Article: Spatio-temporal graph neural networks for dynamic warehouse logistics configuration in cross-border e-commerce Journal: International Journal of Information and Communication Technology (IJICT) 2026 Vol.27 No.86 pp.62 - 86 Abstract: Cross-border e-commerce warehousing and logistics face severe challenges of dynamic and variable demands as well as complex networks. Traditional static planning methods are unable to achieve real-time optimisation. Therefore, this paper proposes a spatio-temporal graph neural network model, which can simultaneously capture the spatial correlations between warehouses and transportation routes in the logistics network and the changing patterns of order flows over time, thereby enabling dynamic and intelligent allocation of warehousing resources. The experiments are based on real cross-border transaction data, and the results show that compared with the optimal benchmark model, this method increases the order fulfilment rate by 8.2% and reduces the average logistics cost by 15.1%. The research indicates that this method provides an effective approach for building an efficient and flexible intelligent supply chain system. Inderscience Publishers - linking academia, business and industry through research

Title: Spatio-temporal graph neural networks for dynamic warehouse logistics configuration in cross-border e-commerce

Authors: Liqin Liu

Addresses: School of Logistics, Henan College of Transportation, Zhengzhou, 451460, China

Abstract: Cross-border e-commerce warehousing and logistics face severe challenges of dynamic and variable demands as well as complex networks. Traditional static planning methods are unable to achieve real-time optimisation. Therefore, this paper proposes a spatio-temporal graph neural network model, which can simultaneously capture the spatial correlations between warehouses and transportation routes in the logistics network and the changing patterns of order flows over time, thereby enabling dynamic and intelligent allocation of warehousing resources. The experiments are based on real cross-border transaction data, and the results show that compared with the optimal benchmark model, this method increases the order fulfilment rate by 8.2% and reduces the average logistics cost by 15.1%. The research indicates that this method provides an effective approach for building an efficient and flexible intelligent supply chain system.

Keywords: spatio-temporal graph neural network; STGNN; cross-border e-commerce; warehousing logistics; dynamic configuration.

DOI: 10.1504/IJICT.2026.155564

International Journal of Information and Communication Technology, 2026 Vol.27 No.86, pp.62 - 86

Received: 25 Feb 2026
Accepted: 31 Mar 2026
Published online: 05 Aug 2026 *

Title: Spatio-temporal graph neural networks for dynamic warehouse logistics configuration in cross-border e-commerce

Keep up-to-date