We examine the resiliency of retail locations of a supply chain network to aid in the recovery of the local community after an extreme event. A two-stage stochastic programming model is used to determine the placement of permanent generators at the retail locations of Stewarts Shops, which distributes both convenience items and fuel in Upstate New York and Vermont, to enhance the resiliency of the supply chain. Our measure of resiliency specifically considers the recovery process of each retail location after the extreme event and its interdependency on other external infrastructure systems. Our computational experiments consider the multiple distinct types of hazards that can affect the retail locations of Stewarts Shops. We empirically explore different stochastic sampling procedures to solve the resiliency model. The results of computational tests indicate that we can converge to an approximate optimal solution quickly. We compare the resiliency efforts when planning for different types of hazards versus all hazards simultaneously as well as the impact of external infrastructure systems on the resiliency efforts. The empirical study identifies that the stores in rural, less densely populated areas, serving a large population should be selected to receive generators.