Artificial intelligence is transforming how military aviation operators manage aircraft maintenance, optimize fleet availability and sustain operational readiness. As armed forces face increasing pressure to maximize the capabilities of existing aircraft and resources, AI-driven maintenance optimization is emerging as an important tool for improving operational efficiency. Christophe Billiottet, Founder and CEO of ADGS, is advancing this approach through DOM AI, a fleet-wide optimization platform designed to analyze billions of maintenance scenarios, anticipate operational constraints and help defense aviation operators improve aircraft availability without necessarily expanding their fleets or maintenance infrastructure.
In an exclusive interview with Global Defense Insight (GDI), Billiottet, who trained as a fighter pilot early in his career, discusses how artificial intelligence can reshape military aircraft maintenance and operational decision-making. He explains the capabilities of DOM AI, its application to military helicopter fleets in Qatar, and how integrating predictive maintenance technology such as TRAKT can help operators transition from anticipating component failures to optimizing maintenance decisions across entire fleets. The discussion also explores human oversight in AI-assisted defense operations, resource optimization, and emerging opportunities for military helicopter and fixed-wing operators, maintenance, repair and overhaul (MRO) organizations, and the wider Gulf defense market.
1. How has your background as a trained fighter pilot shaped ADGS’s approach to AI for defense aviation and fleet operations?
I trained as a fighter pilot early in my career, and although it was for a relatively short period, the experience gave me a very direct understanding of what aircraft availability means from an operational perspective.
For a pilot and for a squadron, availability is not an abstract maintenance KPI. What ultimately matters is whether the right aircraft is mission-ready when it is needed. An aircraft may exist on the inventory, but if it is unavailable because of maintenance, a missing part, insufficient technical resources or poor maintenance synchronization, it has no operational value at that moment.
That perspective has strongly influenced the way we developed DOM. We did not want to build AI simply to automate maintenance planning or generate another schedule. We wanted to address the problem from the operational objective backwards: how can we use the aircraft, technicians, spare parts, infrastructure and time already available to maximize fleet readiness?
My aviation background also reinforced another principle that is fundamental to ADGS: AI should support experienced people, not replace their judgement. In defense aviation, operational reality changes quickly and human expertise remains essential. The role of DOM is to analyze billions of possible scenarios in minutes, anticipate conflicts sometimes years ahead, and give commanders, engineers and maintenance teams better options on which to base their decisions.
So, in many ways, DOM reflects both perspectives: the pilot asking, “How many aircraft can I actually fly?” and the maintenance organization asking, “How do we make that possible with the resources we have?”
2. What does DOM AI do differently from conventional MRO and maintenance-planning systems when managing an entire aircraft fleet?
DOM AI goes beyond conventional MRO and maintenance-planning systems by optimizing the entire fleet as one dynamic system.
Traditional MRO platforms are primarily designed to record, manage and schedule maintenance activities based on predefined rules and individual aircraft requirements. DOM adds an intelligent optimization layer on top of these existing systems.
Instead of optimizing each aircraft independently, DOM continuously evaluates the entire fleet over a multi-year horizon, taking into account maintenance requirements, aircraft configuration, operational priorities, hangar capacity, maintenance slots, technicians and skills, spare-parts availability, tooling and other ground resources.
Its AI-driven optimization engine can then move, group, anticipate or defer maintenance activities within their authorized limits to find the best global maintenance strategy for the fleet.
Most importantly, DOM dynamically re-optimizes when reality changes for example after an unexpected failure, parts shortage, manpower constraint, operational deployment or change in mission priorities.
The objective is not simply to produce a maintenance schedule. It is to continuously determine the best achievable maintenance strategy to maximize aircraft availability and operational readiness while reducing cost and resource pressure without compromising airworthiness or regulatory constraints.
3. ADGS reports that DOM AI helped increase Qatari defense helicopter availability from 37% to 75% by analyzing millions of maintenance scenarios. What operational changes drove that improvement, and how transferable is this model to other military fleets?
The improvement from 47% to 75% availability was not achieved by relaxing maintenance requirements or changing airworthiness rules. It came from fundamentally improving how maintenance was orchestrated across the fleet.
DOM AI uses Genetic Algorithms to analyze billions of possible maintenance scenarios and identified better ways to sequence, group, anticipate and schedule maintenance activities, while simultaneously accounting for aircraft status, maintenance deadlines, hangar and slot capacity, technician availability and skills, spare parts, tooling and operational requirements.
The key change was therefore moving from largely aircraft-by-aircraft and sequential planning to fleet-wide optimization. DOM can identify situations where a relatively small change in the timing or allocation of maintenance resources could prevent several aircraft from becoming unavailable by maintenance overlapping at a later time, sometimes years beforehand. This long term sight significantly reduced unnecessary downtime and maintenance bottlenecks while preserving all mandatory technical and airworthiness constraints.
The model is highly transferable because the optimization methodology is not specific to one helicopter type or one Air Force. DOM can be configured around the maintenance program, fleet composition, resources, operational rules and constraints of each operator.
This is particularly relevant for military fleets, where readiness requirements can change very rapidly. DOM can re-optimize around missions, deployments, conflicts, accidents, unexpected failures, spare-parts shortages or changes in available maintenance personnel, continuously seeking the best achievable mission-capable fleet.
The exact improvement will naturally vary from one fleet to another, but the principle is universal: optimize maintenance at fleet level, under real operational constraints, to maximize the number of aircraft available when they are actually needed.
4. Can DOM help armed forces improve aircraft readiness without adding more aircraft, manpower, or maintenance infrastructure? If so, where do the biggest gains usually come from?
Yes. This is precisely one of DOM’s main objectives: increasing aircraft readiness by making better use of the resources already available, rather than requiring additional aircraft, manpower or maintenance infrastructure.
The biggest gains typically come from eliminating inefficiencies that are difficult to identify through conventional planning: future maintenance overlaps, suboptimal sequencing of maintenance events, poorly synchronized resource allocation, unnecessary downtime, and bottlenecks involving technicians, hangars, tooling or spare parts.
By analyzing the fleet globally over a long-term horizon, DOM can anticipate these conflicts months or even years in advance and determine when maintenance should be grouped, anticipated, rescheduled or redistributed within authorized limits. This helps avoid situations where several aircraft become unavailable simultaneously or where an aircraft remains grounded because one critical resource is missing.
DOM also continuously adapts the plan to the resources actually available. If spare parts, qualified technicians, maintenance slots or infrastructure become constrained, the system can re-optimize the fleet maintenance strategy around those limitations.
In military operations, this becomes particularly valuable because the objective is not simply to maximize average availability. It is to have the right number and type of mission-capable aircraft available when operational demand requires them.
In many cases, therefore, the most cost-effective way to increase readiness is not to add capacity, but to optimize the use of existing capacity across the entire fleet.
5. How does combining DOM with TRAKT move operators from simply predicting a component failure to deciding when and how maintenance should be carried out across the whole fleet?
TRAKT and DOM address two different but complementary questions: TRAKT helps determine what is likely to fail and when; DOM determines what to do about it, when to do it, and how that decision should be coordinated across the entire fleet.
A failure prediction by itself does not necessarily tell an operator when the best time is to intervene. Replacing a component too early wastes remaining useful life, while waiting too long may create an unscheduled grounding and disrupt operations.
By feeding TRAKT’s predictive information into DOM, a predicted failure becomes an additional dynamic input to the fleet-wide optimization process. DOM can evaluate the remaining useful life of the component against scheduled maintenance events, aircraft utilization, mission requirements, spare-parts availability, technician skills, hangar capacity and the maintenance requirements of every other aircraft in the fleet.
It can then determine whether the intervention should be anticipated and combined with an already planned maintenance event, scheduled separately at the most appropriate time, or coordinated with other fleet maintenance activities to minimize operational disruption.
The result is a shift from predictive maintenance to prescriptive fleet maintenance: not simply “this component is likely to fail in 200 flight hours,” but “given everything happening across the fleet, this is the best aircraft, time, maintenance window and resource allocation for addressing that risk.”
TRAKT provides foresight. DOM turns that foresight into an optimized fleet-wide maintenance decision.
6. In mission-critical defense environments, how does ADGS make DOM’s recommendations explainable, auditable, secure, and ultimately subject to human decision-making?
In mission-critical defense environments, DOM is designed as a decision-support system, not an autonomous decision-maker. The final decision always remains with the authorized maintenance, engineering and operational personnel.
Every recommendation produced by DOM is based on identifiable inputs and constraints, maintenance requirements, airworthiness limitations, aircraft configuration, operational priorities, available manpower, spare parts, tooling, facilities and other resources. The system can therefore show why a particular maintenance strategy was recommended, which constraints influenced it, and what alternatives or trade-offs were considered.
This traceability is essential for auditability. Recommendations and planning changes can be recorded together with the underlying assumptions, constraints and data used to generate them, allowing engineering and maintenance teams to review and validate the proposed solution before implementation.
Security is equally fundamental. DOM can be deployed within the operator’s controlled and secure IT environment, with access to sensitive fleet and operational data governed according to the customer’s security requirements. This is particularly important for military operators, where aircraft status, availability and mission planning information may be highly sensitive.
Most importantly, DOM does not override technical authority or operational command. It processes a level of complexity and a number of possible maintenance scenarios that would be extremely difficult for a human planning team to evaluate manually, and presents optimized courses of action for their consideration in minutes.
DOM augments human expertise; it does not replace it. The AI explores the possibilities, explains the recommended strategy, and the authorized human decision-maker retains control.
7. Looking ahead, where do you see the strongest opportunities for DOM AI across military helicopters, fixed-wing fleets, MRO organizations, and the wider Gulf defense market?
We see significant opportunities wherever aircraft availability and operational readiness are critical, particularly across military helicopter and fixed-wing fleets, MRO organizations, and the wider Gulf defense market.
Military helicopters are a particularly strong use case because they typically operate under demanding conditions, with complex maintenance requirements and relatively limited fleet sizes. In this environment, having even a few additional aircraft available can represent a major increase in operational capability. Our existing deployment in the region has already demonstrated the potential for substantial availability gains.
For larger fixed-wing fleets, the opportunity comes from DOM’s ability to optimize increasingly complex combinations of scheduled maintenance, aircraft utilization, manpower, infrastructure, spare parts and operational commitments over several years. The larger and more constrained the fleet, the greater the number of possible maintenance scenarios and therefore the greater the potential value of advanced optimization.
For MRO organizations, DOM can help optimize not only aircraft availability but also hangar utilization, maintenance slots, workforce allocation, tooling and parts, allowing existing infrastructure to support more aircraft and potentially increase throughput without a proportional increase in resources.
The Gulf is particularly well suited to this approach. Armed forces in the region operate some of the world’s most advanced aircraft, often across diverse helicopter and fixed-wing fleets, while maintaining very high readiness requirements. At the same time, operational priorities can change rapidly due to deployments, long-range missions, regional events or unexpected aircraft unavailability.
Our ambition is therefore to make DOM a fleet-wide optimization and decision-support layer across different aircraft types and existing maintenance systems, enabling operators to extract significantly greater operational availability from the assets and resources they already have.
Ultimately, the opportunity is not simply to make maintenance more efficient. It is to convert maintenance optimization into greater operational capability more mission-ready aircraft, available when they are needed, at a lower overall cost.



