Trouble is, in the real world you *can't* guarantee that components would fit together when they arrive from your subcontractors. You can't even guarantee that from your own production lines. This is why you test a sample of components as they arrive, and test the end product once you've combined those components with others.
As for standards - there's no universally accepted standards yet. Seperate organisations have their own standards - at least, the competent ones do - and there is some movement towards universal standards. This is similar to the situation a couple of centuries ago in mechanical and civil engineering. Just look, for example, at the myriad incompatible ways of doing something as simple as making a screw. I don't think you could seriously claim that the pioneers of commercial steam power, or those building canals or cotton mills weren't engineers!