Typically sensitivity analysis was performed using a one-factor-at-a time (OAT) approach [8], i.e., each factor is perturbed in turn while keeping all other factors fixed at their nominal values. When the purpose of sensitivity analysis is to assess the relative importance
of input factors in the presence of factor uncertainty, this approach is only justified if the model is proven to be linear. This approach is naturally limited as it does not permit accounting for simultaneous variation of uncertain parameters and it is carried out with respect
to a reference situation [5], although it is known that parameter interactions are generally important in system safety analysis.