Improve of Harmony Search by Scramble Mutation for Global Optimizations Problems

Abstract

Harmony search (HS) is a new meta-heuristic optimization method imitating the music improvisation process where musicians improvise their instruments pitches searching for a perfect state of harmony. The usage of HS has become a common thing for a variety of numerical and real-world problems. It has several advantages over other meta-heuristics. It considers all existing vectors to generate a new vector. It imposes fewer mathematical requirements. The main disadvantage of HS encompasses its tendency to converge prematurely, which in essence leads to lose diversity during the search. In this study, a new variant of HS, called Scramble Mutation Harmony Search (SMHS), is proposed in this work where concepts from Genetic Algorithm (GA) process are borrowed to enhance the performance of HS. The Scramble Mutation is original step of GA, and is popular with permutation representations. In this, from the entire chromosome, a subset of genes is chosen and their values are scrambled or shuffled randomly. The performance of the SMHS is evaluated and compared with HS (a recently developed variation of HS that is, DLHS, and MHS). The experiments conducted show that the SMHS generally outperformed the other approaches when applied to ten benchmark problems. The effect of the SMHS parameters is analysed. Finally, the results show that cellular approaches seem to be an efficient alternative for optimization problems.