Artificial neural network (ANN) based macro-modeling approach for design and optimization of integrated circuits (ICs) with radio frequency microelectromechanical systems (RF MEMS) is presented. The finite element method (FEM) analysis is utilized for both characterization of the device and generation of training and testing data for the ANN model. Developed ANN based approach has shown excellent agreement with the FEM model and measurements. The integration of the ANN model in an RF circuit simulator shows that the proposed methodology is suitable for a circuit-level analysis, design, and optimization of MEMS devices.