A Modified Spider Monkey Optimization Algorithm Based on Good-Point Set and Enhancing Position Update

Abstract

Spider Monkey Optimization (SMO), developed depending on the behavior of spider monkeys, has recently been added to the swarm intelligence class. SMO is a stochastic meta-heuristic based on population. The spider monkey is classified as an animal whose social structure is based on fission and fusion. In addition to being an excellent tool for solving optimization problems, SMO supplies good exploration and exploitation capabilities. In the current work, we present a modified strategy for improving the performance of the basic SMO by following two directions: (a) the good-point-set method is used instead of a random initial population generation one; and (b) by changing both the local leader and global leader phases, SMO's position update approach was modified to increase global convergence while avoiding local solutions. This work was evaluated on ten popular benchmark functions. The current findings prove that the proposed approach outperforms the standard SMO in terms of result quality and convergence rate.