Events such as Hurricanes Katrina and Rita in 2005 have highlighted the need to understand network disruptions, in particular in the humanitarian supply chain. Humanitarian supply logistics cannot be improvised at the time of the emergency since little can be done after a disruption occurs. Therefore, special attention should be given to preparedness efforts and the design of the supply chain network. Inventory prepositioning is one of the strategies in the preparedness stage that enhances postdisaster response because it reduces the lead time needed to reach the affected destinations once a disaster has occurred. In this paper, the prepositioning of supplies for disasters is modeled as a facility location problem that accounts for the routing of vehicles and possible disruptions in the transportation network. The approach uses a combination of the most reliable path and an integer programming model to find the optimal location of supplies. The problem is solved on an example transportation network. The problem exploits a variant of the shortest-path problem and can be solved efficiently even for medium and large-scale transportation networks.