A Modified Marriage in Honey-Bee Optimization for Function Optimization Problems

Abstract Many researches have implemented a genetic algorithm with real-coded chromosomes to solve a wide variety of problems. Their experimental results suggested that the real-coded genetic algorithms gave superior results to binary-coded genetic algorithm on most of the test problems. Inspired by the above founding, this study aims to (1) propose a modified Marriage in Honey-bee Optimization (MBO) technique (2) compare the performance of the proposed technique to that of the real-coded GA technique. In this study, two main ideas are proposed. Firstly, to handle the real encoding of genotypes, we present a new crossover operator and a new heuristic worker, named the scroll-based worker, for manipulating the real value of genes. Secondly, to reduce the number of user-defined parameters, we provide the original MBO with a self-organizing capability. With a self-organizing capability, the proposed model can automatically determine the proper number of queens itself. The experimental results on five benchmark test functions show that the proposed model is very effective in solving the function optimization problems.