Plant closures have always concerned supply chain purchasers. A supplier plant could close anytime due to natural disasters, strikes, pandemics, or financial distress. Supply chain purchasers must assess potential suppliers' risk when identifying annual contracts and evaluating capacity planning. Previous models in literature determine the likelihood of a supplier’s plant closure within a year. In this paper, we expand beyond likelihood to determine the predicted length of plant closure using linear regression along with cox proportional hazards model and accelerated failure time model. We apply our models to a case study within the Department of Defense (DoD). Our results demonstrate that these predictive models have the potential to aid in the mitigation of supply chain risk, improving capacity allocation, and saving taxpayer dollars.