I couldn't agree more on complex or extremely complex interdependent systems. We certainly wouldn't want to replace an impeller for a precision milled high per

I couldn't agree more on complex or extremely complex interdependent systems. We certainly wouldn't want to replace an impeller for a precision milled high performance turbo pump on a rocket with a 3D printed part as the risk of catastrophic failure would be enormous. But there is still great scope for field expedient replacement because of the simple fact that field expedient repair has occurred since mankind started killing each other in an organised fashion. I think much of the genuine risk of cumulative patchwork repairs increasing risk of catastrophic failure could be mitigated with a smart asset registry/history that could manage "go/no go" for depot level or higher maintenance/refurb. DOD would surely be one of the world's biggest consumers of duct tape. If just in time 3D printing can provide short term sustainment of assets that includes proper asset history documentation, the I would imagine that would be a net plus. Looking further into the future, I could easily imagine both 3D printing and production output auditing for common consumables and wear items as not just field expedient, but becoming the standard. Although I would think this will require a massive R&D/certification effort(so easier said than done), as well as designing this flexibility into future assets/platforms. "For want of a nail" comes to mind. For want of a nail the shoe was lost, for want of a shoe the horse was lost; and for want of a horse the rider was lost; being overtaken and slain by the enemy, all for want of care about a horse-shoe nail. 3D print the nail, and then replace it in deport maintenance.