Analyzing Human Error in Aircraft Ground Damage Incidents


C. A. Wenner and C. G.Drury carried out a study into human error in aircraft ground damage incidents. C.A Wenner and C.G Drury utilised the information collected by the sample airline within the airlines existing error reporting systems. This information was used to determine the common latent failures which contribute to typical ground damage incidents i.e how latent failures are linked to hazards.



130 Technical Operations ground damage incident (GDI) reports were analysed in this study (265 reported incidents).
The sample period covered ground damage incidents in a period from January 1992 to April 1995 at the sample airline with each report describing one GDI.
Next the GDI reports were categorized into twelve distinct patterns covering almost all of the ground damage incident (GDI) reports. These 12 distinct patterns were termed as Hazard pattern after Drury and Brill (1978).
Next to determine the specific failure that contributed to the hazard these GDI reports were then analysed with the failures grouped into: active failures, latent failures and local triggers – all failures that have the propensity to contribute to an incident.
Next the common factors between all of the incidents were shown by developing a scenario for each hazard pattern.
Each hazard pattern was also developed into an event tree to show how each latent failure contributes to the final damage occurrence.


Hazard Pattern 1 - Damage occurs when the aircraft is stationary via equipment striking the aircraft or the aircraft moving (or aircraft part) moves to contact an object.

Hazard Pattern 2 - Damage occurs when the aircraft is being towed or taxied via tug striking aircraft, aircraft not properly configured for towing and/or aircraft contacts fixed/movable object/equipment.

Hardware - This variable is to do with poor equipment, either equipment inappropriate for the task or a mechanical problem.

Environment - Environmental factors such as inadequate space (congested area and ill suited for task) and problems with ramp tarmac guide lines (guide lines do not exist, guidelines do not extend out of the hangar and guidelines not suitable for aircraft type)

Liveware (individual) - Lack of awareness of Risks/Hazards. Maintenance personnel are caught unaware of the possible risks associated with their actions. Then this lack of information contributed to an incident.

Liveware - liveware - Poor communication (poor communication between crew or poor communication between shift), Personnel unaware of concurrent work, Pressures to maintain on time departures and push back policies not enforced.

Thus the Null hypothesis is established "There is no significance between the the occurrence of latent failure and hazards."

Coding Hazard Patterns
HP1 Aircraft is Parked at the Hangar/Gate/Tarmac
HP2 Aircraft is Being Towed/Taxied
Latent Failures 1st Division Categories 2nd Division Categories
H Hardware H1 Poor Equipment H1.1 Inappropriate for Task
H1.2 Mechanical Problem
E Environment E1 Inadequate Space E1.1 Congested Area
E1.2 Ill-suited for Task
E2 Problems with Painted Guidelines E2.1 Do Not Exist
E2.2 Do Not Extend Out of Hangar
E2.3 Not Suitable for Aircraft
L Liveware (Individual) L1 Lack of Awareness of Risks/Hazards --
LL Liveware-Liveware LL1 Poor Communication LL1.1 Between Crew
LL1.2 Between Shifts
LL2 Personnel Unaware of Concurrent Work --
LL3 Correct Number of People Not Used --
LL4 Pressures to Maintain On-Time Departures --
LL5 Pushback Polices Not Enforced --

Table 1. An event tree showing the relationship of main hazard patterns and latent failure categories with simplified coding used for analysis.

Data Analysis

Latent Failures Mean Standard Deviation
Hardware 6.55 8.31
Environment 4.64 9.25
Liveware 12.91 13.45

Table 2. Mean and standard deviation of 3 major latent failures

In Comparison Chi-square test Implication Interpretation of Significance
HP1, HP2 vs H, E, L, LL X2(3)=15.2 p < 0.001 Large
HP1, HP2 vs H1, E1, E2, L1, LL1, LL2, LL3, LL4, LL5 X2(8)=28.4 p < 0.001 Large
HP1, HP2 vs 1.1, 1.2, 2.3 X2(16)=90.6 p < 0.001 Large
Note: No latent failure is generally applicable across all hazard patterns therefore latent failures 1st division and 2nd division categories are analyzed specifically with failures of the most number of incidents

Table 3. Chi-square analysis of the significance of the effect of latent failures on the hazards

  • First comparison shows a chi-square test result of 15.2 with a p-value less than 0.001 when comparing hazard patterns 1 and 2 with hardware, environment, liveware (individual) and liveware-liveware latent failures from the 265 reported incidents
  • Second comparison shows a chi-square test result of 28.4 with a p-value less than 0.001 when comparing hazard patterns 1 and 2 with 1st division categories from the 265 reported incidents
  • Third comparison shows a chi-square test result of 90.6 with a p-value less than 0.001 when comparing hazard pattens 1 and 2 with the main 2nd division categories from the 265 reported incidents

Therefore the Null Hypothesis is rejected as the p-values show there is a significance between the occurrence of latent failures and hazards.


The results show that there is a large significance between latent failures and hazards. There is a positive correlation between latent failures and hazards therefore a more detailed analysis is required to reduce the occurrence of incidents when an aircraft is under tow, an aircraft or part moves to contact objects and equipment strikes a parked aircraft in order to reduce the likelihood of the occurrence of latent failures in order to reduce the likelihood of incidents to cause hazards.

If the actual article could provide adequate data on active failures and local triggers, this would help the entire analysis to be more accurate for comparison of latent versus active failures and how they cause a hazard.

1. Caren A Wenner, Colin C Drury Received 10 June 1997. Analysing human error in aircraft ground damage incidents. Google Scholar, 2000 Elsevier Science B.V.


Authors / Editors

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License