Ahead of the 2024 Predictive Aircraft Maintenance Conference in Dublin on 13-14 November, MRO Management asked David Harper, fleet support director at GE Aerospace, who will be speaking at the event, his thoughts on the landscape of predictive maintenance.
What do you consider to be the most significant advancements in predictive maintenance technology within the industry in recent years; what impact have they had?
The growth in computing system capability from ‘under desk’ to ‘on premise’ to the ‘cloud’ has meant more reliable, scalable and secure systems to protect the fleet. We’ve been able to manage the rapidly growing amount of data, not just from more engines flying, but also because the modern engines have more sensors and provide richer data sets. It is the exponential growth in computing power, combined with the growth of even richer data sets, which continues to expand our capabilities with predictive maintenance.
In parallel with these trends, the analytics themselves have become more powerful, with artificial intelligence (AI) and machine learning (ML) becoming part of the tool kit. The use of AI/ML is enabling engineers to configure systems more readily without requiring computer science expertise.
Finally, we’re also seeing young engineers graduating from universities today with a stronger data science background who recognise the importance of data analytics and are bringing fresh ideas to transform the industry. GE Aerospace has made significant strides in applying AI/ML to predictive maintenance.
Can you highlight the key milestones achieved during this journey, the lessons learned along the way, and the current state of these initiatives?
Indeed, we’ve been using AI/ML to great effect for more than a decade in our engine health monitoring solution. Our ProDAPS software, highlighted at the 2023 PAM conference, can detect issues 60% earlier and reduce false alerts by 50%. The capability of the system has improved over the years with the rapid growth in computing power and the increasingly rich data set from engines around the world, especially the modern engines such as GEnx or CFM LEAP. It has also confirmed that engineering domain expertise is key to success, from choosing the right parameters to monitor and providing supervised learning to determining the best recommendations before we issue an alert to the airline.
As predictive maintenance solutions grow more sophisticated, data volume has surged. How can the industry enhance data quality and accuracy in the face of this rapid expansion?
Indeed, more data is helpful but only if it is of high quality and comes in a timely manner. We’ve seen big strides in the automatic transmission of data in near real-time, which reduces the risk of errors and allows a recommendation to be made earlier. Use of common standards across the industry, such as S5000F, will allow for easier data exchange which is why GE is involved in this effort. Breaking down data silos between legacy systems is important and industry platforms such as Skywise help overcome those barriers. Using standards and platforms will also ensure data is managed consistently in accordance with new laws and regulations coming from governments around the world. Planes fly all over the world and so we need to have a common approach.
To what extent do you believe industry collaboration poses a challenge to the advancement of predictive maintenance? Has GE Aerospace undertaken any initiatives to foster greater collaboration across the aviation ecosystem?
As we saw last year at the PAM conference, collaboration is the key word for better predictive maintenance and GE Aerospace is at the forefront of that.
We work closely with airlines to share information and learnings to enhance our support, always being careful to respect each airline’s data and privacy. Our Software-as-a-Service (SaaS) business is one of the founding partners of the Digital Alliance together with Airbus and Delta, each bringing their expertise to a joint effort to help airlines achieve higher productivity. We are also working with our suppliers to share data where it can help them improve the quality of their products.
What emerging trends in predictive maintenance excite you the most, and how do you envision the landscape evolving over the next 5-10 years?
With the increasing application of AI and ML, we see predictive maintenance becoming more personalised in ways that allow us to optimise maintenance for specific airline operations. It’s somewhat analogous to the notion of personalised medicine in health care, where health care treatments become more tailored to an individual patient.
AI and ML will enable maintenance to become more personalised by improving ways to find anomalous data and to find solutions to more complex multi-variable problems. This will allow for optimising maintenance per individual airline operations.
As we increase the application of AI and ML, we’re more committed than ever to making sure we bring the right domain expertise and right data to inform our predictive maintenance technology. Also, AL and ML will continue to be a way for us to augment and enhance the productivity and quality of service our engineers deliver.
Predictive maintenance also will benefit greatly from future aircraft having even more sensors than today gathering data. This increase in available data is happening as data transmission capabilities also are growing exponentially, as global satellite systems become more accessible. You can see how this impacting In Flight Entertainment (IFE) systems, where huge amounts of data can now be streamed to every passenger. For predictive maintenance, this will mean more data to work with to detect things earlier and take even better actions.
On the other hand, we also need to ensure we manage data properly, respecting ownership, rights of access, and always providing the highest levels of security so that we can all trust each other. As we said last year, industry collaboration is crucial to achieve that level of trust and accelerate adoption of technologies.
This feature was first published in MRO Management – October 2024. To read the magazine in full, click here.
PREDICTIVE AIRCRAFT MAINTENANCE (PAM) CONFERENCE 2024
On 13-14 November, MRO Management and its parent company Real Response Media will host the Predictive Aircraft Maintenance (PAM) Conference 2024 at the Radisson Blu Royal Hotel Dublin, Ireland, bringing together senior level executives from airlines, OEMs, MRO specialists and big data experts to explore the future of predictive maintenance.
Attendees will discover how to get the best out of predictive maintenance while overcoming the greatest challenges. Case studies will highlight best practices and how companies are successfully collaborating with industry partners, while panel discussions will explore the major topics from the perspective of all stakeholders.
Thank you to our Gold Sponsor, GE Aerospace, for supporting PAM 2024.
For more information and to find out how to register to attend, visit predictiveaircraftmaintenance.com