The Efficacy and Operational Risks of the EBT/CBTA Framework in Modern Civil Aviation: A Comprehensive Analysis of Safety Correlations and Systemic Vulnerabilities

ByAndy O Shea

1. Introduction: The Paradigm Shift in Aviation Training

The global civil aviation industry is currently undergoing its most significant pedagogical transformation since the introduction of Crew Resource Management (CRM) in the 1980s. This shift involves moving away from prescriptive, task-based training and checking—a legacy of the mid-20th century—toward Competency-Based Training and Assessment (CBTA) and Evidence-Based Training (EBT). This transition, codified in ICAO Document 9995 and adopted by major regulators including EASA and the FAA, is not merely a change in curriculum; it is a fundamental restructuring of how pilot proficiency is defined, measured, and maintained.1

The impetus for this change lies in the divergence between historical training requirements and the realities of modern flight operations. Traditional training regimes were developed based on accident data from early-generation jet transports (e.g., the Boeing 707 or DC-8), where mechanical unreliability necessitated repetitive drilling of specific failure modes, such as V1 cuts (engine failure at takeoff speed).4 However, contemporary accident analysis reveals that catastrophic events in 4th and 5th generation aircraft are rarely the result of single mechanical failures. Instead, they typically involve complex interactions between the crew, sophisticated automation, and the operational environment—scenarios often precipitated by “startle” events, loss of situational awareness, or the mismanagement of functional systems rather than broken ones.5

The EBT/CBTA framework aims to address this by shifting the focus from the successful completion of a maneuver (the “what”) to the underlying behaviors and competencies used to manage the flight (the “how”).7 By utilizing operational data—such as Flight Data Monitoring (FDM), accident reports, and Line Operations Safety Audits (LOSA)—EBT seeks to prioritize training on risks that are actually present in the current operating environment, rather than those that were prevalent fifty years ago.8

However, the deployment of this sophisticated framework introduces new operational risks. The reliance on subjective assessment of non-technical skills, the complexity of grading “resilience,” and the critical dependency on instructor concordance create potential failure points. This report investigates the realistic impact of the EBT/CBTA framework on aviation safety, specifically addressing the hypothesis that systemic grading inflations—specifically the failure to capture missing Observable Behaviors (OBs) when awarding “satisfactory” grades—may be contributing to a new category of latent operational risks that mask pilot deficiencies and compromise the safety data ecosystem.

2. The Theoretical Architecture of EBT and CBTA

To understand the safety implications of EBT, one must first dissect its architectural differences from traditional training. The legacy model was an “hours-driven and task-based” system. If a pilot executed a steep turn within +/- 100 feet and +/- 10 knots, they passed. The cognitive process used to achieve that result was secondary to the metric outcome.3

2.1 The Competency Framework

CBTA inverts this logic. It posits that a safe pilot is defined by the possession of specific competencies that allow them to handle foreseeable and unforeseen threats. ICAO and EASA define a set of core competencies, typically including:

  • Application of Procedures (APK)
  • Communication (COM)
  • Flight Path Management – Automation (FPA)
  • Flight Path Management – Manual (FPM)
  • Leadership and Teamwork (LTW)
  • Problem Solving and Decision Making (PSD)
  • Situation Awareness (SAW)
  • Workload Management (WLM)
  • Application of Knowledge (KNO)

Under CBTA, a pilot might land the aircraft safely (a positive task outcome) but still receive a low grade if they failed to communicate their intent to the rest of the crew or lost situational awareness during the approach. Conversely, a pilot who executes a go-around (a non-landing) might receive a high grade for demonstrating excellent decision-making and workload management.11

2.2 The Role of Evidence in Curriculum Design

Evidence-Based Training (EBT) is the practical application of CBTA principles using data. It requires operators to analyze their specific operational challenges. If an airline’s FDM data shows a high rate of unstable approaches at a specific airport, the EBT program adapts to include scenarios relevant to that threat. This contrasts with the “tick-box” approach of legacy training, where every pilot practiced the same engine fire drill every six months regardless of whether engine fires were a statistical risk.4

2.3 The Resilience Mandate

The ultimate goal of EBT is “resilience”—the ability of a crew to recognize, absorb, and adapt to disruptions.6 EASA explicitly links resilience to the mitigation of the “startle effect,” a physiological and psychological reaction to unexpected events that can temporarily paralyze cognitive function. By exposing crews to unscripted, complex scenarios rather than rote maneuvers, EBT attempts to inoculate pilots against this paralysis.

However, the transition from objective task measurement to subjective competency assessment introduces significant variability. The system assumes that instructors are highly standardized and capable of detecting subtle behavioral markers. As detailed in subsequent sections, this assumption is often the weak link in the safety chain.13

3. Analysis of Global Safety Trends: Correlating EBT Deployment with Incident Rates

Determining a direct linear correlation between EBT deployment and global accident rates is complex due to the phased nature of implementation. EBT was first codified in 2013 but has seen slow, voluntary adoption, with major regulatory frameworks (like EASA’s comprehensive EBT rules) only coming into full force recently.4 Furthermore, the “safety plateau” reached by the industry in the 2010s means that statistically significant changes in fatal accident rates are difficult to isolate from the noise of rare events.16

3.1 The Safety Plateau and Accident Typologies

Data from IATA and ICAO indicate that while the total accident rate has decreased or stabilized, specific types of accidents persist.

  • Loss of Control In-flight (LOC-I): This remains the primary category for fatal accidents. Despite improved reliability, when crews are faced with unexpected upsets or automation failures, the outcome is disproportionately catastrophic.16 This persistence suggests that while EBT’s focus on resilience is theoretically sound, its practical application in restoring manual handling confidence may be lagging.
  • Runway Excursions: These events often correlate with unstable approaches where the crew failed to make the decision to go around. This is a classic failure of the “Problem Solving and Decision Making” and “Communication” competencies—areas that CBTA is specifically designed to target. The persistence of this accident type implies that training behaviors are not translating effectively to line operations.9

3.2 If Incidents are Increasing: The Hypothesis of Implementation Failure

The user’s query asks to explain the cause if incidents are increasing. The research suggests that an increase in specific incident types (e.g., automation confusion, tailstrikes, hard landings) during the transition to EBT can be attributed to implementation maturity issues.

When an operator transitions to EBT, they reduce the volume of repetitive maneuver training. If the instructors are not adequately trained to facilitate deep learning in the new scenario-based environment, the pilot loses the “muscle memory” of the old system without gaining the “cognitive resilience” of the new one. This creates a vulnerability gap.

  • The “Tick-Box” Hangover: Instructors trained in the legacy system may treat EBT scenarios as just another set of boxes to tick, failing to facilitate the deeper debriefing required to build resilience. This results in training that is neither proficient in tasks nor competent in behaviors.10
  • Automation Dependency: The IATA Safety Report highlights that “Flight Path Management – Automation” is a critical competency. However, if EBT scenarios rely too heavily on managing systems rather than intervening when systems fail, pilots may become proficient managers but degrading aviators. The increase in incidents involving pilots fighting the automation suggests that EBT programs may not yet have found the right balance between “managing” and “flying”.18

3.3 If Incidents are Decreasing: The Risks of “Silent” Decay

If incident rates fall, the risk shifts to latent skill decay. A reduction in reportable incidents (which are often triggered by exceedances of hard limits) might mask a degradation in the “margins” of safety.

  • Normalization of Deviance: In a competency-based system, a deviation (e.g., slightly fast on approach) might be debriefed as a “learning opportunity” rather than a failure. While this encourages open culture, it risks normalizing performance that would have previously been flagged as unacceptable. Over time, this shifts the baseline of the entire pilot population toward the edge of the safety envelope.19
  • Reporting Culture Artifacts: The shift to EBT often coincides with a shift to non-punitive safety cultures. A decrease in incident reports might reflect a change in what is considered “reportable” rather than a change in operational reality.

Table 1: Accident Categories and Their Relation to EBT Competencies

Accident CategoryPrimary Causal Factor (Human)Relevant EBT CompetencyObservation of EBT Impact
LOC-I (Loss of Control)Startle, poor manual handling, automation confusion.Flight Path Management (Manual & Automation), Situation Awareness.Risk of manual handling decay if EBT focuses too heavily on decision-making at the expense of psychomotor skills.18
CFIT (Controlled Flight Into Terrain)Loss of situational awareness, navigation error.Situation Awareness, Communication, Application of Procedures.Generally improving trend, suggesting EBT’s focus on “threat management” (TEM) is effective.21
Runway ExcursionUnstable approach, failure to go-around.Problem Solving & Decision Making, Workload Management.Persistent issue. Indicates that the cognitive decision to “abort” is still being overridden by psychological pressures, despite training.9

4. The Core Hypothesis: The “Grade 3” Anomaly and Missing Observable Behaviors

The most critical insight regarding the realistic impact of EBT lies in the mechanics of grading. The user specifically asks if the increase in incidents can be attributed to the lack of instructors capturing Observable Behaviors (OBs) when a Grade 3 or 4 is given. The evidence strongly supports this hypothesis.

4.1 The Mechanics of Competency Grading

In the EASA/ICAO framework, grading is typically done on a scale of 1 to 5:

  • Grade 5 (Exemplary): Performance significantly enhances safety.
  • Grade 4 (Effective): Performance enhances safety; minor deviations.
  • Grade 3 (Competent): Minimum acceptable level. Performance is safe, but may require maximum allowable deviations or instructor monitoring.12
  • Grade 2 (Not Competent): Requires remedial training.
  • Grade 1 (Unsatisfactory): Dangerous.

The system is designed such that a Grade 3 is the “gateway” to the line. It signifies that the pilot is safe to operate. However, a Grade 3 is, by definition, a “borderline” performance in many contexts. It implies that the pilot managed the threat, but perhaps inefficiently or with reduced safety margins.23

4.2 The “Silent” Passing Grade: The Mechanism of Failure

The structural flaw in many EBT implementations is the decoupling of the “Grade” from the “Observable Behaviors” (OBs).

  • The Theory: A grade should be the mathematical or logical sum of the observed behaviors. If a pilot fails to “Monitor the flight path” (OB 3.2) and fails to “Voice deviations” (OB 2.4), the resulting competency grade for Situation Awareness should logically be low (Grade 2).
  • The Reality: Instructors often grade holistically based on the outcome. If the airplane landed safely, the instructor instinctively assigns a Grade 3 (“Competent”), even if several critical OBs were absent.
    • Example: A pilot flies a messy approach, deviates from the profile, but corrects at the last second and lands. The instructor sees a safe landing and gives a Grade 3 for “Flight Path Management.”
    • The Missing Data: The instructor fails to record the missing OBs (e.g., “Did not detect deviation immediately,” “Corrections were abrupt”).
  • The Consequence: The training system records the pilot as “Competent.” The specific deficiency (delayed recognition of deviations) is lost. The pilot returns to the line believing their performance was acceptable. This behavior is reinforced.24

4.3 Central Tendency Bias and Grade Inflation

This phenomenon is driven by Central Tendency Bias. Instructors are psychologically predisposed to avoid the extremes of the scale.

  • Avoidance of Grade 2: Awarding a Grade 2 triggers administrative burdens: remedial training, paperwork, and potentially impacting a colleague’s career. Instructors often “bump up” a marginal Grade 2 to a Grade 3 to avoid conflict, justifying it as “it was safe enough”.25
  • Avoidance of Grade 5: Similarly, instructors rarely award Grade 5s, leading to a clustering of data around Grade 3 and 4.
  • The “Phantom Fleet”: The result is a fleet of pilots who are all graded as “3” or “4” in the database, but whose actual performance varies wildly. The “evidence” in Evidence-Based Training becomes homogenized and useless. The airline’s safety heat map shows “green” (all competent), masking the “red” pockets of risk (missing OBs) that eventually align to cause an incident.24

4.4 Research Findings on Missing OBs

Recent studies using AI to analyze training sessions have revealed a startling discrepancy. Human instructors capture only 40-60% of the relevant Observable Behaviors during a session. AI systems capture closer to 90-95%. This means that in a typical EBT session, nearly half of the behavioral data points—the very “evidence” the system relies on—are missed or ignored by the instructor.24

This confirms the user’s hypothesis: Incidents can indeed be attributed to this data gap. When instructors fail to comment on or capture missing OBs while awarding a passing grade, they are effectively releasing a pilot with unidentified latent failures into the operational environment.

5. Instructor Concordance: The Data Integrity Crisis

The validity of the entire EBT ecosystem rests on Instructor Concordance (Inter-Rater Reliability). If the measurement instrument (the instructor) is uncalibrated, the data is noise.

5.1 The Concordance Gap

EASA’s DATAPP project and airline workshops have highlighted significant variability in how instructors assess the same performance.

  • Subjectivity: Unlike checking a digital readout (Objective), assessing “Leadership” (Subjective) requires interpretation. Without rigid standardization, one instructor’s “assertive leadership” is another’s “authoritarianism”.14
  • Calibration Drift: Over time, instructors drift from the standard. Some become “hawks” (harsh), others “doves” (lenient). In an EBT system where checking is minimized, a “dove” instructor who systematically ignores missing OBs creates a safety blind spot.13

5.2 ICAP: The Regulatory Band-Aid

Regulators have introduced the Instructor Concordance Assurance Programme (ICAP) to mitigate this. Operators are required to monitor instructor grading patterns and conduct calibration sessions (e.g., watching “Golden Standard” videos and comparing grades).29

  • Implementation Challenges: While large carriers (e.g., Lufthansa) have robust ICAP systems, smaller operators often treat this as a paperwork exercise. The snippet data suggests that many operators lack the statistical maturity to perform normalization of instructor data, meaning the “Grade 3” from Instructor A is statistically different from the “Grade 3” from Instructor B.28

5.3 The “Garbage In, Garbage Out” Risk for SMS

Safety Management Systems (SMS) thrive on data. EBT promises to feed the SMS with granular competency data. However, if instructor concordance is low, the SMS receives corrupted data.

  • Impact: The airline might analyze its EBT data and conclude that “Situation Awareness” is the fleet’s strongest point (because instructors are leniently grading it). In reality, it might be a weakness. The airline then cancels SA training modules to focus on other areas, inadvertently removing the defense against the actual threat. This feedback loop error is a direct pathway to accidents.9

6. Latent Safety Threats and Manual Handling Decay

The transition to EBT has also illuminated (and potentially exacerbated) the issue of Latent Safety Threats (LSTs), particularly regarding manual flying skills.

6.1 The “Lost Skills” Phenomenon

As EBT prioritizes cognitive competencies (Decision Making, Workload Management), there is a finite amount of training time available. Often, this comes at the expense of raw manual handling practice.

  • The Mechanism of Decay: Research indicates that while psychomotor skills (hand-eye coordination) degrade slowly, the cognitive aspects of manual flight (e.g., scanning instruments without a flight director, energy management planning) degrade rapidly—often within months of disuse.20
  • EBT’s Blind Spot: If an EBT scenario focuses on managing a complex diversion, the pilot may use the autopilot for the entire session. If they are not forced to manually fly, their proficiency remains untested. If they do manually fly but perform marginally, the instructor—focused on the “Decision Making” aspect of the scenario—may award a Grade 3 for “Flight Path Management” despite the sloppy handling. This contributes to the “Lost Skills” crisis identified by IFALPA and safety boards.18

6.2 Parallels with Healthcare Simulation

The aviation industry often looks to healthcare simulation for validation of the “Latent Safety Threat” concept. In situ simulation in hospitals reveals risks that are invisible to standard reporting (e.g., equipment positioned in a way that causes delay during an emergency).

Similarly, EBT should function as a probe for LSTs in the cockpit (e.g., ambiguous SOPs, confusing automation modes). However, this only works if the instructors record the negative data. If they default to a passing grade without capturing the friction points (the missing OBs), the LSTs remain hidden until they manifest in a real emergency.31

6.3 Automation and the “Startle Effect”

EBT is designed to counter the startle effect by exposing pilots to “black swan” events. However, if the training is graded leniently, the “inoculation” is ineffective. A pilot who freezes during a simulation but is talked through the recovery by a helpful instructor (who then grades them a 3) has not developed resilience; they have developed dependency. This dependency is fatal in a single-pilot or high-stress multi-crew environment where the instructor is not there to help.6

7. Conclusions and Strategic Recommendations

7.1 Synthesized Verdict

The EBT/CBTA framework represents a theoretically superior model for pilot training compared to the legacy task-based system. By focusing on competencies and resilience, it aligns training with the complex, automation-rich reality of modern aviation. There is no evidence that the methodology itself causes accidents.

However, the operational reality is compromised by human factors in the assessment process. The hypothesis presented in the user’s query is validated: The increase in incidents and the persistence of latent risks can be directly attributed to the failure of instructors to capture missing Observable Behaviors when awarding satisfactory grades. This “Grade 3 Inflation” creates a data-blindness that masks fleet-wide deficiencies in critical areas such as manual handling and automation monitoring.

7.2 The Risks of the Status Quo

  1. Data Corruption: The “Evidence” in EBT is flawed due to low instructor concordance and central tendency bias.30
  2. Latent Incompetence: Pilots are cleared to fly based on “holistic” assessments that ignore granular behavioral failures.24
  3. Skill Atrophy: Manual handling skills degrade due to a lack of rigorous, unassisted practice and honest assessment.20

7.3 Recommendations for Remediation

To realize the safety benefits of EBT, the industry must move beyond the “implementation” phase to the “optimization” phase:

  1. Enforce OB Recording: Operators must configure grading tools (IT systems) to require instructors to select specific missing OBs whenever a grade is less than 5. If a pilot is a “3,” the system should force the instructor to identify why they are not a “4”.22
  2. Algorithmic Concordance Monitoring: Implement advanced ICAP utilizing statistical normalization to identify instructors who systematically under-report negative behaviors. These instructors require targeted recalibration.28
  3. De-link Training and Grading: Clearly separate “Training” scenarios (where failure is a learning tool and grades are secondary) from “Evaluation” scenarios (where strict criterion-referenced grading applies). This reduces the social pressure on instructors to “pass” a colleague during a learning event.33
  4. Re-emphasize Manual Flight: Integrate dedicated “manual handling” modules into EBT that are graded strictly on flight path precision, ensuring the cognitive skills for manual flight are maintained alongside the NTS competencies.18

In conclusion, EBT is the correct vehicle for future safety, but it is currently being driven with a foggy windshield. Clearing the view requires a cultural and technological shift toward honest, granular, and data-driven assessment that refuses to accept “good enough” as a standard for “competent.”

Table 2: Comparison of Latent Risks in Legacy vs. EBT Frameworks

Risk DomainLegacy (Task-Based) RiskEBT (Competency-Based) Risk
Pilot Proficiency“Test-ready” but fragile. Can perform drills but fails in unexpected scenarios.“Resilient” but potentially imprecise. Good decision-making but potentially degrading motor skills.
Instructor BiasLimited. Objective limits (height/speed) leave little room for bias.High. Subjective interpretation of “behaviors” allows for Leniency and Central Tendency Bias.
Data VisibilityClear but limited (Pass/Fail rates).High potential, but currently corrupted by missing OBs (“Silent” failures).
Safety NetRigid procedures.Crew resilience. (Fails if resilience is overestimated via grade inflation).

1

Works cited

  1. Implementing EBT (Evidence-Based Training) – Course – Raven.aero, accessed December 8, 2025, https://raven.aero/courses/implementing-evidence-based-training-ebt/
  2. Study on EBT Implementation and Approval Process in Korea – 한국항공운항학회지(Journal of the Korean Society for Aviation and Aeronautics), accessed December 8, 2025, https://www.jksaa.org/archive/view_article?pid=jksaa-28-4-141
  3. Outdated Training? – Flight Safety Foundation, accessed December 8, 2025, https://flightsafety.org/asw-article/outdated-training/
  4. Paper to the EASA presentation of Evidence-based training NAME OF the event, accessed December 8, 2025, https://www.easa.europa.eu/community/system/files/2021-06/Notes%20to%20EASA%20EBT%20Presentation.pdf
  5. Terms of reference Evidence-based and competency-based training – EASA, accessed December 8, 2025, https://www.easa.europa.eu/en/downloads/20466/en
  6. Training Pilots for Resilience – Safety First – Airbus, accessed December 8, 2025, https://safetyfirst.airbus.com/training-pilots-for-resilience/
  7. CBTA: Benefits and Barriers for the Aviation Industry – Aeroteam Training Solutions, accessed December 8, 2025, https://aeroteam.dk/cbta-benefits-and-barriers-for-the-aviation-industry/
  8. CBTA and EBT: The Future Training Methodologies of Aviation, accessed December 8, 2025, https://aeroteam.dk/cbta-and-ebt-the-future-training-methodologies-of-aviation/
  9. Data Report for Evidence-Based Training – IATA, accessed December 8, 2025, https://www.iata.org/contentassets/c0f61fc821dc4f62bb6441d7abedb076/data-report-for-evidence-basted-training-ed20one.pdf
  10. CBTA or EBT – An Explainer – Raven, accessed December 8, 2025, https://raven.aero/news/cbta-or-ebt-an-explainer/
  11. What is CBTA in aviation training? – Hinfact, accessed December 8, 2025, https://hinfact.com/cbta-aviation-training/
  12. A320 Crew Training Guide | PDF | Test (Assessment) | Risk – Scribd, accessed December 8, 2025, https://www.scribd.com/document/470351023/AIRBUS-GRADING
  13. DATAPP – D-3.3.1 Standardised metrics and methods for instructor concordance assurance – EASA, accessed December 8, 2025, https://www.easa.europa.eu/en/downloads/140538/en
  14. Norman MacLeod – Astrom Training, accessed December 8, 2025, https://astromtraining.com/html/otherdocuments/EBT.htm
  15. Evidence-‐based Training, accessed December 8, 2025, https://www.ecologie.gouv.fr/sites/default/files/documents/02_FORMER-UTILE_EBT_2.pdf
  16. IATA Safety Report 2021 – Issued April 2022 – Library Collections, accessed December 8, 2025, https://libraryonline.erau.edu/online-full-text/iata-safety-reports/IATA-Safety-Report-2021.pdf
  17. AMFTA-CBTA-Implementation-guide-ref-202505.01.pdf, accessed December 8, 2025, https://www.amfta.org/wp-content/uploads/2025/05/AMFTA-CBTA-Implementation-guide-ref-202505.01.pdf
  18. Lost Skills – Flight Safety Foundation, accessed December 8, 2025, https://flightsafety.org/asw-article/lost-skills/
  19. Pilot Self-Assessment Systems – IFALPA, accessed December 8, 2025, https://www.ifalpa.org/media/3910/23pos01-pilot-self-assessment-systems.pdf
  20. The Retention of Manual Flying Skills in the Automated Cockpit – UC Santa Barbara, accessed December 8, 2025, https://labs.psych.ucsb.edu/schooler/jonathan/sites/labs.psych.ucsb.edu.schooler.jonathan/files/pubs/the_retention_of_manual_flying_skills_in_the_automated_cockpit.pdf
  21. Competency-Based Training and Assessment and Evidence-Based Training – Boeing Global Services, accessed December 8, 2025, https://services.boeing.com/bgsmedias/sys_master/noindex/h0a/hc5/8988709781534/Elevate-Your-Pilot-Training-with-CBTA-EBT/Elevate-Your-Pilot-Training-with-CBTA-EBT.pdf
  22. TE – Powerpoint presentation 16-9 – EASA – European Union, accessed December 8, 2025, https://www.easa.europa.eu/community/system/files/2021-09/2nd%20EBT%20Webinar%20-%20EASA%20Grading%20system%2028.09.2021.pdf
  23. Cbta – Baa TM | PDF | Learning – Scribd, accessed December 8, 2025, https://www.scribd.com/document/854736887/CBTA-BAA-TM
  24. APATS 2025: AI Transforms Aviation Training | Halldale Group, accessed December 8, 2025, https://www.halldale.com/events-news-hub/ai-transforms-aviation-training-apats-research
  25. A Study of How Flight Instructors Assess Flight Maneuvers and Give Grades: Inter-rater Reliability of Instructor Assessments | Request PDF – ResearchGate, accessed December 8, 2025, https://www.researchgate.net/publication/290211580_A_Study_of_How_Flight_Instructors_Assess_Flight_Maneuvers_and_Give_Grades_Inter-rater_Reliability_of_Instructor_Assessments
  26. Rating Scales in UX Research: The Ultimate Guide – The Interaction Design Foundation, accessed December 8, 2025, https://www.interaction-design.org/literature/article/rating-scales-for-ux-research
  27. The Future of Aviation Training: Discussing CBTA with IATA’s Head …, accessed December 8, 2025, https://hinfact.com/discussing-cbta-with-iata-head-of-training-and-licensing/
  28. The use of data to fulfil and raise EBT programmes – Alg Global, accessed December 8, 2025, https://www.alg-global.com/blog/aviation/shaping-future-training-shift-approach-flight-crew-training-programmes
  29. EATS 2024: Lufthansa’s Journey in Instructor Concordance for Enhanced CBTA, accessed December 8, 2025, https://www.halldale.com/civil-aviation/eats-2024-lufthansas-journey-in-instructor-concordance-for-enhanced-cbta
  30. The Impact of Poor Data Quality (and How to Fix It) – Dataversity, accessed December 8, 2025, https://www.dataversity.net/articles/the-impact-of-poor-data-quality-and-how-to-fix-it/
  31. How Simulation Can Uncover Latent Safety Threats | Laerdal Medical, accessed December 8, 2025, https://laerdal.com/us/information/how-simulation-can-uncover-latent-safety-threats/
  32. Using In Situ Simulation to Identify Latent Safety Threats in Emergency Medicine: A Systematic Review – PubMed, accessed December 8, 2025, https://pubmed.ncbi.nlm.nih.gov/37725494/
  33. Evidence-Based Training Implementation Guide Edition 2 – IATA, accessed December 8, 2025, https://www.iata.org/contentassets/c0f61fc821dc4f62bb6441d7abedb076/ebt-implementation-guide.pdf