IAG’s AI system cuts aircraft maintenance planning from weeks to minutes

International Airlines Group (IAG) has developed an in-house AI-powered Engine Optimisation System that dynamically reschedules aircraft engine maintenance by running millions of “what-if” scenarios daily. The system, initially implemented with Aer Lingus and set to roll out across IAG’s other airlines by year’s end, addresses the complex challenge of balancing regulatory requirements, parts availability, labor constraints, and operational continuity while potentially saving the industry millions in maintenance costs.

What you should know: IAG’s AI system transforms weeks of manual maintenance planning into minutes of automated optimization, helping airlines avoid costly Aircraft On Ground emergencies and passenger delays.

• The system was built within IAG’s London and Barcelona-based AI Labs to solve the complex scheduling puzzle of engine maintenance across regulatory mandates, parts inventory, and flight operations.
• By updating maintenance schedules dynamically as new data flows in, the system re-plans in real time rather than requiring manual intervention when disruptions occur.
• IAG chose to develop the technology in-house rather than licensing existing solutions, allowing them to retain intellectual property and avoid vendor lock-in.

The big picture: With the aviation industry set to spend over $100 billion annually on maintenance, repair and overhaul (MRO) by 2030, even small efficiency gains translate to massive financial impact.

• McKinsey, a global consulting firm, estimates AI-driven maintenance could cut costs by 20% and eliminate up to half of unscheduled repairs across the industry.
• The system also provides sustainability benefits by reducing last-minute flight repositioning and charter needs, lowering fuel burn and CO₂ emissions.

How it works: The AI system treats engines, maintenance slots, and spare parts as interconnected variables in a complex optimization problem.

• Starting with the CFM56 engine common in narrow-body aircraft, IAG’s team proved the concept before expanding to other engine families across their 700-aircraft portfolio.
• The system doesn’t just predict when engines need service—it prescriptively chooses optimal maintenance slots that minimize ground time across the entire fleet.
• Human oversight remains central, with the system presenting schedules and decision factors for engineer approval rather than automating decisions completely.

Competitive landscape: IAG’s approach differs from other airlines’ AI maintenance initiatives through its focus on prescriptive optimization rather than just predictive analytics.

• Lufthansa Technik’s Aviatar platform focuses on predictive diagnostics for fault detection, serving over 100 airlines.
• Delta Air Lines’ APEX system improved parts-demand accuracy from 60% to 90% through real-time sensor data analysis.
• Air France-KLM is working with Google Cloud to add generative AI tools to its existing “Prognos” analytics platform.

What they’re saying: IAG’s leadership emphasizes the system’s role in empowering human decision-making rather than replacing it.

• “By applying advanced algorithms, we’re making our engine maintenance programme more efficient. We are avoiding unnecessary maintenance delays to ensure that our fleet is available and in service,” explains Ben Dias, IAG’s chief AI scientist.
• “The system gives our people the data and tools they need for smarter planning and better teamwork,” Dias added.

Key challenges: Implementation faces significant hurdles around data quality and organizational change management.

• Aviation data remains messy, with inconsistent maintenance logbook entries, paper-based records, and multiple naming conventions for parts requiring months of cleaning and standardization.
• Change management proves equally challenging, as engineers accustomed to traditional planning methods may resist probabilistic recommendation systems.
• Trust building becomes crucial when planners need confidence to act on AI recommendations for critical safety decisions.

Looking ahead: Future developments could expand the system’s capabilities through federated learning across airlines and integration with broader operational systems.

• Anonymized model insights could be shared across member airlines to improve datasets without exposing commercially sensitive information.
• Long-term integration with live flight operations and crew roster data could create unified disruption management and maintenance planning systems.