Drinking water treatment plants automation becomes more sophisticated, more on-line monitoring systems become available and integration of modeling environments with control systems becomes easier. This gives possibilities for model-based optimization. In operation of drinking water treatment plants, the processes are usually optimized individually on the basis of "rules of thumb" and operator knowledge and experience. However, changes in operational conditions of individual processes can affect subsequent processes and an optimal operation, which can include a number of water quality parameters, costs and environmental impact is different for every operator. Improvement of the operation of a drinking water treatment plant is possible by using an integrated model of the entire water treatment plant as an instrument for operational support and for process control. For this purpose, it is important that explicit objectives are defined for the operation. From the research it is concluded that the objective for integrated optimization of the operation of drinking water treatment should be the improvement of water quality and not a priori reduction of environmental impact or costs. In the research an integrated model for ozonation, including ozone decay, bromate formation, assimilable organic carbon (AOC) formation, E. coli disinfection, CT and decrease in UV absorbance at 254 nm (UVA254) is developed. With the model, different control strategies for ozonation are assessed. The research also describes a newly developed design for ozone installations, the dissolved ozone plug flow reactor, (DOPFR) and the effect of character and removal of natural organic matter (NOM) prior to ozonation. The research was carried out as part of the project Promicit, a cooperation of Waternet, Delft University of Technology, DHV B.V. and ABB B.V. and was subsidized by SenterNovem, agency of the Dutch Ministry of Economic Affairs. Part of the experiments was performed in cooperation with Kiwa Water Research.