The aviation industry is already implementing AI solutions, but it is barely scratching the surface of the value AI can bring, according to Ali Pourshahid, Chief Engineering Officer and Alam Khan, Principal Architect, Solace. The industry is weighed down by diverse and siloed processes, preventing airlines from harnessing the full power of AI, in particular exploiting the undoubted potential of agentic AI. To release the power of agentic AI requires an event-driven integration strategy to connect masses of diverse and disparate data in real-time. Only then will airlines be able to understand the bigger picture and have the tools to transform the way we fly, from the point of customer booking, right down to in-flight emergencies.
Organizations across the aviation ecosystem, from manufacturers to air traffic control and airlines to airports, are already deploying AI. The Airbus Skywise platform leverages AI to analyze vast amounts of data for predictive maintenance and improved operational efficiency. Digital airport traffic management systems (DATMS) use AI to automate air traffic control, minimizing human error and reducing tarmac incidents.
(K illustrator Photo/Shutterstock)
Airline passengers too are seeing the benefits. Airlines are utilizing AI for customer service, as seen with Air India’s “Maharaja” virtual agent handling numerous customer queries, and Etihad Airways’ upcoming AI-powered flight booking chat application. Stansted Airport in the UK employs AI-powered kiosks to address passenger queries and identify commercial opportunities for onward travel and personalized services.
And use cases are growing. In its 2025 Travel Industry Outlook, Deloitte sees AI increasingly being applied across the travel sector to improve passenger experiences, boost efficiency, and drive revenue. But these many point solutions just touch the surface of the value that AI can bring to the aviation industry.
Sprawling Aviation Tech Stacks Impede Getting the True Value of AI
The aviation industry is, by its very nature, disparate and far-reaching. McKinsey outlines the problem: “The global aviation ecosystem relies on interwoven networks shaped by competing stakeholder priorities…Many of the decision-making processes through which airlines establish route maps, schedules, fleet management, airport staffing levels, and so forth remain impeded by siloed communications and outdated technology and metrics.”
While we can see AI already bringing powerful capabilities to aviation, in order to exploit its full value the industry faces fundamental challenges of managing this heterogeneous ecosystem, complete with its wide spectrum of IT systems. Traditional point-to-point integrations, and hub-and-spoke architectures that rely on all nodes being connected to a central server for data exchange and communication, all struggle to handle the real-time, distributed nature of modern aviation operations.
The Whole Is Greater Than The Parts
These AI systems must be integrated into the complex web of existing aviation infrastructure that spans ground operations, aircraft systems, reservation systems, departure control systems, passenger services, and maintenance operations.
(Nicolas Economou
/Shutterstock)
Here’s where AI and, in particular, agentic AI has the power to take on these complex objectives, make decisions and execute tasks with limited human intervention. But for agentic AI to make sense of the mass of AI-enabled events and data exchanges taking place across this ecosystem, the diverse material to work on needs to be integrated in real-time.
Enter the agent mesh, a solution that presents aviation organizations with a real-time, event-driven approach to IT integration.
Behind Every Agent Mesh Is An Event Mesh
The foundation of any agent mesh is an event mesh, a data distribution layer that enables the seamless flow of information across environments, organizations and locations. Agent mesh extends the idea of event mesh by introducing a network of autonomous AI agents that can reason about and act upon the information flowing through the mesh.
Think of it as adding a layer of distributed intelligence to the aviation industry’s digital nervous system. Working together, only the necessary data and events are liberated and orchestrated precisely where they’re needed, enabling autonomous AI agents to make decisions, and take actions either independently or with humans in the loop as needed.
No longer will point solutions have a view limited by siloed data. They will be presented with the bigger picture in real-time through the mesh to deliver much more valuable information for critical decision-making.
The Bigger Picture in Action
These six scenarios are the perfect illustration of how an agent mesh can deliver AI-enabled benefits throughout the travel journey:
- Booking: AI-driven pricing intelligence at scale
In the commercial aviation space, an agent mesh can help companies optimize revenue through real-time market analysis and dynamic pricing using large market models (LMM). With an agent mesh in place, airline systems can process vast amounts of data – competitor pricing, historical booking patterns, real-time demand indicators, and external events – to continuously optimize ticket pricing.
The event-driven nature of the information flow ensures that pricing decisions are instantly pushed across all sales channels to maintain consistency across inventory systems and revenue management rules. When market conditions change, such as a competitor’s pricing adjustment or a sudden surge in demand, the system can instantly respond while considering the broader implications for network-wide revenue optimization.
- At the airport: Dynamically orchestrate the perfect passenger experience
Picture a seamless passenger journey where an agent mesh enables real-time orchestration across all customer touchpoints. When a premium passenger enters the airport, the system immediately recognizes their presence through various sensors and begins orchestrating their experience. The mesh coordinates data from reservation systems, departure control systems (DCS), and check-in systems to create a personalized journey.
For instance, if a flight delay is detected, the system doesn’t just notify the passenger, it proactively coordinates alternatives. The system might automatically adjust the passenger’s lounge access duration, rebook connecting flights, and update ground transportation arrangements, all while keeping the passenger informed through their preferred communication channel using natural language thanks to an LLM.
- Boarding: The voice of safety to automate compliance and streamline pre-flight approvals
Airlines operate in a highly regulated environment, where critical pre-flight checks such as Weight and Balance, Maintenance Release Documentation, and Safety and Security Verifications must be completed to ensure operational safety. Currently, these approvals are often done manually, involving paperwork or digital sign-offs that require human validation, which can delay processes and introduce the potential for errors. Given the tight schedules and high stakes at play, there’s a growing need for a streamlined, secure method to complete these checks efficiently.
An agent mesh can revolutionize these pre-flight processes through Voice-based e-Cert sign-offs. This enables ground and maintenance crews to complete approvals using voice recognition, with AI securely authenticating each individual’s voice. Once verified, these sign-offs are instantly logged into interconnected systems, removing manual bottlenecks and reducing paperwork. Additionally, the agent mesh enables immediate situational awareness across the ecosystem by linking data from maintenance records, operational systems, and load control. AI agents continuously monitor this data, alerting teams to any incomplete tasks or safety concerns, thereby accelerating decision-making, minimizing delays, and enhancing overall efficiency in the airline’s operations.
- Intelligent skies: Optimized, fuel-efficient route checking
Airlines operate on thin profit margins, and with fuel consumption being a major cost driver – often accounting for around 30-35% of operating expenses – optimizing fuel efficiency becomes crucial for profitability. Airlines constantly seek innovative solutions to reduce fuel usage while maintaining operational efficiency.
Weather data from airlines is playing an increasingly important role in forecast models
An agent mesh enables the integration of real-time data from various sources, such as weather conditions, air traffic control updates, and aircraft sensor data. AI agents can analyze this data to adjust flight paths dynamically, reducing fuel burn by avoiding turbulent air or taking more direct routes based on latest conditions.
Furthermore, an agent mesh can synchronize data from ground operations, load control systems and ATC, allowing better load management and takeoff timings, ensuring that aircraft are not idling on the tarmac unnecessarily.
- On arrival: Make baggage handling smarter
Baggage handling requires coordination across multiple systems and stakeholders. An agent mesh can streamline this process into an intelligent, self-optimizing system. The architecture enables real-time analysis of various factors including flight schedules, baggage volumes, gate proximity, and belt occupancy rates.
When a flight arrives early, the agent mesh immediately triggers a cascade of coordinated actions. It automatically redirects baggage handling resources, adjusts belt assignments, and updates staff allocations. The system continuously monitors load distribution to prevent bottlenecks, using predictive analytics to anticipate and prevent potential issues before they occur. Each bag’s status is instantly shared across all relevant systems and stakeholders, creating a transparent and efficient operation.
- Not on the itinerary? Adapt resources to meet any unexpected scenario
Agent mesh enables airports to achieve true situational awareness by integrating real-time data from multiple sources. Consider a scenario where an incoming flight reports a medical emergency. The agent mesh immediately coordinates multiple systems: it alerts medical services, adjusts gate assignments to minimize transit time for emergency services, updates ground handling schedules, and modifies connected flight gate assignments if necessary.
The system continuously processes events from various sources – flight schedules, passenger flow data, security checkpoints, and retail operations – to optimize resource allocation dynamically. When passenger flow increases unexpectedly at security checkpoints, the system can automatically request additional staff, open new lanes, and adjust downstream resources to accommodate the changing situation.
A New Era in Global Aviation
Traditional integration approaches aren’t equipped to handle the fast-paced, distributed needs of modern airline operations. With the advent of AI, it has become even more critical for systems to work seamlessly within this intricate framework, spanning everything from ground operations to aircraft systems, reservations, departures, passenger services, and maintenance.
An event-driven strategy is essential to link together the vast array of diverse data sources. Only with a real-time integration approach will airlines be able to gain a holistic view and the capabilities needed to revolutionize air travel.
About the authors:
Ali Pourshahid is the Chief Engineering Officer at Solace, where he leads the engineering teams and is responsible for the delivery and operation of Software and Cloud services.
Alam Khan is the Principal Architect at Solace. His role is focused on designing scalable, cloud-native and event-driven platforms that enable enterprise transformation and deliver measurable business outcomes.
If you want to read more stories like this and stay ahead of the curve in data and AI, subscribe to BigDataWire and follow us on LinkedIn. We deliver the insights, reporting, and breakthroughs that define the next era of technology.
The post Personalized Flights to Intelligent Skies: How Agentic AI will Reshape the Future of Air Travel appeared first on BigDATAwire.
Go to Source
Author: Alam Khan & Ali Pourshahid
