A multi-utility and dynamic approach for the upgrade of an aged water distribution network

Water Distribution Networks (WDNs) along with other utilities are among the most important assets in every industrialized society. The rehabilitation and upgrade of WDNs refer to a very complex and multi-criteria problem which needs a comprehensive decision support system providing optimal renewal plans for the asset managers. There is a multitude of real-world problems faced with such activities. In this regard, this study aims to propose a novel method to cope with three practical challenges in WDNs rehabilitation activities including 1-budget limitation, 2-hydraulic deficiency and 3-correlation of WDNs with other infrastructures bringing the risk of cascading failures to a multiplex network. This results in a multi-objective optimization problem with three objectives which are cost, hydraulic and decoupled reliabilities. The problem is solved dynamically with the contribution of a nature inspired optimization algorithm, the Non-Dominated Sorting Genetic Algorithm (NSGA-II). The method is applied to a deteriorated water pipe network, and the results are compared with the ones obtained only by the consideration of two objectives, costs and hydraulic reliability showing how much the objectives are conflicted with each other.