An Improved Moth-Flame Optimization algorithm with hybrid search phase

Abstract

In order to solve real-life problems, several metaheuristic optimization algorithms have been developed. The Moth-Flame Optimization (MFO) algorithm is a search algorithm based on a mechanism called transverse orientation. In this mechanism, the moths tend to maintain a fixed angle with respect to the moon. MFO suffers from the degeneration of the global search capability and convergence speed. To overcome these imperfections, an Improved Moth-Flame Optimization (IMFO) algorithm is proposed. The main novelty of the proposed approach is the definition of a hybrid phase between exploration and exploitation. This phase is characterized by a fitness depended weight factor for updating the moths positions. IMFO is tested on selected benchmark functions, CEC2014 test functions and 6 design problems, and compared with recent well-known optimization algorithms. The results show that IMFO achieves the best results with respect to the comparison algorithms in terms of search capability and convergence performances.