In various engineering sectors, optimizing maintenance procedures is crucial for ensuring both financial sustainability and profitability. This is particularly true in transportation, where air, road, and sea operators manage fleets of vehicles, striving to minimize downtime without compromising safety. Traditionally, scheduled-based maintenance, based on a damage-tolerant design, has been the most established approach for inspections and component replacement definition. However, advances in Structural Health Monitoring (SHM) are driving a shift towards condition-based and predictive maintenance strategies. This work presents a digital twin for monitoring a helicopter fleet, enhancing a condition-based maintenance policy, and evaluating the resulting life cycle cost (LCC). Unlike traditional models that typically use a failure rate to predict component degradation, the presented digital twin simulates the gradual degradation of components, providing a more nuanced understanding of the transition from operational to failure states. By simulating the actual degradation process, the developed digital twin enables real-time assessment of helicopter health status and remaining useful life (RUL), facilitating informed decisions regarding inspections and component replacements. Within this methodology, in addition to corrosion and fatigue degradation, stochastic rare events are also considered, as they can significantly impact overall costs and potentially lead to mission failures or the loss of aircraft. The proposed methodology has been successfully applied to a real-world case study and demonstrates potential for broader applications with other kind of machines. This fleet model can also be leveraged to develop algorithms for automatic decision-making on maintenance policies, such as those based on Artifial Intelligence.