NGFT

The Future of AI

When an incident occurs, the pressure is on. This is where NGFT's own aviation- trained LLMs come in.

From a simple report to submission-ready occurrence documentation

Timely and accurate occurrence reporting is critical to aviation safety management. When an incident occurs, operators are often under considerable pressure to analyse the event, document key findings and submit a formal report to the authorities - all within 72 hours. This process can be frustrating and time consuming. This is especially true given the need to comply with regulatory requirements, perform risk assessments, and ensure completeness and accuracy of documentation.

How Large Language Models (LLMs) can help

Recent advances in artificial intelligence, in particular Large Language Models (LLMs), offer an efficient and effective approach to supporting occurrence reporting. By integrating LLMs into the reporting process, safety managers can significantly reduce the time and effort required, while improving consistency and accuracy.

Here's how LLMs can improve the process:

Structured report generation

Rather than starting from scratch, an LLM can take an initial draft report, extract key information, and reformat it into a structured template that meets regulatory requirements. This ensures completeness and helps standardize reporting across cases.

Automated analysis and pre-filled sections

By feeding the LLM with details of the incident, the safety manager can ask it to identify missing information, highlight potential risk factors, and suggest appropriate mitigation measures. The LLM can also pre-fill certain sections, making it easier for safety managers to finalize the document.

Risk assessment and mitigation measures

Finally, LLMs can assist in conducting a structured risk analysis. Based on the data provided, the model can evaluate contributing factors and potential hazards, and recommend appropriate mitigation measures. This helps ensure a proactive approach to safety improvement.

Benefits of using LLM in the evaluation process

Structured report generation

One of the challenges of occurrence reporting is maintaining consistency over time, especially when multiple people are involved in the documentation. By using predefined prompts and workflows, an LLM can generate reports in a standardized manner, reducing variability and ensuring unbiased assessments based on facts rather than subjective interpretations.

Increased efficiency and reduced administrative burden

With AI-powered assistance, safety managers can focus more on decision-making and corrective action, rather than spending hours generating reports. The ability to quickly verify and copy AI-generated content into the final submission format can reduce processing time and ensure compliance with regulatory deadlines without compromising quality.

Increased efficiency and reduced administrative burden

As AI continues to evolve, its role in safety management systems (SMS) will expand beyond occurrence reporting. However, LLMs should not replace human judgment, but rather augment it - providing structured insights, improving efficiency and reducing the manual workload associated with safety reporting. Want to explore how this approach could be integrated into your current safety management processes? Let's discuss how AI-driven reporting can work for your operations and see how our SMS has integrated this approach.

What do YOU think?

Who wants to step up to the plate?

Scroll to Top