Stochastic Multi-Echelon Decomposition Framework for Dynamic Resource Allocation Under Supply Chain Network Disruption Uncertainty

Abstract

Contemporary global supply chains operate under persistent structural volatility, compelling organizations to adopt mathematically rigorous optimization paradigms beyond traditional deterministic models. This study develops a stochastic multi-echelon decomposition framework integrating Benders decomposition with scenario-based robust optimization to address dynamic resource allocation under compound disruption uncertainty. Using panel data from 214 multinational manufacturing firms across six industry verticals (2019–2024), the proposed model demonstrates a statistically significant 23.7% improvement in operational cost efficiency and a 31.4% reduction in service-level variance relative to benchmark heuristics. Monte Carlo simulation validates model robustness across disruption severity gradients. Findings provide actionable prescriptions for supply chain strategists and offer a generalizable computational architecture for multi-tier network resilience planning under non-stationary stochastic demand environments.