In this paper we present a new hybrid finite-element/component-modeling approach that can predict the pull-in and release behavior of MEMS switches orders of magnitude faster, and with significantly more behavioral detail, than traditional finite-element or formula-based approaches. The speed and detail allow exploration of the design space, guidance for design-of-experiments, and insight into process variation that was previously infeasible. For instance, the existence of multiple pull-in or release states is very sensitive to device dimensions and is critical to achieving desired performance and yield. Understanding this sensitivity by varying all possible parameters in a traditional finite-element approach could take weeks of simulation. Using the above methodology, the analysis can be done in minutes. The simulation methodology has been verified by comparing with measured Capacitance-Voltage (CV) relationship and Wyko white light interferometry displacement data for a commercial capacitive switch.