Ten Industry Professionals Weigh In on AI's Practical Value in MRO Software

AviTrader's MRO 360° series examined how artificial intelligence, advanced data analytics and digital integration are reshaping aircraft maintenance, repair and overhaul software. The article names the main challenges: integrating decades of maintenance data, connecting disparate systems and ensuring the reliability of digital tools, in an industry where accuracy and traceability are non-negotiable. It also raises the question of how to separate meaningful innovation from marketing hype.

AviTrader put questions about emerging trends, including where AI already delivers value, to ten industry professionals. Monica Badra, Founder and CEO of Aero NextGen, Inc, said MROs running modern ERP platforms with AI modules are already seeing measurable reductions in AOG events. Those platforms correlate fleet utilisation data, shop findings and OEM reliability reports to flag components before failure. She added that AI-powered extraction tools are shortening the time technicians and engineers spend pulling reference data from AMMs and CMMs, without eliminating human review, which she described as a meaningful labour saving for MROs processing high volumes of AD compliance.

Eric Schech, Director of Product Marketing & Digital Media at Ambry Hill Technologies, said measurable value comes from bounded, high-volume work rather than chatbots diagnosing aircraft or approving maintenance. He cited Vista-Quote, whose AI can recognise a plain-text RFQ email, stage it in a standardised format and link company and part numbers to available history. Configurable rules, not generative AI, handle routing, inventory checks, restrictions and auto-quoting, while people retain authority over exceptions.

Scott Wells, Managing Director at MRO-PRO, said MRO software has historically planned known work well but struggled to anticipate emerging defects, parts, tools, services and additional labour, partly because data fragmentation prevented analysis across broader datasets. He said AI can understand maintenance intent rather than simply matching labels, which increases the volume of usable historical data and strengthens planning probability models. MRO-PRO's MILE™ Maintenance Intelligence & Learning Engine is being developed around this principle.

The three accounts converge on one pattern: the value being claimed sits in constrained tasks where a person still signs off. Reference-data extraction, RFQ intake and flagging components ahead of failure all share that shape, and Schech's description of rules rather than generative AI handling routing and auto-quoting shows how vendors are splitting deterministic logic from AI in a regulated trade.

That split follows from how MRO works. A maintenance record has to be traceable and a compliance finding has to be defensible, so software can assist with preparation and prediction long before it is trusted with approval. The sources frame AI as shortening preparation time and sharpening planning probabilities, while authority over exceptions and compliance stays with certified staff.

The unresolved test is the one the series itself names: whether decades of fragmented maintenance data can be integrated and connected well enough for these models to generalise beyond early adopters on modern ERP platforms. Wells's point that fragmentation has blocked analysis across broader datasets is the clearest statement of why that integration sets the ceiling for everything built on top of it.