In Europe the interest in and the importance of vertical breakwaters is growing. A central point is the optimal geometry, e.g. the width and height of the breakwater caisson chosen such that the total costs over the lifetime of the structure are minimized. Probabilistic design tools provide several methods to determine the probability of failure of a structure. In a design process however, the designer of the structure is faced with the problem of defining the acceptable probability of failure. In general there are three ways to determine the optimal probability of failure: • Consider the probability of dying of an individual due to collapse of the structure (individual point of view); • Consider the probability of occurrence of a certain number of casualties in case of failure of the structure (societal point of view); • Minimize the sum of initial investment and capitalized risk over the lifetime of the structure (economical optimization). In the case of a breakwater without amenities the probability of loss of life due to failure is very small, but the economic losses can be severe. Therefore the application of the economical optimization is suitable. In this study a framework for the optimization of vertical breakwaters is developed. The optimization procedure has been implemented in a numerical model. In this model three failure modes are considered: • Sliding of the caisson over the rubble foundation (ultimate limit state); • Overturning of the caisson (ultimate limit state); • Wave transmission (serviceability limit state). Several calculations have been made with the computer program. The results of the calculations show the following: • The capitalized risk has a large influence on the optimal geometry of the structure; • In general only one mechanism largely determines the probability of failure of an optimal designed vertical breakwater. This mechanism is in general the one which is most expensive to strengthen the breakwater for. In the situation chosen in this stUdy, this is rotation failure of the caisson. The caisson width is the most expensive design variable; • The optimal design is influenced by all random variables used in the design; • Wave transmission influences the optimal geometry of a breakwater towards higher and narrower caissons. In the situation chosen in this study, a caisson height such that no wave transmission occurs seems optimal. The developed model provides a good starting point for the development of more advanced optimization models.