I bring you a weekly bite-sized chunk of the science behind helicopter human factors and CRM in practice, simplifying the complex and distilling a helicopter related study into a summary of less than 500 words.
TITLE:
Enhancing aviation safety with artificial intelligence: A systematic literature review on recent advances, challenges and future perspectives.
WHAT?
A large-scale systematic literature review examining how artificial intelligence (AI) is being applied to improve aviation safety, alongside the key challenges limiting its safe and reliable deployment.
WHERE?
International research, with authors based in Hong Kong, the UK, Taiwan, and Germany. Studies span global aviation operations, accident databases, simulators, and field applications.
WHEN?
Published in 2026, covering AI research in aviation from 2012 to 2024.
WHY?
With air traffic increasing and operations becoming more complex, traditional safety systems are under growing pressure. AI offers tools for managing large datasets, predicting risk, monitoring human performance, and supporting investigations. However, concerns remain around trust, reliability, regulation, and human-AI interaction in safety-critical environments. The review aimed to establish the current state of research in this rapidly developing field.
HOW?
A structured search of Web of Science initially identified 1,794 papers. Through staged screening, 175 peer-reviewed studies were selected.
Content analysis mapped application domains and AI models, while thematic analysis identified major challenges and future research needs.
Studies were grouped into three main areas: human performance, accident analysis and prevention, and AI-driven safety systems.
FINDINGS:
AI use in aviation safety has grown rapidly since 2016, with a surge in LLM-based applications after 2022.
Major application areas include:
- Human performance
- Accident analysis and prediction
- Operational safety systems
LSTM models are being used to spot patterns in data or occurrences over time, for example analysing physiological data, or how workload builds up in flight.
CNN models analyse visual-style data, such as images, signals turned into graphs, or sensor outputs, for example in analysis of heart rate, EEG or flight data.
LLMs are newer AI systems understand and generate text, and are increasingly used to read and summarise accident reports, pull out key causes and safety lessons from databases, and help investigators or pilots with smart digital assistants (like virtual copilots or analysis tools.
Key challenges identified with AI include:
• Lack of explainability and trustworthiness
• Regulatory and certification barriers
• Human-AI teaming issues
• Data quality, representativeness, and privacy concerns
SO WHAT?
The review shows that AI is rapidly transforming aviation safety from monitoring pilot workload in real time to automating accident analysis and supporting decision-making.
Of particular importance for human factors is the growing use of accessible physiological and behavioural data, such as heart rate, EEG, eye tracking, speech, and facial cues, combined with AI to assess workload, fatigue, and situational awareness in realistic environments.
For operational aviation this suggests strong potential for in-the-field human factors monitoring using wearable or non-intrusive sensors paired with AI models, supporting proactive risk management rather than post-incident analysis alone (See H2F BiteSize #28 & H2F BiteSize #32 for examples of such studies).
However, widespread adoption will depend on developing explainable AI, clear certification frameworks, and well-designed human-AI teamwork. Future progress requires AI systems that are not only accurate, but also transparent, trustworthy, and operationally appropriate.
REFERENCE:
Yiu, C. Y., Li, W.-C., Ng, K. K. H., Chi, C.-F., & Schiefele, J. (2026). Enhancing aviation safety with artificial intelligence: A systematic literature review on recent advances, challenges and future perspectives. Advanced Engineering Informatics, 71, 104378. https://doi.org/10.1016/j.aei.2026.104378
competentaviators.com 