Multi-objective optimization for pipeline systems: A maintenance model using NSGA-II considering flow capacity and total cost

Abstract

The maintenance of corroded pipelines poses a significant challenge to ensuring operational safety and efficiency. Effective maintenance strategies must consider various performance metrics with multiple competing objectives. A key issue in optimizing these maintenance strategies is balancing maintenance costs with flow rates. This paper addresses the maintenance challenges of complex pipeline systems affected by corrosion by establishing a multi-objective optimization model, which considers both maintenance costs and system flow rates as optimization objectives. A novel chromosome encoding method is proposed to solve the model using the non-dominated sorting genetic algorithm-II (NSGA II). Compared to traditional empirical strategies, the strategies developed through the proposed multi-objective optimization method reduce total costs by 1.84 %–7.25 % and improve system delivery flow rates by 1.07 %–15.33 %. The effectiveness and universality of the proposed method are demonstrated through case studies of three pipeline systems with different structural complexities, failure probability thresholds and corrosion degradation models. Finally, comparative analyses with other multi-objective optimization methods (SPEA2, PESA-II, MOPSO, and MOEA/D) and sensitivity analyses show that the proposed NSGA II-based strategy exhibits superior performance in terms of convergence and effectiveness.