Pub Date : 2025-11-22DOI: 10.1016/j.trc.2025.105453
Hongxiang Zhang , Andrea D’Ariano , Yongqiu Zhu , Yaoxin Wu , Liuyang Hu , Gongyuan Lu
The train platforming schedule is the crucial plan for guiding trains to travel through a railway station without spatial and temporal conflicts. When trains are delayed in arriving at the station due to disturbances or disruptions, it raises the Train Platforming and Rescheduling Problem (TPRP), one of the hot topics in railway traffic management. It focuses on allocating platforms and time slots for trains to reduce delays and ensure operational efficiency in a station. This paper introduces a novel graph neural network based deep reinforcement learning method to address this problem, named Learning to Reschedule Platforms (L2RP). We formulate the solving process of TPRP as a customized Markov decision process. Meanwhile, we integrate a microscopic discrete-event train operation simulation model to serve as the agent exploration environment, which provides states, executes actions, and completes transitions. Then, we design a hybrid graph neural network based policy network to derive high-quality actions under each graph encoded state.
The policy network is trained with the reward function designed to minimize total train knock-on delays and platform changes. The experiments on real-world instances show that the proposed L2RP method can produce high-quality solutions for instances of various scenarios within stably short solving times.
{"title":"Learning to reschedule platforms: A graph neural network based deep reinforcement learning method for the train platforming and rescheduling problem⁎","authors":"Hongxiang Zhang , Andrea D’Ariano , Yongqiu Zhu , Yaoxin Wu , Liuyang Hu , Gongyuan Lu","doi":"10.1016/j.trc.2025.105453","DOIUrl":"10.1016/j.trc.2025.105453","url":null,"abstract":"<div><div>The train platforming schedule is the crucial plan for guiding trains to travel through a railway station without spatial and temporal conflicts. When trains are delayed in arriving at the station due to disturbances or disruptions, it raises the Train Platforming and Rescheduling Problem (TPRP), one of the hot topics in railway traffic management. It focuses on allocating platforms and time slots for trains to reduce delays and ensure operational efficiency in a station. This paper introduces a novel graph neural network based deep reinforcement learning method to address this problem, named Learning to Reschedule Platforms (L2RP). We formulate the solving process of TPRP as a customized Markov decision process. Meanwhile, we integrate a microscopic discrete-event train operation simulation model to serve as the agent exploration environment, which provides states, executes actions, and completes transitions. Then, we design a hybrid graph neural network based policy network to derive high-quality actions under each graph encoded state.</div><div>The policy network is trained with the reward function designed to minimize total train knock-on delays and platform changes. The experiments on real-world instances show that the proposed L2RP method can produce high-quality solutions for instances of various scenarios within stably short solving times.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"183 ","pages":"Article 105453"},"PeriodicalIF":7.6,"publicationDate":"2025-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145575243","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}
Pub Date : 2025-11-22DOI: 10.1016/j.trc.2025.105451
Yimeng Zhang , Liang Huang , Jiaci Wang , He Lin , Shuyang Zhu , Mi Gan , Xiaobo Liu , Ruixue Ai
We introduce a data-driven approach for optimizing maritime logistics by integrating the construction of maritime transport networks with the routing of the ship fleet. Utilizing data mining techniques, this approach identifies crucial nodes and routes from Automatic Identification System (AIS) data and builds a directed weighted transport network. Based on the obtained transport network, we then optimize ships’ routes using Mixed Integer Programming and Adaptive Large Neighborhood Search. This data-driven method provides a complete solution that improves maritime logistics from data mining to route optimization and enhances the operational autonomy of both autonomous and traditional ships. The results using real-world AIS data illustrate how data mining can be leveraged to develop a detailed transport network that significantly enhances fleet routing optimization. We evaluate our approach against two benchmarks in the literature and demonstrate that it enhances identification accuracy by over 14 %. Furthermore, through numerical analyses under various scenarios, such as route disruptions and varying levels of port congestion, our routing approach proves capable of managing large-scale operations and adapting to transport time variations. Compared to disruptions on routes, severe port congestion notably increases operational costs as it extends loading and unloading times and causes higher delay penalties.
{"title":"Data-driven optimization for maritime logistics: integrating transport network mining with ship fleet routing","authors":"Yimeng Zhang , Liang Huang , Jiaci Wang , He Lin , Shuyang Zhu , Mi Gan , Xiaobo Liu , Ruixue Ai","doi":"10.1016/j.trc.2025.105451","DOIUrl":"10.1016/j.trc.2025.105451","url":null,"abstract":"<div><div>We introduce a data-driven approach for optimizing maritime logistics by integrating the construction of maritime transport networks with the routing of the ship fleet. Utilizing data mining techniques, this approach identifies crucial nodes and routes from Automatic Identification System (AIS) data and builds a directed weighted transport network. Based on the obtained transport network, we then optimize ships’ routes using Mixed Integer Programming and Adaptive Large Neighborhood Search. This data-driven method provides a complete solution that improves maritime logistics from data mining to route optimization and enhances the operational autonomy of both autonomous and traditional ships. The results using real-world AIS data illustrate how data mining can be leveraged to develop a detailed transport network that significantly enhances fleet routing optimization. We evaluate our approach against two benchmarks in the literature and demonstrate that it enhances identification accuracy by over 14 %. Furthermore, through numerical analyses under various scenarios, such as route disruptions and varying levels of port congestion, our routing approach proves capable of managing large-scale operations and adapting to transport time variations. Compared to disruptions on routes, severe port congestion notably increases operational costs as it extends loading and unloading times and causes higher delay penalties.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"183 ","pages":"Article 105451"},"PeriodicalIF":7.6,"publicationDate":"2025-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145567486","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}
Pub Date : 2025-11-21DOI: 10.1016/j.trc.2025.105450
Hyunjoon Kim, Stephane Barde
The Orienteering Problem with Time Windows (OPTW) is a complex combinatorial optimization problem with applications in logistics, tourist route planning, and emergency services. Traditional methods for solving OPTW, including metaheuristics, often struggle with scalability, adaptability, and generalization to new instances. Recently, deep reinforcement learning (DRL) has shown promise in tackling routing problems. However, existing DRL methods typically rely on non-Markovian state representations and handcrafted masking rules, which limit their adaptability and generalization. This paper presents Meta Pointer Network for OPTW (MetaPNet-OPTW), a meta-learning-enhanced DRL framework that combines a Markovian state formulation with OR-based feasibility rules within a pointer network model. We introduce the Meta-Learning enhanced REINFORCE algorithm, which learns across diverse problem instances and enables rapid adaptation to unseen configurations with minimal fine-tuning. During inference, active search with beam search is used to refine solutions dynamically. Extensive experiments show that MetaPNet-OPTW outperforms existing DRL approaches in efficiency and generalization, and notably improves 20 of 33 best-known solutions on the Gavalas benchmark. We further provide a t-SNE analysis of the learned latent space, enriched with spatio-temporal statistics, which explains why the model excels on Gavalas instances while identifying harder clusters such as r2 and c2. This study contributes a scalable DRL framework for OPTW that not only achieves state-of-the-art performance but also provides new interpretability into benchmark difficulty and model adaptability.
{"title":"A meta-learning enhanced deep reinforcement learning approach for generalizing across orienteering problem with time windows","authors":"Hyunjoon Kim, Stephane Barde","doi":"10.1016/j.trc.2025.105450","DOIUrl":"10.1016/j.trc.2025.105450","url":null,"abstract":"<div><div>The Orienteering Problem with Time Windows (OPTW) is a complex combinatorial optimization problem with applications in logistics, tourist route planning, and emergency services. Traditional methods for solving OPTW, including metaheuristics, often struggle with scalability, adaptability, and generalization to new instances. Recently, deep reinforcement learning (DRL) has shown promise in tackling routing problems. However, existing DRL methods typically rely on non-Markovian state representations and handcrafted masking rules, which limit their adaptability and generalization. This paper presents Meta Pointer Network for OPTW (MetaPNet-OPTW), a meta-learning-enhanced DRL framework that combines a Markovian state formulation with OR-based feasibility rules within a pointer network model. We introduce the Meta-Learning enhanced REINFORCE algorithm, which learns across diverse problem instances and enables rapid adaptation to unseen configurations with minimal fine-tuning. During inference, active search with beam search is used to refine solutions dynamically. Extensive experiments show that MetaPNet-OPTW outperforms existing DRL approaches in efficiency and generalization, and notably improves 20 of 33 best-known solutions on the <em>Gavalas</em> benchmark. We further provide a t-SNE analysis of the learned latent space, enriched with spatio-temporal statistics, which explains why the model excels on <em>Gavalas</em> instances while identifying harder clusters such as <em>r2</em> and <em>c2</em>. This study contributes a scalable DRL framework for OPTW that not only achieves state-of-the-art performance but also provides new interpretability into benchmark difficulty and model adaptability.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"183 ","pages":"Article 105450"},"PeriodicalIF":7.6,"publicationDate":"2025-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145567502","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}
Pub Date : 2025-11-20DOI: 10.1016/j.trc.2025.105439
Kai Zhang , Zhiyuan Liu , Honggang Zhang , Yicheng Zhang , Yuk Ming Tang , Xiaowen Fu
Traffic assignment is an essential component of the traditional four-step transportation planning methodology and significantly contributes to the prediction of traffic flow distribution and optimization of traffic planning. Existing algorithms for solving the user equilibrium traffic assignment problem typically rely on equal intervals and random sampling strategies to divide a set of origin–destination (OD) pairs. However, these sampling strategies fail to address the path overlap issue among OD pairs and often depend on sensitivity analyses to partition the OD set, hindering the efficiency of task parallelism. To address this challenge, the OD grouping problem was formulated as a vertex-coloring problem, which was translated into an integer linear programming (ILP) model. The largest degree first algorithm was proposed to solve the OD grouping problem, enabling the identification of OD pairs within each block with minimal path overlap. Thereafter, the results of the OD grouping based on vertex coloring were incorporated into the parallel block coordinate descent (PBCD) method, increasing the number of OD subproblems within each block and enhancing the parallel computation. An adaptive algorithm is further proposed to address the OD-based restricted subproblem depending on the number of paths for a given OD pair. The proposed method is evaluated based on various large-scale transportation networks and compared with existing algorithms, demonstrating its effectiveness in reducing path overlap within blocks and improving the efficiency of solving traffic assignment problems in large-scale networks.
{"title":"A graph vertex-coloring-based parallel block coordinate descent method for solving the traffic assignment problem","authors":"Kai Zhang , Zhiyuan Liu , Honggang Zhang , Yicheng Zhang , Yuk Ming Tang , Xiaowen Fu","doi":"10.1016/j.trc.2025.105439","DOIUrl":"10.1016/j.trc.2025.105439","url":null,"abstract":"<div><div>Traffic assignment is an essential component of the traditional four-step transportation planning methodology and significantly contributes to the prediction of traffic flow distribution and optimization of traffic planning. Existing algorithms for solving the user equilibrium traffic assignment problem typically rely on equal intervals and random sampling strategies to divide a set of origin–destination (OD) pairs. However, these sampling strategies fail to address the path overlap issue among OD pairs and often depend on sensitivity analyses to partition the OD set, hindering the efficiency of task parallelism. To address this challenge, the OD grouping problem was formulated as a vertex-coloring problem, which was translated into an integer linear programming (ILP) model. The largest degree first algorithm was proposed to solve the OD grouping problem, enabling the identification of OD pairs within each block with minimal path overlap. Thereafter, the results of the OD grouping based on vertex coloring were incorporated into the parallel block coordinate descent (PBCD) method, increasing the number of OD subproblems within each block and enhancing the parallel computation. An adaptive algorithm is further proposed to address the OD-based restricted subproblem depending on the number of paths for a given OD pair. The proposed method is evaluated based on various large-scale transportation networks and compared with existing algorithms, demonstrating its effectiveness in reducing path overlap within blocks and improving the efficiency of solving traffic assignment problems in large-scale networks.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"183 ","pages":"Article 105439"},"PeriodicalIF":7.6,"publicationDate":"2025-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145553973","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}
Pub Date : 2025-11-19DOI: 10.1016/j.trc.2025.105428
Yancheng Ling , Zhenlin Qin , Zhenliang Ma
In an era of unprecedented data availability and increasingly complex transportation systems, there is a pressing need for computational paradigms that can unify cross-disciplinary knowledge and systematically deduce new hypotheses. Knowledge graphs (KGs) provides a powerful approach in organizing and connecting fragmented evidence from multiple disciplines into a single, holistic analysis framework for deep scientific discoveries. The challenge is to automate the KG construction process by integrating diverse data sources and, importantly, harmonizing fragmented, incomplete, or even contradictory evidence that arises from multiple domains. Large language models (LLMs), trained in extensive corpora from multidisciplinary data, serve as a vast knowledge repository with advanced cognitive and reasoning capabilities. LLMs lends great opportunity to automate the KGs’ construction and expansion with transdisciplinary data integration and harmonization capabilities. To facilitate the quick adoption of KGs and LLMs in transportation, this paper presents a comprehensive review of LLMs for the construction of KGs with a particular focus on methodological development, including classifications, definitions, and challenges of KG construction tasks, and methodological pipelines and techniques using LLMs for these tasks. Building on these, we propose a LLM-driven pipeline for ontological transportation KG construction harmonizing un-/structured data across disciplines and generic purpose KGs. The graph evolves iteratively through an adaptive graph-refinement process, enabling updates with new findings and data while ensuring logical consistency and theoretical coherence. The transportation ontology system provides the structural backbone for the process, ensuring knowledge alignment to maintain semantic consistency across domains. Finally, we summarize the challenges from aspects of data quality, model capability, and computational costs, and outline future research directions. The study advances the use of LLMs for KG-based knowledge representation, facilitating automated discoveries and innovations in transportation.
{"title":"A review of knowledge graph construction using large language models in transportation: Problems, methods, and challenges","authors":"Yancheng Ling , Zhenlin Qin , Zhenliang Ma","doi":"10.1016/j.trc.2025.105428","DOIUrl":"10.1016/j.trc.2025.105428","url":null,"abstract":"<div><div>In an era of unprecedented data availability and increasingly complex transportation systems, there is a pressing need for computational paradigms that can unify cross-disciplinary knowledge and systematically deduce new hypotheses. Knowledge graphs (KGs) provides a powerful approach in organizing and connecting fragmented evidence from multiple disciplines into a single, holistic analysis framework for deep scientific discoveries. The challenge is to automate the KG construction process by integrating diverse data sources and, importantly, harmonizing fragmented, incomplete, or even contradictory evidence that arises from multiple domains. Large language models (LLMs), trained in extensive corpora from multidisciplinary data, serve as a vast knowledge repository with advanced cognitive and reasoning capabilities. LLMs lends great opportunity to automate the KGs’ construction and expansion with transdisciplinary data integration and harmonization capabilities. To facilitate the quick adoption of KGs and LLMs in transportation, this paper presents a comprehensive review of LLMs for the construction of KGs with a particular focus on methodological development, including classifications, definitions, and challenges of KG construction tasks, and methodological pipelines and techniques using LLMs for these tasks. Building on these, we propose a LLM-driven pipeline for ontological transportation KG construction harmonizing un-/structured data across disciplines and generic purpose KGs. The graph evolves iteratively through an adaptive graph-refinement process, enabling updates with new findings and data while ensuring logical consistency and theoretical coherence. The transportation ontology system provides the structural backbone for the process, ensuring knowledge alignment to maintain semantic consistency across domains. Finally, we summarize the challenges from aspects of data quality, model capability, and computational costs, and outline future research directions. The study advances the use of LLMs for KG-based knowledge representation, facilitating automated discoveries and innovations in transportation.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"183 ","pages":"Article 105428"},"PeriodicalIF":7.6,"publicationDate":"2025-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145553976","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}
Pub Date : 2025-11-19DOI: 10.1016/j.trc.2025.105415
Aoyong Li , Yaotian Tan , Wei Zhang , Kai Wang , Xiaobo Qu
The RoboTaxi service is considered a groundbreaking mode of transportation, offering autonomous ride services to passengers. Once a travel request is placed, the passenger is picked up promptly and transported directly to their destination. Following a trip, the RoboTaxi can immediately serve another customer, remain idle, or relocate within the city to anticipate future demand. The operation of RoboTaxi systems involves two key processes: routing and rebalancing. Routing determines which vehicle will serve a passenger and in what order. Rebalancing involves moving idle vehicles to strategic locations to improve passenger satisfaction for future requests. This relies on short-term demand prediction to ensure efficient resource allocation. While existing research has primarily considered these components independently, this study integrates them into a comprehensive framework designed to improve operator profitability and passenger satisfaction. In addition, different prediction methods are adopted to examine the impact of prediction accuracy on optimization results. The results indicate that the proposed framework achieves an approximate 12 % increase in operator profits and a significant improvement in the acceptance ratio.
{"title":"An integrated framework of routing and rebalancing for RoboTaxi systems","authors":"Aoyong Li , Yaotian Tan , Wei Zhang , Kai Wang , Xiaobo Qu","doi":"10.1016/j.trc.2025.105415","DOIUrl":"10.1016/j.trc.2025.105415","url":null,"abstract":"<div><div>The RoboTaxi service is considered a groundbreaking mode of transportation, offering autonomous ride services to passengers. Once a travel request is placed, the passenger is picked up promptly and transported directly to their destination. Following a trip, the RoboTaxi can immediately serve another customer, remain idle, or relocate within the city to anticipate future demand. The operation of RoboTaxi systems involves two key processes: routing and rebalancing. Routing determines which vehicle will serve a passenger and in what order. Rebalancing involves moving idle vehicles to strategic locations to improve passenger satisfaction for future requests. This relies on short-term demand prediction to ensure efficient resource allocation. While existing research has primarily considered these components independently, this study integrates them into a comprehensive framework designed to improve operator profitability and passenger satisfaction. In addition, different prediction methods are adopted to examine the impact of prediction accuracy on optimization results. The results indicate that the proposed framework achieves an approximate 12 % increase in operator profits and a significant improvement in the acceptance ratio.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"183 ","pages":"Article 105415"},"PeriodicalIF":7.6,"publicationDate":"2025-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145536927","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}
Pub Date : 2025-11-17DOI: 10.1016/j.trc.2025.105385
Junyi Ji , Derek Gloudemans , Yanbing Wang , Gergely Zachár , William Barbour , Jonathan Sprinkle , Benedetto Piccoli , Daniel B. Work
Analyzing stop-and-go waves at the scale of miles and hours of data is an emerging challenge in traffic research. The past 5 years have seen an explosion in the availability of large-scale traffic data containing traffic waves and complex congestion patterns, making existing approaches unsuitable for repeatable and scalable analysis of traffic waves in these data. This paper makes a first step towards addressing this challenge by introducing an automatic and scalable stop-and-go wave identification method capable of capturing wave generation, propagation, dissipation, as well as bifurcation and merging, which have previously been observed only very rarely. Using a concise and simple critical-speed based definition of a stop-and-go wave, the proposed method identifies all wave boundaries that encompass spatio-temporal points where vehicle speed is below a chosen critical speed. The method is built upon a graph representation of the spatio-temporal points associated with stop-and-go waves, specifically wave front (start) points and wave tail (end) points, and approaches the solution as a graph component identification problem. It enables the measurement of wave properties at scale. The method is implemented in Python and demonstrated on a large-scale dataset, I-24 MOTION INCEPTION. Our results show insights on the complexity of traffic waves. Traffic waves can bifurcate and merge at a scale that has never been observed or described before. The clustering analysis of all the identified wave components reveals the different topological structures of traffic waves. We explored that the wave merge or bifurcation points can be explained by spatial features. The gallery of all the identified wave topologies is demonstrated at https://trafficwaves.github.io/.
{"title":"Scalable analysis of stop-and-go waves: Representation, measurements and insights","authors":"Junyi Ji , Derek Gloudemans , Yanbing Wang , Gergely Zachár , William Barbour , Jonathan Sprinkle , Benedetto Piccoli , Daniel B. Work","doi":"10.1016/j.trc.2025.105385","DOIUrl":"10.1016/j.trc.2025.105385","url":null,"abstract":"<div><div>Analyzing stop-and-go waves at the scale of miles and hours of data is an emerging challenge in traffic research. The past 5 years have seen an explosion in the availability of large-scale traffic data containing traffic waves and complex congestion patterns, making existing approaches unsuitable for repeatable and scalable analysis of traffic waves in these data. This paper makes a first step towards addressing this challenge by introducing an automatic and scalable stop-and-go wave identification method capable of capturing wave generation, propagation, dissipation, as well as bifurcation and merging, which have previously been observed only very rarely. Using a concise and simple critical-speed based definition of a stop-and-go wave, the proposed method identifies all wave boundaries that encompass spatio-temporal points where vehicle speed is below a chosen critical speed. The method is built upon a graph representation of the spatio-temporal points associated with stop-and-go waves, specifically wave front (start) points and wave tail (end) points, and approaches the solution as a graph component identification problem. It enables the measurement of wave properties at scale. The method is implemented in Python and demonstrated on a large-scale dataset, I-24 MOTION INCEPTION. Our results show insights on the complexity of traffic waves. Traffic waves can bifurcate and merge at a scale that has never been observed or described before. The clustering analysis of all the identified wave components reveals the different topological structures of traffic waves. We explored that the wave merge or bifurcation points can be explained by spatial features. The gallery of all the identified wave topologies is demonstrated at <span><span>https://trafficwaves.github.io/</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"182 ","pages":"Article 105385"},"PeriodicalIF":7.6,"publicationDate":"2025-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145554543","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}
Pub Date : 2025-11-17DOI: 10.1016/j.trc.2025.105455
Zhenyang Qiu, Xiaowei Hu, Shi An
This study proposes a robust collaborative scheduling approach to maximize resource utilization and reduce costs, thereby addressing the critical challenge of service-demand mismatch across multiple routes. This study focuses on multiple electric bus (EB) routes sharing a hub station. First, we propose a collaborative scheduling optimization model integrating EB configuration, timetable planning, and vehicle and charging scheduling. The model comprehensively considers upward and downward trips, passenger distributions across stations, various EB types, and battery decay rates. Second, by analyzing the interaction across stochastic inter-station speed, uncertain passenger demand, and battery discharge fluctuations, a robust relaxation form of the model is constructed, incorporating both stochastic programming and robust optimization. Finally, we design a hybrid heuristic solution algorithm based on genetic algorithms and neighborhood search and solve the integrated collaborative optimization problem by modularizing the departure interval optimization subproblem and combining the trip link. Case studies of Harbin bus routes demonstrate that the proposed model enhances operational efficiency and service quality while increasing the robustness of the scheduling scheme. Moreover, battery control strategy and the joint effect of stochastic variables significantly impact EB scheduling.
{"title":"Robust collaborative scheduling optimization for multiple electric bus routes under stochastic traffic conditions","authors":"Zhenyang Qiu, Xiaowei Hu, Shi An","doi":"10.1016/j.trc.2025.105455","DOIUrl":"10.1016/j.trc.2025.105455","url":null,"abstract":"<div><div>This study proposes a robust collaborative scheduling approach to maximize resource utilization and reduce costs, thereby addressing the critical challenge of service-demand mismatch across multiple routes. This study focuses on multiple electric bus (EB) routes sharing a hub station. First, we propose a collaborative scheduling optimization model integrating EB configuration, timetable planning, and vehicle and charging scheduling. The model comprehensively considers upward and downward trips, passenger distributions across stations, various EB types, and battery decay rates. Second, by analyzing the interaction across stochastic inter-station speed, uncertain passenger demand, and battery discharge fluctuations, a robust relaxation form of the model is constructed, incorporating both stochastic programming and robust optimization. Finally, we design a hybrid heuristic solution algorithm based on genetic algorithms and neighborhood search and solve the integrated collaborative optimization problem by modularizing the departure interval optimization subproblem and combining the trip link. Case studies of Harbin bus routes demonstrate that the proposed model enhances operational efficiency and service quality while increasing the robustness of the scheduling scheme. Moreover, battery control strategy and the joint effect of stochastic variables significantly impact EB scheduling.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"182 ","pages":"Article 105455"},"PeriodicalIF":7.6,"publicationDate":"2025-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145553999","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}
Given that autonomous vehicles operate in open-world, safety-critical environments, it is essential to rigorously assess their reliability, particularly under worst-case traffic scenarios, to ensure passenger safety. Most existing testing methods fail to adequately evaluate the impact of adversarial behaviors and neglect the compound errors introduced when the subject under test is incorporated into the test scene. To expedite testing and reveal potential vulnerabilities in AV algorithms, we propose a universal adversarial testing framework designed to generate worst-case traffic scenarios focused on prediction and planning, assessed from harm, ambiguity, and rarity perspectives. For harm, we apply noncooperative game theory to strategically disrupt the tested vehicle while ensuring the disruptions remain reasonable via an asymmetric risk field. For rarity and ambiguity, we encourage the adversarial agents to exhibit high levels of aleatoric and epistemic uncertainty by maximizing the k-nearest neighbor distance in the latent space of a surrogate predictor, thereby crafting conditions that diverge from conventional scenarios in the training set. Our adversarial traffic scene generation algorithm is evaluated on the Argoverse 2 dataset and further validated on the NGSIM dataset without requiring retraining. Through comparison with other testing methods and comprehensive ablation studies, we qualitatively and quantitatively demonstrate that our algorithm effectively, efficiently, and reasonably produces highly critical traffic scenarios for interactive AV planning, including optimization-based and learning-based autonomous driving algorithms.
{"title":"Adversarial traffic scene generation considering harm, rarity, and ambiguity for autonomous driving testing","authors":"Yiran Zhang , Shanhe Lou , Baichuan Lou , Haitao Zhang , Chen Lv","doi":"10.1016/j.trc.2025.105426","DOIUrl":"10.1016/j.trc.2025.105426","url":null,"abstract":"<div><div>Given that autonomous vehicles operate in open-world, safety-critical environments, it is essential to rigorously assess their reliability, particularly under worst-case traffic scenarios, to ensure passenger safety. Most existing testing methods fail to adequately evaluate the impact of adversarial behaviors and neglect the compound errors introduced when the subject under test is incorporated into the test scene. To expedite testing and reveal potential vulnerabilities in AV algorithms, we propose a universal adversarial testing framework designed to generate worst-case traffic scenarios focused on prediction and planning, assessed from harm, ambiguity, and rarity perspectives. For harm, we apply noncooperative game theory to strategically disrupt the tested vehicle while ensuring the disruptions remain reasonable via an asymmetric risk field. For rarity and ambiguity, we encourage the adversarial agents to exhibit high levels of aleatoric and epistemic uncertainty by maximizing the k-nearest neighbor distance in the latent space of a surrogate predictor, thereby crafting conditions that diverge from conventional scenarios in the training set. Our adversarial traffic scene generation algorithm is evaluated on the Argoverse 2 dataset and further validated on the NGSIM dataset without requiring retraining. Through comparison with other testing methods and comprehensive ablation studies, we qualitatively and quantitatively demonstrate that our algorithm effectively, efficiently, and reasonably produces highly critical traffic scenarios for interactive AV planning, including optimization-based and learning-based autonomous driving algorithms.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"182 ","pages":"Article 105426"},"PeriodicalIF":7.6,"publicationDate":"2025-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145528441","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}
Pub Date : 2025-11-14DOI: 10.1016/j.trc.2025.105436
Muqing Du , Dongyue Cun , Yu Gu , Anthony Chen
The multimodal traffic network equilibrium problem (MTNEP) is a classical problem that can be modeled as a combined modal split and traffic assignment (CMSTA) problem to include both route and mode consideration. Recent studies were devoted to explicitly considering the combined travel modes in the MTNEP, as a significant portion of the daily travels in the modern urban metropolis are realized using multiple modes. To address the challenges of enumerating combined modes in existing multimodal traffic equilibrium models, this study proposes a novel two-phase approach for characterizing the combined travel modes in a multimodal transportation network. It converts the multimodal transportation network structure into a two-layered network representation, in which the upper-level network captures the mode combinations between the origin/transfer/destination nodes. Based on the two-layered network, we conduct the CMSTA problem by adopting the network generalized extreme value (NGEV) model, which effectively captures both underlying mode similarity and path correlation without explicitly listing all possible combinations of modes and paths. The existence and uniqueness of the proposed model are demonstrated by formulating the MTNEP as a fixed-point problem. Experimental results verify the capability of the two-phase method to avoid same-mode transfers, generate reasonable multimodal routes, and improve convergence efficiency. Particularly, the results show that the two-phase method outperforms the one-phase method which conducts both mode demand and path flow equilibration of all combinations of combined modes directly on the supernetwork. Incorporating the Barzilai-Borwein (BB) step-size strategy, the two-phase method reduces computation time by 32% in the Sioux-Falls network and by 50% in the Anaheim network, while maintaining stable convergence across different network scales.
{"title":"A two-phase approach with a novel network representation for solving the multimodal traffic network equilibrium with multimode combinations","authors":"Muqing Du , Dongyue Cun , Yu Gu , Anthony Chen","doi":"10.1016/j.trc.2025.105436","DOIUrl":"10.1016/j.trc.2025.105436","url":null,"abstract":"<div><div>The multimodal traffic network equilibrium problem (MTNEP) is a classical problem that can be modeled as a combined modal split and traffic assignment (CMSTA) problem to include both route and mode consideration. Recent studies were devoted to explicitly considering the combined travel modes in the MTNEP, as a significant portion of the daily travels in the modern urban metropolis are realized using multiple modes. To address the challenges of enumerating combined modes in existing multimodal traffic equilibrium models, this study proposes a novel two-phase approach for characterizing the combined travel modes in a multimodal transportation network. It converts the multimodal transportation network structure into a two-layered network representation, in which the upper-level network captures the mode combinations between the origin/transfer/destination nodes. Based on the two-layered network, we conduct the CMSTA problem by adopting the network generalized extreme value (NGEV) model, which effectively captures both underlying mode similarity and path correlation without explicitly listing all possible combinations of modes and paths. The existence and uniqueness of the proposed model are demonstrated by formulating the MTNEP as a fixed-point problem. Experimental results verify the capability of the two-phase method to avoid same-mode transfers, generate reasonable multimodal routes, and improve convergence efficiency. Particularly, the results show that the two-phase method outperforms the one-phase method which conducts both mode demand and path flow equilibration of all combinations of combined modes directly on the supernetwork. Incorporating the Barzilai-Borwein (BB) step-size strategy, the two-phase method reduces computation time by 32% in the Sioux-Falls network and by 50% in the Anaheim network, while maintaining stable convergence across different network scales.</div></div>","PeriodicalId":54417,"journal":{"name":"Transportation Research Part C-Emerging Technologies","volume":"182 ","pages":"Article 105436"},"PeriodicalIF":7.6,"publicationDate":"2025-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145528440","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号