Maximizing failure occurrence in water distribution Systems: A Multi-Objective approach considering Reliability, Economic, and environmental aspects

Abstract

National and local regulations mandate water utilities to deliver safe and clean drinking water to consumers. However, due to the aging water infrastructure, they face numerous challenges regarding water loss and water quality impairments within the distribution networks. In this paper, a multi-objective nonlinear model was developed to maximize the buried water mean time to failure with consideration of cost and environmental aspects. The model determined the type of pipe material appropriate for each location with the maximum reliability, minimum environmental impacts, and minimum cost. An exact method (weighted sum method (WSM)) and two meta-heuristic algorithms (multi-objective particle swarm optimization (MOPSO) and non-dominated sorting genetic algorithm II (NSGA-II)) were applied to solve the model in the large dimension. After examining the proposed solution methods across different numerical examples, the best optimal approach was identified. Then, the mathematical model and the best solution approach were validated in a real case study of the drinking water distribution system (WDS) in Iran. Moreover, the inherent uncertainty associated with the total cost objective function were assessed by conducting a sensitivity analysis on key parameters such as the annual interest rate, pipe cost, probability of pipe installation, and installation cost. Therefore, to overcome these uncertainties, a robust optimization method introduced by Soyster was used. Ultimately, the study provided managerial insights drawn from the results of the sensitivity analysis. The findings can aid water utilities and decision-makers in formulating water infrastructure rehabilitation plans that are more effective and characterized by diminished environmental impacts and lower costs.