The reliability assessment of complex adaptive systems requires the identification of dominant input parameters and the quantitative evaluation of the associated effects on the system performance. This can be achieved using experimental and numerical sensitivity analysis methods. In this paper a simulation based approach is presented, assessing the system performance of an active vibration isolation device with respect to parameter variations, such as temperature, load amplitude, material properties and geometry dimensions of the structural elements. The modeling of the active system is described utilizing the Finite Element Method and a Krylov Subspace based model order reduction scheme. The implemented Morris screening technique and variance based sensitivity analysis are discussed. For the example of an active vibration system the sensitivity analysis strategy is outlined and it is shown that a quantitative assessment of the system performance considering large scale parameter variations is provided.© (2009) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.