ESG-Constrained Generative Optimization of Global Supply Chains Under Geopolitical Risk

We propose a generative optimization framework that designs global supply chains subject to Environmental, Social, and Governance (ESG) constraints and geopolitical disruption risk. Our method combines stochastic search with constraint screening and multi-objective scoring to discover resilient, low-emission, and cost-competitive networks. Using a synthetic, multi-region dataset, we show that ESG thresholds and risk budgets reshape the Pareto frontier between cost, emission, and lead-time (latency) while maintaining high service levels. We formalize the objective, provide feasibility conditions, and discuss governance hooks for auditability and compliance.