In this talk, we present the multi-index stochastic collocation (MISC) method, which is the
multi-fidelity counterpart of the classical sparse grid stochastic collocation method
for Uncertainty Quantification. In the first part of the talk we introduce the method,
discuss the available convergence results and showcase its efficacy on a few numerical tests.
Then, we will focus on a recent development, namely, how to efficiently apply MISC
to problems where the system at hand is formed by multiple components (disciplines),
by computing a MISC surrogate model of each component first and then suitably combining
the surrogates together.