The SUA tool we developed performs probabilistic analyses of systems of components each of which are represented by analytic models with uncertainties. The component models are generally legacy physics-based deterministic simulation codes arbitrarily selected by the user. Hence, users may use codes they are familiar with rather than attempt to learn new codes with their associated learning curves. An example problem concerning the Space Shuttle’s main engine illustrates its value. The three LOX/LH2 rocket engines each depend upon 4 primary turbopumps in a complex high-temperature/high-stress environment. In order to maintain adequate thrust, the control system may increase the turbine inlet temperatures to mitigate component performance shortfalls or other deficiencies. But raising the turbine inlet temperature to compensate for such variations also consumes turbine life quickly. Hence, it is important to determine how much impact on turbine life (and therefore risk) is present due to these uncertainties. In the example problem, 13 uncertainties were identified and the deterministic ROCETS code was used in conjunction with SUA to determine the impact of these uncertainties on turbine inlet temperature. The SUA results are presented to the user graphically with CDF and PDF curves that convey the system-level impact of the 13 uncertainties. SUA also displays a probabilistic sensitivity factor chart that identifies the most important uncertainties. In this case, the two most important uncertainties are the high-pressure fuel pump efficiency (ETAMHPFP) and the high-pressure fuel turbine efficiency (ETAHMHPFT). Armed with this knowledge, program managers can take the most effective preventive actions to avoid serious operational risks.