Offshore wind farms play a critical role in achieving global renewable energy targets; however, their remote marine locations and harsh environmental conditions pose significant challenges to efficient operations
and maintenance (O&M). To improve ...
Offshore wind farms play a critical role in achieving global renewable energy targets; however, their remote marine locations and harsh environmental conditions pose significant challenges to efficient operations
and maintenance (O&M). To improve operational efficiency and reduce costs, effective maintenance planning and execution are essential. Nevertheless, many existing models fail to integratively account for
uncertainties and periodic characteristics inherent in metocean conditions and component failure rates, often resulting in inefficient schedules and increased operational costs. To address these limitations, this study
proposes a comprehensive maintenance planning framework that explicitly incorporates such uncertainties. An annual maintenance planning problem is first formulated as a stochastic mixed-integer linear programming
(MILP) model. The annual plan aims to minimize total O&M costs, including downtime losses, by integrating wake-effect constraints and probabilistic scenarios. This enables strategic allocation of maintenance
resources and determination of the required numbers of vessels and maintenance teams over a yearly horizon. While establishing long-term strategic decisions, the framework simultaneously considers short-term
operational conditions influenced by stochastic failures and weather variability. To further enhance planning accuracy, a weekly short-term scheduling model is developed to generate detailed maintenance
schedules. Maintenance activities are dynamically adjusted each week according to realized stochastic conditions, yielding realistic and implementable schedules. The weekly model incorporates preventive
maintenance, corrective maintenance, and opportunistic maintenance, and employs a flexible scheduling strategy that accounts for the operational efficiency of maintenance vessels. Computational experiments demonstrate that the proposed framework effectively reduces downtime losses by 19.0% and recovery delays by 38.2% through scheduling flexibility. Finally, based on the weekly scheduling outcomes, a daily crew transfer vessel (CTV) routing model is introduced. The model represents consecutive maintenance operations using a Work–Pickup node structure with duplicate nodes, accurately capturing real-world operational
procedures by minimizing technician waiting times and pickup-time deviations among vessels. As a result, efficient daily routing schedules are generated under various time-window and offshore accessibility
constraints. The proposed framework successfully integrates annual, weekly, and daily planning into a unified hierarchical maintenance planning system for offshore wind farms.
Keywords: Offshore wind farm, Maintenance scheduling, Mixed-integer program, Stochastic program,
Wake effect, Routing