A jar test is usually applied in order to control the coagulant dose. However, this method is manual and cannot be performed in a continuous manner. Moreover, it implies the presence of a qualified operator. In large water treatment plants a jar test is performed daily, while in small plants the frequency is even lower because of the associated O&M costs. Therefore a significant amount of research has been conducted in the past ten years to propose other alternative solutions for the on-line prediction of the optimal coagulant dose as a function of raw water quality characteristics (such as turbidity, pH, conductivity, etc.). Different methods can be used for on-line control and automatic regulation of the coagulation process. A method for the on-line prediction and control of coagulant dose using raw water quality parameters measurement was developed on the basis of the artificial neural network approach. The full-scale plant on which this study was performed was the 30 million gallons per day Viry water treatment plant situated near Paris, France. Includes 5 references, table, figures.