Particle swarm optimization (PSO) is simulation of the society model. This algorithm is a stochastic global optimization technology, and it finds the optimal regions of complex search space. PSO has the excellent performance, for example, simpleness, realization easily and powerful functions. This paper introduces the PSO, builds the predictive, model of cutting surface roughness, and brings up P SO which is employed in the parameters identification of the predictive model of cutting surface roughness. The simulation experience shows PSO is more suitable I than the linear regression algorithm and other optimization algorithm. When the prediction model and optimal cutting parameters using PSO are compared with using the linear regression algorithm and other optimization algorithm, and it can reduce predictive error.
