A major part of evolutionary computing is developing genetic algorithms to find solutions to computing problems. The authors of this paper propose a crossover operator, which extends previous results. The operator generates offspring that are parent-centric and good for local search strategies. After reviewing some real-coded crossover algorithms, the authors propose a real-coded memetic algorithm (RCMA) that invokes real-parameter crossover hill-climbing (XHC). Two important factors of the RCMA are the population diversity by means of the negative assortative mating strategy and the refinement of solutions carried out by XHC.
The authors also report their test results on six classic nonlinear continuous functions and three recent application problems. These testing results are interesting; however, they do not show any expected performance advantages. On our planet, some species flourish, while others disappear. Nowadays, we can only see the skeletons of dinosaurs in museums. Will evolutionary computing, or more specifically RCMA, prove to be a flourishing species or a dinosaur skeleton? Only time will tell.