This study presents a mathematical model for optimizing warehouse-location decisions in supply-chain network design. Previous studies have modeled the facility-location problem and provided fortification plans and solutions. However, these studies did not consider multi-capacity warehouses. The proposed model identifies optimal warehouse locations, sizes, and branch-warehouse assignments, and it determines fortification measures to mitigate the risk of warehouse failure. The model is formulated as a mixed-integer nonlinear problem and is subsequently linearized using standard linearization techniques combined with a proposed method based on the average-probability approximation. A case study of a Canadian company is used to illustrate the model's efficacy. Sensitivity analysis results demonstrate that when the numbers of fortified and built warehouses are equal, failure probability has no effect on the objective function of the model. Thus, this study presents a comprehensive model that can be used to assist businesses in decision-making regarding their warehouse operations to lower their risk of failure and optimize their use of warehouse infrastructure.