Cooperative model for nature-inspired algorithms in solving real-world optimization problems

Abstract

A cooperative model of eight popular nature-inspired algorithms (CoNI) is proposed and compared with the original algorithms on benchmark set CEC 2011 collection of 22 real-world optimization problems. The results of experiments demonstrate the superiority of CoNI variant in the most of the real-world problems although some of original nature-inspired algorithms perform rather poorly. Proposed CoNI shares the best position in 20 out of 22 problems and achieves the best results in 8 out 22 test problems. Further fundamental points for improvement of CoNI are in selection of topology, migration policy, and migration frequency.