Abstract

The COVID-19 crisis exposed profound structural fragilities in global medical supply chains, revealing how pandemic-induced shocks can propagate rapidly across interconnected production, logistics, and distribution networks. Scientific advances in modeling these disruptions have since accelerated, integrating epidemiological forecasting, network science, system dynamics, and artificial intelligence to better anticipate, quantify, and mitigate systemic vulnerabilities. This study synthesizes recent methodological innovations in modeling pandemic-induced disruptions across global medical supply chains, focusing on predictive analytics, stochastic optimization, agent-based simulation, and digital twin architectures. Emerging models increasingly couple disease transmission dynamics with supply network topology, enabling scenario-based simulations that capture nonlinear demand surges, supplier shutdowns, border restrictions, and transportation bottlenecks. Advanced graph-theoretic approaches identify critical nodes and high-betweenness pathways whose failure disproportionately amplifies shortages of essential medical commodities such as personal protective equipment, ventilators, diagnostics, and vaccines. Machine learning techniques enhance demand forecasting under uncertainty, while reinforcement learning algorithms optimize inventory allocation and multi-echelon distribution strategies in real time. Additionally, hybrid models combining system dynamics with discrete-event simulation provide granular insights into feedback loops between policy interventions, workforce availability, manufacturing capacity, and global trade flows. The integration of blockchain-enabled traceability and real-time data streams further strengthens model calibration and transparency, improving decision support for governments, humanitarian agencies, and private-sector stakeholders. Importantly, resilience metrics have evolved beyond efficiency-based paradigms toward robustness, adaptability, and recovery time optimization, incorporating equity considerations to ensure fair allocation across low- and middle-income regions. Despite these advances, challenges remain in harmonizing heterogeneous data sources, validating models across jurisdictions, and addressing ethical implications of algorithmic prioritization. The findings underscore the necessity of interdisciplinary modeling frameworks that bridge public health intelligence with supply chain engineering to enhance preparedness for future pandemics. By advancing scientifically rigorous, data-driven modeling approaches, policymakers and industry leaders can transition from reactive crisis management to proactive resilience planning, thereby safeguarding global health security and stabilizing critical medical supply networks under extreme uncertainty.