Nuclear reactions are the most versatile probe to study rare isotopes and therefore are an essential part of the experimental program at facilities such as FRIB. The planning and interpretation of reaction experiments rely on theory models.

In this presentation, we review efforts to quantify the uncertainties introduced by these models, through Bayesian analyses

Examples include the calibration and emulation of effective interactions (optical potentials) as well as the impact on direct reaction observables and compound nuclear cross sections.

I also discuss the challenges in performing UQ in the nuclear-reaction domain, as well as opportunities for the near future.