Abstract: In the cement production process it is hard to develop an accurate mathematical model for both f-CaO and stack bed thickness which characterize the key technical indices in the cooling process with a grate cooler. However, it is also hard to control efficiently the process by conventional methods. Combining the case-based reasoning (CBR) with conventional control method in clinker cooling process, a new control method is proposed to optimize intelligently the index setting by CBR, thus stabilizing the interval control of the mass fraction of f-CaO in clinker and stack bed thickness. The intelligently optimized index model can update automatically all basic loop setpoints so as to avoid the subjectivity and randomness due to the arbitrary manual setting. The approach proposed has been successfully applied to the clinker cooling process with grate cooler in a cement plant, and the application results showed its effectiveness.