CAP 1367 Template W Charts

Aircraft Maintenance Incident Analysis
CAP 1367
CAP 1367 Contents
Published by the Civil Aviation Authority, 2015
Civil Aviation Authority,

Aviation House,
Gatwick Airport South,
West Sussex,
RH6 0YR.
You can copy and use this text but please ensure you always use the most up to date version and use it in context so as not to
be misleading, and credit the CAA.
The latest version of this document is available in electronic format at www.caa.co.uk.
Enquiries regarding the content of this publication should be addressed to:
Airworthiness,
Safety and Airspace Regulation Group,
Aviation House,
Gatwick Airport South,
West Sussex,
RH6 0YR
January 2016 Page 1

CAP 1367 Contents
Contents
Contents ..................................................................................................................... 2
Foreword .................................................................................................................... 4
Executive summary .................................................................................................... 6
Chapter 1................................................................................................................... 8
Introduction ................................................................................................................ 8
Chapter 2................................................................................................................. 11
The analysis of incident data .................................................................................... 11
Chapter 3................................................................................................................. 13
MEMS ...................................................................................................................... 13
Background ....................................................................................................... 13
The UK-MEMS programme ............................................................................... 14
The UK-MEMS group ........................................................................................ 15
Chapter 4................................................................................................................. 16
Methodology for the analysis .................................................................................... 16
Background ....................................................................................................... 16
Taxonomy ......................................................................................................... 16
Baseline analysis results ................................................................................... 18
Chapter 5................................................................................................................. 28
Detailed analysis for large aircraft ............................................................................ 28
Large aircraft global statistics ............................................................................ 28
Large aircraft maintenance error types.............................................................. 32
Aircraft type analysis ......................................................................................... 34
Chapter 6................................................................................................................. 40
Detailed analysis for small aircraft ............................................................................ 40
January 2016 Page 2

CAP 1367 Contents
Small aircraft global statistics ............................................................................ 40
Small aircraft maintenance error types .............................................................. 43
Chapter 7................................................................................................................. 46
Detailed analysis for large helicopters ...................................................................... 46
Chapter 8................................................................................................................. 48
Examples of maintenance error ............................................................................... 48
Chapter 9................................................................................................................. 57
Summary .................................................................................................................. 57
Chapter 10............................................................................................................... 59
Conclusions.............................................................................................................. 59
Actions to the CAA ............................................................................................ 59
Recommendations to operators, PART M and PART 145 organisations .......... 59
January 2016 Page 3

CAP 1367 Foreword
Foreword
1. This report was originally intended to be released in 2013. This date was
delayed because of the UK CAA having to carry out a review Offshore
Helicopter operations in the North Sea. The UK CAA published CAP 1145
on the 20 February 2014. Some of the data which had been analysed as
part of this paper was used to produce the Human Factors charts for the
offshore operators in CAP 1145, Annex F. This resulted in CAP 1145,
Action 30 which states that:
The CAA will carry out a further review of Human Factors Maintenance
Error data referred to in this report and publish the results to seek
improvements in this important area.
2. The research in this paper was carried out by the engineering specialist
staff within the Confidential Human Incident Reporting Programme
(CHIRP) using data recorded on the UK-MEMS database, supplemented
by additional data on maintenance error events from reports in the CAA
MOR database. The UK-MEMS database was established by CHIRP with
CAA funding to provide a means of capturing the information from
investigative reports into maintenance error events.
3. Further information on Maintenance Error Management Systems (MEMS)

can be found in CAAIP Leaflet B-160, including the philosophy behind the
reporting of maintenance errors and the actions required by organisation
through their investigative processes. The CAA had also published two
previous papers on Aircraft Maintenance Incident Analysis and these serve
a useful background to this paper. The CAA Papers are available on the
CAA web-site, reference CAA Papers 2007/04 and 2009/05.
4. A limited number of Part 145 organisations (23) have contributed the

reports of their findings over the period to this database, and therefore the
experience of all Part 145 organisations (in the order of 400 plus) is not
necessarily captured by the UK-MEMS database alone. A preliminary
analysis was presented by CHIRP to the CAA in 2009 which gave useful
results but reflected only a partial analysis of the available data. Additional
use has therefore been made of CAA MOR data, which perhaps better
reflects the industry wide experience of maintenance error events.
5. In conjunction with the CAAs high level HF Review, which is focused on

reviewing the CAAs strategy for HF matters across the aviation industry,
and work to review and update the CAA guidance on Maintenance Error
Management Systems, it was felt appropriate to carry out further work with
January 2016 Page 4

CAP 1367 Foreword
a view to publishing a further five year analysis over that contained in CAA
Paper 2009/05.
6. Once again, this paper seeks to provide information on the common

causes where maintenance error has been a contributory factor in incidents
and occurrences reported to the CAA. The aim is to provide industry with a
more up to date set of data which they can review and, where appropriate,
use to complement their own analyses as part of their Safety Management
Systems.
January 2016 Page 5

CAP 1367 Executive summary
Executive summary
1. Maintenance Error continues to be one of the most obvious safety threats

from an engineering or airworthiness perspective. Over the years, the
evolution of aircraft design techniques, the use of new materials and using
the learning from incidents and accidents has seen improvements in
aircraft system design and component reliability. Whilst accidents due to
airworthiness issues do happen they are relatively rare.
2. However, despite those improvements the system of performing

maintenance remains vulnerable to the issues surrounding human factors.
Humans are fallible and therefore errors and mistakes are still going to
happen. It is through good training and competence assessment regimes
and creating a culture that fosters good engineering practice that will
reduce the likelihood of errors. The Organisation also needs to create an
environment where engineers can concentrate on the task at hand without
external pressures compromising the quality and integrity of the work by
putting pressure on individuals to cut corners. It is these violations from
process and procedures where both the individual and the organisation
become exposed to accidents and incidents.
3. Training is essential as engineering skills and good practice does not

materialise out of thin air. Everyone involved in aircraft maintenance should
possess the necessary competence, including the required behaviours and
attitudes to do the job. Procedural control is important as there is a reason
why maintenance has to be done in a specific sequence. The quality of the
work and the rigour with which the function checks are performed verifies
the adequacy of the work that has been done. It is a continuum from start
to finish and having multiple actors within the overall process it requires co-
ordination and attention to detail.
4. The investigation of maintenance events is a valuable tool in the armoury

for safety improvement. It allows us to look at what went wrong and why.
The contributing factors of any incident are important so that we can
identify interventions that will prevent, or at least minimise the chances of it
happening again.
5. Human error is influenced by the individuals personal circumstances,

frame of mind and approach to the job. There are many causes that lead to
someone having a simple lapse in concentration that results in an error or a
mistake. It is important therefore to look at corrective actions that will help
the individual as well as the maintenance system.
January 2016 Page 6

CAP 1367 Executive summary
6. Training on its own is not the solution to an issue. There is clear evidence
that training is not an entirely effective way of eradicating error. It is the
underlying culture and approach to safety that matters. That needs to be
fostered through the organisations values and strategic direction. The
individual has to recognise that they are part of a team and that each has a
role to play and standards to maintain. The unlicensed mechanic plays a
vital role since it is at the point of doing that many errors are made, as
such, organisations need to focus on all members of staff to ensure they
have the tools, training, procedures / processes and have the competence
necessary to complete the assigned task consistently to a good standard.
7. That does not mean that supervision is un-necessary. It is the overall

integration of the particular task and its effect on other tasks that are being
worked in parallel that makes the role of the supervisor vital. It is the
supervisor who manages the workload, task allocation and planning, whilst
protecting individuals from external pressures and influences.
8. Senior management have an important role to create the right environment

and system for people to work in. This includes providing sufficient
resources, workable procedures, creating the right safety culture and a
shared vision managing external pressures, providing serviceable tools and
equipment. Without a system to manage human related risks maintenance
errors will continue to occur.
January 2016 Page 7

CAP 1367 Chapter 1: Introduction
Chapter 1
Introduction
1.1 Following a number of high profile maintenance error events in the early
1990s considerable work was done in looking at the issue of human
factors (HF) and human performance within aircraft maintenance. It
appeared that the growing complexity of aircraft technologies, the
prevalence for carrying out maintenance during the night and the impact of
the increased pressure on the commercial needs of the operation all had
the potential to create an environment where the potential for error could
exist.
1.2 As part of that HF focus, the need to make engineering staff aware of the
potential pitfalls associated with human error and performance gathered
some support. The concept of error investigation took hold and several
schemes and basic investigative tools followed. However, identifying the
root cause was one thing, knowing how to address it was something else.
1.3 This led to the introduction of HF training requirements for all maintenance
staff, at both an initial and continuation training level. These requirements
were introduced into JAR-1451 and remain an essential element of the new
EASA Part 145 rule. In addition, a syllabus of training was developed for
licensed engineers and included as module 9 in Part 662. The Part 145
requirements were similarly enhanced by the provision of typical subjects
for study in the associated guidance. To provide further guidance to JAR-
145 organisations about their responsibilities regarding management of
human factors the CAA published CAP 7163 in January 2002. This gave
information on safety culture, error reporting, error management
programme and training.
1.4 CAA Papers 2007/054 and 2009/045 reported on research into the common
causes or factors associated with incidents attributable to maintenance
error. The work in the earlier 2007 paper covered the periods from 1996 to
2005 with the later 2009 paper extending that analysis to include data from
reports received during 2006.
1
Introduced into JAR-145.A.30 and A.35 in Amendment 5
2
JAR-66 was introduced in 1998 but was subsequently supported by the issue of CAP 715 An introduction to Aircraft
Maintenance Engineering Human Factors for JAR-66, January 2002
3
CAP 716 Aviation Maintenance Human Factors (EASA / JAR 145 Approved Organisations) guidance material on the UK
CAA Interpretation of Part 145 Human Factors and Error Management Requirements, first issued January 2002
4
CAA Paper 2007/05 Aircraft Maintenance Incident Analysis, December 2007
5
CAA Paper 2009/04 Aircraft Maintenance Incident Analysis, July 2009
January 2016 Page 8

1.5 Almost 4000 MORs were analysed in the data set associated with these
studies, primarily for aircraft above 5700 kgs MAUW. Those two analyses
validated the chosen taxonomy and helped identify emerging themes or
trends.
1.6 In 2005, the CAA had sponsored the setting up of a maintenance error
database under the auspices of the Confidential Human Incident Reporting
Programme (CHIRP)6 at their Farnborough offices. The purpose of this
was to record information supplied by a number of pilot Part 145
organisations from their internal investigations into maintenance error
events.
1.7 These investigations were part of the process assigned to the

organisations Maintenance Error Management System (MEMS), where
error events or incidents were analysed in detail to establish the underlying
causal factors, and associated root cause, with a view to using that
learning to help reduce the future potential for similar errors. The whole
purpose of the UK MEMS database was to collate and share the results of
the investigations for the greater benefit of all organisations. In essence,
the MEMS process readily forms a sub-set of an expected risk analysis
and response mechanism under an organisations Safety Management
System (SMS).
1.8 The MEMS requirements are outlined in guidance contained in CAAIP

Leaflet B-160 (previously Airworthiness Notice No. 71)7 and stem from the
need for an approved organisation to have an internal investigation system
under Part 145.A.60. This is a legally required element of the Part 145
approval as it falls under Regulation (EC) 2042/2003, as amended. All Part
145 organisations are therefore required to have such systems, whether
working on whole aircraft or aircraft components. The content of Leaflet B-
160 has recently been amended, in July 2012, to align MEMS policy with
the latest EASA requirements and with the introduction of SMS that is now
taking place.
1.9 In addition, organisations are required to make reports under the provisions
of the CAAs Mandatory Occurrence Reporting (MOR) Scheme. This is
enacted through the legal provisions of the UK Air Navigation Order and
amplified through CAP 382. That document was also amended, in March
2011, to take account and enact into UK legislation the requirements of the
European Directive on Occurrence Reporting, EU Directive 2003/42/EC. It
should be noted that the revised MOR provisions require incidences of
human factors or maintenance error to be reported to the CAA (reference
6
CHIRP is a charitable trust that had its roots in the Aviation School of Medicine and their interest in human performance issue
relating to flight crew, subsequently extended to include aircraft maintenance engineers.
7
The MEMS guidance was first issued in March 2000.
January 2016 Page 9

CAP 382, Appendix B). It therefore lends itself to MEMS becoming a useful
tool to follow up MORs and provide supplementary reports.
1.10 The purpose of this study, which was carried out by CHIRP on behalf of the
CAA, was twofold. The first aim was to collate updated information on
maintenance error related events and data up to the end of 2011 has been
included in the study. The second element was to try and obtain
information on the current safety threats from maintenance error and to try
and determine if the current application of HF training is effective.
1.11 A copy of the database containing the analysed information, coded and
classified by CHIRP, formed a deliverable to the CAA under this project.
CHIRP staff collated and analysed the majority of the data upon which this
paper is based and the CAA expresses its thanks to the CHIRP team for
their efforts and diligence.
January 2016 Page 10

CAP 1367 Chapter 2: The analysis of incident data
Chapter 2
The analysis of incident data
2.1 The CAA MOR data set analysed by CHIRP comprised 2733 maintenance
occurrence reports covering the period from January 2005 to December
2011. This data set contained 2399 reports relating to large aeroplanes, 85
relating to large helicopters and 249 relating to small aircraft (below 5700
kgs). It is not known what proportion of the actual number of events across
the industry this represents as it is wholly dependent upon reports being
submitted to the CAA, despite the reporting of such incidents being
mandated under the MOR scheme.
2.2 The following information for each CAA Occurrence Report was provided to
CHIRP for them to analyse:
Aircraft type;
Occurrence number;
Occurrence grade classification;
Occurrence date;
Operator / maintainer;
Aircraft manufacturer;
Event descriptor;
Pre-title;
Prcis of the event and investigation;
ATA chapter.
2.3 Whilst this data set is slightly different from those used in the research
under the previous CAA Papers, the key elements were still present and
therefore it is believed that there remains a fair degree of consistency in
the methodology for the data analysis and therefore in the results obtained.
2.4 There were notable differences between the data on the CHIRP-MEMS
database and that supplied by the CAA from the MOR records system. In
many cases, the reporting organisation provided basic information on the
incident to the CAA to satisfy compliance with the MOR requirements but
did then go on, at the end of the investigation to supply more detailed
information to CHIRP for their MEMS database and subsequent analysis.
Whilst the perceived need for individuals to restrict information reported to

CAP 1367 Chapter 2: The analysis of incident data
the CAA is understood there is a confidential route available under the

CAA MOR system8.
2.5 In some instances, the MOR system had a record of a report being made
whilst there was no comparable information within the UK-MEMS
database, and vice versa. CHIRP therefore took information from both
sources to better populate the spreadsheet for a more complete analysis.
2.6 In the case of CAA supplied MOR data, the information provided was
subject to the CAAs normal provisions regarding confidentiality of data9.
The use of UK-MEMS data was further enhanced by the CHIRP protocols
on dis-identifying personal data within their database. The data set
therefore provided a suitable level of confidentiality for individuals whilst
retaining enough markers within the data to allow analysis in accordance
with the developed taxonomy.
2.7 Although the analysis and information in this review covers large
aeroplanes, large helicopters and small aircraft the data has been
analysed both collectively and independently so information on common
themes for each sub-set is available. As in the previous CAA Papers this
study concentrates on the larger fixed-wing transport aircraft in order to
maintain a degree of consistency against the earlier data and findings
wherever possible.
2.8 It was felt that the large helicopter population, with the increased
complexity of such rotorcraft, merited study in its own right. Accordingly
some separate analysis on that data set has been made to determine if
there are sector specific trends that can be identified.
2.9 It was felt that small aircraft should be included and this also aligns with the
work that CAA has done in creating a new General Aviation Unit. Statistical
data and analysis for this additional sector has therefore been included in
this paper. However, given the lack of information on GA aircraft, the
results of the analysis can only be indicative, not necessarily wholly
representative of the issues.
8
The MOR confidential route allows the CAA to have access to the reporter in order to follow up on the details of the incident
without the reporters identity being released to third parties. This allows the CAA to investigate the MOR if necessary without
compromising the reporter. Anonymous reports are not accepted under the MOR system.
9
The CAA MOR scheme does have a confidential reporting provision within it where the identity of the reporter is kept
confidential during any investigation. However, many individuals believe that this is not sufficient. In the early stages of the
MEMS programme, the CAA agreed that the database would be held by CHIRP to allay some of the expressed concerns over
confidentiality. The disclosure of data regarding individuals is covered by the CAAs compliance with the Data Protection Act.

CAP 1367 Chapter 3: MEMS
Chapter 3
MEMS
Background
3.1 There is a considerable amount of general literature available on human

error in the public domain and many documents published in recent years
have been more focused towards aircraft maintenance in particular. This
reflects the increasing interest in engineering related incidents.
3.2 The available literature includes developed and mature aircraft

maintenance error taxonomies including the original outline of Boeing
Maintenance Error Decision Aid (MEDA Rankin et al, 2000), Human
Factors Analysis and Classification System Maintenance Extension
(HFACS ME Weigmann and Shappell, 1997) and the baseline ICAO
ADREP 2000 system. The basic philosophies behind these concepts have
been refined in the light of experience and for example, in the case of
MEDA, adapted to suit other aircraft maintenance applications such as
workshops. A number of consultancies have also developed the original
concept to suit a range of different scenarios and organisational
capabilities.
3.3 The US FAA has also remained quite active in looking at HF issues in
aircraft maintenance and their HF web-site,
https://www.faa.gov/about/initiatives/maintenance_hf/, continues to provide
a wealth of material that is worthy of scrutiny by those interested in the
subject.
3.4 The original CAA guidance on MEMS10 referred to the need for a
methodical and objective approach to error investigation, the adoption of a
just culture and the need to report and collate data across the industry to
get the best learning out of the analysis. This guidance was re-issued as
CAAIP Leaflet B-16011. The latest changes provide alignment against the
requirements for organisations to have an SMS and, following the learning
experience since MEMS was first launched, also better indicates what is
expected now within a functional MEMS process.
3.5 It is clear that many organisations have no formal error capture and
investigative mechanisms, even some 10 years after MEMS was initiated.
Whilst it is accepted that many such organisations are small and cannot
10
Originally Airworthiness Notice No. 71.
11
Part of CAP 562 - Civil Aviation Airworthiness Inspections and Procedures (CAAIP)

afford the cost of dedicated staff to the task, it is nave to think that errors
simply do not occur. It is equally important for small organisations to
recognise the safety issues associated with maintenance error as the
company may be carrying out work on safety critical components. There is
a need therefore to perhaps refocus attention among organisations to the
need for policy and procedures in this area.
3.6 Even where organisations have put in place an effective MEMS system the
extent to which detailed error investigations are conducted appears to
have decreased based on recent submissions to UK-MEMS. This suggests
that the industrys commitment to error investigation has waned. This may
be because the benefits that can accrue from the identification of root
cause and putting in place suitable corrective actions are not readily
quantifiable, or industry is becoming complacent.
3.7 However, the risks arising from maintenance error continue and, if left
unaddressed, will inevitably result in an aircraft incident and possibly the
worst case scenario, a fatal accident. It is essential that organisations work
proactively wherever possible to reduce or eliminate the potential for error.
As with any system that does not result in an immediate and identifiable
economic return the organisation has to believe that it is a valuable
exercise. There is some truth in the phrase, if you think safety is
expensive, try an accident.
The UK-MEMS programme
3.8 When the CAA introduced requirements in 2000 for MEMS industry
expressed concerns about the way in which the data may be misused,
particularly if it was possible to identify the organisation, or more
especially, the individuals concerned.
3.9 However, the CAA still saw benefits in the detail of the reports being
collated and analysed periodically so that trends and common themes
could be identified. This would help industry to capitalise their efforts and
target them on coherent solutions.
3.10 The CAA established an agreement with CHIRP for them to set up a
database to collate the data reported from the output of organisation
MEMS investigations. This did not do away with the need to submit MORs
in respect of incidents. It did, however, provide a continuing avenue for the
detail behind the investigations to be dealt with by CHIRP as an
independent body.
3.11 Despite subsequent attempts to roll the programme out across the UK
industry, the UK-MEMS programme has failed to achieve wider penetration

into the UK industry beyond the 23 pilot organisations that have largely
contributed from the outset of the initiative. There are some 460 UK
organisations that hold Part 145 approval and therefore greater scope for
reports to be made and analysed for the wider benefit of industry.
The UK-MEMS group
3.12 In order to support the UK-MEMS initiative within CHIRP, it was decided to
establish an industry group to manage the interface with the database
and the general principles of error management within the industry. This
activity of the group, UK-MEMS, was mainly focused upon the pilot
scheme members. The terms of reference were aimed at helping develop
policy relating to MEMS and engender better understanding of the issues
among the members. A close liaison between Deputy Director -
Engineering CHIRP12 and the UK-MEMS group was also achieved.
3.13 The membership of the UK MEMS group has subsequently been expanded
over the original members and now includes representatives from the
major UK airlines, Qinetiq, the Military Aviation Authority and of course the
CAA.
3.14 The discussions within the group have also expanded out from the original
focus on MEMS to include wider discussions on fatigue, human
performance issues and safety management systems.
3.15 The UK MEMS group does not represent all organisations that are required
to have a maintenance error management system. This is an issue that the
group has considered before, and are reviewing their constitution to widen
their remit.
3.16 It would also be helpful to consider the groups terms of reference vis--vis
being an expert group for MEMS to sit in collaboration with the UK CAA, as
a CAA/industry committee, and offer advice on such matters as may be
agreed. This will provide a more substantive basis for the group to act as
the interface between industry and the regulator on MEMS issues.
12
Deputy Director Engineering at CHIRP maintains the responsibility for Engineering related reports
to CHIRP but also the UK MEMS database and consequently sits on the UK-MEMS Group.

CAP 1367 Chapter 4: Methodology for the analysis
Chapter 4
Methodology for the analysis
Background
4.1 The MOR data supplied by the CAA was collated and put into a
spreadsheet. This ensured consistency of the data within the spreadsheet.
The data was then analysed individually to determine the nature of the
event, the underlying causes and any contributing factors.
4.2 The taxonomy used followed the general categories that had been
developed in the earlier papers. However, in order to focus on certain
particular issues the terminology does differ slightly. Comparisons can still
be drawn between the earlier papers and this review and have been
included where appropriate in this paper.
4.3 It is clear that the amount and quality of information reported to the CAA
varies from organisation to organisation. In many cases an MOR will
provide only superficial details of the event and the initial action taken to
resolve the issue. Longer term actions, such as the subsequent
amendment of procedures may not be reflected in the data filed against
the MOR. This means that the analysis is limited to the information held.
This may be something that needs to be addressed for future analysis.
4.4 It was also noted that the amount of information held under the UK-MEMS
system varied and, more worryingly, has become more superficial in the
last year or so. This may be simply the consequences of increased
pressures on organisational resources committed to MEMS investigations.
It may also reflect a shift in industrys perceived value of the MEMS
process since, to date, there has been little feedback to the industry on
information supplied under the scheme.
Taxonomy
4.5 It was also noted that the amount of information held under the UK-MEMS
system varied and, more worryingly, has become more superficial in the
last year or so. This may be simply the consequences of increased
pressures on organisational resources committed to MEMS investigations.
It may also reflect a shift in industrys perceived value of the MEMS
process since, to date, there has been little feedback to the industry on
information supplied under the scheme.

MOR number from CAA database

Error type (by analysis of the event)
Date of the event
Aircraft type (e.g. Boeing 757)
Operator (where available)
ATA chapter and sub-set (to help analysis by ATA system)
Manufacturer (e.g. Boeing, Airbus)
Primary error / key causal factor
Aircraft category (e.g. large aeroplane, large helicopter)
Pre-title (summary of the event)
Prcis (main content of the MOR report)
4.6 The data was entered into an Excel spreadsheet and set up so that the
data streams, e.g. aircraft type, could be sifted and the information
presented in different ways. This allowed maximum flexibility in carrying
out a partial or selective analysis on the data to identify any particular
themes of interest.
4.7 It was clear from the data held by CAA in its MOR database that there was
a repeat of the experience from the two previous analyses. The low level of
detail in the MOR maintenance occurrence reports determined the extent
to which the data could be analysed. Once again, the use of the MEDA
reporting and investigative tool would provide additional data that could be
beneficial to identifying a more comprehensive approach to identifying
safety interventions to respond to a maintenance error threat.
4.8 It is recognised however that the limited submission of data to UK-MEMS

does not necessarily complement the MOR data. For an analysis to be
comprehensive and reflective of the whole industry experience there needs
to be more data submitted.
4.9 In addition to the core data supplied by the CAA, further parameters were
added to the Excel database to identify the ATA Chapter, aircraft system
and the maintenance error type. The background experience of the Deputy
Director (Engineering) CHIRP ensured that some level of consistency of
approach and categorisation was achieved in looking at the data.
4.10 A number of occurrences were removed from the data set, as in the
previous analyses, as they were considered to fall outside the scope of the
study as it related specifically to maintenance error events.
4.11 One point of interest during the analysis was to see whether the
introduction of new European regulatory requirements in the form of Part M

and Part 14513 had any effect. This introduced some new aspects to
maintenance management procedures and the analysis was able to
capture some events which appeared to be triggered as a result of the new
requirements.
4.12 The previous reports, CAA Papers 2009/05 and 2007/04 had also identified
a number of high risk occurrences. These were not drawn out in the
CHIRP analysis of the data. Further work was however done to identify
some events where there had been a higher safety risk and some of these
have been included in a later section of the report to give an idea on how
maintenance error can manifest itself.
Baseline analysis results
4.13 The resulting data set for the period covered 2733 events and these were
analysed in accordance with the above taxonomy. The breakdown by
aircraft category is shown in figure 1.
Large A/C: 2399, 88%

Small A/C: 249, 9%
Large helicopter: 85, 3%
Figure 1: Overall breakdown of analysis
4.14 It can be seen that with 88% the largest proportion of events relate to large
aeroplanes. This is not surprising as the mandatory elements of the MOR
scheme apply to these aircraft rather than the smaller, more General
Aviation focused sectors.
4.15 Large helicopters, the bulk of which are operated in support of the North
Sea oil industry or for search and rescue purposes made up only 3% of the
data set. The total UK fleet of such helicopters is small in relation to the
equivalent fixed wing fleet. Despite this, the relatively low level of reported
events is surprising given the knowledge that there is a greater intensity of
maintenance activities on such aircraft.
13
Part M and PART145 are included as Annexes 1 and 2 respectively to Regulation (EC) 2042/2003 as amended.

4.16 Small aircraft reported incidents, both aeroplanes and helicopters,

contributed to only 9% of the total. It should be noted that it is not
mandatory to submit reports for many events related to small aircraft so
there are likely to be additional events that are not captured in the MOR
data held by the CAA. The focus of this review is therefore understandably
and predominantly related to large aircraft and those used for commercial
air transport.
4.17 The data set broke down into events related to Part M maintenance
management issues and maintenance error events. The analysis is shown
in figure 2. It can be seen that, whilst there were small variations in the
balance between the two classes of events the proportion was relatively
consistent across the aircraft categories. What is significant is the slightly
larger proportion of Part M maintenance management issues reported for
large helicopters. Since these relate mainly to maintenance overruns on
life limited components this is not surprising given the much larger number
of life controls and overhaul periods prescribed for such aircraft. This can
also be linked to errors introduced during the migration of data to new
computer systems.
100%
90% 365
39
80% 24
70%
60%
50% Part M
1913 Main. Error

40% 133
30% 33
20%
10%
0%
Large A/C Small A/C Large heli.
Figure 2: Breakdown of events, Part M vs maintenance error
4.18 The occurrences of maintenance error can be further analysed at a high

level to show a correlation with the CAA classification of the MOR when
received by the CAA. This is shown in Figure 3. The most significant
MORs, typically involving accidents, are classified as category A with the
lesser reports, usually closed on receipt by the CAA as category D.

2000
1800
1600
1400
1200 D
1000 C
B
800
A
600
400
200
0
Small A/C Large A/C Large Helicopter
Figure 3: Maintenance error-risk distribution, all categories 2005-2011
4.19 It can be seen from this that the bulk of the maintenance related MORs are
classified by the CAA as Category C. That does not mean they are not
significant and many require CAA investigation of the circumstances
surrounding the MOR.
4.20 There are two reasons for this. The first is to ensure that the operator or
maintenance organisation is reacting properly to the occurrence. This will
initially be to rectify the problem in order to return the aircraft to service.
The secondary function will be to further analyse the event and determine
the root cause in order to put in place corrective measures to prevent a
reoccurrence.
4.21 By comparison, the further high level analysis of the Part M related errors
yields the results shown in figure 4.

Large A/C: 365 (16% of total

errors in category)
Small A/C: 44 (20% of total
errors in category)
Large helicopter: 24 (35% of
total errors in category)
Figure 4: Distributions of Part M errors reported 2005-2011
4.22 Figure 4 shows that there were 433 reported Part M errors. Of these 303
(70%) related to overrun of overhaul lives, airworthiness directives or
maintenance inspections called out as part of the Approved Maintenance
Programme, of these approximately 40 reports indicated a non-compliance
with an Airworthiness Directive. This is often a reflection of poor
maintenance planning where due maintenance is not carried out and not
re-planned within the appropriate timescale. In some instances the overrun
is linked to a data entry issues. A more detailed review of the contributing
causes also suggests that there may be some evidence of a conflict
between the accomplishment of the work and commercial pressures to
continue operating the aircraft.
4.23 A further 130 (30%) reports highlight errors in setting up the data in the first
place. These errors are often simple transcription errors between the hard
copy paperwork following maintenance and the electronic database used
to manage the planning function. For example, inputting component life
data following their fitment to the aircraft and forecasting next due tasks.
Such errors can lead to an overrun if not detected by some other
intermediate means, such as unscheduled component replacement or a
detailed audit of the database.
4.24 The final baseline or high level category to pick up on is the number of
reports that are attributable to manufacture and overhaul, see Figure 3.

Overhaul (workshops): 79, 58%.

Small A/C - 18, large A/C - 57,
large heli. - 4
Manufacture: 56, 42%. Small
A/C - 9, large A/C - 38, large
heli. - 9
Figure 5: Distribution of events considered as manufacturer or overhaul errors, data 2005-2011
4.25 These are events which, following analysis of the underlying

circumstances, are issues that have their roots in an error at the time of
manufacture or which are attributed to an overhaul agency and not
therefore a direct error within the operators system. Both of these failures
can involve human error but the fact that they occurred in third party
organisations clouds the available data and the ability to mine the data and
find the underlying root causes.
4.26 Manufacturing error should be captured within the production and

inspection system. Any deviations should be the subject of design
concession but the assembly of complex components is no different from
the maintenance activity that ultimately can give rise to maintenance error,
e.g. mis-assembly, incorrect parts fitted etc.
4.27 It is logical that as much attention should be given to such errors but the
current requirements under Part 2114 do not require an error management
or capture system in the same way that Part 145 does. This is perhaps
something that EASA ought to consider for future amendments to Part 21.
4.28 The manufacturing and overhaul events together constitute 135 events or
approximately 4.9% of the total reported MORs. It is interesting to note that
not all of the manufacture events are down to a physical error. Many are
due to technical authorship errors in the approved data. The distribution
over the period from 2005 2011 is shown in figure 6.
14
Part 21 forms part of Regulation (EC) 16702/2003 as amended.

50
45
40
12
35
30
25 Overhaul
Manufacture
20
15 32
10
8 23
10 16
11
5 4 8 9
0 2 2 1
2005 2006 2007 2008 2009 2010 2011
Figure 6: Error reports attributable to manufacture or overhaul
4.29 It is clear that the inclusion of manufacture or production events in the data
suggests these activities should indeed also be considered in the context
of the application of an error management system. The potential for an
incident or accident is equally as significant as for maintenance activities.
4.30 The overhaul function, since it is another form of maintenance activity, is

already covered by the need to have a MEMS process under Part 145.
However, many organisations do not have the formal investigative tools,
such as MEDA, in place and therefore some additional focus may need to
be made in order to address maintenance error within those organisations.
These issues will be considered further by CAA in conjunction with EASA.
4.31 The number of maintenance related MORs received each year is shown in
figure 7. It can be seen that the overall reporting levels are fairly
consistent.
4.32 No explanation can be made for the apparent decrease in reporting in 2010
although it is recognised that the overall level of industry activity, in terms
of aircraft operation had decreased as a result of the global financial crisis.
The increase in reports for 2011 perhaps reflects increasing commercial
pressure and the resulting impact of manpower reductions increasing the
risk of an error being made.
4.33 The increasing trend noted in Part M reflects the introduction of the Part M
requirements in 2005 and the learning curve associated with the transition
to the new structure. Further detail on the typical error causes is given in
later Chapters.

50
100
150
200
250
300
0
7 - Lifting
50
100
150
200
250
300
350
400
450
500
4.34
10 - Parking
12 - Servicing
CAP 1367
20 - Std Practices
21 - Air Cond
January 2016
22 - Auto Flt
23 - Comms
24 - Electrical
25 - Cabin
MOR Total
26 - Fire Prot
27 - Flt Controls
28 - Fuel
29 - Hyd
30 - Ice/Rain
31 - Indication
32 - Gear
33 - Lights
34 - Nav
35 -Oxy
36 - Pnuematics
Pt 145
38 - Water
Figure 7: Maintenance MOR reporting levels 2005-2011

39 - Inst Panels
49 - APU
Figure 8: Global maintenance error events by ATA chapter

51 - Structures
52 - Doors
53 - Fuselage
Chapter, presented graphically in Figure 8.

54 - Nacelles
55 - Stabs
56 - Windows
57 - Wings
61 - Prop
Pt M
62 - Main Rotor
63 - Pwr Drive
Increasing trend
64 - Tail Rotor
65 - TRD
67 - RotorcrafT FC
71 - P/Plant
72 - Engine
73 - Fuel
74 - Ignition
2011
2010
2009
2008
2007
2006
2005
75 - Air
76 - Eng Contr
The high level analysis allowed identification of maintenance errors by ATA
77 - Eng Indi
Chapter 4: Methodology for the analysis
Page 24
78 - Exhaust
79 - Oil
80 - Starting
4.35 From this it can be seen that the main ATA Chapters that appear to attract
a maintenance error events are:
ATA 25 cabin / safety equipment 11%
ATA 32 landing gear / undercarriage 8%
ATA 27 flight controls 6%
ATA 35 oxygen 4%
ATA 52 doors 3%
ATA 79 engine oil 3%
4.36 That does not mean that errors do not occur in the other areas of
maintenance activity. Most ATA Chapters have seen some incidence of
error. It is interesting to note that Chapter 25 covers a variety of
maintenance tasks, ranging from seat installation and condition to the
installation of safety equipment. There is a significant difference in the
potential safety threat if an error is made fitting the wrong style of cushion
to incorrectly fitting an emergency escape slide to a door. This will be
discussed further in Section 5.
4.37 Figure 9 shows the change in reporting trends over the period analysed.
Although there is some variation year on year the overall figures for these
ATA Chapters is fairly consistent with the exception of ATA Chapter 25,
Cabin/Safety Equipment. There is no obvious explanation identifiable in
the MOR data for the apparent reduction in the number of incidents for
ATA 25. It can only be concluded that the incidents are not being reported
as they ought to be or tasks not read as frequently.

25
20
25 - Equipment
15
32 - Gears
27 - Flt Contr
10 35 - Oxygen
52 -Doors
0
2005 2006 2007 2008 2009 2010 2011
Figure 9: Reporting trends primary ATA chapters as % of total yearly maintenance error MOR
4.38 The more detailed analysis of the root cause of the events is also able to
be derived. This is shown in Figure 10 for the global data set.
4.39 The 100% figure represents 2108 MORs overall, 1890 for large aircraft,
174 for small aircraft and 44 for large helicopters.
100%
7% 11%
14%
90%
12%
5% 13%
80%
70%
28% 33%
60% 25% Poor maint. Practices
50% Servicing error

Use of app data
40%
Installation error
30%
44% 43% 43%
20%
10%
0%
Large A/C Small A/C Large Heli
Figure 10: Maintenance error distribution comparison, common categories 2005-2011
4.40 From these figures it can be seen that the most likely error type is incorrect
installation. The next highest is the use of approved data, or rather the

likely lack of it. Incorrect installation includes failure to fit all required
components (e.g. seals or spacers), incorrect routing of electrical cables
and incorrect applied torque. Use of approved data includes the proper
use of approved data such as the maintenance manual, service
information or repair drawings.
4.41 Again, Figure 11 shows the change in reporting trends for maintenance
error types over the period analysed. Although there is some variation year
on year the overall figures for these ATA Chapters is also fairly consistent
with the exception of Installation Error.
120
100
80 FOD
Misinterpreted Data
60 Poor Insp
Servicing
40 Approved Data
Installation
20
0
2005 2006 2007 2008 2009 2010 2011
Figure 11: Key maintenance error types as % of total each year, all aircraft categories
4.42 There is no obvious explanation identifiable in the MOR data for the
apparent reduction in the number of incidents for installation error other
than the possibility that internal inspection is catching the error before
release to service. This would mean that the events are still happening but,
despite being caught, are not being reported as required under the CAA
MOR scheme and the European Directive on Occurrence Reporting.

CAP 1367 Chapter 5: Detailed analysis for large aircraft
Chapter 5
Detailed analysis for large aircraft
Large aircraft global statistics
5.1 Following on from Section 4, further detailed analysis was carried out on
the large aircraft data. This allowed certain trends and themes to be
identified which clearly suggest some areas to be explored by companies
to address the root cause. The analysis was also carried out such that
some information as regards aircraft type could be collated and presented.
5.2 A mentioned previously, the analysis identified a significant discrepancy

between the number of maintenance event MORs reported to the CAA and
the submission of MEDA data to UK-MEMS. This is understandable given
the fact that not all Part 145 organisations are necessarily aware of UK-
MEMS or the need to contribute to the database. Figure 12 shows the
relevant results. It should be noted that the MOR data set covers the
period 2005 to 2011 whilst the MEDA data (UK-MEMS) shows a wider
period from 1998 to 2006. Whilst this is not wholly representative of a like
for like comparison it does suggest that the UK-MEMS system could have
produced much more information of value to the industry had it received
information on all of the events?
5.3 This represents the MOR total of 1890 as noted in 4.3.27 above. There
were 584 MEDA events.
5.4 Those 1890 MORs can be broken down by ATA Chapter and the results
are shown in Figure 13.

100
150
200
250
300
50
0
7 - Lifting
10 - Parking
total
12 - Servicing
0
5
10
15
20
25
30
35
40
45
50
20 - Std Practices
21 - Air Cond
CAP 1367
22 - Auto Flt Installation Error

23 - Comms
January 2016
24 - Electrical
25 - Cabin
26 - Fire Prot
App data not followed
27 - Flt Controls
28 - Fuel
29 - Hyd
Poor Insp stds
30 - Ice/Rain
31 - Indication
32 - Gear
33 - Lights Servicing error
34 - Nav
35 -Oxy
36 - Pnuematics
38 - Water Poor maint practices
39 - Inst Panels
49 - APU
51 - Structures
52 - Doors Misinterp of app data
53 - Fuselage
54 - Nacelles
55 - Stabs
FOD
56 - Windows
57 - Wings
61 - Prop
Figure 13: Breakdown of maintenance error MORs by ATA chapter, 2005-2011

62 - Main Rotor
63 - Pwr Drive
64 - Tail Rotor
65 - TRD
67 - RotorcrafT FC
71 - P/Plant
72 - Engine
73 - Fuel
74 - Ignition
CAA 05-11
75 - Air
MEDA 98-06
76 - Eng Contr
77 - Eng Indi
Figure 12: Comparison of CAA MOR and MEDA maintenance event analysis, large aircraft, shown as % of
Chapter 5: Detailed analysis for large aircraft
Page 29
78 - Exhaust
79 - Oil
80 - Starting
5.5 Once again, the top six ATA Chapters can be identified and the percentage
of MORs shown against the total maintenance error number. These are:
ATA 25 cabin / safety equipment 14%
ATA 32 landing gear / undercarriage 8%
ATA 27 flight controls 7%
ATA 35 oxygen 5%
ATA 52 doors 4%
ATA 79 engine oil 3%
5.6 This, not unsurprisingly, reflects the data presented in figure 8 and
paragraph 4.3.23 above. The figures for ATA 25 are slightly higher
reflecting the more complex and voluminous cabin configurations on large
aircraft. There is simply more to go wrong. Similarly ATA 35 and ATA 52
are up slightly.
5.7 A closer analysis of ATA 25 shows the following as typical maintenance

errors:
Cabin sidewall trim not fitted properly
Emergency door slide not connected
Emergency door slight transit pins in place
Passenger seats not in correct configuration
Passenger seats not connected to seat rail
Incorrect passenger oxygen masks found fitted
Life raft found time expired
P2 seat harness, incorrect shoulder assembly
Emergency path lights found covered in tape
Mountain bolts found missing from cabin attendants seat
Galley incorrectly installed
Seat belts incorrectly installed
Seat rows partially blocking emergency exit
Roll of masking tape left in oven causing smoke
5.8 For ATA 32 the following were typical of the error events:
Hydraulic brake pipes cross connected
MLG fairing failed due to incorrect rigging
Wrong main wheel bearing grease seal and retainer

Main wheel heat shield bolts missing

Main wheel fitted without spacer
Gear failure messages due to tacho not refitted
Brake pedal bellcrank assemblies incorrectly fitted
Incorrect and excessive grease caused brake fire
Gear failed to retract, pins still fitted
Incorrect tyre type fitted to wheel caused foul on retraction
Incorrect brake steering control unit fitted, steering failure
Incorrect brake steering control unit fitted, steering failure
Incorrect installation of the NLG door connecting rods, fouled
structure
MLG shimmy damper found disconnected with blanks fitted
5.9 For ATA 27 the following were typical of the error events:
Elevator feel pressure pipes damaged / disconnected
Ailerons heavy due to lost motion device not correctly rigged
Elevator manual rigging incorrect leading to uncontrollable dive
Incorrect installation of rudder limiter actuator
Loose screw caused elevators to jam (FOD)
Boeing 747 rudder trim unit fitted to Boeing 767
RH spoiler cable failed. LH cable outside of wear limits
Alternate flap switch safety locked with heavy gauge lock wire
Rudder trim actuator to indicator wiring crossed
Range of elevator movement reduced due to incorrect installation
Stabiliser trim hydraulic motor incorrectly assembled during overhaul
Flap roller failure due to incorrect shimming
Incorrect flap control unit fitted
5.10 In many of the cases reported in paragraphs 5.7 to 5.9 the number of
incorrect installation examples bears out the belief that this is the biggest
error threat. Whilst the ramifications of errors on flying control system are
more readily apparent to the flight crew, the same is not true of items such
as door slides being incorrectly installed. The consequences of a door
slide not operating correctly during an emergency evacuation are equally
of concern as the in-flight problems that errors can lead to.

Large aircraft maintenance error types
5.11 It is clear from the supporting data behind the reports that in many cases
the engineer concerned did not use the latest available manuals or
approved data to carry out the task. This highlights a cultural issue where,
due to the information perhaps not being readily available, the engineer
reverts to basic engineering skills. This does not, however, cater for critical
dimensions or tasks when carrying out rigging etc.
5.12 Figure 14 provides an illustration of the MOR maintenance error types for
the period 2005 2011. This tends to speak for itself.
Installation error - 834, 44%
Approved data not followed -

534, 28%
Servicing error - 222, 12%
Poor troubleshooting standards

- 14, 0.7%
Poor maintenance practices -
147, 9%
Poor inspection standards - 104,
5%
Misinterpritation of approved
data - 41, 2%
Figure 14: MOR maintenance error types 2005-2011
5.13 This clearly shows that installation error and not following approved data
collectively represent around 72% of the reported events. The failure to
use approved data is, of course, a key underlying causal factor in the case
of incorrect installation events.
5.14 Of the 834 events relating to installation error the following list represents a
broad analysis of the associated factors (a few random events are not
included):
Instruction non-adherence 325 events
Poor inspection 158
Wrong part fitted 96
Part not fitted 73

Wrong orientation 54
Cross connection 35
Poor insp. (independent) 33
Poor insp. / test 32
Panel detached in flight 13
Wrong location 10
5.15 This shows that there is a mixture of underlying factors that can contribute
to incorrect installation. In the analysis for this review, it is possible that
there were more than one factor present in the lead up to the error being
made, e.g. multiple elements. However, in order to simplify the
presentation the key underlying factor has been used.
5.16 Of the 534 incidents where approved data was not followed the following
list gives an indication of the key underlying factor that contributed to the
event:
Approved maintenance manual 181 events
Procedures 131
Minimum equipment list 119
Structural repair manual 49
AD / SB 27
Approved maintenance programme 9
Illustrated parts catalogue 6
Wiring diagram manual 6
5.17 This shows that there is some diversity of document that is used to refer to
when carrying out maintenance. It also highlights the different potential
failure paths during line and base maintenance. For example, the Minimum
Equipment List (MEL) is not used much during base maintenance but is
crucial on the line.
5.18 The number of events relating to servicing error totalled 222 incidents. This
represents some 12% of the total and the key ATA Chapters affected are
shown below:
ATA 79 engine oil 43 events
ATA 32 landing gear 23
ATA 35 oxygen 17

ATA 29 hydraulics 11
5.19 The vast majority of engine oil incidents related to overfilling of the engines
during servicing. Whilst this may appear to be innocuous it is known that
this can lead to fumes in the cabin and is a subject that has been of
concern over recent years. Other events include incorrect filters being
installed and magnetic chip detector plugs not being fitted (loose in engine
cowls).
5.20 Poor maintenance practices (147 or 9%) also revealed some indications of
the lack of attention paid by engineers whilst carrying out work or possibly
organisational issues such as distraction.
Foreign objects 78 events
Unrecorded work 14
Aircraft damage 10
5.21 It is surprising the number of foreign objects that are left on aircraft, despite
it being general practice to carry out a loose objects and cleanliness check
after all maintenance. The loose objects found include torches, spanners,
rags and individual bolts or bags of bolts. Where the tool is left adjacent to
a flying control system there is a real risk of a control restriction occurring
at some point so the actual safety threat can be significant. There should
be no reason for work going unrecorded as every engineer, whether
licensed or not has an obligation to record any disturbance or defect found.
Aircraft type analysis
5.22 Whilst the generic analysis of MOR data gives an indication of potential
areas where human error can occur in maintenance the specific design
characteristics of different aircraft can also be quite influential. For
example, flight control design is fundamentally the same but with the
introduction of fly-by-wire systems the supporting software infrastructure
can be quite different as far as operating philosophy goes. It is important
that any analysis also attempts to account for any type specific data that
may suggest an area to pay particular attention to.
Figure 15 provides an indication of the reporting trends as a percentage of

the total maintenance error MORs by aircraft type, for the main aircraft
types in UK service.

30
25
B737
20 B757
B747
15 B767
A319
10 A320
A321
5 EMB145
0
2005 2006 2007 2008 2009 2010 2011
Figure 15: Reporting trends as % of total M/E MOR by type
5.23 It is difficult to understand exactly why the trends spike as shown in figure
14. For example, the Boeing 737 fleet was reduced as the Airbus A319
type was introduced into service. The increasing MOR trend may be a
reflection on the lack of engineering commitment to a type that was being
run down. Likewise the increasing spike for the Boeing 747 in 2010
perhaps reflects the stand-down of some aircraft due to over-capacity or
alternatively a more intensive scheduling without ample time to carry out
maintenance. However, without looking at the specific details against a
particular operator these reasons are purely speculative.
5.24 What may provide a better indication of the prevalence of an aircraft type to
maintenance error is a look at the errors relating to key ATA Chapters for
specific aircraft types. Figure 16 shows the ATA Chapters for each type
where there are 6 errors or more reported. This information has again to
be considered in context. The data covers all aircraft of that type within the
UK register so a more numerous aircraft type such as the Boeing 737 will
give rise to larger numbers of error reports. However, what is of use is to
look at the ATA Chapters for each type to see the proportion of
maintenance errors shown for each ATA Chapter.

200
ATA 79
180
ATA 78
160
ATA 72
140 ATA 71
120 ATA 55
100 ATA 53
ATA 52
80
ATA 35
60
ATA 34
40
ATA 32
20 ATA 28
0 ATA 27
B737 B757 B747 B767 A319 A320 DHC8 B777 A340
Figure 16: Top 9 aircraft types / ATA (large aircraft category) 2005-2011, 6 errors or more identified in
each ATA chapter
5.25 For most aircraft types, the incidence of errors for ATA Chapter 25 is
noticeable. What is also noticeable is that the Boeing 737 appears to suffer
a lot of errors relating to flying controls. Analysis of the actual reports
shows that elevator, flap and configuration warning issues were of note.
The Boeing 757 suffers from errors related to Chapter 79, engine oil. This
reflects a known issue with overfilling of the engines. It has been identified
that a number of organisations were found not to have been following the
manufactures recommendations and had not being topping up the engine
oil correctly. The DHC-8 has seen its fair share of undercarriage issues.
5.26 As noted in paragraph 5.3.4 the raw data is of little value in determining the
relative risk unless it can be standardised. By taking into account aircraft
utilisation data a Maintenance Error rate per flight can be derived. The
resulting figures are shown in figure 17.

0.0014
0.0012
0.001
0.0008
0.0006
0.0004
0.0002
0
A340
DHC8
B767
B747
B777
B757
B737
EMB145
A320/A319
Figure 17: Maintenance MOR rate per flight by aircraft type
5.27 This gives a better, although not perfect, view of the relative likelihood of an
occurrence for various types. However, the information does not account
for the stage of the aircrafts useful life that the graphs represent.
Obviously for an aircraft that is at a mid design life point or nearing end of
useful design life, it will be more maintenance intensive. There will
therefore be a greater likelihood of a maintenance error taking place
unless suitable mitigation is put in place.
5.28 The depth of analysis also precludes identification of any social and
environmental issues that may affect the likelihood of error. Shift patterns
and cumulative fatigue may impact an engineers ability to make the
correct engineering judgements or open up the possibility of a lapse in
attention during critical inspections. Is the engineer really taking in what it
is he or she is seeing?
5.29 From an environment perspective, the organisations circumstances can

determine influencing factors.
5.30 Financial pressures could lead to a shortage of readily available spares, a

need to cut back on recruitment and training. These could compromise the
engineering system, if the risks are not identified and managed
appropriately.
5.31 Aircraft flight scheduling, whilst serving passengers interests in terms of

flight availability can often do nothing to balance out the needs of
engineering to ensure that ample time is given to address airworthiness
issues. A tight schedule often does not give any realistic time between

flights in order to address defects that may arise. The pressures of the
denied boarding compensation requirements simply exacerbate this
pressure to keep the aircraft in service. An intensive day time schedule
also limits the aircrafts availability for maintenance, pushing what can be
intensive engineering checks and inspections into the night time period
where there are different human performance issues to contend with.
5.32 Weather too is a factor not to be dismissed lightly. Whilst some

maintenance can be accomplished on the line environment there are still
times when the prevailing weather is such that logic dictates that some
form of refuge, e.g. hangar etc, is needed in order to avoid the potential
human performance issues with adverse weather.
5.33 To identify the implications of these factors a much more detailed

maintenance error and reporting culture must be fostered. The data
available from the current system simply is inadequate to give that level of
granularity in the analysis.
Type 2005 2006 2007 2008 2009 2010 2011
A319 157610 218469 260852 304849 333897 344435 384032
A320 106970 107015 104290 94239 110227 124035 141918
A340 9476 10643 13368 14220 13523 12959 12981
B 737 292309 259421 254560 230056 152750 119178 111785
B 747 39939 40344 44697 44083 40434 34191 35093
B 757 108357 107301 109916 97774 78387 70800 70160
B 767 32787 33919 36318 30307 27095 28192 28269
B 777 28130 28951 28768 28308 28635 30632 31757
DHC8 26164 30082 18248 13645 15966 13967 6705
EMB145 71327 74607 69838 53812 27449 22834 25201

Table 1: Annual utilisation (flight hours) by selected aircraft type
5.34 Figure 18 uses the cumulative fleet number of flights from 2005 2011 to
present the data. Table 1 show the actual yearly breakdown of flights for
the types. This allows a trend analysis to be developed and this is shown
in Figure 16.

0.0018
0.0016
B737
0.0014 B747
0.0012 B757
B767
0.001
B777
0.0008 A319
0.0006 A320
A340
0.0004
DHC8
0.0002 EMB145
0
2005 2006 2007 2008 2009 2010 2011
Figure 18: Maintenance MORs / flight for selected aircraft types
5.35 This is more useful as it indicates aircraft where a developing trend is

identified, or vice versa, an improving trend. This can often be reconciled
against a detailed understanding of the issue that the aircraft type is
experiencing. By a combination of the two, organisation can identify a
strategy to try and contain the likelihood of maintenance error on that
aircraft type, in relation to the particular ATA Chapter or actual component
where there are known issues. This strategy can then be fed into the
engineering and maintenance system by internal notices to engineers and
staff and through the medium of continuation training.

CAP 1367 Chapter 6: Detailed analysis for small aircraft
Chapter 6
Detailed analysis for small aircraft
Small aircraft global statistics
6.1 As for Section 5, some analysis was carried out on the small aircraft data.
There was more of an issue here regarding the availability of data as the
CAA MOR scheme is generally regarded as not being mandated for
General Aviation (GA) or small aircraft, other than turbine engines.
6.2 The analysis that has been carried out can therefore only be indicative of
what lies within the data set that was looked at. The lack of maintenance
error data and the absence of MEMS processes in many, if not all, GA
oriented organisations maintaining small aircraft means that many events
will go unreported. During the review of MOR data for another project 15, a
review of the Regulatory Approach to Recreational Aviation, it was
identified that inconsistency in the way the data was classified upon receipt
meant that it was difficult to extract accurately the data which may have
been reported that covered maintenance errors. However, without carrying
out a full review of each MOR the inconsistent classification of MORs
cannot be addressed. The following analysis is therefore not
comprehensive or conclusive.
6.3 Figure 19 shows the global data for the small aircraft category. A total of
174 events were analysed ranging across a range of fixed wing aircraft
such as the DHC-1 Chipmunk to BN-2 Islander and smaller rotary wing
types such as the Robinson R22 to the Sikorsky S76. Fixed wing was
responsible for 98 of these events (some 56%) with the remaining 76
events (44%) relating to rotary wing. Proportionally however there are far
more small fixed wing aircraft than rotary wing so it tends to reinforce the
belief that there are fewer reports on large helicopters than there ought to
be.
15
Some 22500 MORs relating to GA/Recreational Aircraft were reviewed as part of the RA2 programme .

Total errors 174
Rotary wing - 76, 44%

Fixed wing - 98, 56%
Figure 19: Global MOR data for maintenance error, small aircraft
6.4 Obviously small aircraft are not as complex as their larger counterparts.
Systems tend to be simpler in design with less redundancy and system
protection. This means that they can be more prone to the effects of
maintenance error.
6.5 Figure 20 shows the breakdown by ATA Chapter. It should be noted that
not all small aircraft manufacturers use the ATA Chapter system for
defining and coding maintenance activity. In the analysis of the data,
CHIRP assigned the relevant ATA codes to the report in order to provide
some consistency of analysis and to provide a comparison against large
aircraft if anyone was interested.

14
12
10
0
25 - Cabin
62 - Main Rotor
71 - P/Plant
21 - Air Cond
29 - Hyd
34 - Nav
79 - Oil
33 - Lights
61 - Prop
64 - Tail Rotor
74 - Ignition
30 - Ice/Rain
72 - Engine
28 - Fuel
55 - Stabs
73 - Fuel
75 - Air
12 - Servicing
24 - Electrical
27 - Flt Controls
32 - Gear
52 - Doors
56 - Windows
57 - Wings
63 - Pwr Drive
65 - TRD
67 - Rotorcraft FC
53 - Fuselage
Figure 20: Breakdown by ATA chapter, small aircraft
6.6 The 6 top ATA Chapters can be shown as a percentage of the total reports
as follows:
Chapter 32 Landing Gear 7% (13 events)
Chapter 27 Flying Controls 7% (13 events)
Chapter 62 Main Rotor 6% (11 events)
Chapter 63 Power Drive 5% (9 events)
Chapter 71 Powerplant 4% (7 events)
Chapter 24 Electrics 4% (7 events)
6.7 Problems and errors on landing gear or undercarriage systems (ATA

Chapter 32) amounted to some 7% of the reports. Looking at the source
data, many of these events were attributable to poor lubrication, incorrect
rigging, excessive wear in retraction mechanism and incorrect tyre fitting
causing it to foul on the structure during retraction.
6.8 Flying Controls (ATA Chapter 27) shows instances of control being
incorrectly installed (reverse sense) despite duplicate inspections having
been performed. The incorrect installation of bellcranks, cable routing and
improper use of tools during installation (leading to subsequent component

failure due damage) are also evident. There were also two incidents where
the flying control cables were excessively worn and two incidences of
loose object (rag and loose screw) where the controls were stiff to operate.
6.9 There were four events where the engine cowl partially detached in flight
as the cowl was not properly fastened.
6.10 On the electrical side (ATA Chapter 24) the issues found were incorrect
routing of wiring, improper installation of battery box cover causing a short
and unapproved parts used in repair/overhaul of starter motor.
Small aircraft maintenance error types
6.11 In a similar manner to the analysis of maintenance error types for large
aircraft, as shown in paragraph 5.2.2. and Figure 14, Figure 21 provides
that information for the small aircraft category.
Installation error - 75, 43%
Approved data not followed -

58, 33%
Poor insp. stds. - 7, 4%
Poor maint. practices - 24, 12%
Figure 21: MOR maintenance error types 2005-2011, small aircraft
6.12 Whilst, based on the limited data available, Installation Error was similar in
proportion to large aircraft, the percentage of Approved Data not Followed
(33% against 28%) and Poor Maintenance Practices (14% against 9%) for
small aircraft is higher than for large aircraft. This may well be a reflection
on the culture of small aircraft maintenance, the lack of organisational
focus on quality control and the absence of training available for new
entrants. However, without carrying out an individual re-assessment of
each MOR the actual reasons cannot be determined.

6.13 Once again a more detailed analysis of the maintenance error types can be
obtained through the analysis. This is shown in Figure 26. In this case, it
appears that the primary error type was approved data not being followed.
This accounted for 58 events out of the total of 174 occurrences or around
33%. These break down into the following causal factors related to non-
adherence:
Procedures 21 events
AMM 20
AD / SB 6
IPC 4
AMP 3
SRM 3
WDM 1
6.14 The lack of adherence to procedures and the maintenance manual (AMM)
are statistically the highest. However, many small organisations do not
have formal procedures as such. This may be a reflection that the MORs
received were submitted by organisations that are large enough to have
developed procedures to manage their maintenance activity.
6.15 Installation error is also a cause for concern, as was the case for large
aircraft. In this case it accounted for 75 events or 43% of the total (for large
aircraft the figure was 834 events or 44% of the large aircraft total). For
small aircraft this breaks down into the following causal factors:
Instruction non-adherence 18 events
Wrong part fitted 14
Poor inspection standards 12
Wrong orientation 10
Poor insp. stds. (duplicate) 9
Cross connection (duplicate 2) 5
Part not fitted 4
Panel detached 2
Wrong location 1
6.16 It is interesting to note the extent to which the wrong part has been fitted
and the incorrect orientation on fitting. However, poor inspection of the final
task and failure of the duplicate inspection process to detect errors are
dominant causes, as is the case for large aircraft.

6.17 With regard to poor maintenance practices, this accounted for 24 events or
some 14% of the total. The detailed data suggested the following were the
key elements:
General standards 13 events
FOD 10
Unrecorded work 1
6.18 The fact that general standards and foreign or loose objects being found
suggest that there is inadequate inspection and supervision within the
maintenance process. This may be attributable to the relatively low number
of licensed engineers involved in the process and the absence of final
inspection procedures.

CAP 1367 Chapter 7: Detailed analysis for large helicopters
Chapter 7
Detailed analysis for large helicopters
7.1 The number of maintenance related MORs for large helicopters was
substantially smaller than it was for large aircraft. This is in part a reflection
of the smaller size of the large helicopter fleet but there is a question as to
the number for reports on record, given the differing technologies and
more complex systems.
7.2 A total of 44 MORs were analysed and the results are presented in this
section. Once again, the analysis by CHIRP correlated the available
information from the UK MEMS database with the CAA MOR record.
7.3 The breakdown of the MORs by ATA Chapter is shown in figure 22.
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0
21 - Air Cond
25 - Cabin
62 - Main Rotor
71 - P/Plant
29 - Hyd
34 - Nav
79 - Oil
26 - Fire
64 - Tail Rotor
72 - Engine
51 - Structure
55 - Stabs
32 - Gear
52 - Doors
63 - Pwr Drive
65 - TRD
67 - Rotorcraft FC
53 - Fuselage
Figure 22: Breakdown of large helicopter maintenance MORs by ATA chapter
7.4 It can be seen that there are a few ATA Chapters that show a higher level
of maintenance error events. These include:
Chapter 52 Doors 9%
Chapter 62 Main Rotor 9%
Chapter 63 Power Drive Train 9%
7.5 At a secondary level of interest we can see issues within:

CAP 1367 Chapter 7: Detailed analysis for large helicopters
Chapter 29 Hydraulics
Chapter 32 Landing Gear
Chapter 64 Tail Rotor
7.6 The remaining incidents are spread over the other primary systems.
7.7 The numbers are not large but do suggest that the greater potential for
error is present where the systems are more complex.
7.8 A further more detailed breakdown of the maintenance error types is shown
in Figure 23.
Procedures 4
Installation error - 19, AMM 3
43%
Approved data not Flt. Manual 3
followed - 11, 25%
IPC 1
Poor inspection stds. -
2, 4%
Poor maintenance
practices - 6, 13%
Figure 23: Large helicopter MOR maintenance error types
7.9 The highest proportion of error as shown in Figure 21 relates to incorrect

installation. This accounted for some 43% or 19 of the 44 events. The
more detailed breakdown of these events was as follows:
Wrong part fitted 5
Instruction non-adherence 4
Wrong orientation 4
Part not fitted 2
Poor inspection standards 2
Panel detached 2
7.10 As can be seen the number of events is not high and therefore has no real
statistical significance. However the underlying causes still have some
similarity with the issues shown for large aircraft and the issues should not
be discounted. The low number of reports may simply be a lack of
reporting under the MOR scheme.

CAP 1367 Chapter 8: Examples of maintenance error
Chapter 8
Examples of maintenance error
8.1 The following examples are typical of reports received under the CAA MOR
system relating to maintenance errors. The MOR report is summarised in
each case along with the CAA closure recommendation. However, that is
only part of the issue. The company that experienced the event is required,
under Part 145.A.60 and their MEMS programme to carry out the
investigation and establish the root cause.
8.2 Having identified the problem and the root cause, the organisation will then
be expected under their Part 145 approval to put in place corrective
actions. This can be one or a combination of actions ranging from better
enforcement of supervision of tasks, additional inspections on complex
tasks to retraining of the staff.
8.3 In the examples shown, the MOR number and aircraft type are not shown
as, for the purposes of this report, the key issue is to encourage engineers
and organisations to think about the issue, what could have caused it and
consider whether they too are exposed to the risk of a similar occurrence.
8.4 It is recognised that any corrective actions that may be proposed will
possibly differ from organisation to organisation. The reason for this is that,
despite the aircraft maintenance manual requirements being the same,
different modification and equipment configuration standards can result in
differences. It is essential therefore that organisations look at how any
event, whether arising internally to their operation or relating to a potential
learning experience in another operator, is considered in the appropriate
context.
Example 1
8.5 Date 07/2011: ATA Chapter 25 - Equipment-Furnishings
Event description: When the aircraft was put in for a routine maintenance
inspection all four escape slide inflation bottles were found to have safety
pins still installed.
Implications: In the event of an evacuation being required, the slides

would not inflate and this would slow down the safe evacuation of
passengers and crew. It also increased the potential risk of injury if
passengers and crew were forced to jump from the doorway.

Investigation: It was clear that upon installation the pins had not been
removed. The organisation looked back into the records to see when the
slides were last removed and installed. From this it was possible to identify
the engineers involved in the task and the certifying engineer who cleared
the job. The event was clearly caused by a maintenance error, a lapse in
concentration. However, if the task was performed by an unlicensed
mechanic then there is a question over that individuals competence in
relation to the task and the need to ensure the appropriate maintenance
instructions were followed. There is also a clear issue about the
supervision and application of the relevant inspection standards by the
certifying engineer. Given the safety significance of the error the engineer
clearing the work should have ensured that the system was fully
operational. With the door covers installed the pins could not be seen.
Corrective actions: The human factors identified in this event can

possibly be attributed to organisational factors regarding staffing levels and
work processes and procedures not being adequately followed. The
mechanics performing the work failed to discharge their obligations to
following the AMM. The certifying engineer failed to ensure the system was
functional before release. Was this a case of blind stamping, certifying
without inspection in the belief that the mechanics would have done the job
properly? A fleet check carried out by the operator did not reveal any other
instances of incorrect installation. In order to try and prevent similar
occurrences in future the operator decided to instigate stage sheets for
multiple installation/removal tasks to allow engineers to document the task
stages.
Commentary: Whilst the job itself is relatively straightforward, the

importance of the final operational status should have been considered by
the engineers performing the work. Adopting a system for secondary
inspection would add another level of assurance that the system was
indeed fully operational on release to service. However, the issue of using
tasks cards that simply calls for slide replacement due, perhaps, to the
slide being time expired opens up a potential area for error, if the task card
is not broken down into the constituent sub-tasks. Ideally, there should be
scope for a clear inspection requirement and final clearance to refit the
cover. In this case, if such a requirement existed the certifying engineer did
not physically inspect the work before clearing the task.
8.6 Date 08/2010 ATA Chapter 78 Exhaust
Event description: During a walk around inspection the flight crew noticed
an anomaly in the bypass duct of the right hand engine. With the C duct
open damage was found to the heat shield, apparently caused by a
boroscope plug being missing.

Implications: The missing plug meant that hot engine gases were not
contained as designed and had the potential to escape and cause heat
damage to components on the engine. Up to the time the issue was noted,
the engines overheat warning system and/or fire warning system did not
activate.
Investigation: The boroscope plug was missing and therefore allowed

gases to escape. Further engineering investigation revealed that the heat
shield was actually missing. Hot gases had impinged upon the thrust
reverser structure causing some heat damage and distortion. Investigation
found that the aircraft had been subject to a boroscope inspection some
three days earlier by a contracted maintenance organisation. The
maintenance organisation carried out a MEDA investigation and found
that, due to other commitments, the primary contracted organisation could
not accomplish the work. The task was therefore contracted by the
operator to another boroscope inspection organisation. This led to some
management issues with regard to check planning and call out of the work.
Planned access to the proposed hangar was not possible due to another
aircraft check overrunning so gaining access to an alternate hangar
resulted in the check commencement being delayed, significantly reducing
the available period for the task to be completed. This also meant that the
usual facilities and support were not available. The work pack, as supplied
by the operator was incomplete and this required the check supervisor to
spend additional time sorting out the issue. Staged inspection sheets were
not provided and as the plug removal and fitment was carried out by two
different mechanics the opportunity to ensure that all disturbed plugs were
refitted was compromised.
Corrective actions: There were several actions proposed following

investigation into this event. These included improvement in check
planning by the operator, the interface with the contracted maintenance
organisation, management and operator/third party interface issues, the
need for stage inspection sheets and improved supervision of third party
work.
Commentary: There are a number of issues clearly identified in the MEDA

investigation and therefore no single root cause. The absence of stage
inspection sheets meant that there was a lack of clarity about what plugs
had been disturbed. The change of location and the consequent delay in
the check commencement brought about pressure to complete the
required work in a shorter allotted time. This, the use of an independent
third party to do part of the work (the boroscope) and the incomplete
worksheets meant that the check supervisor was distracted and not able to
perform as expected.

Example 3
8.7 Date 09/2011 ATA Chapter 32 Landing Gear
Event description: The LH nose wheel tyre was found to be worn to

limits. On jacking the aircraft to replace the wheel the wheel assembly was
found to be excessively loose with an axial and radial movement of about
half an inch.
Implications: The excessive movement appears to have potentially

increased the wear rate of the tyre. However, there was clear potential for
the movement to change the loading of the wheel and bearing assemblies
and this could have led to bearing failure. In similar events previously
recorded this has led to damage to the axle and in the extreme case the
wheel departing the aircraft.
Investigation: On investigation the wheel outer bearing retention circlip

was found to be missing and the wheel nut on removal was found to be
less than hand tight. This suggested that there was an issue regarding the
application of the correct torque when the wheel was fitted. It was not
possible to determine whether the circlip was fitted at the last tyre and/or
bearing replacement.
Corrective Actions: The organisation was able to check back through the
records to the last wheel change. However, it was not possible to prove
conclusively that the wheel had not been fitted properly. Likewise, the
organisation was able to trace back to the last wheel/bearing change but
again there was no conclusive evidence to suggest that the work was not
done properly in the workshop. The only possible action was to highlight
the event to the engineers.
Commentary: This example shows the need to ensure as the end user
that the part to be fitted is indeed serviceable. Checks to ensure that it has
not suffered any transit damage is the most obvious issue to be
considered. However, in many cases wheels are supplied as ready to fit
assemblies and do not require the bearings to be swapped over between
wheels. Although the engineer fitting the wheel should have been able to
rely upon the workshop having done the work properly the final
responsibility for ensuring that the wheel was indeed in the correct
configuration for fitment lay with the engineer installing it on the aircraft.
This shows the importance of having the competence to determine that the
bit to be fitted is what it should be.
Example 4
8.8 Date 11/2011 ATA Chapter 73 Fuel Control

Event description: Following an FCU replacement, on starting during the

engine post-task function checks the right hand engine did not control and
continued to accelerate past normal idle. Engine shut down to prevent
overspeed.
Investigation: The maintenance organisation had changed the FCU on

both left and right hand engines on the night shift. During the following day
shift the organisation carried out the post-task functions checks. The left
engine started and operated normally. The right hand engine continued to
accelerate beyond idle and, showing no signs of stopping, was shut down
when around 80% was achieved. The rigging of the FC was checked and
found satisfactory. A second start attempt was made with the same fault
present. The FCU was removed and inspected whereupon a small plastic
drive shaft was found to be missing. The driveshaft was found still attached
to the old FCU and was transferred. Subsequent engine run carried out
without problem.
Corrective action: The fact that the ground run was carried out and found
the defect is some mitigation against the failure to transfer the driveshaft
during replacement of the FCU. The AMM procedure was clear about the
shaft being in place. However, a simple comparison between the unit being
removed would also have highlighted the anomaly. The issue was followed
up with the relevant staff.
Commentary: This example shows the need for vigilance when carrying
out replacements. Following the AMM procedure will help reduce the
potential for such errors but the mechanic/engineer involved should still be
aware of the need to verify that the relevant seals, shafts and attaching
hardware are available and serviceable. It shows the value in carrying out
the ground runs to verify correct engine operation. In this case, the
engineer was able to shut the engine down without damage.
Example 5
8.9 Date 10/2011 ATA Chapter 27 Flying controls
Event description: After departure a control restriction was felt during a

turn to the right to the required heading. Increased force in the control input
achieved the required turn. Further incidences of control restriction noted
during the next few turns but the aircraft remained controllable. Aircraft
returned.
Implications: There was clear scope for the control restriction to become
worse and affect the controllability of the aircraft.

Investigation: The investigation revealed that the aircraft had a history of

autopilot trimming issues for about a month. A deferred defect had been
raised noting the autopilot as being unserviceable. The post incident
analysis suggested there may be an issue with the aileron servo actuators.
Further analysis of the previous history showed that the previous reports in
the technical log implied that there were similar control restrictions, not
autopilot problems. On this occasion, the restriction was evident on the
ground with the autopilot disengaged. The aileron PCU was changed. The
autopilot roll servo was replaced as a precaution. In addition, the aileron
hinge bearings were replaced. Restriction was no longer evident. No
subsequent reports.
Corrective action: The defect was analysed and components replaced to

re-establish the systems serviceability. Follow up action on aileron PCU
strip revealed no defects. Roll servo actuator strip report likewise showed
no defects. Manufacturer suggested the there was a possibility of ice build
up on a control cable adjacent to a detached recirculation fan. The area
was inspected but no defect with the recirculation fan could be seen. The
operators reliability monitoring system will continue to review the system
operation.
Commentary: Troubleshooting in this case failed to identify a root cause.

The absence of reported defects following receipt of the component strip
reports does not help. The report suggests that the fault, which is
suspected as having been present for some time, may have not been
properly reported in the technical log. This could mean that the opportunity
to resolve the issue at an earlier date may have been missed. This
highlights the importance of clear reporting in the aircrafts technical log of
the symptoms seen when pilots find a defect.
Example 6
8.10 Date 10/2011 ATA Chapter 29 Hydraulic systems
Event description: On selecting flap 1, left hydraulic system pressure

warning illuminated followed by left hand quantity warning. Crew declared
PAN call and following some further checks made an approach and landed
safely.
Implications: Loss of hydraulic fluid, loss of pressure and potential

compromise of certain system operation.
Investigation: On investigation after landing, a split in a hydraulic flexible

hose was found in the left hand main landing gear wheel well. This pipe
ran from the left system filter/case drain module and the relief valve
assembly. It was noted that the hose was poorly supported with a single P

clip and was free to move. A temporary repair was carried out using
flexible hoses and the left system replenished. An ADD was raised to
include re-inspection at each daily check.
Corrective action: Noting the temporary repair carried out the aircraft was
returned to service. The MOR record notes that the operators quality
assurance department appears to have authorised for replacement of the
pipe not to exceed the next A check. Further investigation into the
replacement pipe showed that a rigid pipe should have been installed.
Commentary: Although the temporary repair used flexible pipes as a short

term measure in accordance with a provision in the AMM it seems that the
flexible pipe that failed had been installed for some time. With the
discovery that there should have been a rigid pipe in that position there is a
question as to how the flexible came to be installed in the first place and
why the aircraft records did not subsequently require replacement by the
proper rigid pipe.
Example 7
8.11 Date 01/2010 Chapter 28 Airframe Fuel
Event Description: The aircraft arrived with approximately 16 Tons of fuel

in the number 2 main tank. The cockpit indication showed a hard past
zero empty indication.
Implications: Suspected indication problem giving erratic/incorrect

readings.
Investigation: The initial checks carried out on the wiring confirmed there
was an open circuit in the tank harness. When entry was gained to the
tank to make repairs it became evident that work had been carried out on
the harness. A number of defects were found including incorrect routing of
the wiring resulting in a number of taut cables. The harness was only
partially secured and in addition was really too short placing undue strain
on the wiring and connections. A number of wires, including the fly leads
had been cut but remained attached and there was some wiring that,
despite being redundant had been left in the tank. Following these
discoveries the rest of the aircraft was inspected with a number of further
defects being found.
Corrective Actions: Work was carried out to rectify the defective wiring
and the fuel tank harness was replaced. Detailed review of the aircraft
records failed to reveal where the fuel tank wiring was carried out. It was
not therefore possible to establish the root cause beyond the clear

application of incorrect working practices and failure to follow approved

data and repair data.
Commentary: Given the more recent focus on fuel tank wiring and EWIS
requirements the defects found in this instance are a cause for concern. It
is clear that the work had been done by an approved maintenance
organisation but, as it was not possible to establish when the work had
been done and the aircraft had transferred between registers, it could not
be determined what approval had been involved. However, given that the
manufacturers continuing airworthiness requirements, approved
maintenance manual and the wiring diagram manuals applying equally
under any of the approvals there was no cause for these deficiencies to
have been present. This demonstrates the need to ensure that work
carried out has been done properly, which may mean supervising the
approved maintenance organisation working with an operators
engineering staff. It also shows the need to ensure that maintenance is up
to scratch when an aircraft is brought onto the UK register.
Example 9
8.12 Date 04/2010 ATA Chapter 28 Airframe Fuel
Event Description: A fuel leak was evident from the right hand wing surge
tank panel.
Implications: Fuel leak with the attendant loss of fuel and fire risk.
Investigation: On removal of the access panel for re-sealing, an incorrect

panel was found fitted to in the right hand main fuel tank. The access panel
houses the surge tank over pressure relief valve. With it being fitted in the
wrong position, the surge tank relief valve ended up being fitted in the right
main tank rather than the surge tank. This altered the design of the fuel
vent and surge system with the attendant implications on system
operation. The panels had been transposed during a maintenance check
inspection five months previously. The aircraft had suffered a fuel leak
some two months later. This was rectified by a re-torque of the attachment
screws on the leaking panel. The next day saw a further fuel leak and air
turn back. This was put down to a surge valve problem and the pressure
relief valve was reset.
The aircraft pressure relief valve was replaced the following day and the
surge tank inlet scoop inspected. The float valve was also replaced.
Although the aircraft was then operated for a couple of months a further
fuel leak occurred. This led to the discovery of the incorrectly located
panels.

Corrective Action: The panels were relocated to their correct positions

and fuel system checks carried out. Two investigations were completed by
the airlines safety department and the engineering organisation. A number
of recommendations were made. Documentation was changed including
updating of the AMM by the manufacturer.
Commentary: It is interesting to note that the problem was originally the

result of inadequate control of fuel tank closure and system reinstatement
following the maintenance check. When the original fuel leak was reported
the engineers looked to simply re-torque the panel screws. There was no
consideration of whether the panels were correctly positioned, based most
likely that they had been fitted correctly at the time. It is of concern that
further work was then carried out following the next fuel leak and air turn
back where there was a question over the operation of the pressure relief
valve and the surge tank inlet. This should have, but failed to, detect the
incorrect positioning of the panels. It was not until a further leak some two
months later that the anomaly was discovered.
This event shows the importance of staged inspections and a thorough

check of panel installation following maintenance. In this case, it is
apparent that the leak did not really happen until the fuel system was filled
to a certain level. Given the variable nature of the aircraft operation the
appearance of the defects, e.g. a leak, may not happen until sometime
after the work was done. A similar situation was noted by AAIB in the
investigation of a Boeing 777 centre tank access panel being left
uninstalled following maintenance. This resulted in a heavy fuel leak
following take off. The criticality of panels being installed correctly is also
highlighted in the AAIB report of a Boeing 757 where the flap panels were
left off, leading to lateral control difficulties in flight.

CAP 1367 Chapter 9: Summary
Chapter 9
Summary
9.1 The information above from the analysis of the CAA maintenance error
related MORs gives a clear indication of the potential areas to be
considered. This is useful information for any engineering organisation or
operator wishing to look at potential safety risks and develop safety
strategies to address them.
9.2 The examples given are a snapshot of some MORs giving additional detail
and an indication of what the investigation revealed. It is clear from the
examples that the MOR system does not always have much information
relating to the actions taken or what the company may have done to try
and prevent a reoccurrence.
9.3 Where applicable, under an organisations SMS there is an inherent

obligation to identify risks and put appropriate mitigation in place. More
than that, there is a need to become a learning organisation, looking at the
risks within other organisations that may have been identified in accident
investigations or through reports made under the CAA MOR scheme. An
organisation that looks simply at its internal issues leaves itself open to the
hazards that others have already fallen foul off.
9.4 There are some fundamental issues that need to be borne in mind. The
operator has an obligation to manage the airworthiness and maintenance
requirements of its aircraft fleets. Whilst Part M covers this in part there
may be a need to exercise some form of oversight of the contracted Part
145 organisation that actually performs the maintenance to ensure that
what is expected is what is delivered, in terms of both content and quality.
9.5 The Part 145 organisation has an obligation to ensure the competence
(knowledge, experience, behaviours and attitudes) of the staff performing
the work, whether licensed or not. In addition, it is important that there is
an appropriate balance of supervisory and inspection staff to ensure that
the work is progressed in controlled manner and that the required quality of
work is achieved.
9.6 The analysis shows there is scope for concern about how often
maintenance staff fail to follow the correct procedures, maintenance
manual processes such that incorrect installation or incomplete installation
is the undesirable consequence. This can impact aviation safety as well as
damaging an organisations reputation.

CAP 1367 Chapter 9: Summary
9.7 There is a need for maintenance error management systems (MEMS), not
least because Part 145 requires it, which allows the organisation to look at
continuous improvement but if the organisation pays only lip service to it
the organisation could well be guilty of corporate failings. In the event of an
accident this could compromise any defence the organisation may wish to
put forward. Corporate manslaughter and corporate killing are real
business risks if an organisation fails to meet its obligations16.
9.8 It is hoped that the information in this CAA paper will help organisation to
develop a strategy to minimise the likelihood of any maintenance error and
the resulting compromise of the operators flight safety programme.
9.9 It became clear during the Offshore review that there is still a need to
decrease the number of maintenance errors and also improve the
maintenance standard within organisations. The review of the MOR data
and events showed that errors are still occurring at a concerning rate. It is
therefore clear that a programme of initiatives is needed to reduce the
likelihood of significant events occurring in the future. To address this
issue CAP 1145 Action 31 states:
The CAA will form an Offshore Maintenance Standards Improvement

Team with the offshore helicopter operators with the objective of reviewing
the findings at Annex F to the CAA Strategic Review of the Safety of
Offshore Helicopter Operations and making proposals to achieve a step
change in maintenance standards.
9.10 During the establishment of the Action 31 Improvement Teams, it was clear
that these issues were not specific to the Helicopter maintainers and
management organisations and were equally applicable to the fixed wing
organisations. This review will be published as a separate report.
16
Corporate Manslaughter and Corporate Homicide Act 2007

CAP 1367 Chapter 10: Conclusions
Chapter 10
Conclusions
It is clear from the report and from a number of AAIB reports, during the same
period, that maintenance Human Performance is still a casual/contributory factor, in
a significant number of events. Therefore the following actions/recommendations
are made:
Actions to the CAA
1. The CAA should continue to carry out a periodic review of occurrences to

identify the maintenance error event trends, to follow on from this 2005 to
2011 report.
2. The CAA should work with operators to provide guidance on how to

identify best practice to identify and reduce the likelihood of errors
occurring and the impact on aircraft safety.
3. The CAA should develop a method of ensuring that maintenance staff are
fully aware of the their responsibilities in the following areas:
Correctly recording and signing off work;
Identifying and carrying out safety critical tasks or independent /
duplicate inspections;
The importance of following procedures, maintenance instructions,
reporting and investigating errors;
Improving tool and debris control.
Recommendations to operators, PART M and PART 145

organisations
Organisations should ensure that all of their staff should be made aware of and
discus relevant areas of this report during their continuation training or briefing
sessions that focus on the above actions. They should review their procedures,
working practices and highlight any occurrences where contributory factors have
resulted in an installation error.

CAP 1367 Template W Charts

Uploaded by

Copyright:

Available Formats

CAP 1367 Template W Charts

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CAP 1367 Template W Charts

Uploaded by

Copyright:

Available Formats

Aircraft Maintenance Incident Analysis

Published by the Civil Aviation Authority, 2015

Civil Aviation Authority,

The latest version of this document is available in electronic format at www.caa.co.uk.

Enquiries regarding the content of this publication should be addressed to:

January 2016 Page 1

Executive summary .................................................................................................... 6

The analysis of incident data .................................................................................... 11

The UK-MEMS programme ............................................................................... 14

The UK-MEMS group ........................................................................................ 15

Methodology for the analysis .................................................................................... 16

Baseline analysis results ................................................................................... 18

Detailed analysis for large aircraft ............................................................................ 28

Large aircraft global statistics ............................................................................ 28

Large aircraft maintenance error types.............................................................. 32

Aircraft type analysis ......................................................................................... 34

Detailed analysis for small aircraft ............................................................................ 40

January 2016 Page 2

Small aircraft global statistics ............................................................................ 40

Small aircraft maintenance error types .............................................................. 43

Detailed analysis for large helicopters ...................................................................... 46

Examples of maintenance error ............................................................................... 48

Actions to the CAA ............................................................................................ 59

Recommendations to operators, PART M and PART 145 organisations .......... 59

January 2016 Page 3

3. Further information on Maintenance Error Management Systems (MEMS)

4. A limited number of Part 145 organisations (23) have contributed the

5. In conjunction with the CAAs high level HF Review, which is focused on

January 2016 Page 4

6. Once again, this paper seeks to provide information on the common

January 2016 Page 5

1. Maintenance Error continues to be one of the most obvious safety threats

2. However, despite those improvements the system of performing

3. Training is essential as engineering skills and good practice does not

4. The investigation of maintenance events is a valuable tool in the armoury

5. Human error is influenced by the individuals personal circumstances,

January 2016 Page 6

7. That does not mean that supervision is un-necessary. It is the overall

8. Senior management have an important role to create the right environment

January 2016 Page 7

January 2016 Page 8

1.7 These investigations were part of the process assigned to the

1.8 The MEMS requirements are outlined in guidance contained in CAAIP

January 2016 Page 9

January 2016 Page 10

The analysis of incident data

January 2016 Page 11

the CAA is understood there is a confidential route available under the

January 2016 Page 12

3.1 There is a considerable amount of general literature available on human

3.2 The available literature includes developed and mature aircraft

January 2016 Page 13

The UK-MEMS programme

January 2016 Page 14

The UK-MEMS group

January 2016 Page 15

Methodology for the analysis

January 2016 Page 16