Siyuan Feng

Ride-hailing platforms face significant challenges in optimizing order dispatching and driver repositioning operations in dynamic urban environments. Traditional approaches based on combinatorial optimization, rule-based heuristics, and reinforcement learning often overlook driver income fairness, interpretability, and adaptability to real-world dynamics. To address these gaps, we propose LLM-ODDR, a novel framework leveraging Large Language Models (LLMs) for joint Order Dispatching and Driver Repositioning (ODDR) in ride-hailing services. LLM-ODDR framework comprises three key components: (1) Multi-objective-guided Order Value Refinement, which evaluates orders by considering multiple objectives to determine their overall value; (2) Fairness-aware Order Dispatching, which balances platform revenue with driver income fairness; and (3) Spatiotemporal Demand-Aware Driver Repositioning, which optimizes idle vehicle placement based on historical patterns and projected supply. We also develop JointDR-GPT, a fine-tuned model optimized for ODDR tasks with domain knowledge. Extensive experiments on real-world datasets from Manhattan taxi operations demonstrate that our framework significantly outperforms traditional methods in terms of effectiveness, adaptability to anomalous conditions, and decision interpretability. To our knowledge, this is the first exploration of LLMs as decision-making agents in ride-hailing ODDR tasks, establishing foundational insights for integrating advanced language models within intelligent transportation systems. While the current framework incurs higher computational costs than traditional methods, we show that parallel decomposition and model distillation can reduce latency to production-viable levels for deployment.

Engineering structures are essential in the current society, and generating accurate three‐dimensional Digital Twins (3D DTs) is crucial for their effective Structural Health Monitoring (SHM). However, challenges arise due to the complex geometry and details of large‐scale engineering structures. Accordingly, a Computer Vision (CV) method named Adaptive‐Weight 3D Gaussian Splatting (AWGS) was proposed. The method employs an adaptive weighting strategy to balance reconstruction objectives and utilizes 3D smoothing and two‐dimensional (2D) Mip filters to suppress artifacts caused by inconsistent multiview sampling. The performance of AWGS was validated on two practical engineering structures: Gerrard Hall in the United States and an extra‐large concrete bridge in China. Results show that AWGS generates highly realistic 3D models with accurately preserved structural details. Furthermore, under the fixed training configurations, AWGS exhibited empirically stable sampled GPU‐memory trajectories while maintaining high reconstruction fidelity. These measurements characterize the tested implementation rather than an explicitly enforced algorithmic memory bound. Additionally, robustness analysis indicates that even under severe noise and occlusion, AWGS maintains structural coherence and achieves stable performance. These capabilities are expected to provide significant technical advancements for further intelligent inspection and SHM applications.

Autonomous mobility-on-demand (AMoD) systems face persistent challenges due to the spatio-temporal mismatch between vehicle supply and passenger demand, which results in low fulfillment rates and inefficient fleet utilization. Existing repositioning strategies primarily follow two paradigms. Region-level approaches direct idle vehicles to high-demand areas using coarse-grained policies but often fail to provide effective guidance within the target region. In contrast, route-level methods offer fine-grained control by generating paths on the road network, yet they frequently lack global planning and overlook broader supply-demand dynamics. To address the limitations of both paradigms, we propose a novel top-to-bottom repositioning (T2BR) framework that hierarchically integrates decision-making at multiple levels. At the regional level, reinforcement learning is employed to optimize inter-regional movements of idle vehicles based on long-term platform objectives. At the route level, Monte Carlo Tree Search is utilized to generate context-aware paths that facilitate efficient passenger pickups within target regions. This hierarchical structure allows for dynamic, adaptive, and spatially coordinated repositioning decisions. Comprehensive evaluations using real-world operational data from Manhattan demonstrate that the proposed T2BR framework significantly improves key performance metrics, including order fulfillment rate, platform revenue, and vehicle utilization, when compared to existing baseline methods. These results highlight the effectiveness of our approach in enhancing the operational efficiency of AMoD systems.

This paper introduces Localized Bipartite Match Graph Attention Q-Learning (BMG-Q), a novel Multi-Agent Reinforcement Learning (MARL) algorithm framework tailored for ride-pooling order dispatch. BMG-Q advances ride-pooling decision-making process with the localized bipartite match graph underlying the Markov Decision Process, enabling the development of novel Graph Attention Double Deep Q Network (GATDDQN) as the MARL backbone to capture the dynamic interactions among ride-pooling vehicles in fleet. Our approach enriches the state information for each agent with GATDDQN by leveraging a localized bipartite interdependence graph and enables a centralized global coordinator to optimize order matching and agent behavior using Integer Linear Programming (ILP). Enhanced by gradient clipping and localized graph sampling, our GATDDQN improves scalability and robustness. Furthermore, the inclusion of a posterior score function in the ILP captures the online exploration-exploitation trade-off and reduces the potential overestimation bias of agents, thereby elevating the quality of the derived solutions. Through extensive experiments and validation, BMG-Q has demonstrated superior performance in both training and operations for thousands of vehicle agents, outperforming benchmark reinforcement learning frameworks by around 10% in accumulative rewards and showing a significant reduction in overestimation bias by over 50%. Additionally, it maintains robustness amidst task variations and fleet size changes, establishing BMG-Q as an effective, scalable, and robust framework for advancing ride-pooling order dispatch operations.

Since flight delay hurts passengers, airlines, and airports, its prediction becomes crucial for the decision-making of all stakeholders in the aviation industry and thus has been attempted by various previous research. However, previous delay predictions are often categorical and at a highly aggregated level. To improve that, this study proposes to apply the novel Temporal Fusion Transformer model and predict numerical airport arrival delays at quarter hour level for U.S. top 30 airports. Inputs to our model include airport demand and capacity forecasts, historic airport operation efficiency information, airport wind and visibility conditions, as well as en-route weather and traffic conditions. The results show that our model achieves satisfactory performance measured by small prediction errors on the test set. In addition, the interpretability analysis of the model outputs identifies the important input factors for delay prediction.

The on-demand ride-pooling service, defined as two or more passengers sharing the same vehicle en-route along a proportion of their travel trajectories, offers many benefits, such as discounted trip fares for customers, higher income for drivers, increased profit for ride-sourcing companies, and reduced fuel consumption for environmental protection. Motivated by the potential of ride-pooling, many ride-sourcing companies launch pooling services based on the non-pooling market. By providing pooling and non-pooling services simultaneously, they compete with public transit for passengers. Of particular interest to service providers is the upfront pricing problem for the pooling service. It allows pooling riders to be informed of service prices even before the trip starts, and consequently makes the pooling service more attractive to passengers. However, it remains a challenging issue to obtain the optimal spatial-temporal upfront pricing strategy for the pooling service, considering the heterogeneity, dynamics, imbalance of demand /supply, and differentiation between pooling and non-pooling services. To address this problem, two reinforcement learning frameworks (i.e., single-agent Markov Decision Process (MDP) and multi-agent Markov Decision Process (MMDP)) are implemented to gain the pricing policy with the maximum daily profit of the platform, where the pooling price, as the action, not only directly affects the profit of each pooling request, but also has an influence on the mode splitting among pooling, non-pooling, and public transit service. Two tailored reinforcement learning methods are developed and adopted to solve the MDPs. Through extensive empirical experiments with a well-designed simulator, we show that the proposed multi-agent framework is able to remarkably improve the system performance.