A graph-based monte Carlo framework for multi-tier supply chain disruption Analysis

Global supply chains are increasingly exposed to complex disruptions arising from natural disasters, pandemics, geopolitical tensions, and technological failures. Traditional risk assessment techniques often rely on deterministic assumptions or static network representations, limiting their ability to capture stochastic propagation and cascading effects. This paper introduces a graph-based Monte Carlo simulation framework designed to model multi-tier supply chains as dynamic networks in which nodes represent suppliers, production facilities, or distribution centers, and edges represent transport or contractual relationships with probabilistic attributes such as lead time, capacity, and reliability. The framework integrates stochastic disruption sampling with graph-theoretic propagation rules, allowing the generation of thousands of disruption scenarios. Resilience is evaluated using composite key performance indicators, including recovery time, service level maintenance, and stockout probability. A prototype implementation demonstrates practical utility in an electronics supply chain case study, illustrating how mitigation strategies such as alternate sourcing and buffer stock can reduce expected recovery time and improve service performance. The results suggest that combining Monte Carlo methods with network analysis provides actionable insights for decision-makers seeking to enhance resilience in volatile supply environments. This study offers both a methodological contribution and a practical tool for operational risk management.