A Multiplex Network Model for Resilient Semiconductor Distribution: Integrating Multi-tier Channels, Geopolitical Risks, and Predictive Analytics
摘要
Semiconductor and electronics supply chains are increasingly exposed to geopolitical shocks and cascading failures that cannot be captured by single-layer or stylized flow models. Building on early supply-chain conceptualizations by Oliver and later expansions by Chopra and Meindl, we develop a multiplex framework integrating material flows (OCM \(\rightarrow \) DIST \(\rightarrow \) EMS \(\rightarrow \) OEM) with equipment dependencies, and evaluate resilience under embargo and export-control scenarios. Four strategies are compared: no mitigation, buffers, rewiring, and combined. Results show that baseline resilience stabilizes at approximately 47%, buffers smooth transients without raising the ceiling, rewiring lifts steady-state to around 60%, and combined interventions accelerate recovery. Percolation analysis further demonstrates that multiplex coupling reveals hidden fragility missed by single-layer representations. All research questions are answered, objectives met, and hypotheses validated: high-risk nodes amplify cascades, buffers + rewiring outperform isolated measures, and multiplex analysis identifies systemic vulnerabilities. The study highlights potential vulnerabilities relevant to OEMs, OCMs, distributors, EMS providers, and policymakers, offering preliminary insight into how different levers affect resilience.