Based on classical diffraction theories with modifications and extensions in analytical formulations and numerical implementations, a new code has been developed at NASA for the prediction of aircraft noise shielding, named as Propulsion Airframe Aeroacoustic Shielding Attenuation (PAAShA). The code is developed primarily for aircraft system noise predictions, although it may also be useable in other applications with acoustic shielding. The requirements for this code are driven by the need for a robust, capable code to use with NASA’s Aircraft Noise Prediction Program (ANOPP) for aircraft integration and system noise research. The requirements are met and include capabilities to use a wide range of aircraft geometries, rapid calculation times consistent with aircraft system noise problems, and the flexibility to model realistic noise source characteristics and distributions. The accuracy and robustness of the method are demonstrated in this paper with a set of problems, including a cylinder, a finite plate, a symmetrical two-dimensional airfoil, and a full three-dimensional hybrid wing body aircraft model tested in a wind tunnel. This range of problems demonstrates both smooth and sharp edge diffraction capabilities for a wide range of frequencies and low Mach number flow effects at low angles of attack. Predictions are shown to be accurate to within 1–4 dB over a wide range of the most significant frequencies and directivity angles. This is determined by comparing with data, which have experimental uncertainties, particularly at high frequencies, high angles, and source characteristics. The accuracy diminishes for geometries that include a significant reflection component, which is not calculated by the code. Accuracy can also be somewhat diminished for high azimuthal angles. Accurate modeling of the noise source, particularly its frequency and directivity characteristics, is essential to obtaining accurate results.