Bridge: Retrieval-Augmented Spatiotemporal Modeling for Urban Delivery Demand

2026-05-18

AI Summary

This work addresses the cold-start demand forecasting challenge in newly added urban delivery areas, where historical data are scarce. The authors propose a retrieval-augmented spatiotemporal prediction framework that integrates an inductive spatiotemporal graph backbone with a time-aware memory bank to retrieve future demand patterns from contextually and temporally similar regions. A gated fusion mechanism adaptively combines retrieved information with local representations to enhance prediction accuracy. Notably, the framework introduces a novel future-oriented retrieval objective during training, aligning the retrieved content directly with the forecasting task—an approach that, for the first time, effectively extends retrieval augmentation to spatiotemporal cold-start scenarios. Experiments on four real-world delivery datasets demonstrate that the proposed method significantly outperforms existing approaches under both intra-city cold-start and cross-city partial-observation transfer settings.

Abstract

Forecasting urban delivery demand becomes substantially more challenging when newly added service regions lack historical records. Existing spatiotemporal forecasters effectively model spatial dependence once sufficient node histories are available. Still, they remain parametric and therefore struggle to recover short-term operational dynamics in cold-start regions. Geospatial embeddings help identify where a region is and what function it serves, yet they do not directly reveal how a similar region behaves under a comparable temporal context. We propose Bridge, a retrieval-augmented spatiotemporal graph framework that combines an inductive contextual graph backbone with a time-aware memory of region-time windows. For each target region, Bridge retrieves future demand patterns from the memory using both regional context and recent dynamics, and refines the backbone forecast through a gated fusion mechanism. To align retrieval with forecasting utility, we further train the retriever with a future-aware objective that favors entries whose future trajectories best match the target. Experiments on four real-world delivery datasets show that Bridge consistently improves over competitive spatiotemporal baselines in both within-city cold-start and cross-city transfer with partial observations. The results show that retrieval augmentation provides a useful operational memory for cold-start urban demand forecasting when parametric graph generalization alone is insufficient.

Problem

• urban delivery demand forecasting
• cold-start regions
• spatiotemporal modeling
• retrieval augmentation
• historical data scarcity

Innovation

• retrieval-augmented
• spatiotemporal forecasting
• cold-start
• graph neural networks
• urban delivery demand