In a variety of nuclear applications, it is often necessary to estimate the probability of a rare event occurring, such as the probability of radiation leaking from a fissure in the material, or the probability of radiation hitting a detector from a given source. It is well known that adaptive multilevel splitting (AMS) is a powerful method to estimate such probabilities when considering branchless dynamics. We aim to extend the notion of AMS to fissile systems, generalising the result to estimate quantities such as the probability of k particles hitting a detector. 

Project team: Chris Dean, Emma Horton