Multi-dimensional XAI Framework Revealing Critical Supply Chain Vulnerability Drivers

This article introduces novel Explainable AI (XAI) methodologies tailored for multi-factor supply chain risk models to address the opacity of predictive models in global supply chain management. Traditional risk assessment approaches often function as black boxes, providing risk scores without transparent justification, which hinders proactive mitigation efforts. The article develops context-aware explanation algorithms that move beyond simple feature importance to generate actionable, interpretable insights into the specific drivers of potential disruptions. The multi-dimensional XAI framework incorporates temporal and spatial dimensions alongside causal relationship modeling to pinpoint vulnerabilities such as upstream supplier dependencies, geopolitical instability indicators, and transportation chokepoints. Through rigorous implementation across diverse supply chain typologies and comparison with traditional methods, it demonstrates that these explainable approaches significantly enhance risk driver identification, decision-making timeliness, and mitigation effectiveness. Despite implementation challenges related to data accessibility, computational complexity, and organizational factors, the enhanced transparency enables more targeted interventions, collaborative risk management, and improved operational efficiency. The implications extend beyond supply chain management to establish explainability as a fundamental requirement for responsible AI deployment in business operations.