Satavia
Contact: Dr. Adam Durant, CEO; adam.durant@satavia.com
Feedback - proposal needs to be more self-contained, rather than just "improve our product"
Environmental factors in the atmosphere, like dust, ice, sulphur, and volcanic, accelerate wear of aircraft components. Modern aircraft generate large volumes (several GB) of data per flight and digital predictive analytics technology has great potential to support asset health monitoring, and disrupt traditional maintenance and flight planning processes. The solution combines high spatio-temporal environmental factor analysis with machine-learning to provide advanced risk-based decision-making capability.
Data driven model predictions (in this case Satavia’s environmental factor analyses) inevitably contain some level of uncertainty, due to the reliance on a finite (typically small) sample of direct observational measurement data, or coarse resolution input meteorological data used to drive the NWP model. The project will demonstrate advanced environmental data analytics capability through the application of artificial intelligence (AI) techniques such as reinforcement learning (RL), to develop a framework for sequential decision-making under uncertainty. The objectives will include:
1. Improve the skill of the NWP-based model using training data acquired from a local in situ measurement network; 2. Correlate aircraft environmental factor exposure to component wear rates and proxy data provided by engine health management; 3. Develop autonomous decision-making capability to adjust aircraft flight plans and engine maintenance plans in near-real-time.
Aircraft original equipment manufacturers (OEM), operators, and maintenance repair organisations now increasingly share maintenance liability through ‘power-by-the-hour’ services for the lifetime of the aircraft. Unscheduled maintenance costs the industry billions of dollars each year (e.g., in 2016 sulphurous air pollution cost Rolls-Royce >£65M) and causes disruption to airline operator flight schedules. Customers currently implement highly conservative maintenance schedules at pre-defined maintenance frequency. Predicting when to do the maintenance based on environmental exposure could make huge savings for the industry. Satavia’s data-as-a-service enables aircraft operators to modify aircraft flight plans and scheduling to extend engine lifetime, while engine manufacturers can proactively adjust maintenance plans to minimise unscheduled maintenance. Satavia is currently working on a series of ‘proof-of-value’ projects with a UK-based aircraft engine manufacturer.
