Aviation’s biggest challenge is fragmentation, not innovation

The aviation industry has become highly effective at optimising individual operations, but Raymond Lim (Head of Technical Assurance, Searidge Technologies– a NATS Company) argues that the next major gains will come from connecting them, from aircraft turnaround and runway operations to the wider air traffic network

Aviation does not have a shortage of technology. Airports, airlines, ground handlers and air navigation service providers have invested heavily in systems designed to improve efficiency, increase capacity and strengthen operational performance.

Yet avoidable delays, inefficient resource utilisation and operational disruption remain persistent challenges. Technologies introduced to solve specific problems often operate independently, with limited consideration of their wider consequences.

When local improvements create wider problems

Consider how an airport recovers from disruption. An airline may re-sequence departures to protect its schedule, while a ground handler prioritises aircraft closest to pushback. Meanwhile, the airport operator is maximising stand availability and air traffic controllers are managing runway demand.

Each decision makes sense individually. However, without coordination, an aircraft might leave its stand only to join a longer departure queue, while a stand reassignment creates additional towing or servicing requirements.

One organisation’s performance may improve while the airport becomes less efficient. Individual processes can meet their targets even as delays spread elsewhere.

In my experience, few airports have a genuinely complete, end-to-end operational picture. Each stakeholder holds valuable information but rarely has sufficient visibility to understand how its decisions affect others.

The priority should be identifying where constraints exist, anticipating how they might shift and determining whether an intervention resolves a problem or simply moves it elsewhere.

Connecting decisions, not just systems

Technology integration is part of the answer, but some of the hardest barriers are organisational.

Aviation systems have developed under different ownership structures, regulatory obligations and commercial arrangements. Concerns surrounding security, data sovereignty and accountability complicate information sharing. Even when data is available, organisations may have competing priorities.

Addressing these challenges requires a trusted information foundation, a shared operational model and governance arrangements establishing how stakeholders coordinate decisions and resolve trade-offs.

This does not mean replacing established platforms or consolidating operational information into one enormous system. Existing technologies can continue performing their intended functions while open interfaces and common information structures enable new capabilities to work across them. The opportunity is to build a reusable foundation of data, models and decision tools, so each new application builds on shared capabilities rather than creating another silo.

However, technology cannot determine whose priorities take precedence during disruption or who remains accountable for a shared recommendation. Those questions require leadership and agreed operating relationships.

AI’s greatest value is collective intelligence

Much of the discussion surrounding artificial intelligence in aviation focuses on automating individual activities. While these applications have value, AI’s greater contribution may lie in helping operational teams understand complex interactions and evaluate alternative decisions.

During runway disruption, for example, AI could compare recovery strategies by considering turnaround progress, stand availability, departure sequencing, airspace constraints and airline schedules simultaneously.

Rather than identifying the fastest solution for one stakeholder, it could show which interventions best meet agreed operational objectives, making the trade-offs and consequences visible to everyone involved.

Combining operational data, AI and digital twins makes this increasingly achievable. Data establishes what is happening, while a digital twin dynamically represents the operation and allows teams to examine how conditions might evolve. AI can then explore scenarios, identify patterns and compare potential consequences.

Together, AI and a high-fidelity digital twin could help teams explore a much wider range of possible futures and test interventions before implementation—not simply accelerating analysis, but exposing options and knock-on effects that might otherwise remain unseen.However, AI should support rather than replace operational judgement. Recommendations must explain their objectives, assumptions, limitations and uncertainty, allowing people to understand the reasoning behind them.

Building confidence through experience

Workforce readiness is as important as technological capability. Operational teams have legitimate reasons to question recommendations that conflict with their experience, and organisations should encourage staff to challenge and validate AI-assisted decisions.

One approach is to replay historical events to demonstrate whether a model accurately represents what happened. Teams can then test scenarios before running the technology alongside live operations without influencing decisions.

Advisory applications can follow, with experts retaining authority. Greater automation should only be considered where evidence, assurance and regulation justify it.

Sometimes an operator will possess information the model lacks. On other occasions, AI may reveal consequences invisible from that operator’s position. Combining these perspectives can improve decisions and build confidence.

Measure what matters across the network

Aviation leaders should begin with a shared operational challenge rather than another technology procurement. This could involve adverse weather recovery, turnaround disruption or coordination between arrivals, stands and departures.

Establish baseline performance, identify the decisions involved and agree how improvements will be measured. Individual operational metrics remain important, but should be complemented by measures of overall delay, recovery time, predictability, capacity utilisation and resilience.

The first implementation need not connect an entire airport. It should demonstrate measurable benefits across a meaningful part of the operation, establishing an approach that can extend across airfield, airport and en-route activities.

No single organisation controls the entire aviation operation. Progress therefore depends on partnerships that align responsibilities and improve performance beyond individual organisational boundaries.

Aviation’s next breakthrough will not come from adding another isolated platform. It will come from connecting existing capabilities and helping the entire system perform better together.