A new simple, fast and efficient algorithm for global optimization over continuous search-space problems: Radial Movement Optimization

• We propose a new stochastic global optimization method for continuous search-space.
• We chose ten benchmark functions to evaluate the ability of the proposed algorithm.
• The results are compared with two other methods which are PSO and DE.
• The proposed method obtains proper and fast solution and escapes from local optima.
• Being robust, fast and needing less memory are the main features of the method.

Optimization of non-linear and non-differentiable problems has been considered as an important issue for mathematicians and engineers. A new stochastic global optimization method for non-linear and non-differentiable problems is proposed and extensively explained, in this article. It is a swarm-based method which uses spherical boundaries in a vector search-space to explore for the optimal solution. Having a few numbers of parameter to be adjusted, being robust and fast, needing small memory storage size and capability of escaping from local optima, are the main features of this new algorithm. To analyze and evaluate the capability of this novel method, ten benchmark functions are chosen and the results are compared with two existing optimization algorithms which are Differential Evolution and Particle Swarm Optimization. Comparisons are made based on the consistency in obtaining optimal solutions, computation time and convergence profile. Results show the capability of the proposed method in finding a proper solution in a very short time and also escaping from local optima of the solution-space.