A method of evaluation cylindricity errors based on particle swarm optimization (PSO) was proposed. According to the minimum zone condition, a mathematical model of cylindricity errors together with the optimal objective function and fitness function was developed. The principle and implementation techniques of PSO were introduced. Then the particle swarm optimization algorithm was used to search for the optimal solution. Finally the example showed that PSO method is suitable for the nonlinear optimization problems such as cylindricity error evaluations. In addition, the precision of PSO is better than other methods under minimum zone condition. Thus PSO has the advantages of setting up a few parameters and a rapid convergent velocity. So it can be applied in dealing with the measured data of cylindricity obtained by three-coordinate measuring machine (CMM).