Pilot Training: How AI Can Shape its Future

artificial intelligence pilot training

Follow RAVEN on Social Media! ✈️✈️✈️
Get aviation career insights and a stream of aviation learning content

In the last article I wrote for RAVEN, I briefly outlined how ChatGPT and other Artificial Intelligence (AI) tools could potentially be used to revolutionize the future of pilot training.

In this article, I want to explore the area a little bit deeper by explaining the evolution of pilot training to readers with limited knowledge of the aviation industry – the traditional (legacy) approach to pilot training, the shift into evidence based training (EBT) and competency-based training and assessment (CBTA). As an industry, we need to start exploring how Artificial Intelligence (AI) can be integrated within these constantly evolving flight crew training frameworks.

If you are new to AI, check out our FREE course on “AI for Aviation Applications

I also want to place the application of AI technologies into the context of a wider pilot training environment which is also changing because of new technologies like manned electric Vertical Takeoff and Landing (eVTOL) aircraft.

How does it all fit in? And above all, is it all necessary? Will it improve safety levels? Well, I hope this article will allow you to start putting what feels like a 1500 piece puzzle together.

Pilot Training and Standardisation (of terms)

Before we go into the evolution and future of pilot training, it might be best if we take a philosophical view into a term which is commonly used in pilot training – standardisation – and give it a wider context. In pilot training, when we refer to standardisation, we generally refer to the exercise of guiding your instructors to a common approach to training and assessment which will allow everyone to be trained the same way, each time, every time. However, we could also potentially apply it to the maintenance of a standard, including in the definition of terms. At RAVEN, we have advocated this since the very beginning.

During my formative years in Air Malta, when I was starting to explore air operations regulation, I was fascinated (or rather, perplexed) with the use of multiple terms in EASA regulation to refer to the same approach to learning. Computer-based training, distance learning, computer-aided learning, online course – the list continues. In an industry where standards are the name of the game, I think the time is ripe for a shift in thinking. Why, you may ask? Because unless we move forward from these classical terms, it will become even more complicated to apply new technologies to the next era of our industry.

I will therefore take the liberty of standardising terms and use the term digital learning to move forward. Because the term computer-based is not relevant in an era where educational technologies are moving towards mobile solutions, whether you are at a distance of 1000 miles or in your operator’s training centre does not matter, and a lot of times an online course can be conducted offline.

online aviation course
Digital Learning on RAVEN: which could also be AI powered

Another standard term we must allow ourselves to define is the use of the words ‘aviation training‘. Beyond the traditional framework of aviation, the industry is rapidly evolving to incorporate mobility into the mix. With the rise of vertical takeoff and landing (VTOL) aircraft and the growth of urban air mobility (UAM), we are seeing the emergence of new markets and opportunities for aviation. So whilst reading this article, I want you to mentally move away from any perceived notions we have been accustomed to in pilot training. When we talk about training, it needs to start including any manned aircraft activity within the mobility spectrum.

The Evolution and Future of Pilot Training

The aviation industry has undergone significant changes in recent years, with the introduction of evidence-based training (EBT) and competency-based training and assessment (CBTA), representing a significant departure from the traditional training framework. The traditional approach to pilot training relies on standardised procedures, checklists, and memorisation, with pilots trained to follow a set of prescribed steps (rote learning). This is typically done using ground based instruction (classroom or non-interactive digital learning) and full flight simulators. However, this approach has been criticised for failing to adequately prepare pilots for the complex and dynamic challenges they face in the real world.

Despite this criticism, the legacy training framework remains in place, with many operators continuing to use traditional training methods. The International Civil Aviation Organization (ICAO) allows operators to choose between the traditional approach and the EBT/CBTA model, depending on their individual needs and circumstances.

EBT is designed to provide pilots with a more realistic training environment that replicates the challenges they will face in the real world. Rather than relying on memorisation and standardised procedures, pilots are taught to develop their problem-solving skills by confronting realistic scenarios in a simulated environment. EBT is grounded in research and data analysis, with training scenarios developed based on an analysis of real-world data and feedback from pilots. According to a study conducted by the European Union Aviation Safety Agency (EASA) (2015), EBT can result in up to a 50% reduction in the number of training hours required for a pilot to achieve proficiency.

CBTA takes this approach one step further, focusing on the development of specific competencies required to operate an aircraft safely and efficiently. Rather than relying on hours of training, CBTA provides a structured approach to pilot training, with pilots required to demonstrate competency in specific areas before moving on to the next stage of training. Competencies are identified based on an analysis of the job requirements and are tailored to the specific needs of the operator.

Under CBTA, the aim of the training is to develop the nine pilot competencies and the four Instructor/Evaluator competencies. The training course is also generally composed of theoretical knowledge, ground, and flight training elements. Nevertheless, the training objectives are considered satisfactorily completed when there is sufficient evidence to ensure that the trainee has achieved competency, without any reference to prescribed training time, and that he meets the interim and/or final competency standards. 

White Paper: Competency-Based Training and Assessment (CBTA) Expansion within the Aviation System (IATA)

A survey conducted by the International Air Transport Association (IATA) found that 88% of airlines using CBTA reported improvements in safety, while 78% reported improvements in operational efficiency. So not only an improvement in pilot performance, but also an improvement to the operators’ bottom line.

Under CBTA, Threat and Error Management (TEM) is naturally and fully embedded in the training curriculum. The pilot and Instructor/Evaluator (IE) competencies provide individual and team countermeasures to threats and errors to avoid a reduction of safety margins during training and operations. (Source: IATA)

AI as a Catalyst to Competency-Based Training and Assessment (CBTA)

With the rise of digitalisation and AI, the implementation of CBTA concepts can be catalysed, to provide more effective and efficient aviation training, and encourage those operators still delivering legacy training to shift towards CBTA. Here are a few examples of how.

Personalised pilot training plans based on their performance data

AI algorithms can analyse pilots’ performance data, including their strengths and weaknesses, and develop personalised training plans to improve their skills in areas that need improvement. For example, if a pilot struggles with communication skills, AI can automatically develop targeted training programs that focus on improving these skills – both in digital and simulator training. This personalised approach to training can lead to more efficient and effective training, saving time and improving safety.

Machine learning algorithms to analyze and track pilots’ progress in their training

Machine learning algorithms can analyse pilot communities’ training progress over time, identifying areas where they have improved and areas where they need further development. This can help trainers develop more targeted training programs and provide pilots with feedback on their progress. By analysing training data over time, machine learning algorithms can also identify trends in pilot performance and develop targeted interventions to address any issues.

Scenario-based training that simulates real-life scenarios

AI can be used to develop scenario-based digital learning that simulate real-life scenarios, allowing pilots to practice their skills in a safe environment. These simulations can be tailored to specific scenarios, such as weather-related emergencies or mechanical failures, allowing pilots to gain experience in a variety of situations. This type of training can be especially beneficial for pilots who are new to the industry or who are transitioning to new aircraft types.

Real-time feedback and training for pilots during flights

AI-powered systems can provide pilots with real-time feedback during flights, helping them make better decisions and improving their performance. For instance, if a pilot is making a mistake, AI can alert them to the error and provide guidance on how to correct it. This can help pilots improve their skills in real-time, ultimately leading to safer flights.

AI-powered flight data analysis to identify areas for improvement

AI algorithms can analyze flight data to identify areas where pilots need to improve their skills during line operations, helping them focus their training on areas that need the most attention. For example, AI can analyze data on a pilot’s landing performance and identify areas where they need to improve their approach and touchdown techniques. By feeding back these instances into the training system, pilots can receive targeted training that addresses specific issues.

AI-powered systems that can identify safety risks and provide mitigation strategies

AI-powered systems can be used to identify safety risks during flights and provide pilots with mitigation strategies to avoid accidents. For example, AI can analyse data on weather patterns, aircraft performance, and flight paths to identify potential safety risks, such as turbulence or obstacles in the flight path. This information can then be used to provide pilots with guidance on how to mitigate these risks.

AI can provide pilots with the skills and tools needed to make informed decisions in the cockpit

By analysing vast amounts of data and providing pilots with real-time updates on weather, traffic, and other factors, AI-powered systems can help pilots develop better decision making about flight routes, fuel consumption, and other operational factors. Additionally, by incorporating AI into ground training programs, pilots can learn how to use these tools effectively and efficiently, improving their decision-making skills and overall situational awareness.

Pilot Training, Big Data and the Internet of Things

The introduction of AI in the future of pilot training will only be made possible through the extensive use of Big Data and the Internet of Things, two technologies that are already available and rapidly advancing.

Big Data refers to the large volumes of structured and unstructured data generated by various sources, including sensors, instruments, and social media. The Internet of Things, on the other hand, refers to the network of physical objects, vehicles, buildings, and other items embedded with sensors, software, and connectivity that allow them to collect and exchange data. The Airbus A350 has 50,000 onboard sensors and generates 2.5 terabytes of data per day of operation – a clear example of Big Data, and the Internet of Things (sensors). We have only scratched the surface of how we can use this data to power pilot training. As we look to the future, it is clear that there is still much work to be done. In order to fully leverage the power of these technologies, aviation will need to embrace a culture of innovation and constantly push the boundaries of what is possible. This will require a willingness to experiment with new approaches, to take calculated risks, and to be open to new ideas and ways of thinking.

In order to fully leverage the power of these technologies, aviation will need to embrace a culture of innovation and constantly push the boundaries of what is possible. This will require a willingness to experiment with new approaches, to take calculated risks, and to be open to new ideas and ways of thinking.

Ultimately, the future of aviation will be shaped by those who are able to embrace this spirit of innovation and push the industry forward. By leveraging the latest technologies and taking a forward-thinking approach to pilot training and operations, aviation can continue to evolve and meet the demands of a rapidly-changing world. With the right mindset and a commitment to innovation, the sky truly is the limit.

Airbus A350 landing
The Airbus A350 is a long-range, wide-body twin-engine jet airliner developed and produced by Airbus

Related Articles


Your email address will not be published. Required fields are marked *