Abstract

Global supply chains have become increasingly complex, interconnected, and vulnerable to disruptions arising from geopolitical tensions, natural disasters, pandemics, and market volatility. Traditional disruption management approaches, which are largely reactive, are insufficient in today’s dynamic trade environment. This proposes a predictive framework for supply chain disruption forecasting using network analytics, aimed at enabling proactive resilience planning in global trade. The framework conceptualizes the supply chain as a dynamic, multi-layered network of suppliers, manufacturers, logistics providers, and markets. By modeling entities as nodes and material, information, and financial flows as weighted edges, the approach leverages graph-theoretic metrics—such as betweenness centrality, degree distribution, and community structure—to identify critical nodes whose failure could propagate systemic risk. Temporal network analysis is integrated to capture evolving trade relationships and detect early warning signals of stress within the network. Machine learning models, trained on historical disruption events and enriched with external data sources (e.g., commodity price indices, port congestion metrics, climate data, and political stability indicators), provide probabilistic forecasts of disruption likelihood. Scenario-based simulations enable the assessment of potential cascading effects and the testing of mitigation strategies, such as supplier diversification, buffer inventory optimization, and alternative routing. Results from a series of case studies in sectors including electronics, automotive, and pharmaceuticals indicate that the proposed network analytics framework improves disruption lead time prediction by 15–25% compared to baseline statistical models. Furthermore, centrality-based risk scoring provides actionable insights for prioritizing resilience investments. The findings underscore the potential of combining network science with predictive analytics to transform supply chain risk management from reactive recovery to proactive prevention, enhancing the robustness and adaptability of global trade systems in the face of rising uncertainty.