An Analysis of the Interaction between Water Distribution System Design and Operations in Infrastructure Planning

<p dir="ltr">Water utilities increasingly seek flexible and adaptive strategies that align infrastructure interventions with the evolving nature of system uncertainties. Staged, flexible, and adaptive design approaches respond to this need. Nevertheless, the irreversibility of infrastructure investments inherently limits adaptive capacity. In contrast, operational reconfigurations can be executed more frequently and at reduced costs, offering a highly responsive mechanism to address evolving uncertainties. This work analyses how operational and design decision variables interact within Water Distribution Systems (WDS) planning problems. We propose a methodological framework to explicitly model and optimise operational adaptive behaviour within the WDS planning process. Using a multi-objective evolutionary algorithm, we optimise several bi-objective formulations of the Anytown benchmark problem, each employing distinct strategies for handling design variables and operational decisions. Our results demonstrate that WDS design optimisation problems incorporating operational decisions require sequential re-optimisation of the operations and multiple optimisation runs with varied operational settings to identify the global Pareto front. Only integrated optimisation of operational and design decision variables can reveal synergistic potential(up to 29% higher objective space coverage), as independently fixing either type inherently constrains the discovery of optimal solutions in the other domain.</p><p dir="ltr">This paper was presented at the 21st Computing and Control in the Water Industry Conference (CCWI 2025) at the University of Sheffield (1st - 3rd September 2025).</p>