Abstract: A heuristic optimization method is applied to reduce the vibration amplitude of a rotor system magnetically supported. The rotor system is supported by two Active Magnetic Bearings (AMB) that are operated by commonly used PID controllers. Applying the Ant Colony Optimization (ACO) method, the control parameters of both AMBs are optimized for minimizing the rotor vibrations during the run-up. The necessary steps needed to develop an ACO algorithm are described and the factors that influence the PID control parameters are discussed. Numerical simulations were performed to demonstrate the applicability and efficiency of the proposed procedure.