Improvement of Particle Swarm Optimization: Application of the mutation concept for the escape from local minima

In this paper, the improvement of Particle Swam Optimization (PSO) is proposed. PSO is one of the evolutionary computation methods to obtain the approximate optimal solution for the continuous optimization problem with multi-peak objective function. It can be applied to solve many optimization problems arising in the Supply Chain Management, such as Facility Location Problem, Inventory Portfolio Problem or Dynamical Lot Sizing Problem. One of the major weaknesses of PSO is trapped into local optima. To overcome this weakness, in this paper, the introduction of concept of mutation in Genetic Algorithm (GA) to PSO is proposed. In the computational experiment, the three benchmark problems are tested in order to validate the effectiveness of proposed method.