Collaborative Research: Enhancing Stochastic Supply Chains via Cascading Simulation and Adjustable Optimization

Global supply chains are essential to national economic stability, public health, and the availability of critical goods, yet they remain vulnerable to disruptions from demand and supply fluctuations, labor shortages, natural disasters, and other extreme events. These risks are difficult to manage because disruptions can spread across interconnected suppliers, transportation networks, and regions. At the same time, the structure of many supply chains is only partially visible because firms often share limited information with one another. This award supports research that will develop data-driven methods to identify hidden vulnerabilities and design practical mitigation strategies before disruptions cause widespread impacts. This research advances the national interest by improving the resilience of supply chains that support critical goods, advanced manufacturing, and industrial production, strengthening preparedness for future disruptions, and enhancing the ability of organizations to make informed decisions under uncertainty. The project will also train students through coursework, research projects, and outreach activities focused on data analytics, optimization, and supply chain resilience. The research will develop an integrated framework combining stochastic network learning, cascading simulation, and adjustable optimization. First, the project will develop new methodologies to construct probabilistic models of supply chain networks from incomplete data by integrating known supplier relationships, uncertain supply and demand, geographic information, production capacity, and inferred hidden connections. Second, the project will use these models to develop simulation-optimization methods that characterize how single-site and multi-site disruptions propagate through multi-tier supply chains while accounting for the operational responses of supply chain stakeholders. This approach will reveal vulnerabilities that may remain hidden when disruptions are analyzed individually or when stakeholder responses are oversimplified. Third, the project will develop adaptive mitigation policies for alternative and backup suppliers using interpretable decision rules that adjust procurement, production, and inventory decisions as new information becomes available. The framework will be evaluated across representative supply chain settings to assess its scalability and its ability to improve supply chain resilience. Expected contributions include new modeling frameworks, computational algorithms, and analytical insights that improve the ability to identify hidden vulnerabilities, evaluate cascading disruptions, and design adaptive mitigation strategies for supply chains operating under uncertainty. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.