Electrified-Autonomous Flexible Transit (E-AFT) represents a promising paradigm for on-demand mobility, necessitating the integration of routing and energy management to ensure viable operations. This study develops a two-stage optimization model for dynamic vehicle routing and charging scheduling, formulated as a Mixed-Integer Nonlinear Programming (MINLP) framework designed to maximize overall system profit. In the first stage, an Adaptive Large Neighborhood Search (ALNS) algorithm determines routes to maximize operation profit, with energy consumption and time constraints explicitly linking to the second stage Variable Neighborhood Search (VNS) which optimizes charging schedules to minimize total charging costs. This sequential ALNS-VNS procedure is embedded within a Rolling Horizon Control (RHC) strategy, effectively tackling the computational challenges of large-scale, real-time demand through iterative subproblem resolution. Validation using real-world urban network case studies demonstrates the model’s effectiveness: the ALNS-VNS approach achieves near-optimal solutions with superior computational efficiency, and the RHC framework reveals the significant impact of horizon interval and battery capacity on service reliability and economic feasibility, offering valuable insights for E-AFT system design.
{"title":"Two-stage optimization approach for dynamic routing and charging scheduling in electrified-autonomous flexible transit","authors":"Haoran Jiang , Shaozhi Hong , Kenan Zhang , Jian Yuan , Qing Yu","doi":"10.1016/j.tre.2025.104600","DOIUrl":"10.1016/j.tre.2025.104600","url":null,"abstract":"<div><div>Electrified-Autonomous Flexible Transit (E-AFT) represents a promising paradigm for on-demand mobility, necessitating the integration of routing and energy management to ensure viable operations. This study develops a two-stage optimization model for dynamic vehicle routing and charging scheduling, formulated as a Mixed-Integer Nonlinear Programming (MINLP) framework designed to maximize overall system profit. In the first stage, an Adaptive Large Neighborhood Search (ALNS) algorithm determines routes to maximize operation profit, with energy consumption and time constraints explicitly linking to the second stage Variable Neighborhood Search (VNS) which optimizes charging schedules to minimize total charging costs. This sequential ALNS-VNS procedure is embedded within a Rolling Horizon Control (RHC) strategy, effectively tackling the computational challenges of large-scale, real-time demand through iterative subproblem resolution. Validation using real-world urban network case studies demonstrates the model’s effectiveness: the ALNS-VNS approach achieves near-optimal solutions with superior computational efficiency, and the RHC framework reveals the significant impact of horizon interval and battery capacity on service reliability and economic feasibility, offering valuable insights for E-AFT system design.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"207 ","pages":"Article 104600"},"PeriodicalIF":8.8,"publicationDate":"2025-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145823197","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 : 2025-12-23DOI: 10.1016/j.tre.2025.104634
Biao Chen , Songhua Hu , Xiangdong Xu , Guangchao Wang , Der-Horng Lee , Zhengbing He
The multi-modal urban transportation network serves as a cornerstone in fostering urban economic vitality and facilitating daily human travel. However, its vulnerability to emergencies, including natural disasters, infrastructure breakdowns, and terrorist attacks, will cause severe disruptions, resulting in profound societal and economic losses. A resilient urban transportation system thus becomes critical for maintaining city-wide and societal stability. This paper initiates a thorough investigation within the framework of “network modeling − resilience evaluation − resilience optimization”, with particular attention to emerging modes such as shared mobility and urban air mobility. This review is structured around three dimensions: 1) Transportation network modeling, which provides a comprehensive survey of methods for mono-modal and multi-modal urban transportation network in constructing network models and analyzing network topological properties through the lens of the complex network theory; 2) Network resilience and evaluation, which summarizes the definitions and evaluation methods of transportation network resilience, with particular emphasis on the unique aspects and evaluation complexities specific to multi-modal networks; 3) Resilience optimization, which synthesizes strategies for resilience optimization considering network structure, facility maintenance, and system operation. It underscores the benefits of cooperative operation within multi-modal transportation systems on resilience optimization. In each section, the paper provides a comparative analysis of mono-modal versus multi-modal transportation networks across these three dimensions and critiques the advantages and disadvantages of existing approaches. The challenges inherent in current research and potential future research directions are also identified to strengthen the resilience of multi-modal urban transportation networks against a backdrop of increasing urban challenges.
{"title":"A survey of multi-modal urban transportation network resilience: modeling, evaluation, and optimization","authors":"Biao Chen , Songhua Hu , Xiangdong Xu , Guangchao Wang , Der-Horng Lee , Zhengbing He","doi":"10.1016/j.tre.2025.104634","DOIUrl":"10.1016/j.tre.2025.104634","url":null,"abstract":"<div><div>The multi-modal urban transportation network serves as a cornerstone in fostering urban economic vitality and facilitating daily human travel. However, its vulnerability to emergencies, including natural disasters, infrastructure breakdowns, and terrorist attacks, will cause severe disruptions, resulting in profound societal and economic losses. A resilient urban transportation system thus becomes critical for maintaining city-wide and societal stability. This paper initiates a thorough investigation within the framework of “network modeling − resilience evaluation − resilience optimization”, with particular attention to emerging modes such as shared mobility and urban air mobility. This review is structured around three dimensions: 1) Transportation network modeling, which provides a comprehensive survey of methods for mono-modal and multi-modal urban transportation network in constructing network models and analyzing network topological properties through the lens of the complex network theory; 2) Network resilience and evaluation, which summarizes the definitions and evaluation methods of transportation network resilience, with particular emphasis on the unique aspects and evaluation complexities specific to multi-modal networks; 3) Resilience optimization, which synthesizes strategies for resilience optimization considering network structure, facility maintenance, and system operation. It underscores the benefits of cooperative operation within multi-modal transportation systems on resilience optimization. In each section, the paper provides a comparative analysis of mono-modal versus multi-modal transportation networks across these three dimensions and critiques the advantages and disadvantages of existing approaches. The challenges inherent in current research and potential future research directions are also identified to strengthen the resilience of multi-modal urban transportation networks against a backdrop of increasing urban challenges.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"207 ","pages":"Article 104634"},"PeriodicalIF":8.8,"publicationDate":"2025-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145823195","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 : 2025-12-22DOI: 10.1016/j.tre.2025.104604
Yangchun Xiong , Geng Wang , Chunyu Xiu , Xinyue Wang , Hugo K.S. Lam , Rachel W.Y. Yee
It is unclear whether adopting virtual reality technologies in a firm’s manufacturing processes is a risky decision or serves to reduce risk. Our study answers this question by combining the Mahalanobis distance matching with the difference-in-differences (DID) analysis to quantify the effect of virtual reality-enabled manufacturing practices (VRMPs) on firm risk. We also consider how external market environments, in terms of market competition, dynamism, and munificence, influence the relationship between VRMPs and firm risk. Our DID analysis, based on 74 treatment firms that adopted VRMPs and 72 matched control firms without such adoption, suggests that VRMP adoption helps reduce firm risk. Moreover, the risk-reduction effect is more significant for firms operating in highly competitive and dynamic markets, but less so in the contexts of high market munificence. These results demonstrate VRMPs’ potential to reduce firm risk and highlight the important role of external market environments in shaping this relationship.
{"title":"How does virtual reality adoption affect firm risk? The role of external market environments","authors":"Yangchun Xiong , Geng Wang , Chunyu Xiu , Xinyue Wang , Hugo K.S. Lam , Rachel W.Y. Yee","doi":"10.1016/j.tre.2025.104604","DOIUrl":"10.1016/j.tre.2025.104604","url":null,"abstract":"<div><div>It is unclear whether adopting virtual reality technologies in a firm’s manufacturing processes is a risky decision or serves to reduce risk. Our study answers this question by combining the Mahalanobis distance matching with the difference-in-differences (DID) analysis to quantify the effect of virtual reality-enabled manufacturing practices (VRMPs) on firm risk. We also consider how external market environments, in terms of market competition, dynamism, and munificence, influence the relationship between VRMPs and firm risk. Our DID analysis, based on 74 treatment firms that adopted VRMPs and 72 matched control firms without such adoption, suggests that VRMP adoption helps reduce firm risk. Moreover, the risk-reduction effect is more significant for firms operating in highly competitive and dynamic markets, but less so in the contexts of high market munificence. These results demonstrate VRMPs’ potential to reduce firm risk and highlight the important role of external market environments in shaping this relationship.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"207 ","pages":"Article 104604"},"PeriodicalIF":8.8,"publicationDate":"2025-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145813836","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 : 2025-12-22DOI: 10.1016/j.tre.2025.104631
Yu Xia , Bolin Wang , Yunlong Yang
Autonomous vehicles (AVs) have the potential to reshape the ride-hailing industry by lowering operating costs. While many platforms view AV introduction as a key strategic move, its profitability implications remain uncertain. This study develops a stylized game-theoretic model of two competing ride-hailing platforms to examine the strategic role of AV introduction. The model incorporates key market factors, including consumer-side and driver-side competition as well as the balance between supply and demand. Our analysis shows that the effects of AV introduction on platforms are determined by the trade-off between two concurrent effects: a vehicle supply effect, which alleviates competition in the driver market; a consumer competition effect, which intensifies competition in the consumer market. When only one platform introduces AVs, the competitor is not necessarily worse off if driver-market competition is fierce. When both platforms introduce AVs, profits may rise under intense driver competition but decline under strong consumer competition. Moreover, the strategic interaction in platforms’ choices regarding AV introduction may lead to a prisoner’s dilemma in which both platforms introduce AVs but end up worse off than in the absence of adoption. Finally, we highlight the first-mover advantage, showing that delayed AV introduction after a competitor can reduce profitability.
{"title":"For better or worse? The impacts of autonomous vehicles on competitive ride-hailing platforms","authors":"Yu Xia , Bolin Wang , Yunlong Yang","doi":"10.1016/j.tre.2025.104631","DOIUrl":"10.1016/j.tre.2025.104631","url":null,"abstract":"<div><div>Autonomous vehicles (AVs) have the potential to reshape the ride-hailing industry by lowering operating costs. While many platforms view AV introduction as a key strategic move, its profitability implications remain uncertain. This study develops a stylized game-theoretic model of two competing ride-hailing platforms to examine the strategic role of AV introduction. The model incorporates key market factors, including consumer-side and driver-side competition as well as the balance between supply and demand. Our analysis shows that the effects of AV introduction on platforms are determined by the trade-off between two concurrent effects: a vehicle supply effect, which alleviates competition in the driver market; a consumer competition effect, which intensifies competition in the consumer market. When only one platform introduces AVs, the competitor is not necessarily worse off if driver-market competition is fierce. When both platforms introduce AVs, profits may rise under intense driver competition but decline under strong consumer competition. Moreover, the strategic interaction in platforms’ choices regarding AV introduction may lead to a prisoner’s dilemma in which both platforms introduce AVs but end up worse off than in the absence of adoption. Finally, we highlight the first-mover advantage, showing that delayed AV introduction after a competitor can reduce profitability.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"207 ","pages":"Article 104631"},"PeriodicalIF":8.8,"publicationDate":"2025-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145823203","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 : 2025-12-22DOI: 10.1016/j.tre.2025.104605
Zhiwei Yin, Bin Jia, Xiao-Yong Yan, Yitao Yang, Hao Ji, Ziyou Gao
{"title":"Corrigendum to “Classification of the freight trip purpose of heavy trucks using trajectory data and waybill data”. [Trans. Res. Part E: Logist. Trans. Rev. 206 (2026) 104584]","authors":"Zhiwei Yin, Bin Jia, Xiao-Yong Yan, Yitao Yang, Hao Ji, Ziyou Gao","doi":"10.1016/j.tre.2025.104605","DOIUrl":"https://doi.org/10.1016/j.tre.2025.104605","url":null,"abstract":"","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"2 1","pages":""},"PeriodicalIF":10.6,"publicationDate":"2025-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145813833","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 : 2025-12-21DOI: 10.1016/j.tre.2025.104601
Yongjian Yang , Chenglong Li , Dujuan Wang , Yunqiang Yin , T.C.E. Cheng
The facility location problem and routing problem, two critical components of humanitarian operations management, are integrated into the location-routing problem. Given that disasters are characterized by an exceptionally high degree of uncertainty, this study focuses on addressing supply-side uncertainties, such as the disruption probability of relief facilities, and receiver-side uncertainties, including demand fluctuations in affected areas and deadlines for receiving relief supplies. To develop a reliable location scheme that mitigates potential disruptions, a two-stage adaptive robust optimization formulation is constructed. It requires determining location, allocation, and routing plans across a set of disaster scenarios. A hybrid exact algorithm incorporating column-and-constraint generation, integer Benders decomposition, and branch-and-price algorithms is developed, along with advanced acceleration strategies to improve the solution process. Comprehensive numerical studies using randomly generated datasets evaluate the algorithm’s performance, demonstrating its superiority over both the CPLEX solver and an algorithm without column-and-constraint generation. Additionally, the study examines the influence of key model parameters on the solution structure and performance metrics. Furthermore, a real-world case study in Ya’an City, Sichuan Province, validates the model’s applicability, showing that it outperforms conventional deterministic and stochastic optimization models.
{"title":"A two-stage adaptive robust optimization model for the location-routing problem with drone delivery and uncertainty in humanitarian relief","authors":"Yongjian Yang , Chenglong Li , Dujuan Wang , Yunqiang Yin , T.C.E. Cheng","doi":"10.1016/j.tre.2025.104601","DOIUrl":"10.1016/j.tre.2025.104601","url":null,"abstract":"<div><div>The facility location problem and routing problem, two critical components of humanitarian operations management, are integrated into the location-routing problem. Given that disasters are characterized by an exceptionally high degree of uncertainty, this study focuses on addressing supply-side uncertainties, such as the disruption probability of relief facilities, and receiver-side uncertainties, including demand fluctuations in affected areas and deadlines for receiving relief supplies. To develop a reliable location scheme that mitigates potential disruptions, a two-stage adaptive robust optimization formulation is constructed. It requires determining location, allocation, and routing plans across a set of disaster scenarios. A hybrid exact algorithm incorporating column-and-constraint generation, integer Benders decomposition, and branch-and-price algorithms is developed, along with advanced acceleration strategies to improve the solution process. Comprehensive numerical studies using randomly generated datasets evaluate the algorithm’s performance, demonstrating its superiority over both the CPLEX solver and an algorithm without column-and-constraint generation. Additionally, the study examines the influence of key model parameters on the solution structure and performance metrics. Furthermore, a real-world case study in Ya’an City, Sichuan Province, validates the model’s applicability, showing that it outperforms conventional deterministic and stochastic optimization models.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"207 ","pages":"Article 104601"},"PeriodicalIF":8.8,"publicationDate":"2025-12-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145796025","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 : 2025-12-20DOI: 10.1016/j.tre.2025.104603
Chaojing Li , Minyu Shen , Qiaolin Hu , Li Zhen , Feng Xiao
Bus corridors serving multiple lines are critical in urban transportation but face challenges such as severe queuing at stops and irregular headways under high passenger demand. Existing literature primarily focuses on regulating headway and often neglects queuing issues, especially in multi-line corridors. In this study, we consider a congested multi-line bus corridor with severe queuing phenomena and propose a dynamic holding method based on deep reinforcement learning (DRL) method. The dynamic hold problem is formulated as a markov decision process (MDP), and we develop a graph-aware deep deterministic policy gradient (GADDPG) method to optimize bus holding strategies. GADDPG employs a graph-based state representation, utilizing graph attention network (GAT) to accommodate variable numbers of buses and capture their temporal relationships. This state representation relies solely on high-frequency GPS data, ensuring practicality. We evaluate our approach using real-world data from the Guangzhou BRT corridor. Our results demonstrate a significant finding: implementing holding control in a busy bus corridor yields dual benefits - it improves headway regularity while simultaneously reducing total bus delays. This reduction occurs because the introduced holding delay offsets and reduces more delays (queueing and in-berth delays), resulting in total bus delay savings. Additionally, our performance evaluation shows that the proposed GADDPG method outperforms benchmark holding control methods, achieving complete dominance on the Pareto frontier when optimizing for both headway regularity and reducing bus delays.
{"title":"Dynamic bus holding control for multi-line busy bus corridors: Mitigating bus queues and improving headway regularity via graph-aware deep reinforcement learning","authors":"Chaojing Li , Minyu Shen , Qiaolin Hu , Li Zhen , Feng Xiao","doi":"10.1016/j.tre.2025.104603","DOIUrl":"10.1016/j.tre.2025.104603","url":null,"abstract":"<div><div>Bus corridors serving multiple lines are critical in urban transportation but face challenges such as severe queuing at stops and irregular headways under high passenger demand. Existing literature primarily focuses on regulating headway and often neglects queuing issues, especially in multi-line corridors. In this study, we consider a congested multi-line bus corridor with severe queuing phenomena and propose a dynamic holding method based on deep reinforcement learning (DRL) method. The dynamic hold problem is formulated as a markov decision process (MDP), and we develop a graph-aware deep deterministic policy gradient (GADDPG) method to optimize bus holding strategies. GADDPG employs a graph-based state representation, utilizing graph attention network (GAT) to accommodate variable numbers of buses and capture their temporal relationships. This state representation relies solely on high-frequency GPS data, ensuring practicality. We evaluate our approach using real-world data from the Guangzhou BRT corridor. Our results demonstrate a significant finding: implementing holding control in a busy bus corridor yields dual benefits - it improves headway regularity while simultaneously reducing total bus delays. This reduction occurs because the introduced holding delay offsets and reduces more delays (queueing and in-berth delays), resulting in total bus delay savings. Additionally, our performance evaluation shows that the proposed GADDPG method outperforms benchmark holding control methods, achieving complete dominance on the Pareto frontier when optimizing for both headway regularity and reducing bus delays.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"207 ","pages":"Article 104603"},"PeriodicalIF":8.8,"publicationDate":"2025-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145786097","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 : 2025-12-20DOI: 10.1016/j.tre.2025.104635
Wenyuan Wang , Huakun Liu , Yun Peng , Zhen Cao , Pengxi Yu , Zanxin Lu
In container terminal yards, operational efficiency is significantly hindered by the mismatch between stochastic workload variations and static configurations of Yard Cranes (YCs) across multiple blocks. The real-time YC scheduling problem (YCSP-r) aims to develop adaptive, instantaneous scheduling policies that dynamically respond to workload fluctuations. In this paper, we propose a multi-agent reinforcement learning (RL) method to address the YCSP-r. Specifically, the YCSP-r is formulated as a Markov Decision Process (MDP) within an asynchronous timestep framework. Considering the non-negligible redeployment cost of YCs in real-time operations, the MDP is designed to balance redeployment costs with overall operational efficiency. A general simulator for the YC scheduling system is developed to execute action decisions and provides performance feedback. Proximal Policy Optimization (PPO) is employed to train the scheduling policy. A multi-agent shared-policy framework and a global–local mixed state structure is tailored to mitigate the challenges posed by high dimensional state and action spaces, thereby enhancing both convergence and training stability. To evaluate the solution quality, a mixed integer programming model for YCSP-r is developed and solved by a commercial solver as a benchmark for comparison. The proposed approach is further compared with other advanced RL and heuristic methods. Experimental results demonstrate that the proposed PPO-based approach is able to provide high-quality solutions in real time—typically within seconds—meeting the practical demands of container terminal operations. Notably, compared to a static YC deployment strategy, our scheduling strategy achieves a substantial 12.29% reduction in operational costs. We believe that our study provides valuable insights for port managers in developing practical and reliable YC scheduling solutions.
{"title":"Yard crane real-time scheduling among multi-block at terminal: A reinforcement learning based proximal policy optimization approach","authors":"Wenyuan Wang , Huakun Liu , Yun Peng , Zhen Cao , Pengxi Yu , Zanxin Lu","doi":"10.1016/j.tre.2025.104635","DOIUrl":"10.1016/j.tre.2025.104635","url":null,"abstract":"<div><div>In container terminal yards, operational efficiency is significantly hindered by the mismatch between stochastic workload variations and static configurations of Yard Cranes (YCs) across multiple blocks. The real-time YC scheduling problem (YCSP-r) aims to develop adaptive, instantaneous scheduling policies that dynamically respond to workload fluctuations. In this paper, we propose a multi-agent reinforcement learning (RL) method to address the YCSP-r. Specifically, the YCSP-r is formulated as a Markov Decision Process (MDP) within an asynchronous timestep framework. Considering the non-negligible redeployment cost of YCs in real-time operations, the MDP is designed to balance redeployment costs with overall operational efficiency. A general simulator for the YC scheduling system is developed to execute action decisions and provides performance feedback. Proximal Policy Optimization (PPO) is employed to train the scheduling policy. A multi-agent shared-policy framework and a global–local mixed state structure is tailored to mitigate the challenges posed by high dimensional state and action spaces, thereby enhancing both convergence and training stability. To evaluate the solution quality, a mixed integer programming model for YCSP-r is developed and solved by a commercial solver as a benchmark for comparison. The proposed approach is further compared with other advanced RL and heuristic methods. Experimental results demonstrate that the proposed PPO-based approach is able to provide high-quality solutions in real time—typically within seconds—meeting the practical demands of container terminal operations. Notably, compared to a static YC deployment strategy, our scheduling strategy achieves a substantial 12.29% reduction in operational costs. We believe that our study provides valuable insights for port managers in developing practical and reliable YC scheduling solutions.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"207 ","pages":"Article 104635"},"PeriodicalIF":8.8,"publicationDate":"2025-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145785138","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 : 2025-12-20DOI: 10.1016/j.tre.2025.104585
Christian Truden , Mike Hewitt
We present a method for sizing a vehicle fleet in operational contexts in which fleet performance is measured along multiple dimensions. One premise of the method is that decision-makers regarding fleet composition are interested in fleets that peform well across multiple seasons and in changing demand markets. Another premise is that daily operations are dynamic and stochastic in that vehicle routing decisions must be determined with incomplete information of future customer requests for service that day. A third premise is the existence of a solver for the daily dynamic planning problem that the method can use in a black-box fashion. Based on these premises, we present a heuristic framework for generating a predictive model of fleet performance. The method involves sampling operational settings across seasons and demand markets and executing a black-box planning tool for different fleet compositions to generate solutions and corresponding performance metric values. These settings and values are used to establish a training data set to which a prediction model is fitted. Once fitted, the decision-maker can use such a prediction model to quickly determine the fleet size and attributes that are likely to perform as desired on the performance measures of interest. To demonstrate the effectiveness of the proposed method we constructed a carefully curated data set from publicly available data sources to simulate the operational context of a grocery home delivery service. In that context, fleet vehicles can have multiple compartments to support the transportation of different food products that require storage at different temperate ranges. Thus, fleet sizing decisions involve both the number of vehicles and the size of each compartment within a vehicle. With an extensive computational study and analysis we illustrate that the proposed heuristic approach produces prediction models, both regression models and neural networks, that exhibit strong predictive power and can effectively inform fleet sizing decisions.
{"title":"Multi-objective, multi-attribute fleet sizing in a dynamic and stochastic environment: A data-driven approach","authors":"Christian Truden , Mike Hewitt","doi":"10.1016/j.tre.2025.104585","DOIUrl":"10.1016/j.tre.2025.104585","url":null,"abstract":"<div><div>We present a method for sizing a vehicle fleet in operational contexts in which fleet performance is measured along multiple dimensions. One premise of the method is that decision-makers regarding fleet composition are interested in fleets that peform well across multiple seasons and in changing demand markets. Another premise is that daily operations are dynamic and stochastic in that vehicle routing decisions must be determined with incomplete information of future customer requests for service that day. A third premise is the existence of a solver for the daily dynamic planning problem that the method can use in a black-box fashion. Based on these premises, we present a heuristic framework for generating a predictive model of fleet performance. The method involves sampling operational settings across seasons and demand markets and executing a black-box planning tool for different fleet compositions to generate solutions and corresponding performance metric values. These settings and values are used to establish a training data set to which a prediction model is fitted. Once fitted, the decision-maker can use such a prediction model to quickly determine the fleet size and attributes that are likely to perform as desired on the performance measures of interest. To demonstrate the effectiveness of the proposed method we constructed a carefully curated data set from publicly available data sources to simulate the operational context of a grocery home delivery service. In that context, fleet vehicles can have multiple compartments to support the transportation of different food products that require storage at different temperate ranges. Thus, fleet sizing decisions involve both the number of vehicles and the size of each compartment within a vehicle. With an extensive computational study and analysis we illustrate that the proposed heuristic approach produces prediction models, both regression models and neural networks, that exhibit strong predictive power and can effectively inform fleet sizing decisions.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"207 ","pages":"Article 104585"},"PeriodicalIF":8.8,"publicationDate":"2025-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145786096","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