If artificial intelligence has not yet transformed your aviation MRO, it soon will
GenAI offers to partner in measures which boost safety, enhance efficiency, improve performance, automate routine tasks, minimize human errors, and stimulate creative endeavor.
By Don Van Dyke
ATP/Helo/CFII. F28, Bell 222
Pro Pilot Canada Technical Editor
Using content derived from AI (eg, text, audio, imagery, synthetic data), Generative AI (GenAI) uses new or substituted data to innovate insights and perspectives (pattern creation) for advanced analysis, prediction, and provision of advice, guidance, or recommendations.
GenAI is particularly well suited to knowledge-based and data-intensive endeavors like aviation maintenance, repair, and overhaul (MRO), which relies on the analysis of differently formatted and sourced information.
In simulating superior learning, analytical, and decision-making skills of the human brain, the promise of GenAI is virtually without boundaries.
However, for high-consequence, low-probability endeavors like aviation, exacting diligence in installation, governance, and control is crucially important in confidently employing GenAI to produce new and relevant content.
MRO market
Aviation MRO comprises 4 main segments: airframe, engine, components (including avionics, hydraulics, electrical systems), and line maintenance (including minor repairs). Management consultancy Oliver Wyman reports that the global spend on MRO in 2023 reached approximately $104 billion and is expected to grow at an annual rate of 1.8% to $124 billion by 2034.
An additional and separate segment has regard to upgrades and modifications to improve performance or to comply with new regulations and may be treated in a future article.
Goldman Sachs suggests that the economic potential of GenAI could boost global labor productivity by more than 1% per year in the decade following widespread usage.
While the market value of aviation-specific GenAI is not usually identified separately, the overall trend indicates significant growth and investment in this technology.
GenAI Drivers
Identifying how GenAI can be applied effectively to these challenges is fundamental to justifying its application to the operational, regulatory, and market demands made on MRO.
Globally, aviation MRO pursues several key goals while facing a number of challenging trends. Table 1 presents a range of traditional MRO segment goals together with a summary of ways in which AI/GenAI can be applied.
Regulatory compliance. The web of global and regional regulations is increasingly complex. MRO providers must ensure strict adherence to safety and quality requirements while adhering to evolving regulatory frameworks.
Growth and demand. Current MRO growth and backlog management is driven primarily by fewer aircraft retirements and unexpectedly rapid air transport traffic recovery.
Concurrently, industry pressure seeks to improve aircraft availability and reduce maintenance intervals safely.
To achieve these goals, reliability engineering is trusted to ensure that aircraft operate without failing. Traditionally, reliability engineers evaluate vast quantities of unstructured maintenance records to recognize failure patterns, predict adverse outcomes (out-of-service time), and develop strategies for their avoidance.
GenAI can achieve the same result almost instantaneously and, moreover, can generate new patterns not yet experienced for analysis and identification of potential threats.
Cost management. GenAI can be used to contain rising cost trends by improving workforce planning, eliminating paper processes, and optimizing maintenance programs.
Skilled workforce shortage. Oliver Wyman reports that maintenance workforce costs rose by an average 7.3% worldwide last year, exacerbated by a global shortage of skilled and experienced labor.
McKinsey & Company forecasts that by 2033, 20% of all MRO positions will be unfilled, rising to a shortage of 70,000 by 2033. MRO talent and labor strategies will require skill sets including decision-making and advanced technical knowledge.
As experienced professionals retire, attracting new talent is essential. Investment in training and development programs is critically important to building a durable and competent workforce.
Promising MRO mitigating strategies may also include the introduction of GenAI virtual maintenance and repair experts (so-called “VTechs”).
Technical staff could engage with these digital assistants to diagnose faults or research unstructured and underutilized information sources such as data from previous repairs to aid in troubleshooting.
VTechs may also be used to complete maintenance reports, and generate and submit work/purchase orders, integrate records of newly acquired aircraft into enterprise resource planning (ERP) systems (an effort currently requiring extensive review and migration of an aircraft’s maintenance history).
McKinsey estimates that use of such virtual experts could improve maintenance technician capacity by 15–35% during a typical workday.
Sustainability. Responsible MRO providers proactively promote both operating and environmental sustainability.
Operating sustainability involves high-quality airframe, engine, and component MRO procedures which simplify, standardize, and speed processes. Achieving this strategy can reduce total costs by 8–15%, maintain a competitive advantage, and promote a culture of continuous improvement.
GenAI may be used in continuous research into methods and materials supporting these goals, as well as improvising strategic use of alternatives in MRO, including such examples as:
• renewable energy purchases
• solar panels
• reusing materials
• recycling
• Increasing use of recycled water or an effluent treatment plant to treat water from operations
• widespread use of electric vehicles
• teardowns
• reducing energy consumption with technologies like LED lighting
Many operators now expect their MRO providers to proactively pursue and achieve environmental sustainability goals. This must be done in a way that balances growth and environmental impact. GenAI may be used to adopt flexible scheduling, optimize APU utilization and reduce running hours, carbon emissions, and noise.
Supply chain and stores. Recent global events have highlighted the vulnerability of MRO supply chains. Routine replenishment, inventory management, and fluctuations in demand require that supply chains be made robust through strategic stockpiling, diversification of suppliers, and adoption of digital tools to anticipate and resolve potential disruptions.
GenAI can analyze disparate communication and delivery patterns to automatically identify early signs of supply and delivery challenges. Moreover, GenAI may offer supply chain analysts strategies to mitigate these prospects proactively and enable parts management to meet rising demand without compromising service quality.
Back office. Another application of MRO cost mitigation is GenAI transaction risk analysis (TRA), which evaluates uncertainty factors that may affect the expected return from a deal or transaction.
TRA can include (but is not limited to) foreign exchange risk, commodity and time risk in procurement, human resources (HR) planning, financial models, and administration.
Important AI/GenAI skills to learn
While the potential rewards of integrating AI/GenAI into the evolving aviation MRO industry are impressive, its use and governance will require learning and using certain key skills.
Data analysis and interpretation. Understanding how GenAI analyzes and interprets data from diverse sources, such as sensors, maintenance logs, and operational records, is critically important. This includes familiarity with statistical analysis, data visualization, and the use of specialized MRO software.
Predictive maintenance. Skills in predictive analytics to anticipate maintenance needs and reduce downtime are increasingly valuable. These involve knowledge of machine learning (ML) and how maintenance data is used to train models, evaluate their performance, and deploy them in a production environment.
Decision-making. Especially considering that the maintenance technician is the final authority on use of GenAI recommendations, it is imperative that decisions be taken based on clear understanding of regulatory requirements, ethical standards, penultimate safety needs, and the processes leading to technically robust and informed decision-making.
Cybersecurity. As MRO relies increasingly on digital tools and networked assets, understanding cybersecurity principles to protect sensitive data and systems from cyberthreats is essential.
Communication and collaboration. Effective communication skills are essential to collaborate with cross-functional teams, including engineers, data scientists, and operations managers.
Guidance material on AI/GenAI in aviation
International. The International Civil Aviation Organization (ICAO) is exploring the potential of GenAI to revolutionize various aspects of aviation, including AI-generated scenarios to improve the skills and readiness of maintenance personnel. (See ICAO Doc 10098).
Europe. In 2023, the European Union Aviation Safety Agency (EASA) released its updated AI Roadmap 2.0 which is designed to ensure that AI “is developed and used in a way that is human-centric, trustworthy, and safe” and includes key ethical requirements to ensure dependable AI deployment.
Key aspects of the EASA roadmap include:
• aviation safety and security
• AI assurance to establish trustworthiness
• human factors considering the impact of AI on human roles and responsibilities
• ethical considerations related to AI deployment
EASA plans to certify the first AI integration in aircraft systems by 2025.
United States. The Federal Aviation Administration (FAA) issued its first Roadmap for Artificial Intelligence Safety Assurance in July 2024.
In harmony with the EASA Roadmap, this 31-page document emphasizes a human-centric approach to integrating AI in aviation, recommends AI safety assurance strategies, and presents 5 related key points:
• guiding principles which leverage extensive experience
• safety assurance involving incremental demonstration of AI technologies before their application in aviation
• collaboration of regulators and industry stakeholders to establish priorities for the safe introduction of AI in aviation
• a strategic framework of oversight to balance opportunities for AI innovation with maintenance of high safety standards
• regulatory guidance emphasizing the need for detailed guidance supporting certification and safe deployment of AI/ML technologies in aviation
Future trends
ChatGPT and other GenAI models can access all information publicly available on the Internet, but cannot access the valuable private intellectual property held internally by companies. It has been suggested that allowing access to that highly protected information would make available a world of new use cases for AI/GenAI to benefit users across the industry.
While ChatGPT is a language-learning model that outputs only text, GenAI can also create images and 3D models such as a digital twin, which is a virtual representation of an object or system designed to reflect a physical object accurately.
Digital twins represent a paradigm shift in innovation and MRO. Simulations operate in entirely virtual environments isolated from the external world. Digital twins are fitted with sensors that update their virtual counterparts continuously in real time with granular, high-quality data.
Aircraft developers and MRO providers often rely on these sophisticated models to design, build, operate, and monitor component lifecycles.
MRO technicians can use digital twins for predictive maintenance and to detect anomalies by comparing real-world sensor data with data generated by digital twins.
GenAI is poised for a significant transformation of the aviation MRO industry. Overall, the integration of GenAI is expected to improve the reliability and availability of aircraft, enhance operational efficiency, and reduce costs.
MRO areas in which future GenAI is expected to have the greatest impact include predictive maintenance, enhanced decision-making, efficiency and productivity, supply chain optimization, training, and support.
Aviation will benefit greatly from widespread use of this technology.
Don Van Dyke is professor of advanced aerospace topics at Chicoutimi College of Aviation – CQFA Montréal. He is an 18,000-hour TT pilot and instructor with extensive airline, business and charter experience on both airplanes and helicopters. A former IATA ops director, he has served on several ICAO panels. He is a Fellow of the Royal Aeronautical Society and is a flight operations expert on technical projects under UN administration.