This work examines whether a vehicle-centered dynamic graph with route awareness with explicit route state improves ETA prediction, and presents DSTRA-GNN (Dynamic Spatio-Temporal Route-Aware GNN), which integrates vehicle-level dynamics and route intent in a Mixture-of-Experts framework.
A Spatio-Temporal Graph Convolutional Network (STGCN) for forecasting route-level travel times on a simulated Nashville, Tennessee road network with 1,037 junctions and 1,601 road segments is presented, demonstrating the potential of the model as an efficient surrogate for transportation resilience screening, incident...
Abhilasha J. Saroj, Natalie Myers, Hao-Ran Niu· 0 citations
This study presents a deep learning–enabled framework for real-time dynamic route optimization in logistics systems, addressing fundamental limitations of traditional static routing and heuristic-based decision approaches.
A. Rakha, Mohammed S. A. Elsersy· Informatica· 0 citations
This work shows that embedding physically meaningful structure into learning objectives is an effective strategy for traffic surrogate modeling, yielding models that maintain competitive predictive accuracy while substantially improving directional behavioral consistency.
Blessing Itoro Afolayan, Arka Ghosh, Santhanakrishnan Narayanan et al.· Communications in Transporta...· 0 citations
Results confirm that explicitly coupling traffic prediction with online trajectory replanning enables more adaptive and efficient navigation under time-varying traffic conditions.
Hassan Haghighi, Maamar El Amine Hamri, D. Asadi et al.· Data Science for Transportat...· 1 citation
A novel TSE method named HGCN-VA is proposed, enabling a real-time inductive inference based on a graph neural network (GNN), and extensive experiments demonstrate that HGCN-VA outperforms baselines in TSE accuracy.
Xiwen Lou, Zhengfeng Huang, Hang Yang et al.· Sustainability· 0 citations
Results across the reported 15 and 30 min settings indicate that event-conditioned topology and lag-aware heterogeneous attention can improve traffic-flow forecasting on this hybrid real–simulation benchmark, indicating potential for pilot-zone applications rather than confirming real-world deployment performance.
