The inverse transform method has to do with simulating a random variable given a uniform distribution (which is built in to many computer packages by default). Here is some more information:
Inverse transform sampling - Wikipedia, the free encyclopedia
Basically, it means that if you can simulate U (a standard uniform) then you can simulate another continuous random variable by using Y = F^(-1)(Y) which will return a point for the Y distribution that corresponds to a realization.
What it does is it finds the point where the cumulative probability matches the realization of the uniform distribution and you "work backwards" to find what point on your complex distribution it corresponds to.