I think the issue would be that whereas this I think can get away with just a really fast rigid body simulation, the exhausts and so on would need to simulate acoustics where you have feature sizes in the same ballpark as the wavelength of the sound i.e. your ear will almost definitely be a harsh critic.
I mean there are models in acoustic physical modelling for e.g. throat simulation that break the throat up into a series of cylinders and frustrums that work quite ok. Not sure about real time performance tho.
