Abstract: A multi-sections particle swarm optimization algorithm(MSPSO) is proposed.The new(algorithm) assumes that the search area is divided into several sections.The best position of each section is found using particle swarm optimization algorithm and a new swarm that consists of the best positions of sections searches for the global best position by particle swarm optimization algorithm again.Then,both MSPSO and PSO are used to resolve five widely used test functions' optimization problems.Results show that MSPSO has quicker convergence velocity and better optimization performance than PSO.