Multi-objective maintenance decision-making of corroded parallel pipeline systems

Abstract

Reasonable maintenance of pipeline systems is essential for ensuring energy transportation and achieving a trade-off between safe operation and maintenance costs. However, previous studies have barely considered the system structure, system availability, and multi-stage corrosion. To fill this gap, a multi-objective optimization framework was established. This framework can balance the cost and downtime caused by maintenance to achieve the desired trade-off between the maintenance cost and transportation continuity of a system with multi-stage corrosion defects and multi-level reliability constraints. By combining the NSGA-II algorithm with the general structure of pipeline systems, chromosome coding was designed to improve the efficiency of the evolutionary algorithm, and a series of preventive maintenance strategies were proposed, including decisions regarding repaired segments, time point for maintenance, and the type of maintenance mode. The effectiveness and accuracy of the methodology were illustrated by considering the pipeline systems with different structures and parameters as examples. A comparison of multi-objective programming with empirical decision and single-objective decisions demonstrates the rationality and universality of the methodology. The proposed methodology can help adjust the corresponding maintenance strategies for diverse requirements and provide reasonable maintenance schemes for corroded pipeline systems.