Spatio-Temporal Demand Prediction for Food Delivery Using Attention-Driven Graph Neural Networks

2025-07-21

AI Summary

Urban food delivery faces significant forecasting challenges due to highly heterogeneous and dynamically fluctuating spatiotemporal demand patterns. To address this, we propose Attention-driven Spatio-Temporal Graph Neural Network (ASTGNN), a novel framework that models urban regions as dynamic graph nodes. ASTGNN integrates multi-scale temporal convolution with an adaptive spatial attention mechanism to explicitly capture non-stationary spatial dependencies and complex temporal evolution. Furthermore, it employs a dynamic neighborhood weighting strategy to enhance responsiveness to sudden order surges. Extensive experiments on multiple real-world food-delivery datasets demonstrate that ASTGNN consistently outperforms state-of-the-art baselines, achieving 12.6%–18.3% lower MAE. The model exhibits high prediction accuracy, strong generalizability across diverse urban settings, and excellent scalability. These properties make ASTGNN particularly suitable for supporting critical operational decisions—including proactive fleet pre-allocation, optimal resource configuration, and real-time order dispatching—in large-scale urban logistics systems.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Graph-based Machine Learning
• Search and Optimization: Algorithm Configuration

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Accurate demand forecasting is critical for enhancing the efficiency and responsiveness of food delivery platforms, where spatial heterogeneity and temporal fluctuations in order volumes directly influence operational decisions. This paper proposes an attention-based Graph Neural Network framework that captures spatial-temporal dependencies by modeling the food delivery environment as a graph. In this graph, nodes represent urban delivery zones, while edges reflect spatial proximity and inter-regional order flow patterns derived from historical data. The attention mechanism dynamically weighs the influence of neighboring zones, enabling the model to focus on the most contextually relevant areas during prediction. Temporal trends are jointly learned alongside spatial interactions, allowing the model to adapt to evolving demand patterns. Extensive experiments on real-world food delivery datasets demonstrate the superiority of the proposed model in forecasting future order volumes with high accuracy. The framework offers a scalable and adaptive solution to support proactive fleet positioning, resource allocation, and dispatch optimization in urban food delivery operations.

Problem

• Predict food delivery demand with spatial-temporal accuracy
• Model urban zones as graphs for demand forecasting
• Optimize fleet and resource allocation using neural networks

Innovation

• Attention-based Graph Neural Network framework
• Dynamic weighting of neighboring zones influence
• Joint learning of spatial-temporal demand patterns