Both practitioners and researchers working in after-sales service logistics have gained interest in additive manufacturing (AM), as the technology is promising in reducing spare part lead times through local part production. We focus on using AM at remote locations to produce spare parts that are used as a temporary fix until a regular, conventionally manufactured part arrives. Although our model fits various contexts, we focus on a military context with a mission in a remote, uncertain and isolated area. The most downstream location is in the mission area, which can cause threats that may influence the possibility to use AM. The next higher location in the supply chain is typically more safe, but if threat levels are high, it may be difficult to ship a part into the mission area. Our aim is to determine if AM capabilities should be installed, and if so where. We assess various printing locations and various AM machine types, by combining scenario analysis with discrete-event simulation. We apply our model on a case study of the Royal Netherlands Army. We find that the location and type of AM capability influence vehicle availability and the total operating costs, depending on the remote context.