A numerical and analysis method for optimizing multilayer supermirrors is developed based on the combination of the power-law method and the local optimization method of simplex algorithm. The parameters in the power-law formula are optimized by genetic algorithm. This allows a global minimization of the merit function and a many-fold decrease of the computing time. Several groups of X-ray supermirrors with the energy extended to 30 keV are successfully designed using this optimization method for a hard X-ray telescope. Tungsten and boron carbide are chosen as the multilayer materials. High reflectivity and high effective area are obtained, indicating that this numerical and analysis method is an effective tool to design hard X-ray supermirrors.© (2010) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.