Real-Time Matching and Dispatching for Urban Freight Transportation: A Hierarchical Reinforcement Learning Through Actor-Critic and H3 Spatial Partitioning
Bibliographic record
Abstract
Real-time freight matching and dispatching present complex challenges due to the dynamic nature of supply and demand, high-dimensional decision spaces, and the need for rapid response under operational constraints. In this paper, we propose a novel Hierarchical Reinforcement Learning (HRL) framework that jointly optimizes matching and dispatching processes in freight transportation, offering enhanced responsiveness, modularity, and coordination under real-time constraints. To mitigate computational demands without sacrificing matching accuracy, we present two efficient pre-filtering algorithms, PAMA (Pre-filtering Algorithm for Matching Agent) and PADA (Pre-filtering Algorithm for Dispatching Agent), which enhance the H3 hexagonal geospatial partitioning system. A GIS-based simulation of freight flow in Montreal provides realistic validation. Experimental results show that our framework improves successful match rates by 1.73%, reduces vehicle idle time by 5.7%, and maintains low empty mileage. Compared to state-of-the-art baselines including Deep Q-Network and clustering-based methods, our HRL model achieves superior reward efficiency, scalability, and adaptability in dynamic freight environments. These findings underscore the potential of HRL to enhance the operational efficiency of smart freight platforms.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".