In this paper, a new structure of predictive fuzzy PID controller is proposed. The new controller is robust and effective in controlling higher order and time-delayed complex processes. The proposed predictive fuzzy PID controller combines the fuzzy PID and generalized predictive control (GPC) ideas together, and is equipped with optimization capability that minimizes a cost function. These features make the new controller more effective than individual fuzzy PID and GPC controllers in handling nonlinear systems. Computer simulations are shown for various types of process stabilization and set point tracking problems.