In input i set number of primaries to be 12500. I created 4 spawns. And ran 3 cycles. After that in output plots i see info that number of primaries is 187500(3512500). I though that Fluka runs some number of cycles and than somehow avereges them to one. Can you please explain to me how this cycles work?
Please take a look to the following example:
Consider 3 parallel processes (spawn), and each of them running 5 cycles, that are run in series. We will then have a total of 3x5=15 independent sets of results. In each of the cycles, FLUKA would simulate the number of primaries requested by the user, for instance 1000. As a result, we would have 15 estimations of the same observable, each of them calculated using 1000 primaries.
Usually, one would be interested in a single (and more accurate) set of results for the total amount of histories simulated (15x1000=15000) and the associated statistical uncertainties. To do this, we need to merge the 15 estimations in a single one, which can be done by means of few programs (scoring-specific) that can be found in the FLUKA directory. When we process some results using Flair, it is making use of these programs to merge the different estimations into one final result and provide its statistical uncertainty.
In your particular case, provided all your cycles finished successfully, you should have 4x3=12 sets of results, each of them calculated using 12500 primaries. After you successfully merge all your data, the final results should correspond to a total number of primaries of 12x12500=150000. In Flair, when you go to the tab Run->Data in order to process your results, you should be able to see the different detectors that you are about to process and the different files that would be merged by doing it.
Hope it helps.