Global sensitivity analysis in the context of imprecise probabilities (p-boxes) using sparse polynomial chaos expansions.

•Global sensitivity analysis is extended to the context of imprecise probability theory (parametric p-boxes).•Imprecise Sobol’ indices are defined and computed from a sparse polynomial chaos expansion (PCE) of the computational model.•The sparse PCE is built in an augmented space of input parameters and associated hyperparameters.•A smart experimental design using phantom points is devised to reduce the number of runs of the computational model.