De Backker, ThomasVercammen, AnouckBoute, Robert2025-09-302025-10-082025-10-0820250925-5273https://repository.vlerick.com/handle/20.500.12127/7747We present a data-driven approach to assess supply chain disruption risk at the component level. Our ‘Component Risk Matrix’ categorizes components based on their predicted stockout frequency and severity (in terms of average stockout duration). Our predictive models employ an XGBoost model with historical disruption data and each component’s unique characteristics. Our approach enables prioritizing components for resilience measures by quantifying their component risk and identifying the key drivers behind this risk. We validate our methodology on 1,239 components from an original equipment manufacturer, demonstrating its practical applicability and providing insights toward risk mitigation. This data-driven approach empowers companies to strategically build supply chain resilience in designing their products and supply chains.Attribution-NonCommercial-NoDerivatives 4.0 Internationalhttp://creativecommons.org/licenses/by-nc-nd/4.0/Supply Chain ResilienceDisruption RiskComponent RiskRisk MatrixInternational Journal of Production Economics1873-7579102358