During the past years, there has been an increasing interest in the application of multi-criteria optimisation models to sustainable urban drainage system design. Generally, these models result in crowded multidimensional Pareto-fronts, which embody a large number of efficient non-dominated solutions. However, since the human brain can only analyse a limited amount of information at a time, multi-criteria, and especially many-objective, optimisation models are usually complemented by difficulties on the way to a final decision. This study marks the first attempt to solve this issue by reducing the number of non-dominated solutions obtained by many-objective optimisation of a sustainable drainage system. To this end, a soft clustering algorithm is implemented, which discovers similarities between the solutions, partitions the front accordingly, and picks representative solutions while preserving the distributional structure of the front. Two decision support tools are also introduced with a sense of long-term interest and objectivity while satisfying decision-makers’ explicit preferences. These measures exemplify overall performances of the portfolios in the reduced front and illustrate their future pathways in terms of rainfall scenarios with various intensities. In this way, while decision-makers’ primary requirements are fulfilled, they can make consistent decisions within a decision-making approach that is repeatable for future projects. According to the obtained results, the implemented clustering algorithm is able to narrow down the resultant Pareto-front with thousands of solutions into a handful of solutions as the front representatives. Results also show the merit of the proposed decision-making methodology in providing high-level information about the Pareto-optimal solutions with a significantly smaller amount of information.
