A friend sent me an email asking about incidents and accidents occasioned by turbulence. The member found it odd that turbulence (an every-day industry occurrence) would yield an incident or accident report. The query identifies one of the problems with our industry’s reliance on the accident-incident determination as the basis for identifying whether there may be damage to an aircraft part that affects its airworthiness.
Facts
The fact pattern involves an aircraft that was flown in a non-US country, by that country’s flag operator.
My friend obtained a “non-accident / incident” statement from the operator indicating that there were two occurrences in the history of the aircraft. One occurrence was an incident and one was an accident and they were almost eight years apart. The statement specifies that both occurrences were classified by the government (as incident and accident, respectively) because of serious injuries to passengers and/or crew. The statement also specifies that each serious injury arose in the context of severe turbulence.
The operator’s “non-accident / incident” statement specifies that it is based on International Civil Aviation Organization (ICAO) Annex 13.
The concern raised by the ASA member was whether turbulence represents a reportable incident. Focusing on the turbulence, though, is focusing on the wrong element.
Rules and Standards
Our industry uses terms that are defined in the ICAO Standards and Recommended Practices (SARPs). The terms that are relevant to our inquiry today are found in Annex 13, which is the annex for Aircraft Accident and Incident Investigation.
Under ICAO Annex 13, the term “accident” includes an occurrence involving a death or serious injury to a person on the aircraft. Serious injuries (generating an “accident” label) can include fractures and lacerations. So is a passenger suffers a broken arm as a consequence of turbulence, then the broken arm is reported as an accident (the turbulence in this case would merely be the environment in which the accident arose).
Under ICAO definitions, a serious incident is any circumstance in which “there was a high probability of an accident.” ICAO Annex 13 clarifies that incapacitation of a flight crew member during flight is an example of a serious incident under the ICAO standards. So if the turbulence in our example incapacitated a flight crew member during flight, then it would reflect a serious incident under the ICAO standards.
An incident (not a serious one) is any “occurrence, other than an accident, associated with the operation of an aircraft which affects or could affect the safety of operation.” This is a catch-all term for anything that falls below the threshold of an accident, but might nonetheless be interesting to an accident investigation body.
The terms “accident” and “incident” are both defined to arise only in the context of operation of an aircraft while people are on board.
Note that these terms tend to have very similar (but not always identical) meanings enshrined in national law, because the national laws are often based on the ICAO standards. In the US, the National Transportation Safety Board (NTSB) regulations have very similar language in their definitions.
You can see from these definitions that it is quite easy to have an incident, a serious incident, or even an accident, that has absolutely no effect on the integrity of the aircraft parts.
It is important to remember that these terms are defined in the context of investigations. They help define when a state has authority to investigate an occurrence, and when a state has an obligation to investigate an occurrence. The purpose of this is to foster a uniform process of investigation that will allow the world to learn from accidents and incidents, and to better prevent them in the future. They are not defined in terms of the occurrence’s likely effects on aircraft parts, and they were never originally intended to be used for the ‘accident-incident statement’ purposes that our industry has applied to them. So it is only natural that these terms would fail to be 100 percent useful in some circumstances when we are using them for something other than their intended purpose.
Analysis
The member asked whether instances of turbulence could be cause for a shop to treat these units as incident-related or accident-related. The short answer is “no, turbulence alone is typically not an incident, nor an accident” but in this fact pattern it was not the turbulence that was the basis for the incident and accident labels … it was the fact that passengers and flight crew were injured during the turbulence.
As a threshold matter, it is important to remember why we identify units as incident-or-accident related. The reason is because where an incident or accident may have caused damage to a part, we want a repair facility to assess whether such damage exists – usually through an appropriate form of hidden damage inspection. Microscopic cracks/fissures can propagate in to larger ones, so hidden damage assessment is intended to identify potential safety problems before they become actual safety problems.
There are two ways to analyze this situation – a legal answer and a practical answer.
Legally, a repair station has an obligation to develop a procedure for “inspecting all articles that have been involved in an accident for hidden damage before maintenance, preventive maintenance, or alteration is performed.” 14 C.F.R. 145.211(c)(1)(iii). An FAA repair station must follow that procedure. 14 C.F.R. 145.211(b).
The FAA regulations do not impose a specific response for incidents, although a repair station is free to develop its own procedure for responding to known incidents. When the repair station is faced with a component that has been installed on an incident-related aircraft, the regulation impose no additional burden on the repair station – any decision to perform hidden damage assessment will be based on customer requests for inspection, maintenance manual provisions (some manuals require hidden damage assessment in all overhauls, regardless of whether there is an accident or incident history), and good safety practice (performing the assessment where circumstances suggest a need for such an assessment, even where the law might not require it).
However, where the occurrence has been designated as an accident, the analysis changes. When an article is identified as having been involved in an accident, then the repair station has a legal obligation to follow its hidden damage assessment procedure (which might require hidden damage assessment). Thus, if turbulence resulted in an accident (e.g. there was a serious injury to a person, like a fracture), then this could drive a legal obligation for hidden damage assessment under the repair station’s quality system. One way to avoid this, where it becomes an inappropriate waste of inspection resources, might be for the repair station to write into its procedure a clause that permits it to waive the hidden damage inspection where the facts indicate that hidden damage to the particular unit in question was not reasonably possible from the reported event.
At a practical level, though, it is wise to ignore the semantic labels applied to a part, and consider whether the history of the part makes it plausible that there might be hidden damage. In the case of a complete aircraft, turbulence that resulted in an incapacitating crew injury might be a serious incident (of the sort that the government wants to track in order to reduce such circumstances), but it is an expected environment for aircraft operation. If there was no other basis for calling the occurrence an incident and is not be the sort of incident that could reasonably cause hidden damage to the aircraft, nor to any specific component from the aircraft, then hidden damage inspection might not be reasonably necessary. This does not change the fact that the NTSB wants to track injuries in order to find ways to protect against them.
On the other hand, there might be other occurrences that do not even rise to the level of incident that might nonetheless drive a practical need for a hidden damage assessment. For example, if an aircraft is improperly chocked during a windstorm and gets pushed by the wind into a building, then this could cause stresses on the affected structure that might lead to cracking. This is the sort of circumstance where hidden damage assessment of the affected structures might be appropriate. But because no one was on the aircraft and it was not being operated, it is likely not defined as an accident nor as an incident, as those terms are defined by ICAO (and, in the US, as those terms are defined by the NTSB).
I advise companies that hidden damage assessments should be applied as appropriate to the physical realities of possible hidden damage, and also as called-for in the appropriate maintenance manuals. An incident involving a catering truck hitting the aft portion of the fuselage is unlikely to cause hidden damage to avionics in the cockpit, but it may be reasonable to perform assessments of impact damage to the affected aircraft structures.
I also feel that once a hidden damage assessment has ruled out the possibility of hidden damage, there is no longer any practical need to continue to convey that accident-incident history (the FAA issued a private letter ruling confirming this, as a matter of FAA policy, in 1999). But a company must be careful to ensure that it is accurate in its commercial representations (for example, do not say that a part has never been installed on an incident-related aircraft if it has previously been so-installed; but it is acceptable to say that post-incident hidden damage assessment confirmed no hidden damage from the incident).
The Aviation Suppliers Association (ASA) continues to work on parts-specific language that allows a discloser to uniformly identify that any past possible damage to an article has been assessed and cleared through hidden damage assessment. For more on this topic, be sure to attend the ASA Quality Committee meeting in December!