Train-the-Trainer by DataHomeLab
Concept

Evidence-based training

EBT is a recurrent, competency-based programme for commercial aeroplane pilots that uses operational and training data to decide what to train and how often, then develops and assesses a finite set of pilot competencies rather than ticking a fixed catalogue of historical events. Scenarios and manoeuvres remain the vehicle; competencies are the target. The programme exists because early-generation accident lists and cumulative "add another event" regulation saturated recurrent training while modern fleets fail for reasons the old catalogue never covered, often with no technical malfunction at all.

Why it exists

Industry analysis over decades found that roughly 70 to 80 percent of accidents and serious incidents have a crew-performance contributing factor: decision-making, communication, leadership, flight-deck management. Parallel progress in flight data analysis (FDA), line observation, and safety reporting made the operational threat picture visible in a way it was not when national recurrent requirements were first written. Those requirements still largely reflect early-jet hull losses and the assumption that repeating a worst-case event in the simulator is enough mitigation. As aircraft reliability rose, accident classes such as controlled flight into terrain often involved aeroplanes operating without malfunction; situation awareness and related competencies, not another procedure item, were the missing countermeasures.

EBT's answer is resilience through competency: the next accident may be unforeseen, so mastering transferable competencies (rather than a prescribed event list) is the only scalable defence. The data both established the need for reform and supplied the content of the baseline matrices (threats, errors, manoeuvres, and frequencies by aeroplane generation).

Structure of a recurrent module

flowchart LR EVAL[Evaluation
observe baseline] --> MT[Manoeuvres training
coach to standard] MT --> SBT[Scenario-based training
develop competencies] SBT --> DEB[Facilitated debrief
grade and transfer] DEB --> SMS[Training metrics / SMS
refine programme] SMS -.-> EVAL

A typical EBT module is delivered in a qualified FSTD and has three phases:

Phase Purpose Instructor stance
Evaluation (EVAL) Line-oriented scenario(s) that capture baseline competence, feed training-system data, and identify individual needs Observe; minimise interruption; play external parties as needed
Manoeuvres training (MT) Practise critical handling and procedures that demand proficiency from a competent crew; skill, not LOFT Active coaching; reposition the device; relaxed, non-test climate
Scenario-based training (SBT) Largest phase: develop competencies while managing generation-critical threats and errors in real-time line-oriented scenarios Intervene to teach; tailor to weaker competencies seen in EVAL

Across a three-year cycle, ICAO design guidance assumes on the order of 48 hours of FSTD time per pilot so that assessment and training topics appear at the frequencies in the generation matrices (every module, alternate modules, or at least once in the cycle). Assessment runs continuously: observe behaviours, record effective and ineffective examples, classify against observable behaviours (OBs), determine root-cause competency, then grade. Facilitation is the primary debrief technique; showing and telling fill gaps when the crew lack the knowledge to discover the point themselves.

Baseline versus enhanced; CBTA parentage

EBT is a specialisation of competency-based training and assessment (CBTA). An operator builds an adapted competency model from the ICAO aeroplane-pilot framework (competencies, descriptions, OBs, performance criteria, grading). Implementation may be staged: apply CBTA grading to an existing programme; mixed EBT elements under current regulation; full baseline EBT when the authority permits licence and rating maintenance through EBT. Enhanced EBT further customises content from that operator's own safety management system (SMS), FDA, training metrics, and network-specific threats. In both cases the evidence loop continues after go-live: training metrics and operations data refine the curriculum, and instructor concordance keeps grades usable as data.

Instructor use

  • Treat the session as train-to-competency after a fair EVAL, not as a continuous check.
  • In EVAL, do not teach through the scenario; note competency gaps and OBs for later MT and SBT.
  • In MT, coach technique to a defined performance standard; do not turn manoeuvres into a LOFT.
  • In SBT, use scenarios as means: vary difficulty, protect surprise at programme-design level, and target the competencies that actually failed or need stretch.
  • Grade after facilitated debrief; announce pass or additional-training outcome early; ground low grades in observed behaviour and TEM outcome, not gut feel.
  • Explain to sceptical crews why a scenario is on the plan: generation data and operator evidence, not "the regulator wants a tick."

Connections

Sources