No One-Model-Fits-All: Uncovering Spatio-Temporal Forecasting Trade-offs with Graph Neural Networks and Foundation Models

2025-11-07

Proceedings of the 12th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation

AI Summary

Existing IoT spatiotemporal forecasting approaches lack joint optimization of sampling frequency, spatial coverage, and model architecture. Method: We systematically evaluate VAR, GRU, Transformer, STGNN, and foundational time-series models (Chronos, TimesFM, Moirai) under varying sensor densities and sampling rates using real-world wireless sensor network temperature data. Results: Moirai achieves superior performance in high-frequency, dense-sensor regimes by autonomously capturing cross-sensor dependencies; STGNN demonstrates greater robustness under sparse deployments and moderate sampling rates; TSFMs exhibit significant performance degradation as spatial coverage decreases. Our analysis uncovers the intrinsic coupling among spatial granularity, temporal resolution, and model inductive bias, leading to the design principle that “no single model is universally optimal” for spatiotemporal forecasting. All code and datasets are publicly released.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

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: Location- and context-aware Web and WoT applications and services
• Web Mining and Content Analysis: Models for Web evolution

Abstract

Modern IoT deployments for environmental sensing produce high volume spatiotemporal data to support downstream tasks such as forecasting, typically powered by machine learning models. While existing filtering and strategic deployment techniques optimize collected data volume at the edge, they overlook how variations in sampling frequencies and spatial coverage affect downstream model performance. In many forecasting models, incorporating data from additional sensors denoise predictions by providing broader spatial contexts. This interplay between sampling frequency, spatial coverage and different forecasting model architectures remain underexplored. This work presents a systematic study of forecasting models - classical models (VAR), neural networks (GRU, Transformer), spatio-temporal graph neural networks (STGNNs), and time series foundation models (TSFMs: Chronos Moirai, TimesFM) under varying spatial sensor nodes density and sampling intervals using real-world temperature data in a wireless sensor network. Our results show that STGNNs are effective when sensor deployments are sparse and sampling rate is moderate, leveraging spatial correlations via encoded graph structure to compensate for limited coverage. In contrast, TSFMs perform competitively at high frequencies but degrade when spatial coverage from neighboring sensors is reduced. Crucially, the multivariate TSFM Moirai outperforms all models by natively learning cross-sensor dependencies. These findings offer actionable insights for building efficient forecasting pipelines in spatio-temporal systems. All code for model configurations, training, dataset, and logs are open-sourced for reproducibility: https://github.com/UIUC-MONET-Projects/Benchmarking-Spatiotemporal-Forecast-Models

Problem

• Investigating how sampling frequency and spatial coverage affect forecasting model performance
• Analyzing trade-offs between different model architectures under varying sensor network conditions
• Evaluating classical models, neural networks, STGNNs and foundation models for spatiotemporal forecasting

Innovation

• STGNNs leverage graph structure for sparse sensor data
• TSFMs excel at high-frequency forecasting with dense sensors
• Moirai TSFM learns cross-sensor dependencies for best performance