Text this: Modification and improved implementation of the RPD method for computing state relaxations for global dynamic optimization.