8
$\begingroup$

Is there an encoding system that replaces the vagueness of natural language when it comes to logging aircraft problems or repairs that took place at a specific site on the airplane?

So, instead of someone having to enter a description like: "Hydraulic fluid change at system ABC" Which would be in their own native language and filtered by their own personal level of awareness, they could enter something like: "GF65Z ABC"

Therefore, internationally, GF65Z could correspond to exactly that type of intervention with some hydraulic system and that would also make further investigation, in the case of some problem, easier.

Anywhere I can find more information about something like this?

$\endgroup$
0

2 Answers 2

14
$\begingroup$

Natural language is concrete enough for this purpose, the logs aren't meant for machine parsing but instead for pilots, inspectors and other mechanics.

Having such a code system requires mechanics to memorize all of the codes or have a massive list handy when filling out the logs.

Also making such an international code system requires agencies of all countries/manufacturers to sit together and set those codes for all possible maintenance actions. There is a very high chance an action will be missed so a natural language box for "additional comments" is still required. It also needs to be rolled out to the actual mechanics filling out the logs.

$\endgroup$
2
  • 5
    $\begingroup$ While it's not possible to code every repair or maintenance action, some standard language does exist. For example annual inspections are often signed off with a stamp: I certify that this AIRCRAFT has been inspected in accordance with an ANNUAL inspection and was determined to be in an airworthy condition. -- Similarly airworthiness directives & Service Bulletins are usually logged as CW SB 2014-01-01, next due xxxx ("Complied With") or PCW AD 2014-01-01 by replacement of widget ("Permanently Complied With"), and implicitly reference the AD or SB for actions taken. $\endgroup$
    – voretaq7
    Nov 6, 2014 at 17:17
  • 2
    $\begingroup$ Also, a list of opaque alphanumeric codes would make data entry mistakes very common. $\endgroup$ Nov 6, 2014 at 20:50
3
$\begingroup$

To go along with @ratchet freak's answer, machine parsing comes more into play when reporting maintenance status and readiness statistics, and for trend analysis - in other words, high-level aggregation and dissection of user-inputted data. At this level, managers are more concerned with the number of aircraft available for use rather than the last time a particular aircraft's tires were kicked changed. Typically, this is all done using pricy electronic logbook software, and would be used by large organizations with lots of equipment to track and the money to employ such a system.

Take, for example, the United States Army. Its program of record for aviation maintenance and logistics is called Unit Level Logistics System-Aviation, or ULLS-A for short. This software suite automates a lot of the manual forms that the Army uses. At the most basic level, aircraft maintenance is recorded using DA Form 2408-13-1, according to DA Pam 738-751. The time and date of each status is compiled into a daily statistic, which is then rolled up into a monthly report, often only detailing (aside from the hours that aircraft was available for mission) what subsystem was responsible for the majority of each aircraft's downtime. Here's what a blank maintenance form looks like: DA Form 2408-13-1

As you can see, there are a multitude of blocks, each with its own list of codes. For instance, here is a list of the codes for WHEN DISC (when discovered), from Table 1-6 of DA Pam 738-751:

B   Handling
D   Sustainment facility (depot) level repair/overhaul/rebuild
E   Storage
G   Flight
H   Phase maintenance inspection (PMI, PPM, PMS2/3 and Reset)
J   Calibration
K   Unscheduled maintenance
L   Maintenance operational check
M   Maintenance test flight and/or functional check flight
N   AOAP results
O   Special inspection, scheduled maintenance
P   Diagnostic test (health and usage monitoring device and/or sytem)
Q   Servicing
R   Rearmament
S   Reconfiguration
T   Preflight inspection
U   Thru flight inspection
V   Post-flight inspection
W   Acceptance inspection
X   Daily/PMS/PMS1 inspection
Y   Intermediate inspection
Z   Periodic inspection  

However, not all blocks are required to be filled out when using paper logbooks (believe it or not, still in use in some places); in addition to WHEN DISC, HOW REC, MAL EFF, WUC, ACTION CODE, and CAT are optional entries, and in practice never get used. In contrast, these are required fields in ULLS-A, and so the system provides a number of helps to the user so he doesn't have to refer to the manual all the time.

Overall, only these codes (and not the free text fault/remarks block) will be machine-parsed, for the purpose of high-level analysis to determine trends and/or systemic issues affecting the entire fleet.

$\endgroup$
2
  • 2
    $\begingroup$ Thank you very much for the comprehensive response. I work in healthcare which shares similar risks in terms of loss of life with aviation and encoded information is used everywhere (e.g. apps.who.int/classifications/icd10/browse/2015/en). It was natural to ask myself if there was something that aviation might be doing much more efficiently. To an extent, aviation and healthcare probably involve a similar amount of surprising realisations of how things are actually managedin the field. $\endgroup$
    – A_A
    Mar 24, 2015 at 10:07
  • $\begingroup$ "Optional codes and never get used". The NAMP is like the mirror-universe version of this. $\endgroup$
    – fectin
    Oct 11, 2022 at 1:07

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .