Grid Partition-Based Dynamic Spatial–Temporal Graph Convolutional Network for Large-Scale Traffic Flow Forecasting

Gao, Lifeng; Chen, Liujia; Qiu, Agen; Wang, Qinglian; Wang, Jianlong; Chen, Cai; Zhang, Fuhao; Ou’er, Geli

doi:10.3390/ijgi14050207

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

Grid Partition-Based Dynamic Spatial–Temporal Graph Convolutional Network for Large-Scale Traffic Flow Forecasting

by

Lifeng Gao

^1,2

,

Liujia Chen

³,

Agen Qiu

^1,2,*

,

Qinglian Wang

^1,2

,

Jianlong Wang

⁴,

Cai Chen

⁵,

Fuhao Zhang

^1,2 and

Geli Ou’er

⁶

¹

Key Laboratory of Surveying and Mapping Science and Geospatial Information Technology of MNR, Chinese Academy of Surveying and Mapping (CASM), Beijing 100036, China

²

Chinese Academy of Surveying and Mapping (CASM), Beijing 100036, China

³

National Center for Public Credit and Geospatial Information, Beijing 100045, China

⁴

Changjiang Spatial Information Technology Engineering Co., Ltd., Wuhan 430010, China

⁵

School of Geomatics and Urban Spatial Informatics, Beijing University of Civil Engineering and Architecture, Beijing 102616, China

⁶

Geomatics Technology and Application Key Laboratory of Qinghai Province, Xining 810001, China

^*

Author to whom correspondence should be addressed.

ISPRS Int. J. Geo-Inf. 2025, 14(5), 207; https://doi.org/10.3390/ijgi14050207

Submission received: 22 January 2025 / Revised: 15 May 2025 / Accepted: 16 May 2025 / Published: 19 May 2025

(This article belongs to the Special Issue Advances in AI-Driven Geospatial Analysis and Data Generation (2nd Edition))

Download Versions Notes

Abstract

Accurate forecasting of city-level large-scale traffic flow is crucial for efficient traffic management and effective transport planning. However, previously proposed traffic flow prediction methods model dynamic spatial correlations across entire traffic networks, leading to high computational complexity, elevated memory usage, and model overfitting. Therefore, a novel grid partition-based dynamic spatial–temporal graph convolutional network was developed in this study to capture correlations within a large-scale traffic network. It includes the following: a dynamic graph convolution module to divide the traffic network into grid regions and thereby effectively capture the local spatial dependencies inherent in large-scale traffic topologies, an attention-based dynamic graph convolutional network to capture the local spatial correlations within each region, a global spatial dependency aggregation module to model inter-regional correlation weights using sequence similarity methods and comprehensively reflect the overall state of the traffic network, and multi-scale gated convolutions to capture both long- and short-term temporal correlations across varying time ranges. The performance of the proposed model was compared with that of different baseline models using two large-scale real-world datasets; the proposed model significantly outperformed the baseline models, demonstrating its potential effectiveness in managing large-scale traffic networks.

Keywords: graph convolutional network; spatiotemporal forecasting; large-scale; global-local spatial dependency; traffic flow

Share and Cite

MDPI and ACS Style

Gao, L.; Chen, L.; Qiu, A.; Wang, Q.; Wang, J.; Chen, C.; Zhang, F.; Ou’er, G. Grid Partition-Based Dynamic Spatial–Temporal Graph Convolutional Network for Large-Scale Traffic Flow Forecasting. ISPRS Int. J. Geo-Inf. 2025, 14, 207. https://doi.org/10.3390/ijgi14050207

AMA Style

Gao L, Chen L, Qiu A, Wang Q, Wang J, Chen C, Zhang F, Ou’er G. Grid Partition-Based Dynamic Spatial–Temporal Graph Convolutional Network for Large-Scale Traffic Flow Forecasting. ISPRS International Journal of Geo-Information. 2025; 14(5):207. https://doi.org/10.3390/ijgi14050207

Chicago/Turabian Style

Gao, Lifeng, Liujia Chen, Agen Qiu, Qinglian Wang, Jianlong Wang, Cai Chen, Fuhao Zhang, and Geli Ou’er. 2025. "Grid Partition-Based Dynamic Spatial–Temporal Graph Convolutional Network for Large-Scale Traffic Flow Forecasting" ISPRS International Journal of Geo-Information 14, no. 5: 207. https://doi.org/10.3390/ijgi14050207

APA Style

Gao, L., Chen, L., Qiu, A., Wang, Q., Wang, J., Chen, C., Zhang, F., & Ou’er, G. (2025). Grid Partition-Based Dynamic Spatial–Temporal Graph Convolutional Network for Large-Scale Traffic Flow Forecasting. ISPRS International Journal of Geo-Information, 14(5), 207. https://doi.org/10.3390/ijgi14050207

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Grid Partition-Based Dynamic Spatial–Temporal Graph Convolutional Network for Large-Scale Traffic Flow Forecasting

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI