arXiv subjects

Simon Hu

Publications and source records attributed to Simon Hu.

8 recordsLinked to original sources

arXiv2026

Metropolis-Scale Resilient and Trustworthy Traffic Flow Inference Using Multi-Source Data

Inferring network-wide traffic states from sparse observations with high accuracy and trustworthy uncertainty quantification is essential for intelligent transportation systems, yet it remains challenging due to the underdetermined nature of the problem, multifaceted disturbances in sensing networks, and the inherent conflicts among multiple inference sub-tasks when modeled jointly. We propose the Task-Aware Attentive Neural Process (TA-ANP), a unified probabilistic framework for resilient and trustworthy global traffic state inference (GTSI) by fusing floating car data (FCD) with sparse fixed-detector measurements. By casting GTSI as a stochastic process, TA-ANP leverages the meta-learning properties of neural processes to adapt rapidly to changes in sensing configurations without retraining. A task-aware multi-query attention module with distinct spatiotemporal inductive biases is introduced to jointly handle three GTSI sub-tasks, while mitigating cross-task interference. For uncertainty quantification, we combine neural processes with Monte Carlo Dropout to capture both aleatoric and epistemic uncertainty. To support metropolis-scale evaluation, we construct the Metropolitan Multi-Source Traffic Dataset (MMTD), integrating fixed-loop sensor measurements, FCD statistics, and OpenStreetMap road-network data over an urban network of 2,371 road segments. Experiments on MMTD show that TA-ANP achieves state-of-the-art performance across all sub-tasks under deterministic and probabilistic metrics. The resulting well-calibrated uncertainties enable more efficient fixed-sensor placement with fewer sensor deployments. Under a Damage-Repair-Addition sensing lifecycle, TA-ANP demonstrates superior resilience in terms of disturbance absorption, performance recovery, and adaptability to unseen sensing configurations.

cs.LG↗

arXiv2025

ORThought: Benchmarking and Automating Logistics Optimization Modeling

Optimization modeling stands as the engine of scientific decision-making in logistics and transportation, yet its adoption is hindered by a steep expertise threshold and the latency of manual workflows. Automating this process via Large Language Models (LLMs) offers a potential solution, but current approaches face critical bottlenecks: (i) a lack of high-quality, complex benchmarks; (ii) methodological inefficiencies in autonomous multi-agent frameworks, which often exhibit instability and redundant computation; and (iii) evaluations that lack diagnostic depth. In this work, we address these challenges from the following three aspects. First, we introduce LogiOR, a diverse logistics benchmark with rigorous annotations, and enrich existing datasets with the same annotation standard to support community utilization. Second, we propose ORThought, a structured dual-agent framework. By incorporating expert-level modeling principles via chain-of-thought reasoning, ORThought eliminates the redundancy of uncontrolled autonomous agents. Third, extensive empirical evaluations demonstrate that ORThought consistently outperforms state-of-the-art baselines by 9-17 percentage points, exhibiting distinct advantages in handling complex constraints while maintaining high token efficiency. Building on these results, we further conduct a multidimensional error analysis, which identifies key failure modes and success factors, providing actionable insights for future research. The dataset and code are available at https://huggingface.co/datasets/LabMem012/LogiOR and https://github.com/ZJU-TSELab/ORThought, respectively.

cs.AI↗

arXiv2025

Network-wide Freeway Traffic Estimation Using Sparse Sensor Data: A Dirichlet Graph Auto-Encoder Approach

Network-wide Traffic State Estimation (TSE), which aims to infer a complete image of network traffic states with sparsely deployed sensors, plays a vital role in intelligent transportation systems. With the development of data-driven methods, traffic dynamics modeling has advanced significantly. However, TSE poses fundamental challenges for data-driven approaches, since historical patterns cannot be learned locally at sensor-free segments. Although graph representation learning shows promise in estimating states at locations without sensors, existing methods typically handle unobserved locations by filling them with zeros, introducing bias to the sensitive graph message propagation. The recently proposed Dirichlet Energy-based Feature Propagation (DEFP) method achieves State-Of-The-Art (SOTA) performance in unobserved node classification by eliminating the need for zero-filling. However, applying it to TSE faces three key challenges: inability to handle directed traffic networks, strong assumptions in traffic spatial correlation modeling, and overlooking distinct propagation rules of different patterns (e.g., congestion and free flow). We propose DGAE, a novel inductive graph representation model that addresses these challenges through theoretically derived DEFP for Directed graph (DEFP4D), enhanced spatial representation learning via DEFP4D-guided latent space encoding, and physics-guided propagation mechanisms that separately handle congested and free-flow patterns. Experiments on three traffic datasets demonstrate that DGAE outperforms existing SOTA methods and exhibits strong cross-city transferability. Furthermore, DEFP4D can serve as a standalone lightweight solution, showing superior performance under extremely sparse sensor conditions. The code of this work is publicly available at: https://github.com/ZJU-TSELab/DGAE.

cs.LG↗

arXiv2025

MoGERNN: An Inductive Traffic Predictor for Unobserved Locations

Given a partially observed road network, how can we predict the traffic state of interested unobserved locations? Traffic prediction is crucial for advanced traffic management systems, with deep learning approaches showing exceptional performance. However, most existing approaches assume sensors are deployed at all locations of interest, which is impractical due to financial constraints. Furthermore, these methods are typically fragile to structural changes in sensing networks, which require costly retraining even for minor changes in sensor configuration. To address these challenges, we propose MoGERNN, an inductive spatio-temporal graph model with two key components: (i) a Mixture of Graph Experts (MoGE) with sparse gating mechanisms that dynamically route nodes to specialized graph aggregators, capturing heterogeneous spatial dependencies efficiently; (ii) a graph encoder-decoder architecture that leverages these embeddings to capture both spatial and temporal dependencies for comprehensive traffic state prediction. Experiments on two real-world datasets show MoGERNN consistently outperforms baseline methods for both observed and unobserved locations. MoGERNN can accurately predict congestion evolution even in areas without sensors, offering valuable information for traffic management. Moreover, MoGERNN is adaptable to the changes of sensor network, maintaining competitive performance even compared to its retrained counterpart. Tests performed with different numbers of available sensors confirm its consistent superiority, and ablation studies validate the effectiveness of its key modules. The code of this work is publicly available at: https://github.com/ZJU-TSELab/MoGERNN.

cs.LG↗

arXiv2022

A Four-stage Heuristic Algorithm for Solving On-demand Meal Delivery Routing Problem

Meal delivery services provided by platforms with integrated delivery systems are becoming increasingly popular. This paper adopts a rolling horizon approach to solve the meal delivery routing problem (MDRP). To improve delivery efficiency in scenarios with high delivery demand, multiple orders are allowed to be combined into one bundle with orders from different restaurants. Following this strategy, an optimization-based four-stage heuristic algorithm is developed to generate an optimal routing plan at each decision point. The algorithm first generates bundles according to orders' spatial and temporal distribution. Secondly, we find feasible bundle pairs. Then, routes for delivering any single bundle or multiple bundles are optimized, respectively. Finally, the routes are assigned to available couriers. In computational experiments using instances from open datasets, the system's performance is evaluated in respect of average click-to-door time and ready-to-pickup time. We demonstrate that this algorithm can effectively process real-time information and assign optimal routes to the couriers. By comparing the proposed method with existing the-state-of-the-art algorithms, the results indicate that our method can generate solutions with higher service quality and shorter distance.

eess.SY↗

arXiv2020

Identifying Critical Fleet Sizes Using a Novel Agent-Based Modelling Framework for Autonomous Ride-Sourcing

Ride-sourcing platforms enable an on-demand shared transport service by solving decision problems often related to customer matching, pricing and vehicle routing. These problems have been frequently represented using aggregated mathematical models and solved via algorithmic approaches designed by researchers. The increasing complexity of ride-sourcing environments compromises the accuracy of aggregated methods. It, therefore, signals the need for alternative practices such as agent-based models which capture the level of complex dynamics in ride-sourcing systems. The use of these agent-based models to simulate ride-sourcing fleets has been a focal point of many studies; however, this occurred in the absence of a prescribed approach on how to build the models to mimic fleet operations realistically. To bridge this research gap, we provide a framework for building bespoke agent-based models for ride-sourcing fleets, derived from the fundamentals of agent-based modelling theory. We also introduce a model building sequence of the different modules necessary to structure a simulator based on our framework. To showcase the strength of our framework, we use it to tackle the highly non-linear problem of minimum fleet size estimation for autonomous ride-sourcing fleets. We do so by investigating the relationship of system parameters based on queuing theory principles and by deriving and validating a novel model for pickup wait times. By modelling the ride-sourcing fleet function in the urban areas of Manhattan, San Francisco, Paris and Barcelona, we find that ride-sourcing fleets operate queues with zero assignment times above the critical fleet size. We also show that pickup wait times have a pivotal role in the estimation of the minimum fleet size in ride-sourcing operations, with agent-based modelling to be a more reliable route for their identification given the system parameters.

eess.SY↗

arXiv2018

Day-to-day dynamic traffic assignment model with variable message signs and endogenous user compliance

This paper proposes a dual-time-scale, day-to-day dynamic traffic assignment model that takes into account variable message signs (VMS) and its interactions with drivers' travel choices and adaptive learning processes. The within-day dynamic is captured by a dynamic network loading problem with en route update of path choices influenced by the VMS; the day-to-day dynamic is captured by a simultaneous route-and-departure-time adjustment process that employs bounded user rationality. Moreover, we describe the evolution of the VMS compliance rate by modeling drivers' learning processes. We endogenize traffic dynamics, route and departure time choices, travel delays, and VMS compliance, and thereby captur their interactions and interdependencies in a holistic manner. A case study in the west end of Glasgow is carried out to understand the impact of VMS has on road congestion and route choices in both the short and long run. Our main findings include an adverse effect of the VMS on the network performance in the long run (the 'rebound' effect), and existence of an equilibrium state where both traffic and VMS compliance are stabilized.

math.OC↗

arXiv2018

Nonlinear Decision Rule Approach for Real-Time Traffic Signal Control for Congestion and Emission Reductions

We propose a real-time signal control framework based on a nonlinear decision rule (NDR), which defines a nonlinear mapping between network states and signal control parameters to actual signal controls based on prevailing traffic conditions, and such a mapping is optimized via off-line simulation. The NDR is instantiated with two neural networks: feedforward neural network (FFNN) and recurrent neural network (RNN), which have different ways of processing traffic information in the near past, and are compared in terms of their performances. The NDR is implemented within a microscopic traffic simulation (S-Paramics) for a real-world network in West Glasgow, where the off-line training of the NDR amounts to a simulation-based optimization aiming to reduce delay, CO2 and black carbon emissions. The emission calculations are based on the high-fidelity vehicle dynamics generated by the simulation, and the AIRE instantaneous emission model. Extensive tests are performed to assess the NDR framework, not only in terms of its effectiveness in reducing the aforementioned objectives, but also in relation to local vs. global benefits, trade-off between delay and emissions, impact of sensor locations, and different levels of network saturation. The results suggest that the NDR is an effective, flexible and robust way of alleviating congestion and reducing traffic emissions.

math.OC↗