Prioritizing water distribution pipelines rehabilitation using machine learning algorithms

Abstract The majority of water pipelines are subjected to serious deterioration and degradation challenges. This research examines the application optimized neural network models for estimating condition in Shaker Al-Bahery, Egypt. proposed hybrid compared against classical network, adaptive neuro-fuzzy inference system, group method data handling using four evaluation metrics. These metrics are; Fraction Prediction within a Factor Two (FACT2), Willmott's index agreement (WI), Root Mean Squared Error (RMSE), Bias (MBE). results show that trained Particle Swarm Optimization (PSO) algorithm (FACT2 = 0.93, WI 0.96, RMSE 0.09, MBE 0.05) outperforms other machine learning models. Furthermore, three multi-objective swarm intelligence algorithms applied determine near-optimum intervention strategies, namely PSO salp optimization, grey wolf optimization. performances aforementioned evaluated Generalized Spread (GS), (Δ), Generational Distance (GD). yield (GS 0.54, Δ 0.82, GD 0.01) exhibits better when algorithms. obtained solutions ranked new additive ratio assessment relational analysis decision-making techniques. Finally, overall ranking is approach based on half-quadratic theory. aggregated obtains consensus trust level 0.97.