Transportation Research Part E-Logistics and Transportation Review最新文献

{"title":"Optimizing driver’s discount order acceptance strategies: A policy-improved deep deterministic policy gradient framework","authors":"Hanwen Dai ,&nbsp;Chang Gao ,&nbsp;Fang He ,&nbsp;Congyuan Ji ,&nbsp;Yanni Yang","doi":"10.1016/j.tre.2025.104628","DOIUrl":"10.1016/j.tre.2025.104628","url":null,"abstract":"<div><div>The rapid expansion of platform integration has emerged as an effective solution to mitigate market fragmentation by consolidating multiple ride-hailing platforms into a single application. To address heterogeneous passenger preferences, third-party integrators provide Discount Express service delivered by express drivers at lower trip fares. For the individual platform, encouraging broader participation of drivers in Discount Express services has the potential to expand the accessible demand pool and improve matching efficiency, but often at the cost of reduced profit margins. This study aims to dynamically manage drivers’ acceptance of Discount Express from the perspective of an individual platform, incorporating the spatiotemporal demand-supply patterns. The lack of historical data under the new business model necessitates online learning. However, early-stage exploration through trial and error can be costly in practice, highlighting the need for reliable early-stage performance in real-world deployment. To address these challenges, this study formulates the decision regarding the proportion of drivers accepting discount orders as a continuous control task. In response to the high stochasticity, the opaque matching mechanisms employed by third-party integrator, and the limited availability of historical data, we propose an innovative policy-improved deep deterministic policy gradient (pi-DDPG) framework. The proposed framework incorporates a refiner module to boost policy performance during the early training phase, leverages a convolutional long short-term memory network to effectively capture complex spatiotemporal patterns, and adopts a prioritized experience replay mechanism to enhance learning efficiency. A customized simulator based on a real-world dataset is developed to validate the effectiveness of the proposed pi-DDPG. Numerical experiments demonstrate that pi-DDPG achieves superior learning efficiency and significantly reduces early-stage training losses, enhancing its applicability to practical ride-hailing scenarios.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"208 ","pages":"Article 104628"},"PeriodicalIF":8.8,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145940519","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":"Discriminatory order assignment and payment-setting of on-demand food-delivery platforms: A multi-action and multi-agent reinforcement learning framework","authors":"Zijian Zhao ,&nbsp;Sen Li","doi":"10.1016/j.tre.2025.104653","DOIUrl":"10.1016/j.tre.2025.104653","url":null,"abstract":"<div><div>This paper studies the discriminatory order assignment and payment-setting strategies for on-demand food-delivery platforms. We consider an on-demand food-delivery platform that coordinates customers, couriers, and restaurants to maximize the profit. It determines how to bundle orders, assign orders to couriers, and set payments to couriers in real-time. These decisions are made in a personalized manner, depending on the historical data collected from each of the couriers, such as the order acceptance and rejection rates under distinct scenarios of order assignment and payment values. A Markov Decision Process is formulated for the courier, capturing the decisions of the platform (including differentiated order assignment/bundling strategies and the discriminatory payment-settings decisions) while considering its dependence on the personalized work-related data of each individual courier. To derive the optimal policies, we propose a novel multi-action and multi-agent deep reinforcement learning framework, where a double Deep Q-Network is employed to develop discrete order assignment strategies, and double Proximal Policy Optimization is utilized to determine continuous payment decisions. Within this learning framework, we introduce a novel neural network architecture that leverages the Query-Key attention mechanism to transform multiplicative time complexities into additive computation complexity for order assignment, and we adopt a variable-length Bi-LSTM module that compresses variable-length order sequence into a fixed-dimensional feature space to enhance scalability. The proposed neural network and algorithmic framework was validated in a case study using real-world food-delivery data from Hong Kong. By comparing the proposed method with a vanilla MLP-based neural network architecture, we find that the proposed neural network architecture significantly enhances platform performance: it increases the number of orders served by 5.25%, reduces platform expenses by 10%, and improves the overall reward of the platform by over 50%. Additionally, our results reveal that couriers with higher order rejection rates receive more orders during peak hours but earn lower wages. This counterintuitive finding is attributed to a strategic approach by the platform to differentiate order allocation: instead of simply allocating fewer orders to couriers with higher rejection rates, the platform preferentially assigns longer-distance trips to couriers with a higher likelihood of order acceptance. These findings expose the implicit biases in the discriminatory algorithms used by the profit-maximizing platform and highlight potential areas for governmental regulatory intervention. The code of this paper is provided at <span><span>https://github.com/RS2002/Discriminatory-Food-Delivery</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"208 ","pages":"Article 104653"},"PeriodicalIF":8.8,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145940520","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":"Inventory-constrained online learning for revenue management with delayed feedback","authors":"Sheng Ji","doi":"10.1016/j.tre.2025.104649","DOIUrl":"10.1016/j.tre.2025.104649","url":null,"abstract":"<div><div>Delayed feedback is a prevalent challenge in modern logistics and transportation systems, especially on digital retail platforms. This paper investigates an online learning and pricing problem characterized by aggregated and anonymous delays. In this setting, neither demand nor revenue is immediately observable following a pricing decision; instead, these metrics become available to the retailer only after some stochastic delay. The retailer also faces an initial inventory constraint, creating a complex exploration-exploitation trade-off among learning demand, generating revenue, and managing inventory. To address this challenge, we propose a novel batch-based learning algorithm, referred to as Bandits with Dual Mirror Descent (BUD for short), which integrates mirror descent with bandit control. The algorithm employs a carefully designed batch structure to isolate the impact of delayed feedback, while combining Upper Confidence Bound (UCB) for pricing with dual updates for inventory management. Our theoretical analysis shows that the regret (defined as the revenue gap between the optimal policy and the learning algorithm) of BUD grows sublinearly with the selling horizon and matches the known lower bounds in both bandit with delays and online pricing problems. We conducted numerical experiments to demonstrate that the regret of BUD converges to 0 in various scenarios.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"208 ","pages":"Article 104649"},"PeriodicalIF":8.8,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145940516","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Joint optimization of flood water routing and congestion-aware evacuation scheduling","authors":"Sina Bahrami ,&nbsp;Mehdi Nourinejad ,&nbsp;Matthew J. Roorda ,&nbsp;Yafeng Yin","doi":"10.1016/j.tre.2025.104645","DOIUrl":"10.1016/j.tre.2025.104645","url":null,"abstract":"<div><div>Urban flood emergencies pose significant risks to human safety and infrastructure operability, particularly in smart cities with interdependent systems. This study proposes an integrated optimization model for coordinating water and transportation networks during flood evacuations. The model simultaneously determines optimal reservoir discharge rates and dynamic vehicular evacuation schedules to maximize the number of evacuees within the limited warning time. Water flow is modeled using the Muskingum-Cunge flood-routing method to simulate flood propagation through a river-reservoir system, while traffic flow is captured via the Cell Transmission Model, which accounts for congestion dynamics and road capacities. The problem is formulated as a nonlinear program and solved through a linear relaxation using generalized Benders decomposition. A case study of the Town of High River, Canada, illustrates the model’s practical utility. Results show that the integrated strategy extends warning times, reduces congestion, and lowers the number of individuals exposed to flood risks compared to uncoordinated approaches. By enabling real-time, infrastructure-aware evacuation planning, the proposed framework offers a scalable decision-support tool for emergency managers. This work contributes to the growing body of research on the management of city infrastructures under disruption and supports the development of resilient and coordinated evacuation strategies in smart urban environments.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"208 ","pages":"Article 104645"},"PeriodicalIF":8.8,"publicationDate":"2026-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145886032","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":"Offline operations strategies of bike-sharing platforms: pure profit or beyond profit?","authors":"Dongliang Guo ,&nbsp;Zhi-Ping Fan ,&nbsp;Minghe Sun","doi":"10.1016/j.tre.2025.104644","DOIUrl":"10.1016/j.tre.2025.104644","url":null,"abstract":"<div><div>Bike-sharing platforms can adopt two different offline operations strategies, i.e., independent operations (Strategy I), where each platform independently manages its offline operations, and outsourcing (Strategy O), where one platform outsources its offline operations to another competing platform. With the rise of the “beyond profit” management doctrine, many bike-sharing platforms have begun to pursue dual purposes, i.e., both profits and consumer surpluses, instead of the single purpose, i.e., “pure profit”. Given these facts, this work examines the equilibrium offline operations strategies of two bike-sharing platforms in a duopoly market based on the Hotelling framework and analyzes the platform profits and consumer surplus when the platforms pursue a single purpose or dual purposes. Several important results are obtained. When the platforms engage in intensive competition, the equilibrium operations strategy of the two platforms is Strategy O, and pursuing dual purposes can harm their respective profits. Under weak platform competition, both the investment synergy effect and the investment efficiency of offline operations can significantly affect the platform equilibrium offline operations strategies, and the platforms can obtain higher profits when pursuing dual purposes than pursuing a single purpose if they give low attention weightings to consumer surplus. Additionally, consumer surplus can always be higher when the platforms pursue dual purposes than when pursuing a single purpose, but Pareto improvement may be achieved by the platforms and consumers regardless of the platform competition intensity and the adoption of Strategy I or O.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"208 ","pages":"Article 104644"},"PeriodicalIF":8.8,"publicationDate":"2026-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145876989","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":"Spatiotemporal Features-Aware relocating for idle vehicles using spatial mean field deep Q network reinforcement learning","authors":"Zhiju Chen ,&nbsp;Kai Liu ,&nbsp;Jiangbo Wang ,&nbsp;Jintao Ke","doi":"10.1016/j.tre.2025.104651","DOIUrl":"10.1016/j.tre.2025.104651","url":null,"abstract":"<div><div>The cruising behavior of idle ride-hailing vehicles in search of passengers is a key influencing factor that restricts the spatiotemporal balance between online ride-hailing supply and passenger demands. This paper aims to simulate the strategy of transferring idle vehicles in multiple hexagonal partitions to adjacent grid partitions by proposing a spatiotemporal features-aware relocating approach (STFAR) that integrates spatiotemporal features of ride hailing into deep reinforcement learning. Specifically, spatial clustering algorithm and time series clustering algorithm are used to identify the spatiotemporal pattern of ride-hailing demand in each hexagonal partition. In addition, the direction of central hot spot is determined by accurately predicting the future short-term travel demand of each hexagonal partition. Finally, a spatial mean field deep Q network (SMFDQN) reinforcement learning method which regards the hexagonal partition as limited and fixed numbers spatial multi-agents is proposed to optimize the efficiency of idle vehicle transfer. STFAR improves the SMFDQN method by integrating the above spatiotemporal features into state space and action space designs and effectively improves the supply and demand balance in the entire region. Experiments based on Didi Chuxing order data during a certain time period in Chengdu showed that STFAR increases the cumulative order revenue by 3.64%, increases the completion rate of demand by 4.03%, and increases the dispatched rate of idle vehicles by 2.98% compared with the state-of-the-art algorithms.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"208 ","pages":"Article 104651"},"PeriodicalIF":8.8,"publicationDate":"2026-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145876990","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 flood fighting problem: A basic model and construction heuristics","authors":"Karolin Eisele,&nbsp;Alf Kimms","doi":"10.1016/j.tre.2025.104636","DOIUrl":"10.1016/j.tre.2025.104636","url":null,"abstract":"<div><div>Natural disasters such as floods occur more and more frequently due to climate change and claim many victims. If protective measures such as floodplains and dams are not sufficient or are damaged, emergency services must be deployed. In order to be able to deploy them as effectively as possible, we present a model for emergency services planning in the event of flooding. The mathematical model is based on the idea that the area of interest is subdivided into cells and snapshots of the situation are considered at discrete time periods. This way, we can model the spread of water over time taking the specific profile of the terrain into account. Also, the locations and the movement of the emergency teams can be described with user–specified granularity. Since solving such models optimally is out of the scope of today’s computational capabilities, we discuss several variants of so–called construction heuristics. Such methods run fast and produce results that help to assess a flood situation and about what can be achieved over time by fighting the floods. Such insights may not only help after the occurrence of an event, but also in advance in order to be prepared better. In a computational study the performance of heuristics based in simple priority rules is studied.</div></div>","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"208 ","pages":"Article 104636"},"PeriodicalIF":8.8,"publicationDate":"2026-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145886031","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":"","authors":"","doi":"","DOIUrl":"","url":null,"abstract":"","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"207 ","pages":"Article 104650"},"PeriodicalIF":8.8,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146469862","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":"","authors":"","doi":"","DOIUrl":"","url":null,"abstract":"","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"209 ","pages":"Article 104762"},"PeriodicalIF":8.8,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146498821","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":"","authors":"","doi":"","DOIUrl":"","url":null,"abstract":"","PeriodicalId":49418,"journal":{"name":"Transportation Research Part E-Logistics and Transportation Review","volume":"209 ","pages":"Article 104700"},"PeriodicalIF":8.8,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146498830","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}