Hurricanes are a type of natural disaster for which it is possible to plan for prepositioning of supplies to improve the efficiency of the post-disaster relief effort. This paper develops a model for prepositioning supplies in such a setting. Our model has a distinguishing feature the possible destruction of supply points during the disaster event. To gain insight into our model, we develop a series of theoretical properties. To test the applicability of our model a series of computational tests are performed. From such tests we conclude that it is possible to solve relatively large instances of the problem utilizing standard optimization software. A methodology based on creation of clusters of demand points is proposed for solving even larger problems. Finally we study sensitivity of the results with respect to key parameters. These investigations provide important policy implications. ► Disaster relief modeling. ► Planning for a Hurricane disaster. ► Modeling destruction of supply points. ► Clustering to solve large-scale problems.