Daniel and Personalized Evidence-Based Training (PEBT)

ByCedric Paillard

Cedric Paillard, CEO of the Airline Pilot Club, is the chief architect of the Amelia AI algorithms which are driving data analysis for pilot selection, training and matching.

We can do a lot of things that we didn’t think we would be able to do a year ago, but it actually works fairly well.

Amelia has evolved into the Evidence-Based Training (EBT) and Personalized EBT Optimization Suite.

Daniel was very surprised when we gave him, pretty much 10 minutes after entering all the data, 5 things he could work on for improvement. Daniel is a candidate evaluated in the Avianca pilot recruitment project conducted by APC and Symbiotics.

Daniel’s identified challenges:

  • Situational Awareness (SAW) – enhancing his ability to maintain awareness of the operational environment, especially under stress or when faced with unexpected events.
  • Problem Solving and Decision Making (PSD) – further developing his ability to make timely and effective decisions, particularly when information may be incomplete or rapidly changing.
  • Communication (COM) – improving aviation-specific communication skills, including standard phraseology and clarity in ATC communications.
  • Workload Management (WLM) – developing strategies to manage workload effectively, especially in complex or high-pressure situations.
  • Leadership and Teamwork (LTM) – While already a strength, continuing to develop these skills will be beneficial, focusing on effective collaboration and leadership within the cockpit and with ground operations.

We even recommended to him a couple of training partners.

Now we can actually customize things. We can provide feedback to the candidate. And for training managers at airlines or ATOs. Amelia can detect very important information – the shortfalls, the pros, the cons of candidates, whatever you want to ask.

The AI can also show the consistency or inconsistency of an assessor. We are able to identify, when an instructor makes a comment, if that comment is different from the numerical value that has been assigned to the exercise, and report why it’s being analyzed like this.

But please note, none of those AI reports is generated without a human intervention behind it to validate the information.

The APC EBT/PEBT Optimization Suite, publicly launched in a presentation at the World Aviation Training Summit (WATS) in May 2024, is a comprehensive analytics and development platform designed to utilize Generative AI, combining aviation training experts and aviation psychology insights to analyze, evaluate and continuously improve EBT and PEBT programs. This suite enhances training effectiveness, operational safety, data-driven decision-making, cost efficiency, regulatory compliance, talent attraction and retention, adaptability and resilience.

Key features include: 

  • AI-powered analysis: Advanced algorithms analyze instructor and pilot performance, identifying strengths and areas for improvement.
  • Expert insight integration: Combines AI analysis with insights from seasoned aviation psychologists for a multi-dimensional perspective on training effectiveness.
  • Dynamic reporting tools: Customizable reports detail candidate assessments, training outcomes, and improvement suggestions. 
  • Continuous improvement mechanism: Feedback loops and continuous monitoring ensure training programs evolve with industry standards. 
  • Training customization recommendations: Provides actionable recommendations for addressing competency gaps and leveraging strengths. 

Safety and efficiency focus: Enhancements prioritize flight safety and operational efficiency, aligning with regulatory standards. 

How the EBT / PEBT Optimization Suite Functions

Enhanced Training Effectiveness

  • AI-Driven Analysis: Utilizes AI algorithms to analyze pilot and instructor performance, identifying strengths and areas for improvement with precision.
  • Expert Insights: Integrates insights from aviation psychologists to enhance training programs, focusing on human factors and cognitive aspects critical to pilot performance.
  • Group Insights: Analysis of classes or groups, showing group strengths and weaknesses, enables management to nuance training programs to meet bespoke group needs.

Operational Safety Improvements

  • Data-Driven Decision Making: Provides dynamic reporting tools and customizable dashboards for real-time monitoring of training outcomes and safety metrics.
  • Continuous Improvement: Incorporates feedback loops and continuous monitoring to ensure training programs evolve with changing aviation standards, enhancing safety practices.

Cost Efficiency

  • Resource Optimization: Optimizes the use of instructors, aircraft, simulators and other training resources through AI-driven analytics and real-time data syncing.
  • Predictive Maintenance: Uses predictive analytics to forecast maintenance needs and optimize resource allocation, reducing downtime and operational costs.
  • On-time Training Delivery: Continuous student improvement through personalized training prompts, preventing areas of weakness from hindering progress to successful training outcomes.
  • Reduced Additional Training: Early identification of weaknesses and continuous personalized prompts to improve these areas using low-capital training resources reduces instances of additional training in aircraft or simulator.

Regulatory Compliance

  •  Compliance Tracking: Standard reporting tools are available for assessment and compliance tracking, ensuring adherence to regulatory requirements and standards.
  •  Customization for Standards: Supports customization capabilities tailored to specific aircraft types and operational roles, aligning training programs with regulatory standards.

Talent Attraction and Retention

  • Training Mentor: Acts as an always-available mentor to a student, using deep insights into student characteristics and performance metrics to act as an impartial, supportive personal trainer throughout the course.
  • Personalized Training: Provides AI-driven predictive analytics for personalized training and performance forecasting, enhancing the learning experience and career development opportunities.
  • Support and Training: Offers dedicated support and consultancy services for users, including technical support and training, fostering a supportive environment for pilot development.

Adaptability and Resilience

  • Multi-Device Access: Supports multi-device access, including tablets and mobile devices, for on-the-go training and assessments, promoting flexibility and accessibility.
  • Customization and Integration: Allows full customization capabilities for integration with airline-specific IT infrastructure, adapting to organizational needs and enhancing operational resilience.

Safety Improvements

  • Data-Driven Decision Making: Amelia leverages AI to analyze vast amounts of training data in real-time, providing dynamic reporting tools and customizable dashboards for continuous monitoring of training outcomes and safety metrics. This helps in identifying potential safety issues early and addressing them proactively.
  • Enhanced Feedback Mechanisms: Incorporates continuous feedback loops where data from training sessions are constantly reviewed and used to adjust training programs. This ensures that any identified weaknesses or gaps in training are promptly addressed, leading to a continuous improvement cycle and mitigating potential safety risks.

EXCERPTED from The Robot in the Simulator: Artificial Intelligence in Aviation Training – NEW BOOK – AI in Aviation Training – AVIATION VOICES by Rick Adams, FRAeS