The optimization process is concerned with the finding of the best solution from all possible solutions for a given problem. In this paper, the focus is on the importation step of HS algorithm using Scramble Mutation. The essential idea of new type of modification on variable. The selection method of Scramble mutation hybridized onto harmony search. The resulting variant of HS is called the Scramble Mutation Harmony Search algorithm (SMHS). For optimization problems to avoid premature convergence problem due by using one of the mutations to increase the number of solutions proposed, thus improving the performance of the Harmony Search (HS) algorithm, by maintaining diversity in the use of different rates to modify the proposed solutions. Compare the SMHS with HS and other existing methods to validate the efficiency of the SMHS. The results obtained by comparing the SMHS with basic HS and two other methods (i.e. MHS and DLHS), using ten benchmark functions illustrated that SMHS outperformed the basic HS for majority of the functions. It is therefore concluded that the SMHS algorithms is highly sensitive for the HMCR and obtains the best results at high value of HMCR. For harmony memory size, the HS performs better when HMS is relatively small. In the same vein, the SMHS performs better when the number of domain ( ) is compatible with HMS.
Optimizations, Evolutionary Algorithm (EA), Harmony Search (HS), Genetic Scramble Mutation Harmony Search (SMHS)