How can AI achieve connected, intelligent aviation operations?
Airports are complex operational environments with multiple stakeholders working on shared processes. Large airports maintain and manage more than 500,000 different assets, thousands of flights and hundreds of thousands of passengers every day. Working at this scale requires the airport, airlines, ground handlers, refueling providers, catering firms, government stakeholders and many more organizations to co-operate.
This is also where AI’s real value in aviation will be created. Not through isolated tools or standalone pilots, but by connecting trusted operational data, embedded business logic and the many partners involved in the airport journey. As travel becomes increasingly AI-enabled, airports need a secure and neutral execution layer that can help intelligence move across the ecosystem — from insight to action.
What are the barriers preventing optimal performance?
Today, industry cooperation is constrained by fragmented systems, limited data sharing and manual processes that prevent airports from achieving optimal performance. Barriers to intelligent decision-making include:
Key takeaways
- By connecting an airport’s own applications with its stakeholder ecosystem, we can deliver a data foundation that’s ready for AI.
Agentic AI provides a pro-active intelligence layer that unifies situational awareness, provides faster access to insights and context-aware recommendations.
Embedding intelligence optimizes capacity and asset utilization, reduces cost and improves disruption management.
Airports are becoming intelligent ecosystems where stakeholders work more closely based on common data and embedded AI to achieve more.
Orchestration unlocks value for the ecosystem.
- Fragmented operational systems: a large airport may run hundreds of different operational applications. These systems are rarely connected.
- Limited data sharing: airport stakeholders haven’t had the tools, governance frameworks or mindset to exchange data that can provide a common operational view.
- Legacy protocols and manual processes: it’s still common for data to be exchanged using type B and type Y Teletype messages – an expensive, outdated protocol from the 1960s.
Agentic AI in airport operations: the intelligence layer
Agentic AI changes the game for airports and their stakeholders. Quite simply: this technology is mind-blowingly powerful, and it’s ready to go in aviation.
With the right, trusted data foundation in place, grounded in integrated workflows and clear operational guardrails, agentic AI delivers two primary benefits:
- It allows any airport stakeholder to quickly and securely access the information they need to make decisions faster.
- It provides pro-active, intelligent, context-rich recommendations based on the complete picture – supporting not only faster, but also higher-quality decisions.
In an airport environment where teams are accustomed to inaccurate data that’s spread across multiple systems, often requiring validation via manual calls using a walkie talkie – the impact of AI is significant.
Embedding a proactive intelligence layer across the airport
Imagine a terminal manager being able to ask, “why is security slow this morning?” and “which flights are at risk due to queues?” via a chat session on their mobile.
The AI agent retrieves the relevant data from multiple systems, including the Airport Operational Database (AODB), analyzes the situation and explains the six flights that are at risk of delay.
This marks a huge improvement on today’s process that requires an analyst to run a query to generate a data visualization, which the airport manager must then analyze. We’re talking about orders of magnitude faster time to insight with the information provided in a way that everyone can easily understand.
But what if you didn’t need to ask? What if the AI monitored your operational environment to identify current and future bottlenecks and pro-actively alerted you to them? What if that alert didn’t only describe the issue but also recommended the optimal fix.
This is where the technology stands today. At Amadeus we’re working with customers to embed AI agents within operational workflows.
Establishing your own data foundation
The value an airport experiences from AI is directly proportional to the scale, reliability and quality of the data it can access. That’s why successful AI projects don’t focus on analytics, they focus on data readiness.
For airports, the first step is ensuring data generated by passenger, operational and flight related systems is centralized, normalized and ready to power a team of AI agents. This integration work involves establishing data pipelines, data quality, transformation and governance.
Exchanging data to create an intelligent airport ecosystem
Aviation involves complex processes with multiple stakeholders and by exchanging data the industry can unlock productivity, cost reduction and traveler experience benefits.
Consider an aircraft turnaround. For example, if an inbound flight is delayed by 20 minutes, shared data can automatically alert the ground handler, baggage team, gate staff, cleaning crew and connecting-flight teams. Each party can adjust resources and sequencing before the aircraft arrives, rather than discovering the change through calls, radio messages or separate system updates.
By sharing real-time operational data, they can respond collectively to disruption, optimize resources and keep passengers informed. The result is faster turnarounds, lower operating costs and a more reliable journey.
At Amadeus, we're currently building the first data exchange infrastructure for aviation, designed to help stakeholders collaborate in a neutral, secure and governed way. The platform allows stakeholders to discover datasets that might be helpful, to provision connections and easily agree legal and governance conditions. By drawing on cutting edge techniques, stakeholders can share data insights without revealing the underlying data – unlocking new use cases while preserving privacy and control.
Shared aviation processes should not rely on disconnected data. This infrastructure links airline, airports and their partners' information, creating a common operational foundation for better efficiency.
Orchestrating an intelligent airport ecosystem
It’s increasingly clear that AI agents will become available everywhere across an airport’s operations. Looking ahead, we see the APOC evolving into an environment where coordinated teams of AI agents continuously support operations. Each agent would oversee a specific operational domain such as disruption management, baggage journeys or passenger-flow optimization—monitoring live conditions, detecting emerging issues and alerting the right teams early.
As the availability of AI agents increases, the ability to orchestrate and govern them, grows in importance. Consider flight disruption. Individual agents can advise on availability of aircraft, crew, gates and passengers that have connections which must be managed. But how can we organize the agents to find the optimal solution that considers an airline and airport’s broader objectives and regulatory responsibilities?
The answer is orchestration and tight guardrails based on the airport’s underlying rules and business logic. As a platform provider to our airport and airline customers, this is the role Amadeus aspires to play. We connect AI agents to the data and business logic contained in the industry’s underlying systems so the intelligence layer has the context to solve even the most complex challenges.
The answer we’ve been seeking
For years the aviation industry has accepted structural weakness in its technology architecture. Operational systems are siloed. Airlines and airports don’t exchange data. Decision making is slow while data is collated and analyzed. Skilled humans have compensated for these gaps with manual workarounds, making the best of a fragmented environment.
When deployed on top of a strong data foundation, agentic AI can overcome these challenges. And it’s easy to access with AI agents available on everyone’s smartphone to answer queries, provide proactive alerts and make smart recommendations based on the complete picture.
Embedding an intelligence layer helps airport stakeholders achieve common situational awareness, gain insights faster and make better decisions that improve operational performance.
Frequently asked questions
A data exchange platform helps airports, airlines and partners share operational data securely and consistently. It is a neutral governed platform through which airlines, airports, ground handlers and their ecosystem partners can publish, contract for and exchange operational data.
AI in aviation depends on reliable, timely and connected data. High-quality data helps AI tools generate more accurate insights, recommendations and alerts.
Aviation organizations can prepare for AI by connecting key data sources, improving data quality and setting clear governance rules. The aim is to make trusted data available when and where it is needed.
Major barriers include fragmented systems, legacy data formats, manual processes and and trust deficit due to unclear governance. These challenges make it harder to build a shared, real-time view of airport operations.
Data governance sets the rules for how aviation data is accessed, shared and protected. It helps stakeholders collaborate while respecting security, privacy and legal and regulatory requirements.
Real-time data sharing helps aviation teams respond quickly to delays, disruptions and operational changes. It can improve turnaround times, resource planning and passenger communication.
Agentic AI is an AI “doer” that combines perception, reasoning and action. It’s given a mission (instructions) and it uses context, knowledge, AI models, and tools (including possibly cooperating with other AI agents) to accomplish tasks. Agentic AI can monitor airport operations, interpret changing conditions and recommend actions within defined guardrails. It supports faster insight and reduces decisions and execution latency for airport teams.
TO TOP