Available online at www.sciencedirect.

com

ScienceDirect
Procedia Engineering 80 (2014) 293 302

3rd International Symposium on Aircraft Airworthiness, ISAA 2013

Individual-related factors and Management-related factors

in Aviation Maintenance
Mengyao Baoa,b, *, Shuiting Dinga,
a

Airworthiness Technologies Research Center, BeiHang University, Beijing 100191, China,

Department of Aircraft Airworthiness Certification,Civil Aviation Management Institute of China, Beijing 100102, China

Abstract
Aviation safety relies heavily on maintenance, yet improper maintenance contributes to a significant
proportion of safety threats that cause aviation accidents or incidents. In order to fully understand the
maintenance errors and contributing factors, this paper investigated a sample of 3,783 Aviation Safety
Reporting System (ASRS) incident reports submitted in the period of January 1st 2008 to December 31st 2008,
which are formed by maintenance personnel or non-maintenance personnel. The Maintenance Error Decision
Aid (MEDA) and Correspondence Analysis (CA) methods are used to analysis maintenance error, its
contributing factors and the relationship between them. The result shows both maintenance personnel and nonmaintenance personnel considered human error accounts for a large proportion of maintenance error, and
individual-related factors and management-related factors are the most frequent reasons for maintenance error.
The outcomes also indicate that non-maintenance perspective should not be ignored because it can provide
abundant information which is not included in maintenance personnel reports and helpful to reduce error.
Meanwhile, the management-related factors are as crucial as individual-related factors, which should be paid
more attention.
2014 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
2013 Published by Elsevier Ltd. Selection and peer-review under responsibility of
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
ENAC and peer-review under responsibility of Airworthiness Technologies Research Center, Beihang University/NLAA.
Selection
Keywords: aviation safety; maintenance error; individual-related factor; management-related factor; ASRS

*Doctor, Department of Aircraft Airworthiness Certification,Civil Aviation Management Institute of China,HuaJiaDi East
Road No. 3, ChaoYang District; bomeya@gmail.com.
Professor, National Key Laboratory of Science and Technology on Aero-Engine & Aero-Thermodynamics, School of Jet
Propulsion, XueYuan Road No. 37, HaiDian Distric; dst@buaa.edu.cn (Corresponding Author).

1877-7058 2014 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Selection and peer-review under responsibility of Airworthiness Technologies Research Center, Beihang University/NLAA.
doi:10.1016/j.proeng.2014.09.088

294

Mengyao Bao and Shuiting Ding / Procedia Engineering 80 (2014) 293 302

1.

Introduction

Aviation safety relies heavily on maintenance. Without maintenance, systems will deteriorate
and performance will decline. In aircraft, maintenance is crucial to ensure continued safe operation.
In previous research[1], maintenance was found to be related to 4.1% of all aviation accidents and
28.6% of maintenance-related fatalities from 1999 to 2008, indicating that it has a slightly higher
fatality rate than accidents overall. Unfortunately, although all aviation accidents have fallen into
the response to aircraft technology, and airline traffic safety have been generally improved over
time, the number of maintenance-related accidents did not decreased commensurately in past five
decades.
Many previous research have investigated the causes of maintenance-related accidents or
incidents. In general, studies on maintenance-related factors can be divided in to two groups,
examining either:(1) Individual-related factors as the reason to the maintenance error, or (2)
management-related factors as the reason to the maintenance error. The first group includes studies
that have concentrated on the direct-maintenance safety, most of studies focusing on maintenance
error discuss maintainer performance as a cause of the accident or incident. Reason and Maddox[2]
indicated more attempts to improve aviation safety have focused on reducing inspector and repair
personnel error, not only maintenance operation but also technical knowledge and skills. Hobbs and
Williamson[3] and Rankin and Sogg[4] supported this viewpoint by discovering that individualrelated factors such as pressure and fatigue contribute a significant proportion to the incomplete
installation and uninstalled parts maintenance error.
The second group of studies on maintenance-related accidents or incidents examines
management-related factors as the reason to the maintenance error. Management-related factors
have been an important component of maintenance to increase the safety reliability of aviation
system since it became a major theme in the 1995 National Aviation Safety conference[5]. Another
Hobbs paper[6] presented that maintenance error was not only because of individual failures, but
reflected the interaction of personnel, workplace and organizational factors. There are several
evidences to prove the importance of management-related factors. In a study of the normal job
performance of 286 aircraft mechanics, McDonald et al.[7] found 34% mechanics admitted the
performing of their recent task had contravened the formal procedures and organizational workflow
to some extent. Similarly, Taylor[8] indicated organizational characteristics (e.g., definition of
work groups, operation of pressures, and issues of trust and authority) could influence performance
at the individual level, and further affected patterns of work in aviation maintenance operations.
Later, Taylor[9] and Rogan[10] both emphasized human error in a major airline carrier's
maintenance facility is also influenced by characteristics such as planning, scheduling and
communications. A detailed analysis of 46 ASRS reports by Patankar et al.[11] also revealed a
variety of factors contributing to procedural errors in maintenance, especially document
deficiencies and user errors.
To conclude, above studies demonstrated that aviation maintenance errors are not solely due to
maintenance personnel individual mistakes, but also induced by inadequate management,
leadership/supervision or communication from across departments, which finally affect the quality
of maintenance. Focusing back on the identification of maintenance error promoting factors, the
Maintenance Error Decision Aid (MEDA) is a major process to investigate factors that contribute
to various maintenance-related accidents and incidents[12]. Although the MEDA has been widely
accepted by maintainers as a means that promoted such safety occurrences in aviation industry by
, its information is
not widely available to the research community. The incident databases maintained by the NASA
Aviation Safety Reporting System (ASRS) can be a useful source of information on the nature and

Mengyao Bao and Shuiting Ding / Procedia Engineering 80 (2014) 293 302

origins of maintenance errors. (e.g. Hobbs and Williamson, 2003[3]; Rankin and Sogg, 2003[4];
Schmidt, Schmorrow, and Hardee, 1998[13]). The previous analysis of maintenance error and
. However, the
perspective of non-maintenance staff (e.g. Captain, Pilot, and Dispatcher etc.) cannot be
overviewed.
2.

Scope and aim of this study

Our studies are built on Marais and Robichaud previous work[1]and on that of Hobbs[6]. In this
research, we collect 3783 ASRS incident reports which were submitted by both maintenance
personnel and non-maintenance personnel. The MEDA approach is launched to analyze errors and
contributing factors, while the Correspondence Analysis (CA) is applied to analyze relationship
between error and contributing factors.
ctors,
aviation safety agencies, and manufacturers. We define maintenance and non-maintenance
ples. In both reference groups, if
the occurrence of a certain incident is caused by at least one maintenance-related error or
contributing factor, this incident is considered as a maintenance-related incident.
Besides, we define 5 types of error - Personal Injury, Inspection, Installation, Repair and
Servicing as human error because they are direct human related problems. The rest 3 types of
error - Foreign Object Damage (FOD), Airplane/Equipment Damage and Others are defined as
non-human error . In terms of contributing factors, the D)Job/Task, E)Technical Knowledge/Skills
and F) Individual Factors are defined as individual-related factor , because these factors refer to
human characteristics and behaviors. The H) Organizational Factors, I) Leadership/Supervision and
J) Communication are defined as management-related factors , since they are not direct problems
on maintenance performance but the indirect problems existing in maintenance chains. Because this
paper mainly focuses on individual-related factors and management-related factors, the rest
contributing factors - A) Information, B) Equipment/Safety Equipment, C) Aircraft
Design/Configurate/Parts and G) Environment/Facility are roughly defined as other factors.
This paper makes two primary contributions. First, it demonstrates the differences in how
maintenance personnel and non-maintenance personnel report errors and contributing factors to
maintenance-related incidents . Second, through discussions of errors, contributing factors and their
relationships, it explains the likelihood of factors to trigger maintenance-related incidents according
to the two reference groups.
3.

Data and Method

3.1 Data
3.1.1

Data Source

The Aviation Safety Reporting System (ASRS), which is a voluntary, confidential and nonpunitive system, is the largest database of voluntary aviation incident reports in the world. The
ASRS is an online submission system of confidential safety information reports that submitted by
anics and others. Since lunched in
1976, the ASRS has received more than 988122 incident reports that contain very rich and detailed
information which is helpful to determine the reasons and premonitions in aviation safety incidents.
Although the ASRS database significantly benefits research of aviation incidents, considering its
voluntary trait, there are still some limitations of application in aviation maintenance. First, the data

295

296

Mengyao Bao and Shuiting Ding / Procedia Engineering 80 (2014) 293 302

and information are self-selected and then may result in underreporting. In the most extreme case, a
reporter may report every incident, while others may report noting. Second, the depth and quality of
reporting are diversified. A long-form report may represent rare information, while a short one may
be more valuable. Thirdly, the ASRS is still a largely unexploited source of information on
maintenance errors, it requires researchers to cost more time and have advanced skill to extract
worthy information.
3.1.2

Data Collection

We collected 3783 ASRS reports covering Federal Aviation Regulation Part 121 1 scheduled
airline operations for the period January 1, 2008 to December 31, 2008. We removed 20 duplicates,
bringing the total number of reports to 3763.
3.2 Method
3.2.1

Maintenance Error Decision Aid (MEDA)

The Maintenance Error Decision Aid (MEDA) is a structured process used to investigate events
caused by maintenance and/or inspector performance. It was developed by Boeing Customer
Services, the United States Federal Aviation Administration and other industry partners
e, 1994) [14]. The MEDA presents a list of eight types of errors that
may result in incidents, and ten types of contributing factors which are expected immediate factors
contributing to maintenance errors, as shown in Table 1 and Table 2. Furthermore, the contributing
factors are divided into 85 sub-contributing factors which are more detailed and precise. The
MEDA provides a better way to understand aviation maintenance problems. For example, the
communication problems in aviation maintenance can be collected to contributing factor category J)
Communication, and further classified to sub-contributing factors J1) Between departments J2)
Between mechanics, J3) Between shifts, J4) Between maintenance crew and lead, J5) Between lead
and management, J6) Between flight crew and maintenance or J7) Others. These sub-contributing
factors display a comprehensive context of detailed reasons that may finally lead to errors [14].
Table 1. Categories of maintenance error in MEDA
Number
1
2
3
4
5
6
7
8

Category
Installation
Servicing
Repair
Inspection
Foreign Object Damage (FOD)
Airplane/Equipment Damage
Personal Injury
Other

Table 2 Categories of contributing factor in MEDA

Number
A

Information
Information

1
Recall that Federal Aviation Regulation Part 121 is the section of the Code of Federal Regulations (CFR), which refers to
"scheduled or non-scheduled passenger operations that adhere to regulations that limit operations to controlled airspace and
controlled airports for which specific weather, navigational, operational, and maintenance support are available".

Mengyao Bao and Shuiting Ding / Procedia Engineering 80 (2014) 293 302

B
C
D
E
F
G
H
I
J

3.2.2

Equipment/Tools/Safety Equipment
Aircraft Design/Configuration/Parts
Job/Task
Technical Knowledge/Skills
Individual Factors
Environment/Facilities
Organizational Factors
Leadership/Supervision
Communication

Correspondence Analysis (CA)

The taxonomy of maintenance errors and contributing factors provided by MEDA can be used to
analyze errors and contributing factors independently. However, the relationship between errors
and contributing factors cannot be identified by the MEDA approach. Therefore, the
Correspondence Analysis (CA) is applied to reveal the error-contributing factors relationship.
The CA is a statistical visualization method that is used to determine associations within a large
dataset. It converts elements into points on a visual diagram to provide insight of relationships
between elements. Hobbs and Kanki [15][16] has used CA to analyze specific links between
contributing factors and errors caused by maintenance personnel from 2003 to 2008. To our best
knowledge, this is the only instance of CA being used in aviation maintenance research, in which
there are still some defects. First, in Hobbs' CA analysis, the errors categories were seven based on
extended Reason model [17] and the contributing factors categories were nine which adapted from
the one used by Feyer and Williamson [18]. The taxonomy of errors and contributing factors were
rough and mainly concentrated on individual behavior aspect. Second, the essential data in Hobbs'
CA analysis were only collected from maintenance personnel's reports, the perspective of nonmaintenance personnel were not involved.
In this paper, the CA method is used to examine the relationships between contributing factors
Each type of error and contributing factors is directly illustrated as a point in a two-dimensional biplot by chi-square distance. A short distance between points indicates a strong association, while a
long distance indicates a low association. The dimensional scale units are arbitrary, but the
intersection of two dimensions represents the center of plot. Categories near this central point occur
more frequently than categories that are plotted near the dimensional extremes.
4.

Results

4.1 MEDA Results

4.1.1

Preliminary error type analysis

Initially, we focus on the types of maintenance error reported by maintenance and nonmaintenance personnel. There were 198 maintenance related incidents reported by maintenance
personnel and 375 maintenance related incidents reported by non-maintenance personnel. It must
be clear that more than one error or contributing factor may be listed in one incident, and about
22% incidents in maintenance personnel reports were caused by several errors, while in nonmaintenance personnel reports the percentage is 14%.

297

298

Mengyao Bao and Shuiting Ding / Procedia Engineering 80 (2014) 293 302

Fig. 1. Proportion of maintenance errors in maintenance personnel and non-maintenance personnel incident reports from
ASRS in 2008

Fig 1 illustrates the proportion of maintenance errors in both maintenance and non-maintenance
accounts to 91% in total, the most significant types of error are inspection and installation, which
account for 33% and 32% respectively. From nonnumber
of human error accounts to 95%, servicing error is the most likely at 42%. The high percentage of
servicing error may because non-maintenance personnel rarely participate in specific maintenance
work such as inspection and installation; hence it is difficult for them to give out detailed
classification, and inadequately ascribed problems to servicing error.
4.1.2

Preliminary contributing factor type analysis

The distribution of individual-related factors and management-related factors are quite different
in maintenance personnel and non-maintenance personnel incident reports, as shown in Fig 2.

Fig. 2. Proportion of contributing factors in maintenance personnel and non-maintenance personnel incident reports
from ASRS in 2008

perspective, the individual-related factors are the most important

and account for 47% of all contributing factors, while the percentages of management-related
factors and other factors are 20% and 33% respectively. Specifically, the most common

Mengyao Bao and Shuiting Ding / Procedia Engineering 80 (2014) 293 302

contributing factor is F) individual factors at 26%, followed by E) Technical Skills (12%) and H)
Organizational Factor (10%). From nonrelated factors are the most significant and account for 39% of all contributing factors, while the
percentages of individual-related factors and others are 34% and 27% respectively. Specifically, the
most common contributing factor is H) Organizational Factor at 25%, followed by F) individual
factors (19%) and J) Communication (12%).
Through comparing the proportion of contributing factors in two reference groups as show in
Fig 2, it is clear that maintenance personnel selected more individual-related factors, while nonmaintenance personnel preferred to select more management-related factors. The difference of
contributing factors distribution in two reference groups is because of two reasons:
First, maintenance personnel are directly engaged in maintenance work and concentrate more on
their own task at low level, while non-maintenance personnel participates maintenance chain from
task making at high level to task allocation at low level. Although they are not directly engaged in
maintenance work, non-maintenance personnel obtain more information of maintenance such as
organizational structure, management and task collaboration. Therefore, maintenance personnel are
more likely to blame maintenance errors to direct reasons such as individual-related factors; while
non-maintenance personnel prefer to ascribe maintenance errors to management-related factors.
Second, non-maintenance personnel's tasks are more specific during work process. They work
under several management mechanisms and translate organizational level requirement (e.g., those
imposed by regulatory agencies and manufacturers) from high level to low level. Therefore, nonmaintenance personnel selected more communication and organizational factors. In addition, after
maintenance finish, non-maintenance personnel can find out some undiscovered errors during
operation process.
4.2 CA Results
Fig 3(a) and Fig 3(b) shows the relationship between errors and contributing factors in both
maintenance and non4.2.1

Qualitative relationships between errors and contributing factors in maintenance

personnel reports
0.8

Othe r

FOD

0.6

0.4

0.2

J) Communication
Se rvice

D) Job/Task
F) Individual Factors

G) Environment/Facility

R e pa ir

A) Information

-0.2

I nsta lla tion

H) Organizational Factors
I) Leadership/Supervision
I nspection

B) Equipment/Tools/Safety Equipment

-0.4

E) Technical Knowledge/Skills
Airplane /Equipm e nt D a m a ge

-0.6
C) Aircraft Design/Configurate/Parts

-0.8

-1

Pe rsona l I njury

Factor
Error

-1.2
-0.5

0.5

1.5

2.5

3.5

Fig. 3(a). Correspondence plot of Contributing factors and Errors in maintenance personnel reports from ASRS in 2008

299

300

Mengyao Bao and Shuiting Ding / Procedia Engineering 80 (2014) 293 302

Fig 3(a) illustrates that for maintenance personnel, the Installation and Inspection are the most
frequent error types since they are close to the (0, 0) point. In addition, these two errors are closely
clustered with H) Organizational Factors and I) Leadership/Supervision, which means these two
contributing factors are most likely to lead to Installation and Inspection errors. Similarly, the
Servicing is strongly associated with J) Communication, and Repair errors with A) Information and
F) Individual Factors, the Airplane/Equipment damage with E) Technical Knowledge/Skills.
The results shows tha
shows individual-related factors are the most selected contributing factors by maintenance
personnel, the most frequent error types are mainly caused by management-related factors, which
means the prevalence of contributing factors and errors are not corresponding . Second, individualrelated factors do not only lead to human errors, but also cause non-human related errors, such as E)
Technical Knowledge/Skills are most likely to arouse Airplane/Equipment damage error in
maintenance process.
4.2.2

Qualitative relationships between errors and contributing factors in non-maintenance

personnel reports
0.6
C) Aircraft Design/Configurate/Parts

Installation

0.4

B) Equipment/Tools/Safety Equipment

0.2
D) Job/Task
E) Technical Knowledge/Skills
I) Leadership/Supervision
Inspection
H) Organizational Factors
Service
J) Communication
A) Information

F) Individual Factors

0
Repair

-0.2
Airplane /Equipment Da ma ge

Other

-0.4

-0.6

Personal Injury
Factor
Error

G) Environment/Facility

-0.8
-1.2

-1

-0.8

-0.6

-0.4

-0.2

0.2

0.4

0.6

Fig. 3(b). Correspondence plot of Contributing factors and Errors in non-maintenance personnel reports from ASRS in 2008

Fig 3(b) illustrates the relationship between errors and contributing factors in non-maintenance
and inspection are closely associated with management-related factors H) Organizational factor, J)
Communication I) Leadership/Supervision. Second, installation error is clustered by B)
Equipment/Safety Equipment.
The result clearly demonstrates that in nonts, the most frequent
contributing factors- management-related factors are most likely to cause the most frequent errorsservicing and inspection. Moreover, B) Equipment/Safety Equipment is neither individual-related
factor nor management-related factor, but it may lead to human error-installation. It is because from
non-maintenance personnel perspective, the installation error is normally caused by direct factors

Mengyao Bao and Shuiting Ding / Procedia Engineering 80 (2014) 293 302

such as equipment damage or defects, which indicates it is difficult for non-maintenance personnel
to find out potential reasons of installation error.
5.

Conclusion

This paper analyzes the maintenance error, individual-related factors and management-related
factors in 3783 ASRS reports submitted by both maintenance personnel and non-maintenance
personnel. The MEDA and CA methods are applied. The result shows ASRS databases can be a
useful source of information on the nature and origins of maintenance issues, and both groups
considered human error accounts for a large proportion of maintenance errors, but because of the
content of work are different, maintenance personnel selected more inspection and installation
errors, while non-maintenance personnel selected more servicing error.
The analysis of contributing factors demonstrates that for two groups, individual-related factors
and management-related factors are the most frequent reasons for maintenance error. From
-related factors lead to more maintenance errors,
but the most frequent errors inspection and installation are mainly caused by management-related
errors. From non-related errors play a more
important role and lead to the most frequent errors servicing.
In fact, management-related factors can influence performance at the individual level, including
operations. For example, the F) Individual Factors include pressure and fatigue which may be
caused by management-related factors such as unreasonable task allocation, miscommunication or
unsuitable time plan. In many cases, maintenance personnel only reported individual-related factors
because they were unable to identify and recognize latent management-related factors. Meanwhile,
the importance of management-related factors indicates that non-maintenance personnel is also
important for maintenance safety. Fist, Well-designed schedule and workload, reasonable task
assignment and supervision by non-maintenance personnel can improve maintenance task quality.
Second, favorable communication between departments, leader and management crew, flight crew
and maintainer can raise efficiency of maintenance chain. Third, non-maintenance personnel should
build efficient work process/procedure, establish reasonable company policies to support
maintenance work. All of above non-maintenance personnel actions can improve maintainers'
efficiency and performance, and further reduce aviation maintenance error.
Therefore, the analysis of maintenance error and contributing factors from non-maintenance
personnel perspective can provide more information and not be ignored. And the managementrelated factors are as crucial as individual-related factors, which should be paid more attention.
Acknowledgements
This paper is based on previous research done by Dr. Karen B. Marias at School of Aeronautics
and Astronautics, Purdue University. The authors would like to acknowledge Dr. Marias for her
kindly suggestion.

301

302

Mengyao Bao and Shuiting Ding / Procedia Engineering 80 (2014) 293 302

References
[1] Marais KB, Robichaud MR. Analysis of trends in aviation maintenance risk: An empirical approach. Reliability
Engineering & System Safety 2012; 106, 104-118.
[2] Reason J, Maddox ME. Human error, In: Human factors guide for aviation maintenance, U.S. Department of
Transportation, Washington, DC (Chapter 14);1995.
[3] Hobbs A, Williamson A. Associations between errors and contributing factors in aircraft maintenance, Human
Factors 2003; 45, 186 201.
[4] Rankin W, Sogg S. Update on the Maintenance Error Decision Aid (MEDA) process, Proceedings of the
MEDA/MEMS Workshop and Seminar Gatwick, UK; 2003.
[5] Federal Aviation Administration. Zero accidents: a shared responsibility. Aviation Safety Action Plan. U.S.
Department of Transportation, Washington, DC; 1995.
[6] Hobbs A. An overview of human factors in aviation maintenance. ATSB Transport Safety Report 2008; Aviation
Research and Analysis Report-AR-2008-055.
[7] Mcdonald N, Corrigan S, Cromie S, Daly, C. An organisational approach to human factors, in Hayward BJ, Lowe
AR (eds). Aviation resource management Vol. 1 (Aldershot: Ashgate); 2000, 51-61.
[8] Taylor JC. Organizational context for aircraft maintenance and inspection. In: Proceedings of the Human Factors
34th Annual Meeting. Human Factors and Ergonomics Society, Santa Monica, CA, 1990; 1176-1180
[9] Taylor JC, Christensen TD. Airline maintenance resource management: Improving communication. Warrendale, PA:
Society of Automotive Engineers; 1998.
[10] Rogan E. MESH: Managing engineering safety health. Human Factors 2, 5}6, 1995; British Airways Engineering,
London.
[11] Patankar K, Lattanzio D, Kanki, BG. A content analysis of representative procedural errors in ASRS maintenance
reports, Proceedings of the Safety Across High-Consequence Industries Conference 2004; 167-172.
[12] Rankin W, Hibit R, Allen J, Sargent, R. Development and evaluation of the maintenance error decision aid (MEDA)
process, International Journal of Industrial Ergonomics 26 (No. 2) 2000; 261-276.
[13] Schmidt JK, Schmorrow D, Hardee M. A preliminary human factors analysis of naval aviation maintenance related
mishaps (SAE Tech. Paper No. 983111), Warrendale, PA: Society of Automotive Engineers; 1998.
[14]

Boeing

Company,

1994,

From:http://www.hf.faa.gov/opsmanual/assets/pdfs/MEDA_guide.pdf . Retrieved on Dec 12, 2012.

[15] Hobbs A, Kanki BG. A correspondence analysis of ASRS maintenance incident reports. Proceedings of the 12th
International Symposium on Aviation Psychology 2003; Dayton, OH.
[16] Hobbs A, Kanki BG. Patterns of error in confidential maintenance incident reports. International Journal of
Aviation Psychology 2008; 18: (1) 5-16.
[17] Reason J. Human error. Cambridge University Press, New York; 1990.
[18] Feyer AM, Williamson AM. A classification system of causes of occupational accidents for use in preventive
strategies. Scandinavian Journal of Work 1991; 17:302-11.