A two level control strategy that stabilizes and optimizes the production of an industrial copper solvent extraction process is presented. The stabilizing layer consists of a multi-input-multi-output controller or two single-input-single-output controllers with additional four feedforward compensators that regulate the flow rates in the copper solvent extraction process. The optimization layer consists of an optimizer that maximizes the production of the copper solvent extraction process and gives setpoints to the controllers at the stabilizing level. The mechanistic plant models, verified with industrial data, are linearized by identifying first and higher order transfer function models from simulated PREIS data. On the basis of the linear models, the interactions of the controlled variables, and the pairing of the controlled and manipulated variables are studied and the optimizer and the controllers designed. The control strategy employing two PI-control loops or a model predictive controller and additionally four feedforward control loops is successfully tested against simulated disturbances and setpoint changes. The control strategy is also compared to the data collected from the industrial plant under manual control. With this two level control strategy the production of the copper solvent extraction process is increased by 3-5% and the process variation is decreased by 70-90% compared to the manual operation of the case industrial plant. The results gained in simulation environment are successful and encouraging for further testing in an industrial plant. (C) 2008 Elsevier Ltd. All rights reserved.