AI-Enhanced What-If Scenario Analysis in Supply Chain Digital Twins: A Multi-Objective Trade-Off Perspective on Cost, Resilience, and Carbon Efficiency

Abstract

Global supply chains face compounding disruptions driven by geopolitical tensions, tariff volatility, climate events, and demand uncertainty. Digital twin technology, which creates dynamic virtual replicas of physical supply chain networks, has emerged as a promising instrument for proactive decision-making through simulation and scenario analysis. This paper presents an analytical framework examining how artificial intelligence enhances What-If scenario analysis within supply chain digital twin environments, with a particular focus on multi-objective trade-offs among cost, resilience, service level, and carbon efficiency. By synthesizing recent literature, publicly available datasets, and U.S. federal supply chain digitalization initiatives, the study delineates the mechanisms through which AI enables multi-source data integration, automated scenario generation, cascading disruption prediction, and Pareto-optimal decision evaluation within digital twin architectures. Rather than proposing a new computational model or system implementation, this work offers a structured analytical perspective that bridges the gap between isolated single-module optimization studies and the integrated cross-functional decision-making that modern supply chains demand. The findings highlight a convergent relationship between resilience optimization and carbon reduction, and identify key technical and organizational challenges for practical implementation at the enterprise level.