and uncertainty. Here, we model the bike usage continuity at the site with a zero-inflated negative binomial (ZINB) distribution, estimating its parameters via graph neural networks. Large Language Models (LLMs) are integrated to enhance urban context understanding. Next, a simple rebalance model is established based on the predicted usage continuity The proposed approach is examined via real-world bike/e-bike sharing datasets in representative cities of China (Taizhou) and the USA (New York City), respectively. The results demonstrate the model’s effectiveness in terms of prediction accuracy, uncertainty quantification, and rebalancing performance, especially in scenarios with high demand dispersion. By discussing the trade-off between capturing demand stochasticity and farsighted decision-making, practical insights are offered into the operation of shared micro-mobility systems.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"185 ","pages":"Article 105572"},"PeriodicalIF":7.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146192414","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sky highway: An air traffic structure for low-altitude heterogeneous VTOL aircraft","authors":"Rao Fu ,&nbsp;Yazan Safadi ,&nbsp;Quan Quan ,&nbsp;Jack Haddad","doi":"10.1016/j.trc.2026.105531","DOIUrl":"10.1016/j.trc.2026.105531","url":null,"abstract":"<div><div>The development of a new mode of air transport, i.e., low alti tude aircraft, rises with the advancement of aviation and communication technologies. This promotes the concept of advanced air mobility (AAM) and increases engagement from both industry and academia to tackle the futuristic air traffic management challenges in the shared low-altitude airspace. Current research either focuses on airspace design for safety purpose or swarm control to maximize the airspace performance. To achieve an effective balance between airspace safety and productivity, a sky highway structure that consists of airways and intersections is proposed, where traffic network, route, and swarm control design are all considered. The corresponding controller design principles are raised to achieve real-time conflict resolution among aircraft while improving traffic efficiency. In the sky highway, each aircraft will have a designated route, and an airway, similar to a highway road, can accommodate multiple aircraft performing free flight simultaneously. The geometrical structure of the proposed sky highway with corresponding flight modes to support large-scale operation is studied briefly one by one. In this paper, the proposed sky highway structure is investigated and presented, and different analyses are conducted to evaluate the effectiveness of the airspace structure.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"185 ","pages":"Article 105531"},"PeriodicalIF":7.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146192415","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Data-driven optimization of energy-efficient metro timetables accounting for operational deviations","authors":"Yongji Luo ,&nbsp;Yushi Tang ,&nbsp;Lu Liu ,&nbsp;Andrea D’Ariano ,&nbsp;Tommaso Bosi ,&nbsp;Shoushuai Zhang ,&nbsp;Feng Xue","doi":"10.1016/j.trc.2026.105551","DOIUrl":"10.1016/j.trc.2026.105551","url":null,"abstract":"<div><div>Train traction energy constitutes a substantial portion (typically 40–60%) of total energy consumption in metro systems, making energy-efficient timetable optimization a crucial strategy for sustainable urban rail operations. Although existing studies have demonstrated theoretical potential in this field, their practical impact remains limited due to the common neglect of deviations between scheduled and actual timetables. This study presents a novel three-phase data-driven framework to bridge this gap. First, we establish a machine learning-enhanced simulation system to accurately reproduce actual timetables from scheduled timetables by incorporating dwell time estimation models. Second, we propose a data-driven energy calculation methodology to precisely quantify total energy consumption under real-world operating conditions. Third, a simulation-based optimization algorithm is designed to improve the energy efficiency of the operator-provided timetable through iterative refinement. Unlike prior stochastic models, our approach directly leverages real-world operational data for both timetable deviations and per-section energy profiles. Numerical experiments on a northern Chinese metro line demonstrate a simulated traction energy reduction of 5.2% (5,735 kWh), with field implementation confirming actual energy savings of 3.04% (2,646 kWh). The study provides metro operators with a replicable framework for sustainable timetable optimization, demonstrating both methodological innovation and practical energy savings.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"185 ","pages":"Article 105551"},"PeriodicalIF":7.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146152640","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep reinforcement learning-based multi-agent negotiation framework for conflict-free U-Space scenarios","authors":"Zhiqiang Liu,&nbsp;Juan José Ramos Gonzalez,&nbsp;Ender Çetin,&nbsp;Jose Luis Munoz-Gamarra","doi":"10.1016/j.trc.2026.105553","DOIUrl":"10.1016/j.trc.2026.105553","url":null,"abstract":"<div><div>Very-Low Level urban airspace is expected to accommodate large numbers of simultaneous drone missions in the near future, posing acute challenges for the strategic planning and deconfliction services mandated by U-space regulations. The current strategic planner algorithm relies on the first-come-first-served or batch policy that rejects nearly half of the submitted flight plans in large-scale scenarios. This paper presents a novel study that integrates a Deep Reinforcement Learning-based Multi-Agent Negotiation Framework into the U-space strategic planning and deconfliction service. The framework reallocates rejected flight plans through an iterative sealed-bid auction mechanism in a fully competitive auction environment. Four role-specific drone operator agents were constructed: Achiever, Optimizer, Economizer and Normalizer, to learn bidding policies with the Proximal Policy optimization algorithm. Each round’s reward combines individual profit, bidding cost and a global utilisation signal, guiding agents towards system-efficient yet self-interested behaviour. The simulation experiments show that the Deep Reinforcement Learning-based Multi-Agent Negotiation Framework lifts the mission re-accommodation ratio to a notable level while converging to a final allocation within a short strategic planning time. Analysis of reward distribution and bidding patterns confirms that agents adapt their strategies to heterogeneous operational objectives without explicit coordination. These findings indicate that learning-enabled auctions can reconcile operator competition with network-level efficiency, offering a scalable path towards conflict-free, high-density U-space operations.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"185 ","pages":"Article 105553"},"PeriodicalIF":7.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146192416","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Personalised incentives for demand management of congested public transport systems: A reverse-engineering approach and application","authors":"Xia Zhou ,&nbsp;Daniel D. Harabor ,&nbsp;Mark Wallace ,&nbsp;Zhenliang Ma","doi":"10.1016/j.trc.2026.105566","DOIUrl":"10.1016/j.trc.2026.105566","url":null,"abstract":"<div><div>To reduce congestion, public transport service providers can offer incentives that encourage passengers to choose alternative routes and travel times (RTs). To the best of the authors’ knowledge, no existing study on incentive design evaluates scheme performance by explicitly quantifying the deviation of the incentivised system from the exact system-optimal (SO) benchmark. To fill this gap, we develop RE-ESO: a Reverse-Engineering framework using the Exact SO solution for incentive design in public transport. Our algorithm systematically determines specific incentive amounts for each RT combination by iteratively analysing the discrepancies between the current incentivised assignment flows and the exact SO assignment flows. In particular we show, for the first time, incentives that as nearly as possible result in SO passenger choices. The effectiveness of RE-ESO for reducing congestion costs is demonstrated through a case study on the Hong Kong Mass Transit Railway network. In our experiments, RE-ESO achieves a 32.74% reduction in congestion costs, substantially outperforming two comparative baselines: incentive optimisation without a globally optimal target (IO-NGT), which appears popularly in the literature (bi-level method; 22.18% cost reduction), and time-based shifting, which is popular with the industry (off-peak fare reward; 10.90% cost reduction). Notably, our result approaches the theoretical maximum of 36.35% congestion reduction indicated by the exact SO system. Another key finding is that departure time shifting accounts for 82% of the total congestion relief, indicating broad potential for applicability in transit networks like Hong Kong and Stockholm, where route-shifting options are limited.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"185 ","pages":"Article 105566"},"PeriodicalIF":7.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146191407","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Linking string stability and traffic hysteresis: implications for oscillation absorption in vehicle longitudinal control","authors":"Haozhan Ma ,&nbsp;Linheng Li ,&nbsp;Xu Qu ,&nbsp;Bin Ran","doi":"10.1016/j.trc.2026.105565","DOIUrl":"10.1016/j.trc.2026.105565","url":null,"abstract":"<div><div>While the fundamental mechanisms of traffic flow remain elusive—especially under non-steady or oscillatory conditions—existing research has shown that Automated Vehicles (AVs) have the potential to suppress traffic oscillations. However, this often depends on aggressive deceleration and large headways, raising concerns about real-world applicability. To address this gap, this study theoretically analyzes the relationship between traffic hysteresis and string stability in AV control laws. A concise condition is derived to determine the direction of traffic hysteresis, offering a practical indicator of control law stability. Based on this, we propose the Hysteretic Parameter Adaptation Framework (HyPAF), which converts time-invariant control laws into time-varying systems. HyPAF reduces oscillation propagation by slowing ego vehicle responses to upstream fluctuations, while constraining parameter adjustments within stability-preserving bounds. Simulation results demonstrate that the relationship identified in this study between traffic hysteresis and string stability depends solely on the control law, implying that time delays simultaneously influence both phenomena. Moreover, simulations confirm the theoretical findings and show that, under both periodic oscillations and real-world trajectories, HyPAF can significantly reduce oscillation amplitudes and reduces energy consumption, at the cost of a slight increase in risk that remains within acceptable safety margins. These findings offer new insights into the evolution of non-equilibrium traffic flow, provide a practical guideline for parameter tuning in vehicle control systems, and may offer a new perspective for understanding human driving behavior under non-steady conditions.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"185 ","pages":"Article 105565"},"PeriodicalIF":7.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146152648","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Towards unified estimation and calibration in transport models: Integrating micro-level behaviour and macro-level performance","authors":"Ali Najmi ,&nbsp;Michel Bierlaire ,&nbsp;Travis Waller ,&nbsp;Taha H. Rashidi","doi":"10.1016/j.trc.2026.105514","DOIUrl":"10.1016/j.trc.2026.105514","url":null,"abstract":"<div><div>Disaggregated models, such as activity-based and random utility-based frameworks, play a central role in travel behaviour analysis and urban transport planning. However, conventional modelling practices often follow a sequential estimation-calibration process that introduces challenges such as error propagation, inconsistent value-of-time estimation, and poor alignment between micro-level behavioural outputs and macro-level system performance. This paper addresses these issues by proposing an integrated modelling framework that simultaneously estimates discrete choice parameters and calibrates system-level constraints, such as observed traffic counts, OD flows, and reference Value of Time (VoT) targets, within a unified optimisation structure. The proposed approach embeds macro-level calibration objectives directly into the estimation of mode, destination, and route choice models, enabling coherent behavioural interpretation while ensuring system-wide consistency. We implement and evaluate this framework on synthetic networks with varying complexity, employing both global optimisation and Bayesian calibration methods. The results demonstrate that the developed model variants consistently outperform traditional log-likelihood-based models in replicating key system metrics, while maintaining plausible and stable behavioural parameters.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"185 ","pages":"Article 105514"},"PeriodicalIF":7.6,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146135141","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}