As the EUMETSAT Polar System-Second Generation (EPS-SG) is being prepared, a new generation of the hyperspectral Infrared Atmospheric Sounding Interferometer (IASI) has being designed. The IASI New Generation (IASI-NG) will measure at 16921 wavelengths in each sounding pixel, benefiting of a spectral resolution and a signal-to-noise ratio improved by a factor 2 compared to its predecessor [Crevoisier et al. (2014)]. Measurement precision will be improved as well starting from the 1 K in temperature and 10% in humidity IASI precision. The IASI-NG characteristics will lead to huge improvements in detection and retrieval of numerous chemical species and aerosols, and in thermodynamic profiles retrievals, as well as providing, like its predecessor, a huge contribution to Numerical Weather Prediction (NWP). The high amount of data resulting from IASI-NG will present many challenges in the areas of data transmission, storage and assimilation. Moreover, the number of individual pieces of information will be not exploitable in an operational NWP context and the choice of an optimal data subset will be needed. For all these reasons, an appropriate IASI-NG channel selection is going to be performed aiming to select the most informative channels for NWP.