Joint Estimation and Prediction of City-wide Delivery Demand: A Large Language Model Empowered Graph-based Learning Approach
This work addresses a pressing issue in urban logistics for e-commerce and delivery services by enabling better demand forecasting and cross-city generalization, though it is incremental as it builds on existing graph-based and LLM methods.
The paper tackles the joint estimation and prediction of city-wide delivery demand and generalization to new cities by proposing a transferable graph-based spatiotemporal learning approach that integrates large language models for geospatial knowledge, achieving significant improvements in accuracy, efficiency, and transferability over state-of-the-art baselines on datasets from eight cities in China and the US.
The proliferation of e-commerce and urbanization has significantly intensified delivery operations in urban areas, boosting the volume and complexity of delivery demand. Data-driven predictive methods, especially those utilizing machine learning techniques, have emerged to handle these complexities in urban delivery demand management problems. One particularly pressing issue that has yet to be sufficiently addressed is the joint estimation and prediction of city-wide delivery demand, as well as the generalization of the model to new cities. To this end, we formulate this problem as a transferable graph-based spatiotemporal learning task. First, an individual-collective message-passing neural network model is formalized to capture the interaction between demand patterns of associated regions. Second, by exploiting recent advances in large language models (LLMs), we extract general geospatial knowledge encodings from the unstructured locational data using the embedding generated by LLMs. Last, to encourage the cross-city generalization of the model, we integrate the encoding into the demand predictor in a transferable way. Comprehensive empirical evaluation results on two real-world delivery datasets, including eight cities in China and the US, demonstrate that our model significantly outperforms state-of-the-art baselines in accuracy, efficiency, and transferability.