Transportation Research Part C-Emerging Technologies最新文献

{"title":"Deconstructing driving behaviors in interactions with pedestrians at uncontrolled crosswalks: an imitation learning method","authors":"Tao Wang ,&nbsp;Minh Kieu ,&nbsp;Chengmin Li ,&nbsp;Wenqiang Chen ,&nbsp;Ying-En Ge","doi":"10.1016/j.trc.2025.105508","DOIUrl":"10.1016/j.trc.2025.105508","url":null,"abstract":"<div><div>The objective of this paper is to deconstruct driving behaviors in interactions with pedestrians at uncontrolled crosswalks. Trajectory data are used to extract variables describing driver–pedestrian interactions, including position, acceleration, velocity, yaw rate, and interaction risk. Driving behavior is modeled as utility-driven, intelligent, and rational decision-making within the framework of a finite-state Markov decision process (MDP). The vanilla generative adversarial imitation learning (GAIL) framework is improved to reconstruct a human-like driving behavior model where the utility function is defined as the deviation between the agent’s behavior distribution and that of human drivers. Maximizing this utility through a deep reinforcement learning (RL) approach drives agents to progressively clone the behavioral policies of human drivers in the real world. The behavioral policy is formulated as a pre-trained driving behavior model and validated on a simulation platform for its ability in reproducing human driving behavior. Experimental results show that the model successfully reproduces the rationality of human drivers and generates human-like interaction trajectories in the simulation environment. Transfer experiments further demonstrate the generalizability of the pre-tained behavioral model. The interaction policy map and the state-value map are visualized to elucidate the generative mechanisms underlying human-like trajectories by revealing risk- and context-dependent layered patterns and latent behavioral preferences. This work contributes to the advancement of human-like behavioral models, thereby enhancing the fidelity of traffic microsimulation and improving behavior modeling in complex driver–pedestrian interactions.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"184 ","pages":"Article 105508"},"PeriodicalIF":7.6,"publicationDate":"2026-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146014568","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":"Modeling proactive avoidance behaviors in pedestrian flows considering congestion anticipation","authors":"Pei-Yang Wu ,&nbsp;Ying-En Ge ,&nbsp;Zhuanglin Ma ,&nbsp;Ren-Yong Guo","doi":"10.1016/j.trc.2026.105532","DOIUrl":"10.1016/j.trc.2026.105532","url":null,"abstract":"<div><div>This paper investigates proactive avoidance behaviors in pedestrian flows by means of real-world experiments and a potential field model. Typical movement patterns of pedestrians related to the proactive avoidance behaviors are provided. Based on the observed behaviors, a potential field model is proposed to combine tactical-level and operational-level modeling frameworks. This model allows pedestrians to select and move toward temporary destinations instead of moving directly to their final destinations. Pedestrian movements are guided by the potential values associated with different positions in the focused space. Three types of sub-potentials are involved to reflect the effects of route attributes, spatiotemporal congestion anticipation, and potential conflicts on pedestrian movements. Numerical experiments demonstrate that the proposed model can effectively reproduce pedestrians’ proactive avoidance behaviors. The model is validated by comparing the simulation results with existing experimental data and our own experimental data. This investigation provides an alternative explanation for pedestrian movement mechanisms.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"184 ","pages":"Article 105532"},"PeriodicalIF":7.6,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146014571","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":"Integrated optimization for vehicle trajectory reconstruction under cooperative perception environment","authors":"Tianheng Zhu,&nbsp;Wangzhi Li,&nbsp;Yiheng Feng","doi":"10.1016/j.trc.2026.105522","DOIUrl":"10.1016/j.trc.2026.105522","url":null,"abstract":"<div><div>Vehicle trajectories provide detailed information about vehicle movements and interactions, which are essential for various transportation applications. However, collecting complete vehicle trajectory data requires high costs. Reconstructing complete vehicle trajectories from partial observations is thus a more cost-effective alternative. Previous studies on trajectory reconstruction primarily focused on vehicle longitudinal behaviors, usually neglecting lane-change (LC) maneuvers. This study proposes an integrated optimization-based vehicle trajectory reconstruction model that considers LC and overtaking behaviors under a cooperative perception environment with very low market penetration rates (MPRs) of connected and automated vehicles (CAVs) and varying packet loss rates (PLRs) of vehicle-to-everything (V2X) communication. A Mixed Integer Linear Programming (MILP) problem is constructed with the objective of minimizing the errors between reconstructed trajectories and observed trajectories, which converts the trajectory reconstruction problem into a joint trajectory generation problem. Moreover, this study considers a cooperative perception environment where partial observed trajectories are collected from CAV perception sensors. Different from other studies that implemented oversimplified detection models to generate observed trajectories without considering the real-world complexity and variability of detection patterns from perception sensors, in this study, we adopt distance-dependent true positive rates (TPRs) as detection performance metric to mimic CAV detection, computed using BEVFusion detection outputs on the nuScenes dataset. The proposed formulation streamlines the entire process and can be applied to various road geometries and traffic conditions. Numerical studies using both NGSIM highway and urban arterial datasets demonstrate the model’s effectiveness in reconstructing vehicle trajectories under 2%-5% CAV MPRs with varying PLRs. Additional sensitivity analysis was conducted to evaluate the impact of 1) vehicle occlusion in CAV detection model; 2) varying traffic conditions (i.e., demand levels); and 3) weights of different terms in the objective function on the trajectory reconstruction accuracy. Under similar reconstruction rates of unobserved trajectories and road segment lengths, the proposed method outperforms existing studies by a significant margin in terms of both longitudinal position accuracy and LC time prediction. The source code is publicly available at <span><span>https://github.com/Purdue-CART-Lab/CP-TrajRecon-Opt</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"184 ","pages":"Article 105522"},"PeriodicalIF":7.6,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145978353","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":"Distributionally robust optimization of sailing speed, bunkering, and fuel switching for dual-fuel liner services","authors":"Ping He ,&nbsp;Lingxiao Wu ,&nbsp;Jian Gang Jin ,&nbsp;Shaorui Zhou ,&nbsp;Frederik Schulte","doi":"10.1016/j.trc.2026.105528","DOIUrl":"10.1016/j.trc.2026.105528","url":null,"abstract":"<div><div>To reduce CO₂ and SO₂ emissions, shipping companies have started deploying LNG or methanol dual-fuel ships on liner services. Unlike traditional container ships, these dual-fuel ships can use multiple types of fuels during a voyage, allowing them to comply with emission regulations while reducing operational costs through fuel switching and speed optimization. Given the significant fluctuations in bunker prices across different ports, decisions regarding fuel switching, refueling, and sailing speeds must account for price uncertainty. We develop a distributionally robust chance-constrained programming model based on the Wasserstein uncertainty set to minimize operating costs under this uncertainty. We divide each port-to-port sailing leg into sub-legs, considering regional emission requirements or canal segments. This segmentation enables the optimization of fuel usage proportions, sailing speeds, and refueling strategies for each sub-leg. The model is then reformulated as a tractable mixed-integer second-order conic programming model. We validate the model using real-world data from COSCO Shipping. Numerical experiments demonstrate that the model can identify optimal solutions for real-scale instances within practical computational time. Furthermore, the robust solutions significantly outperform those obtained using the traditional sample average approximation method. Our results suggest that the joint optimization of fuel management and sailing speeds for dual-fuel ships can effectively reduce operating costs without increasing emissions.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"184 ","pages":"Article 105528"},"PeriodicalIF":7.6,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145978354","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":"Decarbonizing freight transportation: Joint optimization of intermodal service scheduling and cargo routing","authors":"Zeyu Liu","doi":"10.1016/j.trc.2026.105513","DOIUrl":"10.1016/j.trc.2026.105513","url":null,"abstract":"<div><div>With the rapid growth of freight transportation, Greenhouse Gas (GHG) emissions will increase explosively due to the heavy reliance on trucking. To reach the carbon-neutral goal by 2050, it is crucial to fully utilizing intermodal capabilities, which reduces GHG emissions and economic costs for large volumes of cargo across long distances. Yet, existing optimization models for intermodal transportation lack operational details, typically using broad averages of emission and cost metrics or relying on inflexible scheduling, leading to suboptimal results. In this study, we propose a holistic mixed integer model to optimize container-level transportation in a multi-layered intermodal network, with routing and scheduling decisions of individual containers and vehicles. To address the computational challenges, we establish structural properties and develop novel decomposition methods using variable duplication, relaxation, and symmetry breaking. Real-world intermodal network data in the United States are collected to enable comprehensive experiments. Model behaviors and algorithm performances are investigated through sensitivity analyses and benchmarking. The proposed algorithm leads to more than 50% and 60% improvements in solution quality and efficiency, respectively. Additionally, we compute large-scale scenarios to render future projections of the United States freight transportation sector from 2025 to 2050. The joint effect of synchromodality and clean energy technology foresees up to 154 million tons of reductions in GHG emissions by 2050.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"184 ","pages":"Article 105513"},"PeriodicalIF":7.6,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145978355","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":"A joint, context-aware neural network-based travel demand and scheduling model","authors":"Joel Fredriksson,&nbsp;Anders Karlström","doi":"10.1016/j.trc.2025.105512","DOIUrl":"10.1016/j.trc.2025.105512","url":null,"abstract":"<div><div>Recent advancements in machine learning, and neural networks in particular, have introduced new opportunities for activity-based travel demand modeling and scheduling, providing data-driven alternatives to traditional theory-driven methods. While previous machine learning-based scheduling models have integrated combinations of activity, destination, and mode choice as separate sub-models, none have yet, to the best of our knowledge, unified these components into a single, jointly learned framework.</div><div>This paper introduces Skyline-NNjoint, a novel fully neural network-based scheduling model that jointly predicts an agent’s activity, destination, and mode choice decisions at each discrete time step throughout the day. To capture substitution effects and interdependencies among alternatives, the model introduces a Global Context Module (GCM) that enables each alternative to adjust its attractiveness based on the context of all others. While similar context-based approaches have been used in other domains, this is, to the best of our knowledge, the first application of such a mechanism in travel demand modeling. This provides a data-driven approach to relax the Independence of Irrelevant Alternatives (IIA) assumption inherent in multinomial logit models. The effectiveness of the GCM is evaluated by comparing Skyline-NNjoint to a baseline version without it, isolating its contribution to model performance.</div><div>The model is trained on travel survey data from Stockholm and evaluated using both cross-entropy loss and simulated daily activity–travel trajectories. Cross-entropy loss results confirm that the GCM improves predictive performance. Simulation results show that Skyline-NNjoint produces patterns of activity participation, trip timing, and mode choice that closely match observed data. Notably, the model accurately reproduces mode distributions across activity purposes, highlighting its capacity to capture interdependencies in joint decision-making.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"184 ","pages":"Article 105512"},"PeriodicalIF":7.6,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145978362","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":"Large language model guided deep reinforcement learning for safe autonomous vehicle decision making","authors":"Hao Pang,&nbsp;Zhenpo Wang,&nbsp;Guoqiang Li","doi":"10.1016/j.trc.2025.105511","DOIUrl":"10.1016/j.trc.2025.105511","url":null,"abstract":"<div><div>Deep reinforcement learning (DRL) has shown promising potential for decision-making in autonomous driving. However, it requires extensive interaction with the environment and generally has low learning efficiency. To address these challenges, this paper proposes a novel large language model (LLM) guided deep reinforcement learning (LGDRL) framework for the decision-making problem in autonomous driving. Leveraging the powerful reasoning capabilities of LLMs, an LLM-based driving expert is designed to provide intelligent guidance in the DRL learning process. Subsequently, an innovative expert policy constrained algorithm and a novel LLM-intervened interaction mechanism are developed to efficiently integrate the guidance from the LLM expert to enhance the performance of DRL decision-making policies. Extensive experiments are conducted to evaluate the performance of the proposed LGDRL method. The results demonstrate that our proposed method effectively leverages expert guidance to enhance both learning efficiency and performance of DRL, achieving superior driving performance. Moreover, it enables the DRL agent to maintain consistent and reliable performance in the absence of LLM expert guidance, which is promising for real-world applications. The supplementary videos are available at <span><span>https://bitmobility.github.io/LGDRL/</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"184 ","pages":"Article 105511"},"PeriodicalIF":7.6,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145961770","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":"The dynamic system optimal departure time choice problem for a bottleneck with a stochastic capacity: Model formulation and solution algorithm","authors":"Yao Li ,&nbsp;Jie Wang ,&nbsp;Zijun Wu ,&nbsp;Jiancheng Long","doi":"10.1016/j.trc.2025.105510","DOIUrl":"10.1016/j.trc.2025.105510","url":null,"abstract":"<div><div>This paper concerns a novel dynamic system optimal departure time choice (DSO-DTC) problem that takes the capacity uncertainty at a bottleneck into account. The existing traditional analytical methods are unable to yield satisfactory results without stringent conditional assumptions. We innovatively discretize and reformulate it with linear programming (LP) and nonlinear programming (NLP), respectively. While the LP problem can be solved exactly with a standard solver like CPLEX, its complexity grows dramatically when the underlying discretization becomes a little finer. Therefore, we propose a sensitivity analysis-based (SAB) algorithm for the NLP problem instead, and further refine this algorithm with sophisticated strategies. Our experimental study demonstrates that the algorithm not only achieves superior solution efficiency and quality but also exhibits enhanced scalability in terms of discretization accuracy when compared to the benchmark solver CPLEX. Besides, the model enables us to efficiently study the impact of bottleneck capacity uncertainty on the performance of the bottleneck and on the efficiency of tolling strategies, which can be hardly achieved by traditional bottleneck model analysis methods.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"184 ","pages":"Article 105510"},"PeriodicalIF":7.6,"publicationDate":"2026-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145927802","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":"Dynamic on-demand delivery with spatial divisions of labor","authors":"Yue Yang ,&nbsp;André van Renssen ,&nbsp;Mohsen Ramezani","doi":"10.1016/j.trc.2025.105509","DOIUrl":"10.1016/j.trc.2025.105509","url":null,"abstract":"<div><div>The rapid evolution of on-demand delivery services has significantly influenced traditional logistics, particularly in urban areas where there is a surge in customer demand for timely and efficient delivery of small to medium-sized parcels. This paper investigates the concept of <em>spatial divisions of labor</em> in on-demand deliveries where the delivery network is partitioned into several operational regions, each managed by designated couriers. To facilitate parcel transshipment between regions, a set of accessible lockers is strategically placed at the shared borders of adjacent regions. We introduce and address a multi-hop delivery with spatial divisions of labor (MDSDL) problem, which involves dynamic parcel-courier dispatching and routing to minimize total operational costs. Within a rolling-horizon decision framework, the MDSDL problem is decomposed into two interdependent subproblems: (i) Region-Level Path Optimization (RLPO) that determines the coarse-grained, multi-hop path each parcel should take through the network, from its pickup region to its destination region, via the lockers, with the objective of minimizing delivery lateness. This path specifies the sequence of service regions a parcel must traverse to reach its final destination. (ii) Courier Route Optimization (CRO) that manages fine-grained, intra-region dispatching and routing by assigning incoming pickup and drop-off tasks to local couriers, who each has a continuously updated schedule. Subsequently, we develop a novel heuristic approach to dynamically solve RLPO and CRO in real time considering a rolling-horizon formulation. Extensive comparative experiments are conducted to demonstrate the advantages of the proposed approach. Implementing <em>spatial divisions of labor</em> not only enhances system efficiency and reduces operational costs but also improves the customer experience by reducing lateness and shortening ready-to-pickup times.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"184 ","pages":"Article 105509"},"PeriodicalIF":7.6,"publicationDate":"2026-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145927801","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-reservoir traffic dynamics: Outflow network MFD and state estimation with sparse traffic data","authors":"Omid Mousavizadeh,&nbsp;Mehdi Keyvan-Ekbatani","doi":"10.1016/j.trc.2025.105504","DOIUrl":"10.1016/j.trc.2025.105504","url":null,"abstract":"<div><div>This study introduces an innovative framework for traffic state estimation in multi-reservoir networks, tackling the challenges posed by sparse data in urban traffic networks. By integrating Floating Car Data (FCD) and Loop Detector Data (LDD), the framework estimates traffic dynamics, such as inflows, outflows, transfer flows, and accumulation, without requiring detailed trip information or path flow distribution methods. A step forward in this study is a method for the estimation of internal outflow and transfer flows, enabling the direct estimation of outflow Network Macroscopic Fundamental Diagrams (outflow-NMFDs) which overcomes the shortcomings of outflow-NMFD estimation through the full set of sensors or indirect estimation via the production-NMFD. As argued, the outflow-NMFD is deemed the more robust foundation for the accumulation-based modelling in multi-reservoir systems. Accordingly, the estimated outflow-NMFD is used to build the accumulation-based model, in which the model outputs are combined with sparse real-time measurements, improving the model’s alignment with actual traffic conditions. The framework’s adaptability allows it to function effectively under varying levels of probe vehicle penetration, making it suitable for real-world scenarios where data availability can be inconsistent or limited. Simulation results validate the model’s robustness in capturing both steady-state and dynamic traffic behaviours, maintaining high accuracy even during abrupt demand changes. Furthermore, the framework performs reliably under stochastic conditions, demonstrating its resilience to daily traffic fluctuations. By reducing dependence on widespread sensor deployment across the network or its boundaries, this cost-effective approach offers a practical solution for real-time traffic monitoring and management in multi-reservoir systems, even when the boundaries of the system are not fixed in the network.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"184 ","pages":"Article 105504"},"PeriodicalIF":7.6,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145927800","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}