Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
April 29, 2026Transportation Science

Data-Driven Optimization for Meal Delivery: A Reinforcement Learning Approach for Order-Courier Assignment and Routing at Meituan

View Full Paper
Ask AI
Bookmark
Share

Authors

RARamon AuadFLFelipe LagosTLTomás Lagos

Discussion

Loading...

Member takes

Overview

Hybrid framework improves operational efficiency by 12% through reinforcement learning in meal delivery, indicating enhanced order-routing strategies.

Key Points

  • To improve meal delivery logistics using a hybrid reinforcement learning approach for order assignment and routing.
  • Integrated reinforcement learning with hyper-heuristic optimization
  • Developed a simulation environment using real data from Meituan
  • Employed n-step state-action-reward-state-action with value function approximation
  • Achieved 12% cost reduction through strategic order postponement
  • Largest improvements noted during high-demand periods with courier shortages
  • A 10% increase in courier availability is more beneficial than algorithmic enhancements alone

Cite This Study

Auad et al. (2026) studied this question.

synapsesocial.com/papers/69f154f9879cb923c4945403https://doi.org/10.1287/trsc.2025.0129
View Full Paper
Ask AI
Bookmark
Share