Michael apparently really we mean different things.
I mean here that: the simulation of complex and interrelated systems there - sometimes - the need to change their state so that all elements of the system were changed "as a whole".
Once again I give as an example of this video: http://www.youtube.com/watch?v=9vU2MMc-woM

For example, when modeling such a robot, I want to change the position of a moving part which is connected with the other moving part, and that - the third ... and so on.
In a typical approach is necessary to move all the details manually, while calculating what position will take each of the parts after moving.
Here I am about anything!

An even more obvious example:
http://gerald-massey.org.uk/Railway/images/Development/Walschaerts.gif

Suppose I modeled all the elements of the propulsion unit and are arranged according to some of its provisions.
But.
Then I needed - for example - change the position of the drive wheel. What happens then?
I need to change the layout of all components in accordance with the new position of the wheel, right?

And I'm trying to understand whether it is possible to realize (in principle) any improvement to solve this problem.