This submission belongs to the session S2. Urban Water Modelling and Management of the event The 10th International Electronic Conference on Water Sciences
Published date
19 Nov, 2026
Academic Editor
Abbas Roozbahani
Citation
Laura Valencia-Marín, Pedro L. Iglesias-Rey, F. Javier Martínez-Solano, Daniel Mora-Melia, Hybrid GA–NSGA-II Optimization for District Metered Area Scheduling in Intermittent Water Supply Networks., in Proceedings of The 10th International Electronic Conference on Water Sciences, 24 November–27 November 2026, MDPI: Basel, Switzerland
Share
Email
Facebook
Twitter
LinkedIn
Hybrid GA–NSGA-II Optimization for District Metered Area Scheduling in Intermittent Water Supply Networks.
Laura Valencia-Marín 1
Pedro L. Iglesias-Rey 1
F. Javier Martínez-Solano 1
Daniel Mora-Melia 1
1. Department of Hydraulic Engineering and Environment, Universitat Politècnica de València, Valencia, 46022, Spain
Abstract
Intermittent operation of sectorized water distribution networks poses a complex decision problem: determining when to open and close each sector to simultaneously minimize total supply volume deficit and maximize water distribution equity. NSGA-II (Non-dominated Sorting Genetic Algorithm II) is well suited to address these competing objectives by identifying a set of non-dominated solutions.
This study develops a hybrid strategy combining single-objective genetic algorithms (GAs) with NSGA-II to optimize the operating schedules of district metered areas (DMAs). Solutions obtained by optimizing each objective separately are used to construct a guided initial population, whose performance is compared against a randomly initialized population. Each operating alternative is simulated in EPANET using pressure-driven analysis (PDA), ensuring that delivered nodal demand depends on available pressure throughout the simulation horizon. The formulation uses a sector-based chromosome encoding adapted to the simulation horizon and hydraulic time step, incorporating dynamic constraints for minimum nodal pressure, minimum operating tank levels, and tank-level recurrence.
The guided initialization, built from the single-objective solutions, broadened the coverage of the Pareto front and significantly reduced the number of generations required for NSGA-II to reach convergence, compared with random initialization. These results show that incorporating objective-specific information into the initial population improves search efficiency and the diversity of the non-dominated solution set, providing quantitative and reproducible criteria to support operational decision-making in water distribution networks with intermittent supply.
Keywords
Intermittent water supply
District Metered Areas
Multi-objective optimization
Pareto front
Pressure-driven analysis
Water supply equity
Scenario-Based Performance Assessment of Water Distribution Networks under Perturbations: A Replicable Multi-Dimensional Framework
From Defense to Containment: Historic Enclosure and Flood Vulnerability in the Castle of Ioannina