This research introduces a novel graph-based approach empowered by large language models to jointly estimate and predict delivery demand across entire cities, addressing growing challenges in urban logistics.

• Combines LLM capabilities with graph neural networks to process complex urban delivery patterns
• Enables more accurate city-wide demand forecasting for delivery operations
• Creates a unified framework that handles both estimation and prediction tasks simultaneously
• Provides practical solutions for optimizing delivery resources in complex urban environments

For Engineering teams, this approach offers a significant advancement in managing delivery systems at scale, allowing for more efficient resource allocation and improved operational planning in increasingly complex urban logistics networks.

Joint Estimation and Prediction of City-wide Delivery Demand: A Large Language Model Empowered Graph-based Learning Approach