Revealing implicit competition and functional similarity in trade-logistics networks

Analyzing interaction relationships and patterns among participants within large-scale complex trade logistics networks can reveal role differentiation, interaction strategies, and latent risks inherent in network systems. Current research predominantly focuses on explicit direct relationships, lacking sufficient understanding of the implicit indirect relationships driven by shared nodes. Such indirect relationships in trade logistics networks carry dual connotations: (1) potential competition between two trade entities sharing the same market, and (2) a functional similarity between them arising from serving the same market. However, existing research is often limited to a singular competition perspective, and its quantification metrics, relying on absolute flows, struggle to measure the true strategic dependence between trade entities. To address this gap, this study introduces the concept of the mediator subgraph as a structured network unit. It proposes a novel statistic based on relative importance to quantify these two indirect relationships and assesses their statistical significance via a permutation test. Building on this, four fundamental interaction patterns are identified from the combination of the strength and significance of these two relationships. The framework’s effectiveness is validated through synthetic data experiments and then applied to the 2013–2023 global crude oil trade network. Results show that the evolution of the indirect relationship network, guided by key players as shared nodes (e.g., Russia and China), highly corresponds with concurrent geopolitical dynamics. This research provides a novel perspective and analytical tool for understanding complex implicit interactions among trade entities, contributing to more precise guidance for trade policy formulation and supply chain optimization.