Within the Bayesian framework a non-intrusive uncertainty quantification is performed to assess the uncertainty of ion–solid interaction simulations. For this we employ a reduced-order spectral expansion approach which is capable of confining the number of model runs to a feasible size. Moreover, the method facilitates sensitivity examinations regarding to input parameters of the model. It is applied to the ion–solid simulation program SDTrimSP with several uncertain but normally distributed input parameters, i.e., impact angle α , projectile energy E 0 , and surface binding energy E s b . Consequently, the otherwise hardly accessible model parameter E s b can be estimated in combination with recently acquired experimental data.