Predictive Maintenance in Aviation Using Digital Twins

Kurzfassung

The way in which aviation maintenance is carried out today is largely determined by rather strict schedules for maintenance intervals, which are based on flight hours, flight cycles and age. This very conservative approach has proven to establish aviation as an extremely reliable mode of transport. However, it is also inherently inefficient. Aircraft vary in their specific configuration and the way they are actually used, thus resulting in very individual degradation. Current maintenance requirements do not account for this sufficiently. In order to align maintenance more closely to the actual and individual development and degradation of a component during its usage new concepts like predictive maintenance are being explored. A continuous monitoring of a component should allow to quickly identify relevant changes and their causes and predict future developments based on an intended usage. This does not only apply to the component itself, but also in the context of other components, which influence each other. Being able to perform such a continuous monitoring requires the capability of comprehensively representing real and individual components digitally using digital twins. These do not only represent a component as it is. They connect different sources of data, such as on-board monitoring and operational data, maintenance data, operator data, data on external influences, e.g. weather, and much more. Beyond that they link the real component and its associated data with operational planning, maintenance planning, models, configurations, requirements and processes. In the context of aviation maintenance there are many stakeholders involved throughout the life cycle of a component. Each stakeholder has different knowledge about a component. Some of this can be shared, some is a strongly protected part of the respective intellectual property and some might be offered directly or indirectly as a service. All these aspects have a significant impact on the design and potential of digital twins, their application and corresponding business models.