The Growing Automation in Air Traffic Control

The International Civil Aviation Organisation (ICAO), a specialized agency of the United Nations tasked with establishing standards and oversight, is spearheading the development of a global air traffic management system through the Global Air Navigation Plan.

The implementation of automated air traffic control systems is designed to enhance the operational safety and efficiency of air traffic controllers. However, these automated systems may also introduce unforeseen risks.

Globally, air navigation service providers deliver a variety of services, including air traffic control. Many of these providers are modernizing their communication, navigation, surveillance, and air traffic management systems in alignment with the global plan. This modernization aims to accommodate the increasing volume of flights while enhancing efficiency, decreasing costs, and minimizing the environmental footprint of aviation.

South Africa is participating in this global transformation. The South African Air Traffic and Navigation Service has integrated various automated systems to improve operational efficiencies, adhering to international standards.

Throughout my tenure as an air traffic services officer in Johannesburg, I offered flight information services to aircraft, coordinated international flights with neighboring nations, engaged in high-frequency radio communications with aircraft in South African oceanic airspace, supported search and rescue operations, and contributed to replacing several outdated air traffic control systems.

A significant portion of the work involved deploying automated technologies intended to enhance the performance of air traffic controllers and ultimately increase capacity and efficiency in airspace management.

However, a marked increase in aviation incidents within South Africa was recorded over the past four years. In 2021-2022, there was a 69% rise in general aviation accidents and a 77% rise in incidents compared to the prior year. Additionally, reports of incidents and confidential hazard disclosures from air traffic controllers saw a substantial rise in 2023-2024.

These rising figures prompted an important question: if the technology was designed to enhance safety, why were incidents on the rise?

My doctoral research tackled this question by investigating the impact of automation on air traffic controllers and examining the international, regional, and national regulatory frameworks surrounding these technologies.

The findings suggest that the issue is not solely with the technology, but rather with the intricate dynamics among air traffic controllers, automated systems, and regulations.

Key insights from a survey identified factors such as complacency among air traffic controllers, overreliance on automated systems, and a decline in manual skills, which were further exacerbated by increased workloads due to automation failures.

While participants expressed distrust in artificial intelligence (AI) handling their tasks, none were aware of how AI was currently utilized within their operations.

The legislative analysis revealed various deficiencies at international, national, and operational levels.

Technology Changes Jobs

Automation is frequently touted as a solution to reduce human error and workload. In air traffic management, automated systems can manage extensive data, track aircraft movements, and provide assistance to air traffic controllers.

However, the introduction of automation does not eliminate the need for human input; it merely alters the nature of that input.

Air traffic controllers are transitioning from actively separating aircraft to overseeing automated systems and intervening when necessary.

Though this may alleviate workload in certain instances, it can also introduce novel cognitive challenges.

The study of human factors investigates human strengths and limitations to understand the origins of incidents. It has highlighted several risks associated with highly automated environments.

Air traffic controllers might develop an excessive trust in automated systems, a phenomenon referred to as automation bias. An air traffic controller’s situational awareness—their mental representation of the activities on the radar and in the vicinity—can diminish, resulting in heightened complacency.

This occurs when controllers spend excessive time monitoring automated systems rather than actively managing air traffic. Skills that are infrequently practiced may deteriorate over time.

These challenges do not necessarily signify subpar performance; rather, they reflect expected human reactions to changes in work organization and execution.

New Technology Can Create New Risks

As part of my research, I surveyed air traffic controllers from the United States, Canada, Europe, the United Kingdom, New Zealand, the United Arab Emirates, and South Africa.

One identified risk was that training programs did not sufficiently prepare air traffic controllers for the practical realities of operating automated systems.

Procedures established for older systems often did not align with the capabilities and limitations of the modern automated systems. Civil aviation authorities may not possess the necessary skills to recognize the gradual, unnoticed shift in daily operations away from safety protocols, termed operational drift, following implementation.

Another risk is that failures in automated systems can significantly elevate air traffic controllers’ workloads. The timing of this increased workload is critical.

If it coincides with a period of overreliance on these systems, controllers may find themselves lacking the requisite experience to manage air traffic effectively.

In some instances, the automated system operates exactly as intended, yet controllers interact with it in ways that were not anticipated by developers.

Electronic flight strips serve as a prime example of how air traffic controllers often prefer to write information on paper rather than type it, as typing can disrupt their efficiency during peak periods.

The majority of surveyed air traffic controllers indicated that most automated systems were not interoperable and that they often employed temporary, nonstandard procedures or manual interventions to circumvent system limitations.

In other words, they resorted to “workarounds.” Consequently, incidents occurred, such as reduced aircraft separation that could lead to collisions.

These new risks prompt another critical inquiry: who would bear responsibility for an accident if an air traffic controller had to depend on these “workarounds” to manage air traffic?

Regulation Has Struggled to Keep Pace

The challenges extend beyond operational environments to international, regional, and national legislative bodies as well.

The International Civil Aviation Organisation acknowledges the swiftly expanding role of automation and AI. However, it has been unable to keep up with the pace of change. Establishing standards is a lengthy process.

Another limitation is the absence of a clear international standard mandating automated air traffic control systems and AI to be certified by national regulators.

As recently as 2023, the European Union Aviation Safety Agency proposed the creation of a new regulatory framework for certifying automated systems.

In South Africa, the South African Civil Aviation Authority adheres to international standards and thus does not require automated air traffic control systems to be certified.

The authority reported a shortage of skilled technical inspectors, necessitating the need for inspectors knowledgeable in automation and AI to certify systems in the future.

As automation advances, the responsibility for safety is increasingly becoming shared among system designers, regulators, service providers, and air traffic controllers.

A Lesson for the Age of AI

Technology does not function in isolation. Its success relies on the individuals who utilize it, the organizations that implement it, and the regulatory frameworks that govern it.

When incidents arise, the reasons are seldom as simplistic as attributing blame to either automation or the human operator. The interactions across the entire system culminate in safety outcomes.

As government entities and industries turn to automation and AI to resolve intricate challenges, the aviation sector illustrates that the introduction of AI transcends mere technical hurdles. It encompasses human, organizational, and regulatory dimensions.

Aviation regulators will need to revise existing legislation to define how they will certify and oversee automation, regulate AI, clarify air traffic controller liability, and reevaluate their role within this intricate system.

Cindy Hendrikse, PhD in aviation law, University of Cape Town

This article is republished from The Conversation under a Creative Commons license. Read the original article.