Joint Estimation and Prediction of City-wide Delivery Demand: A Large Language Model Empowered Graph-based Learning Approach
arXiv:2408.17258 · doi:10.1016/j.tre.2025.104075
Abstract
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.
References in corpus (29)
- Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- LLaMA: Open and Efficient Foundation Language Models
- Large Language Models are Zero-Shot Reasoners
- SAITS: Self-Attention-based Imputation for Time Series
- A Time Series is Worth 64 Words: Long-term Forecasting with Transformers
- Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
- TSMixer: An All-MLP Architecture for Time Series Forecasting
- Dynamic Spatiotemporal Graph Convolutional Neural Networks for Traffic Data Imputation with Complex Missing Patterns
- ImputeFormer: Low Rankness-Induced Transformers for Generalizable Spatiotemporal Imputation
- Kriging Convolutional Networks
- Towards General Text Embeddings with Multi-stage Contrastive Learning
- Deep trip generation with graph neural networks for bike sharing system expansion
- Language Models Represent Space and Time
- Correlating sparse sensing for large-scale traffic speed estimation: A Laplacian-enhanced low-rank tensor kriging approach
- Filling the G_ap_s: Multivariate Time Series Imputation by Graph Neural Networks
- GeoLLM: Extracting Geospatial Knowledge from Large Language Models
- INCREASE: Inductive Graph Representation Learning for Spatio-Temporal Kriging
- Learning to Reconstruct Missing Data from Spatiotemporal Graphs with Sparse Observations
- Where Would I Go Next? Large Language Models as Human Mobility Predictors
- Spatial Aggregation and Temporal Convolution Networks for Real-time Kriging
- Large Language Models are Geographically Biased
- Large Language Models as Urban Residents: An LLM Agent Framework for Personal Mobility Generation
- Multivariate Time Series Forecasting with Latent Graph Inference
- Spatiotemporal Implicit Neural Representation as a Generalized Traffic Data Learner
- Large Language Models for Mobility Analysis in Transportation Systems: A Survey on Forecasting Tasks
- Taming Local Effects in Graph-based Spatiotemporal Forecasting
- LaDe: The First Comprehensive Last-mile Delivery Dataset from Industry
- Graph-based Virtual Sensing from Sparse and Partial Multivariate Observations
- Data Mining in Transportation Networks with Graph Neural Networks: A Review and Outlook