Abstract—This paper presents a Novel Hybrid Self Organizing Migrating Algorithm with Mutation for Global Optimization

(M-SOMA). The proposed algorithm includes the hybridization of Self Organizing migrating Algorithm (SOMA) and Non uniform mutation. SOMA is very effective population based algorithm among evolutionary algorithms. Though its convergence is very fast but there are lots of chances to trap in to local optima. As no new points are generated during the search only positions are updated. So to maintain the diversity of the search space and prevent premature convergence it is hybridized with Non Uniform mutation. The proposed algorithm is tested on 15 benchmark unconstrained test problems and its efficiency is compared with SOMA and GA results. On the basis of comparison it is concluded that the presented algorithm shows better performance in terms of function mean best. The graphical results also show that the presented algorithm perform better in terms of efficiency, reliability and accuracy