A safe, adequate and accessible supply of drinking water is one of the basic necessities of any human being. The most efficient and effective way to transport drinking water is through a network of pipelines.

Therefore, the water distribution network is one of the most vital elements of a society's infrastructure, providing people with high-quality drinking water in sufficient quantities and at an adequate pressure. The (re)construction of these networks requires major investments. Hence, an efficient network design is of crucial importance.

Optimally designing a water distribution network consists of finding the optimal diameter for every pipe in the network, such that the total network cost is minimized and all pressure and velocity constraints are satisfied. Traditionally, such design decisions are made on the basis of expert experience. When networks increase in size, however, rules of thumb will rarely lead to optimal decisions. For a small network of 200 pipes and 14 possible diameters, e.g., the number of possible solutions is roughly equal to the amount of chocolate m&ms that could fit into the sun. Finding the optimal solution is equivalent to finding the one peanut m&m hidden inside.

Optimization algorithms can support a decision maker by improving the efficiency and effectiveness of the design decisions. Over the past thirty years, a large number of techniques have been developed to tackle the problem of optimally designing a water distribution network. This PhD thesis provides a critical overview of the current state of the art, and offers some points of improvement. Another contribution of this PhD thesis is HydroGen, a tool that is developed to generate virtual water distribution networks for algorithm testing purposes. Moreover, two iterated local search algorithms are developed to tackle the basic version of the problem and an extended version of the problem that takes into consideration a multi-period water demand patterns and imposes an additional velocity constraint.