Shot noise is a big concern for EUV and electron beam lithography. The concern is based on the fact that electrons or photons arriving from the source randomly possess a Poisson statistical number distribution, in which the standard deviation is the square root of the mean.

For example, an EUV dose of 5 mJ/cm^2 corresponds to 5e-14 mJ = 312.5 eV
~ 3 photons per square nanometer. In a 30 nm X 30 nm area, a little over 3060 photons are projected to be incident. The square root of this number is 55, so 3s/avg = 3*55/3060 = 5.4% There is about 5% natural Poisson or shot noise in the dose. This only considers the random arrival of the photons or primary radiation quanta.

If you consider that exposure depends really on the secondary electrons produced by the primary radiation, it gets even more tricky, since there is no way of knowing how many such electrons are generated in the area of interest. One can estimate how many are produced by directly ionization (dividing primary energy by ionization energy), but not how many are produced in subsequent generations of electrons (tertiaries, etc.). Furthermore these electrons can move outside the area of interest (and others can move in) before finally slowing down enough to initiate chemistry.

Another fundamental consideration ignored...