A novel coding strategy for GA-based numerical optimization

The existing coding strategies for GA-based numerical optimization have their respective benefits. Based on the analysis upon them, and combining their characteristics, a novel strategy named the floating-point binary code is proposed. The strategy covers the representation as well as corresponding operators. The experiments show that the performance of the implementations adopting the proposed strategy were better than those employing either the real coding or the binary coding strategies for given problems.