Pilot Fatigue Risk Analysis: Empirical Study at Flight Operations of

Garuda Indonesia’s Boeing 737 Pilots

F A Zein1 and M Hamsal2

1
Department of Management, Faculty of Economics and Business, Universitas Indonesia,
Kampus UI Depok, West Java, Indonesia 16424
2
Department of Management, Faculty of Economics and Business, Universitas Indonesia,
Kampus UI Depok, West Java, Indonesia 16424

E-mail: Faisal.adrian@ui.ac.id

Abstract

Fatigue is essential in the aviation field because it affects many passenger’s safety. Garuda
Indonesia is a National Airline which operates as a Flag Carrier of the largest archipelago
national, which cover three different time zones. The purpose of this study is to analyze factors
that affect airline pilot fatigue at Boeing 737 fleet of Garuda Indonesia, which fly domestic and
regional flights. While most of the research related to human fatigue in the aviation industry
has focused on long-haul pilots since the exposure of their duties towards jet-lag, short-haul
pilots also experience elevated levels of fatigue caused by their flight operation. As the nature
of the Boeing B737 aircraft operation, this study concentrates on short to medium haul, less
than 8 hours flight time per flight sector, multiple flight sectors a day, with the set of two pilots
for each rotation pattern. Fatigue is classified into physical decline, mental decline, and rest
defects. This study finds that pilot fatigue in Boeing 737 pilot of Garuda Indonesia is affected
by variables of Duty Assignment, Personal Lifestyle, Work Environment, Rest Environment,
Unresolved Stress, and Illness.

Keywords: Fatigue; Airline Pilot; Fatigue management; Flight Operation; Human Factors

Analisis Resiko Kelelahan Pilot: Studi Empiris pada Operasional

Penerbangan Pilot Boeing 737 Garuda Indonesia

Abstrak

Kelelahan dibidang penerbangan mempengaruhi keselamatan banyak orang. Garuda Indonesia,

maskapai penerbangan nasional yang beroperasi sebagai Flag Carrier, yang juga merupakan
BUMN penyedia transportasi udara dari negara kepulauan terbesar di dunia, yang mencakup
tiga zona waktu. Tujuan penelitian ini adalah untuk menganalisis faktor-faktor yang
mempengaruhi kelelahan penerbang armada Boeing 737 di Garuda Indonesia, yang
menerbangkan penerbangan domestik serta regional. Sementara sebagian besar penelitian
terkait dengan kelelahan manusia pada industri penerbangan lebih berkonsentrasi terhadap
penerbang rute jarak jauh, dikarenakan tugas mereka rentan terpapar jet-lag, penerbang rute
jarak pendek juga mengalami peningkatan tingkat kelelahan yang disebabkan oleh karakter
operasional. Sesuai dengan kemampuan operasional dari pesawat Boeing 737, studi ini
berkonsentrasi pada penerbang jarak pendek hingga menengah, dengan waktu penerbangan
kurang dari 8 jam per-sektor penerbangan, atau beberapa sektor penerbangan per-hari, dengan
komposisi dua penerbang. Kelelahan diklasifikasikan menjadi penurunan kemampuan fisik,
penurunan kemampuan mental, dan kurang istirahat; dan kelelahan pilot pada pilot Boeing 737
Garuda Indonesia dipengaruhi oleh tujuh variabel - Tugas, Gaya Hidup, Lingkungan Kerja,
Lingkungan Istirahat, Rekan Kerja, Stres yang Tidak Terselesaikan, serta Penyakit Medis.
Identifikasi dari variabel serta fenomena faktor-faktor yang berkontribusi terhadap kelelahan
pilot di pilot Boeing 737 Garuda Indonesia akan dapat berkontribusi untuk manajemen
kelelahan penerbang.

Introduction

From 1959 to 2015, the number of commercial aircraft of departures (713 million), and more
significantly, the number of flight hours (1,321 million) increased (Boeing, 2016). Reflected
on Flight Safety Foundation’s report regarding fatal airliner accident per year, the number of
accidents from 1946 until 2017 is significantly reduced, however, the number of fatalities
compares to the number of aircraft accidents is considerably high (Boeing, 2016; Flight Safety
Foundation, 2018). Over the last 20 years, human factors are identified as one of the potentials
to compromise the safe operation of the flight, while fatigue has been identified as a causal
factor (Roach et al., 2012). Human factors are recognized as one of the critical determinants
for managing and improving flight safety (NASA, 1999). The “Swiss cheese” model of human
error, by James Reason (2000), describes four levels of human failure, each influencing the
next. The unsafe acts of pilots can be directly linked to nearly 80% of all aviation accidents,
while one of the principal subdivisions of unsafe aircrew conditions were developed
substandard conditions of the pilots, including adverse mental/ physiological states, and
physical/ mental limitation (Shappell & Wiegmann, 2000). Fatigue, subjective or
physiological, consider as a substandard condition, which degrades a person’s ability to stay
awake, alert, and attentive to the demands of controlling their vehicle safely; also, fatigue
impairs our ability to judge just how fatigued we are (NTSB, 2016). This phenomenon should
be expected, given that, today the pilot’s technical and non-technical competencies required to
operate a modern airliner are both susceptible to fatigue-related impairment (ICAO, 2013).

Referring to the non-stop flight time, flight operations are categorized as short-haul flight
(<8hrs) and long-haul flight (≥8 hours) (Roach et al., 2012). The specific characteristic of long-
haul flight operation includes more than three time-zone crossing per flight sector. early-
morning arrival, and low work tempo. While on short-haul flight operation, the characteristic
includes, multiple flights per duty day, highly dynamic environment, and constant high tempo
workload (Gander et al., 1998). Hence, per duty period, short-haul pilots conduct more take-
offs and landings, the most critical stages of flight which involves the most significant
workload, than long-haul pilots, per duty period (Roach et al., 2012).

Majority of airline operator in Indonesia operates in short-haul operation, and since most of the
fatigue researches are focusing on long-haul flight operation, this research aims to contribute
more empirical research towards short-haul operations (Jackson & Earl, 2006). As the only
state-owned airline company in Indonesia, Garuda Indonesia serves to ensure same day
connection through air transportation to the corners of the archipelago with the capital city of
Jakarta. The flight from Jakarta, Garuda’s pilot home base to Jayapura, Papua, the furthest
domestic direct flight (3,778 km), crosses two different time zones. Garuda also operates
International flights to countries in Asia, Australia and Europe. One of the aircraft types which
becomes the workhorse of Garuda Indonesia is the Boeing 737. Garuda operates their 82
Boeing 737 for around 300 domestic and regional flights sectors per day (Garuda Indonesia,
2018). The Boeing 737 is one of the popular models by the Boeing Commercial Airplanes
Industry, a short to medium range (5000+ km) twinjet narrow-body airliner which can travel
at around 0.8 times the speed of sound (Boeing, 2018). Meaning, it can fly from Jakarta to
Jayapura in about 6 hours while carrying 160+ of passengers.
This research aims to identify factors that contribute to pilot fatigue which existed in Garuda
Indonesia’s Boeing 737 fleet flight operations, which is the suitable and popular type for short-
haul flight operation and determine which contributing factors have the more or the least
significant impact on pilot fatigue in Garuda Indonesia’s Boeing 737 fleet flight operations.
This research uses factors analysis to reduce the dimension and linear regression analysis to
determine how the independent variable affect the dependent variable, with and without
moderating variable.

This paper structured as follows. Section 1 present the introduction. Section 2 is a brief
literature review background. Section 3 describes the research methods and framework, while
Section 4 presents findings and the analysis. Section 5 concludes the paper. Section 6 consists
of some recommendations.
Background

Attributing accidents solely to pilot error is like a medical doctor telling the patient that they
are “sick” without examining the causes of further defining the illness (Shappell & Wiegmann,
2000). To understand more about human error, we must be familiar with human factors. Human
Factors is about people in their living and working situations; about their relationship with
machines, with procedures and with the environment about them; and, about their relationships
with other people, which model introduced in 1975 by Frank Hawkins as SHEL model; the
name is derived from the initial letters of its components, Software, Hardware, Environment
and Liveware (ICAO, 1998). Liveware, in the center of the model, is a person, the most critical
as well as the most flexible component in the system. People are subject to considerable
variations in performance and suffer many limitations. An understanding of the characteristics
of Liveware is essential, including physical needs, information processing, and input-output
characteristics (ICAO, 1998).

Human Factors will impact two broad areas of the system’s effectiveness; safety and efficiency,
and the well-being of operational personnel which jointly interrelated overlap and affect each
other (ICAO, 1998). Area of the personnel’s physical and mental well-being, are affected by
fatigue, disturbed circadian rhythm, sleep deprivation, temperature, noise, humidity, light,
vibration, workspace design and space (Dumitru & Boscoiau, 2015). The unsafe, or
substandard, aircrew conditions are identified as adverse mental states, adverse physiological
states and physical/ mental limitations (Shappell & Wiegmann, 2000).

Fatigue is identified as a subjective experience and classified into the dimension of physical
decline (Dawson & McCulloch, 2005; Dinges et al., 1996), mental decline (Lee & Kim, 2018;
Marcora et al., 2008), and rest defect (Dinges et al., 1996; Lee & Kim, 2018; Van Cauter et al.,
2007). While Pilot fatigue in long-haul air transport operations in the aviation industry has
drawn more attention by most of the field-based researchers, rather than short-haul operations
(Darwent et al., 2010; Petrilli et al., 2006; Signal et al., 2005; Spencer et al., 1991). Previous
studies of short-haul operations indicate that short-haul pilots can experience relatively high
levels of fatigue and that the major causes of fatigue are long duty periods and early start times
(Bourgeois-bougrine et al., 2003; Gander et al., 1998; Jackson & Earl, 2006; Powell et al.,
2007). Pilots would obtain less sleep, and have higher levels of fatigue, for duty periods that
started earlier in the morning compared to those that started later in the morning. (Ingre et al.,
2008; Roach et al., 2003; Sallinen et al., 2003), Additionally, short-haul pilots allegedly suffer
from pre-hypertension diastolic blood pressure condition of 81-84 mmHg as an added risk
towards their fitness level (Yuliawati et al., 2018). Fatigue is inevitable since the human’s
physical and mental capability can only function optimally with unrestricted sleep at night.
Therefore, as fatigue cannot be eliminated, it must be managed (ICAO, 2017). Hours of service
guidelines alone are not sufficient to minimize fatigue and optimize on the job alertness in
safety-sensitive contexts (Lerman et al., 2012). Consideration must be given towards; the type
of job being performed, the time of day at which the job is being performed, the work
environment, the amount of pre-duty or off duty sleep, the quality of sleep, accumulation of
chronic sleep debt, the extent of continuous wakefulness, the amount of time on task, and the
level of individual fatigue resistance (FAA, 2012).

Based on the literature review, preliminary interview conducted towards seven subject matter
experts consists of five active Garuda Indonesia Pilots and two of Garuda Indonesia’s in-house
Flight Surgeon Medical Doctors to elicit practical experiences and opinions. Lee and Kim
(2018) on their paper which published in the Journal of air transport, titled Factors contributing
to the risk of airline pilot fatigue, offer a research model with seven independent variables.
Those variables are; flight direction, crew scheduling, partnership, aircraft environment, job
assignment, ethnic difference, and hotel environment. Hence, this paper offers a slightly
different model. Since this paper focused on short-haul flight and considering typical flight
operations in Indonesia, several independent variables, such as flight direction considered as
insignificant, while crew scheduling and job assignment considered as one variable.
Additionally, this paper introduces crew pairing partnership as a moderating variable.

Research Methods and Framework

Research model as depicted in Figure 1 is constructed based on the literature review analysis
and developed from the model created by Lee and Kim in their journal of Factors contributing
to the risk of airline pilot fatigue, which published in 2018 in Journal of Air Transport
Management, with the addition of moderating variable.
 Figure 1. Research Model

Pilots are confronted continuously with multiple take-offs and landings, long duty days, early
departures, late arrivals, and non-standard work hours that include night duty and rotating
schedules (Caldwell, 2005; Bourgeois-Bougrine et al., 2003; Powell et al., 2007; Yuliawati et
al., 2015). Even within regulated time limits, duty assignments are known to affect sleep and
subsequent alertness (Thomas et al., 2015; Bourgeois-Bougrine et al., 2003).
Hypotheses 1a. The Duty Assignment has a positive effect on Physical Decline.
Hypotheses 1b. The Duty Assignment has a positive effect on Mental Decline.
Hypotheses 1c. The Duty Assignment has a positive effect on Rest Defect.

The effects of lifestyle on health are indisputable to clinicians. Individuals with sleep
complaints often exhibit unhealthy lifestyles, including obesity, lack of physical exercise,
strenuous activities and cigarette smoking (Phillips & Danner, 1995; Caldwell, 2016; Goodman
& Gilman, 1966). The lifestyle of practicing heavy physical exercise considered as strenuous
activity, which also contributes to fatigue (Reinhart, 2007). The time at which a person sleeps
can affect the structure of the sleep cycle, and creates a potential safety hazard on the flight
deck (Caldwell, 2016).
Hypotheses 2a. The Personal Lifestyle has a positive effect on Physical Decline.
Hypotheses 2b. The Personal Lifestyle has a positive effect on Mental Decline.
Hypotheses 2c. The Personal Lifestyle has a positive effect on Rest Defect.

Pilot’s working environment, with the maximum permissible cabin altitude of routine
commercial flight at high altitude of 8000 feet, when exposed for several hours may result in
noticeable fatigue and potentially more significant medical problems, which are probably
induced by the combination of mild hypoxia, expansion of abdominal gas, lack of movement
and seated posture with additional stress of continuously pay attention to automatic digital
instruments (Rood & James, 2016; Salazar & Caldwell, 2010). In the same time, vibration
interferes with neuromuscular control and lead to fatigue (Stott, 2016), and noise, defined
mainly as unwanted sound, may also increased levels of stress and fatigue (Rood & James,
2016).
Hypotheses 3a. Undesired Working environments have a positive effect on Physical Decline.
Hypotheses 3b. Undesired Working environments have a positive effect on Mental Decline.
Hypotheses 3c. Undesired Working environments have a positive effect on Rest Defect.

Unresolved stress is common for everyone. It has been said that the eternal hanging of an
incomplete task is the most fatiguing. Personal problems which not resolved drains the body
of energy and also associated with depression (Reinhart, 2007). Unresolved stress chronic, it
is something that continues for an extended period such as financial difficulties or inter-
personal relationship problems (CAA, 2014). This condition, which also includes family
matters and preoccupations, have a particular contribution towards fatigue (Gawron, French,
& Funke, 2001)(Gawron et al., 2001; Bor et al., 2002).
Hypotheses 4a. Unresolved Stress has a positive effect on Physical Decline.
Hypotheses 4b. Unresolved Stress has a positive effect on Mental Decline.
Hypotheses 4c. Unresolved Stress has a positive effect on Rest Defect.

An uncomfortable or unfamiliar environment, noise, light, heat and cold will interfere with
sleep, as will a bed that is either perceived as too hard or too soft, a pillow that does not support
the head and neck properly, or otherwise disturb the tranquility of the sleep setting, which
includes the room condition, distance from activity places, and neighborhood (Caldwell, 2016;
Salazar & Caldwell, 2010; ICAO, 2017; Rudari et al., 2016).
Hypotheses 5a. Rest Environment has a negative effect on Physical Decline.
Hypotheses 5b. Rest Environment has a negative effect on Mental Decline.
Hypotheses 5c. Rest Environment has a negative effect on Rest Defect.

Some pilots may perceive some illness as prevalent, such as common cold, and well-managed
asthma, allergies or a sleep-disturbing pathological condition, including obstructive sleep
apnea, restless leg syndrome, certain medications, depression, insomnia, and chronic pain
which also contribute to fatigue significantly when relapse (Reinhart, 2007; Salazar &
Caldwell, 2010). Sedation or drowsiness is the side effect when the body fights illness through
its metabolic defenses, the fatiguing effects of the illness itself impaired alertness (Reinhart,
2007; Salazar & Caldwell, 2010).
Hypotheses 6a. Illness has a positive effect on Physical Decline.
Hypotheses 6b. Illness has a positive effect on Mental Decline.
Hypotheses 6c. Illness has a positive effect on Rest Defect.

In the major airline crew rotation planning processes like in Garuda Indonesia, a set of
anonymous crew members are paired to cover, at minimum cost, the flights scheduled to be
operated by a given aircraft type over a whole month (Saddoune et al., 2013). Cultural values
predispose each person to have a preferred particular method of handling issues that arise
during teamwork (Smith, 1999). In flight operations, pilots require to work collaboratively with
persons whose personal backgrounds differ from their own. However, when individual values
are mixed in a single flight crew, the various crew members may have trouble in exchanging
information and knowledge, this situation can affect flight safety, and furthermore, it can cause
crew fatigue and accidents. Crew pairing partnership which considered as significant factors
Include extreme age differences, values, and occupation (e.g., Instructor, chief pilot, inspector)
(Lee & Kim, 2018). Proper Crew Resource Management (CRM) application between crews
considered as an essential factor for monitoring each other regarding fatigue condition
(Yuliawati et al., 2015).
Hypotheses 7a. Crew Pairings Partnerships moderate the impact of Duty Assignment on
Physical Decline.
Hypotheses 7b. Crew Pairings Partnerships moderate the impact of Duty Assignment on
Mental Decline.
Hypotheses 7c. Crew Pairings Partnerships moderate the impact of Duty Assignment on Rest
Defect.
Hypotheses 8a. Crew Pairings Partnerships moderate the impact of Personal Lifestyle on
Physical Decline.
Hypotheses 8b. Crew Pairings Partnerships moderate the impact of Personal Lifestyle on
Mental Decline.
Hypotheses 8c. Crew Pairings Partnerships moderate the impact of Personal Lifestyle on Rest
Defect.
Hypotheses 9a. Crew Pairings Partnerships moderate the impact of the Working Environment
on Physical Decline.
Hypotheses 9b. Crew Pairings Partnerships moderate the impact of the Working Environment
on Mental Decline.
Hypotheses 9c. Crew Pairings Partnerships moderate the impact of the Working Environment
on Rest Defect.
Hypotheses 10a. Crew Pairings Partnerships moderate the impact of Unresolved Stress on
Physical Decline.
Hypotheses 10b. Crew Pairings Partnerships moderate the impact of Unresolved Stress on
Mental Decline.
Hypotheses 10c. Crew Pairings Partnerships moderate the impact of Unresolved Stress on Rest
Defect.

This research use probability samples to make unbiased inferences about the population of
interest (Levine et al., 2008). The formula used for calculating sample size is:
n = (p (100-p) z2) / E2
Where; n is the required sample size, p is the percentage occurrence, E is the percentage
maximum error required, and z is the value corresponding to the level of confidence required.
In management and social research, the confidence level of 95% (z=1.96) considered as typical,
5% margin of error considered as acceptable, and p estimate of 50% considered as suggested
for maximization of variance and produce maximum sample size (Bartlett et al., 2001;
Taherdoost, 2016). From this calculation using the formula, with a population of 600 pilots,
the minimum sample would be 234 respondents.

Survey to Garuda Indonesia’s B737 pilots conducts afterward, and the result analyses to find
out which factor contributes a more significant effect than the other. There are 13 questionnaire
clusters of questions, consists of 41 statements with six-points likert scale answers about
respondent perception regarding fatigue contributors, and the dimension of fatigue, which the
participants must answer.

The questionnaires constructed as an e-form, and can be accessed through any web browsers
on any computers, smartphone or electronic tablets, to accommodate easy access by the
respondents. Information and distribution of the survey were assisted by the office of Boeing
737 Chief Pilot office and Asosiasi Pilot Garuda (APG).
From 236 respondents who participated voluntarily, the majority of them are 20-29 years old
age (47.5%), already work for Garuda Indonesia for 5 to 15 years (39.8%), with accumulated
flight time around 4,000 to 10,000 hours (38.6%), never fly other than the Boeing 737 type of
aircraft in Garuda Indonesia (69.1%), with the rank of First Officer (50.4%).

Results

This research uses SPSS 20 as the tool to do all statistical test and analysis. All respondent’s
answer shows a tendency towards agreeing in all statements which represent the independent
and dependent variables, while answers for the moderating variable tend to be neutral.

Table 1. Results of Factor analysis of Independent and Moderating Variables

DA PS WE RE US IL CP
DA1 0.843
DA2 0.813
DA3 0.744
DA5 0.734
PS1 0.864
PS2 0.864
WE2 0.720
WE3 0.826
WE4 0.837
WE5 0.830
WE6 0.790
RE1 0.758
RE2 0.854
RE4 0.815
US1 0.879
US2 0.907
US3 0.819
US4 0.778
IL1 0.809
IL2 0.862
IL3 0.811
CP1 0.756
CP2 0.845
CP3 0.818
CP4 0.774
CP5 0.718
Cronbach’s
0.794 0.662 0.860 0.736 0.868 0.770 0.842
alpha

The gathered data was tested to determine whether it is suitable for factor analysis. Methods
used are: The Bartlett test of sphericity, to check the correlation of each variable, Kaiser Mayer
Olkin (KMO) test, which value of 0.05 or above is deemed acceptable to continue the factor
analysis, and Measure of Sampling Adequacy (MSA), a value of more than 0.5 means that the
variable can be predicted by other variables (Usman & Sobari, 2013).
 Table 2. Results of Factor analysis of Dependent Variables

PD MD RD
PD1 (F1) 0.860
PD2 (F2) 0.898
PD3 (F3) 0.721
MD2 (F5) 0.940
MD3 (F6) 0.940
RD2 (F8) 0.935
RD3 (F9) 0.935
Cronbach’s
0.763 0.868 0.855
alpha

Reliability check also performs, as Cronbach's alpha assesses how well the items in a set are
positively correlated with one another, it was used to test the internal consistency (Lee & Kim,
2018). Cronbach alpha with values of 0.60 deemed the lower limit of acceptability (Hair et al.,
2013). The factor analysis was used to see if the loading is more than 0.6. All factor loading
less than 0.6 were cut off. The results are depicted in table 1 and table 2.

Figure 2. Research Model without and with Moderating Variable

Figure 3 and Table 3 shows the results of the multiple regression analysis on Physical Decline.
Duty Assignment, Working Environment, Unresolved Stress, and Illness have significant
effects on Physical Decline. Hypotheses 1a, 3a, 4a, 5a, 7a, 9a, 10a are supported.
 Figure 3. The result of hypothesis testing (Physical Decline)

Crew Pairing Partnership contributes as pure moderator for unresolved stress since the
correlation of the moderating variable is not significant while the moderation with the
independent variable is significant, and as potential moderator for all other moderated
independent variables since the correlation of moderating variable and the moderation with
those respected variables are not significant (Sharma et al., 1981).

Table 3. The result of regression analysis (Physical Decline)

Unstandardized Standardized
Model Coefficients Coefficients t Sig.
B Std. Error Beta
1 (Constant) 12.835 0.140 91.455 0.000
FAC_DA 0.458 0.161 0.173 2.842 0.005
FAC_PS 0.079 0.147 0.030 0.536 0.592
FAC_WE 0.358 0.166 0.135 2.157 0.032
FAC_RE -0.141 0.152 -0.053 -0.931 0.353
FAC_US 0.789 0.163 0.298 4.826 0.000
FAC_IL 0.572 0.158 0.216 3.618 0.000
2 (Constant) 12.901 0.148 87.189 0.000
FAC_DA 0.521 0.169 0.197 3.087 0.002
FAC_PS 0.052 0.145 0.020 0.357 0.721
FAC_WE 0.425 0.171 0.161 2.483 0.014
FAC_RE -0.110 0.151 -0.042 -0.728 0.468
FAC_US 0.596 0.170 0.225 3.500 0.001
FAC_IL 0.516 0.159 0.195 3.251 0.001
FAC_CP 0.202 0.156 0.076 1.297 0.196
FAC_DACP 0.212 0.154 0.087 1.376 0.170
FAC_PSCP -0.116 0.146 -0.043 -0.797 0.427
FAC_WECP 0.202 0.156 0.080 1.294 0.197
FAC_USCP -0.514 0.164 -0.206 -3.141 0.002

Figure 4 and Table 4 shows the results of the multiple regression analysis on Mental Decline.
Duty Assignment, Unresolved Stress, and Illness have significant positive effects on Mental
Decline. Hypotheses 1b, 4b, 5b, 7b and 10b are supported.
 Figure 4. The result of hypothesis testing (Mental Decline)

Crew pairing partnership contributes as pure moderator for unresolved stress since the
correlation of moderating variable is not significant while the moderation with the independent
variable is significant, and as potential moderator for all other moderated independent variables
since the correlation of the moderating variable and the moderation with the variable are not
significant (Sharma et al., 1981).

Table 4. The result of regression analysis (Mental Decline)

Unstandardized Standardized
Model Coefficients Coefficients t Sig.
B Std. Error Beta
1 (Constant) 8.725 0.105 82.942 0.000
FAC_DA 0.345 0.121 0.182 2.851 0.005
FAC_PS -0.021 0.110 -0.011 -0.187 0.852
FAC_WE 0.029 0.124 0.015 0.230 0.818
FAC_RE -0.155 0.114 -0.082 -1.359 0.176
FAC_US 0.593 0.123 0.314 4.843 0.000
FAC_IL 0.402 0.119 0.213 3.391 0.001
2 (Constant) 8.726 0.112 78.149 0.000
FAC_DA 0.420 0.127 0.222 3.296 0.001
FAC_PS -0.025 0.110 -0.013 -0.225 0.822
FAC_WE 0.071 0.129 0.037 0.549 0.584
FAC_RE -0.127 0.114 -0.067 -1.111 0.268
FAC_US 0.465 0.129 0.246 3.621 0.000
FAC_IL 0.379 0.120 0.200 3.161 0.002
FAC_CP 0.109 0.118 0.058 0.929 0.354
FAC_DACP 0.223 0.116 0.128 1.920 0.056
FAC_PSCP -0.016 0.110 -0.008 -0.147 0.883
FAC_WECP 0.146 0.118 0.081 1.241 0.216
FAC_USCP -0.312 0.124 -0.175 -2.523 0.012

Figure 5 and Table 5 shows the results of the multiple regression analysis on Rest Defect. Duty
Assignment, Personal Lifestyle, and Unresolved Stress have significant positive effects on
Mental Decline. Hypotheses 1c, 2c, 4c, 5c, 7c, 8c and 10c are supported.
 Figure 5. The result of hypothesis testing (Rest Defect)

Crew Pairing Partnership contributes as predictor moderator for all moderated independent
variables, since the moderating variable have a significant correlation with the dependent
variable, while none of the moderation with those respected independent variables is
significant, in other words, the moderating variable contributes as an independent variable
(Sharma et al., 1981).

Table 5. The result of Regression Analysis without Moderating Variable (Rest Defect)

Unstandardized Standardized
Model Coefficients Coefficients t Sig.
B Std. Error Beta
1 (Constant) 7.114 0.147 48.52 0.000
FAC_DA 0.420 0.169 0.174 2.491 0.013
FAC_PS -0.287 0.154 -0.119 -1.868 0.063
FAC_WE 0.314 0.173 0.131 1.814 0.071
FAC_RE -0.057 0.159 -0.024 -0.359 0.720
FAC_US 0.448 0.171 0.186 2.625 0.009
FAC_IL 0.136 0.165 0.057 0.825 0.411
2 (Constant) 7.127 0.152 46.91 0.000
FAC_DA 0.382 0.173 0.159 2.204 0.029
FAC_PS -0.310 0.149 -0.129 -2.078 0.039
FAC_WE 0.253 0.176 0.105 1.440 0.151
FAC_RE -0.067 0.155 -0.028 -0.430 0.667
FAC_US 0.222 0.175 0.092 1.272 0.205
FAC_IL 0.167 0.163 0.069 1.023 0.307
FAC_CP 0.638 0.160 0.265 3.989 0.000
FAC_DACP 0.137 0.158 0.062 0.867 0.387
FAC_PSCP -0.109 0.150 -0.044 -0.728 0.467
FAC_WECP 0.130 0.160 0.057 0.812 0.418
FAC_USCP -0.230 0.168 -0.101 -1.368 0.173

As depicted in table 6, without intervention from Crew Pairing Partnership, the Physical
Decline can be explained by the Independent Variables by 35.4%, while the Mental Decline by
28.7%, and Rest Defect by 14.7%. With intervention from Crew Pairing Partnership, the
Physical Decline can be explained by the Independent Variables and Moderating Variable by
38.9%, while the Mental Decline by 31.8%, and Rest Defect by 22.2%.
 Table 6. Summary of R Square

Conclusion

Physical Decline significantly affected by Duty Assignment, Working Environment,

Unresolved Stress. Crew Pairing Partnership contributes as pure moderator for unresolved
stress and as potential moderator for all other moderated independent variables. Personal
Lifestyle and Rest Environment do not have a significant impact on Physical Decline. The
result of a fatigue-perception is a very biased one because the Personal Lifestyle variable is
expressed individually in working and leisure patterns in activities, attitudes, interest, opinions,
and values, while the data gathered from the self-perception research questionnaire. The
insignificant outcome is therefore comprehensible. Homogeneity of the respondents is
responsible for the insignificant result of Rest Environment impact on Physical Decline, and
the accommodation policies of Garuda Indonesia provide standard hotel conditions.

The next Dependent Variable, Mental Decline, affected significantly by Duty Assignment,
Unresolved Stress, and Illness. Crew Pairing Partnership contributes as pure moderator for
unresolved stress, and as potential moderator for all other moderated independent variables.
The same reasons for the insignificant results of Personal Lifestyle and Rest Environment
effects towards Physical Decline. The working environment is not significant, as the condition
is mentally perceived as usual or normal.

The third Dependent Variable is Rest Defect, which significantly affected by Duty Assignment,
Personal Lifestyle, and Unresolved Stress. Crew Pairing Partnership contributes as predictor
moderator for all moderated independent variables; in other word, the moderating variable
contributes as an independent variable. The insignificance of Work Environment and Rest
Environment caused by the same reason as the impacts towards other Dependent Variables,
and Illness not perceived as causing Rest Defect, hence more towards Physical and Mental
Decline.
It is difficult for a person to measure their performance; it is easier for others to measure.
Therefore, the analysis result of the gathered data from the questionnaire shows Personal
Lifestyle variable is not significant towards Physical and Mental Decline. Personal Lifestyle
becomes significant after moderated by Crew Pairing Partnership towards Rest Defect,
although the beta coefficient is negative, meaning, Rest Defect Perception consider as
decreased, since the atmosphere of the activity considered as more comforting.

Before moderated by Crew Pairing Partnership, Unresolved Stress is not significant, since the
crew has not got any information about who is the pairing partner, and this condition may add
some stress.

Rest Environment is not significant towards Physical Decline, Mental Decline and Rest Defect,
caused by the homogeneity of the sample since Garuda Indonesia has an adequate
accommodation standard for air crewmember on duty.

The R Squared on the Determination Analysis indicates that there are more contributors, other
than identified by this research.. However, the coefficient determination shows that the Crew
Pairing Partnership increases the contribution of affected independent variables towards
respected dependent variable.

The most influencing contributor to Physical Decline is Unresolved Stress, a very personal
matter. Hence, after moderated by Crew Pairing Partnership, the beta coefficient decreased.
Means, the right Pairing Partnership can ease the effect of Unresolved Stress. Duty
Assignment, after moderated by Crew Pairing Partnership, becomes the number two of the
most influencing contributor towards Physical Decline, Mental Decline, and Rest Defect.
Management can do some intervention of adjustment with regards to this matter, as it is more
important to mitigate the risk of fatigue, rather than only managing the fatigue. As shown by
the result, significant fatigue contributors for Garuda Indonesia’s Boeing 737 Pilots prioritize
to be managed for operations planning efficiency and effectivity are: (1) Unresolved Stress, (2)
Duty Assignment, (3) Illness, (4) Working Environment, (5) Crew Pairing Partnership, and (6)
Personal Lifestyle.
Recommendation

1. Pilot Fatigue needs to be studied more thoroughly, especially as input for pilot scheduling,
flight operation planning, duty assignment, and control. The study considered essential for
supporting continuity of improvement to promote safe, effective, and efficient flight
operations.
2. The results of this study are expected to be able to direct or shift the Pilot Fatigue
management from curative aspects, which is remedy after flight assignment, to become
preventive aspects, which is planned before flight assignment.
3. This research focuses on cognitive perception. Further research can be enriched by
incorporating physiological aspects with more accurate measurement indicators.
4. The nature of Unresolved Stress considered as subjective and personal, but significantly
affects the Pilot Fatigue, a mitigating alternative must be found such as periodic counseling
programs.
5. A system or application needs to be made that supports the prognosis or the estimated level
of anticipation of Pilot Fatigue, and for further enhance the ability to mitigate the risk of
Pilot Fatigue.
6. Crew Pairing Partnership moderates the negative influence of Personal Lifestyle, Work
Environment, and Unresolved Stress on pilot fatigue, which shows that the CRM (Crew
Resources Management) program is starting to run effectively. However, it needs to be
studied more deeply and comprehensively.
7. The findings of this research could be the original starting point for the formal and efficient
management of Operation Safety and Health (OS&H) at Garuda Indonesia.
8. Although not directly related, the results of this study are expected to be the initial input,
or preliminary study, for Garuda Medical Center in dealing with pilot fatigue, both
reported and those that have not been reported so far. The pilot fatigue is a proactive
consideration of whether the pilot is fit to fly or not.
9. Further research can include other variables that affect pilot fatigue. The different types of
aircraft (fleet), flight times (night/ day), airlines (possibly in one group), flight areas
(Eastern, Central, and Western Indonesia) may also be considered, and also subsequent
research for long haul flights, and international routes can be carried out.
References
Bartlett, J. E., Kotrlik, J. W., & Higgins, C. C. (2001). Organizational research: Determining
appropriate sample size in survey research appropriate sample size in survey research.
Information Technology, Learning, and Performance Journal.
https://doi.org/10.1109/LPT.2009.2020494
Boeing. (2016). Statistical Summary of Commercial Jet Airplane Accidents Worldwide
Operation 1959-2015. Seattle, WA. Retrieved from
www.boeing.com/news/techissues/pdf/statsum.pdf
Bourgeois-bougrine, S., Cabon, P., Mollard, R., Coblentz, A., Speyer, J., V, U. R. D., &
Industrie, A. (2003). Fatigue in aircrew from short- haul flights in civil aviation : the
effects of work schedules. Human Factors and Aerospace Safety, 3(2), 177–187.
Bourgeois-Bougrine, S., Carbon, P., Gounelle, C., Mollard, R., & Coblentz, A. (2003).
Perceived fatigue for short- and long-haul flights: A survey of 739 airline pilots.
Aviation Space and Environmental Medicine, 74(10), 1072–1077.
Caldwell, J. A. (2005). Fatigue in aviation. Travel Medicine and Infectious Disease, 3(2), 85–
96. https://doi.org/10.1016/j.tmaid.2004.07.008
Caldwell, J. A. (2016). Aviator fatigue and relevant fatigue countermeasures. In D. P.
Gradwell (Ed.), Ernsting’s Aviation and Space Medicine (5th ed., pp. 583–600). Boca
Raton: CRC Press.
Civil Aviation Authority (CAA). (2014). Flight-crew human factors handbook. Sussex: Civil
Aviation Authority (CAA). https://doi.org/10.1557/opl.2012.941
Darwent, D., Dawson, D., & Roach, G. D. (2010). Prediction of probabilistic sleep
distributions following travel across multiple time zones. Sleep.
https://doi.org/10.1093/sleep/33.2.185
Dawson, D., & McCulloch, K. (2005). Managing fatigue: It’s about sleep. Sleep Medicine
Reviews, 9(5), 365–380. https://doi.org/10.1016/j.smrv.2005.03.002
Dinges, D. F., Graeber, R. C., Rosekind, M. R., & Samel, A. (1996). Principles and
Guidelines For Duty and Rest Scheduling in Corporate and Business Aviation. Flight
Safety Digest, 16 No.2(February).
Dumitru, I. M., & Boscoiau, M. (2015). Human Factors Contribution to Aviation Safety. In
International Conference of Scientific Paper AFASES 2015 (pp. 22–24). Brasov.
Retrieved from http://www.afahc.ro/ro/afases/2014/socio/henter_indreica.pdf
Federal Aviation Administration (FAA). 14 CFR Part 117 - Flight and Duty Limitations and
Rest Requirements: Flightcrew members, Pub. L. No. Docket No. FAA-2009-1093, 77
FR 398 (2012). U.S.A. Retrieved from www.faa.gov
Flight Safety Foundation. (2018). Aviation Safety Network. Retrieved August 28, 2018, from
https://www.aviation-safety.net/graphics/infographics/Fatal-Accidents-Per-Year-1946-
2017.jpg
Gander, P. H., Gregory, K. B., & Graeber, R. C. (1998). Flight crew fatigue II: Short-haul
fixed wing air transport operations. Aviat Space Env. Med, 69(9), B8–B15.
Gawron, V. J., French, J., & Funke, D. (2001). An overview of fatigue. In P. A. Hancock &
P. A. Desmond (Eds.), Stress, workload, and fatigue (1st Editio, pp. 581–595). Boca
Raton: Taylor and Francis.
Hair, J., Black, W. C., Babin, B. J., & Anderson, R. E. (2013). Multivariate Data Analysis:
Pearson New International Edition. Prentice-Hall, Inc.
https://doi.org/10.1038/259433b0
Ingre, M., Kecklund, G., Åkerstedt, T., Söderström, M., & Kecklund, L. (2008). Sleep length
as a function of morning shift-start time in irregular shift schedules for train drivers:
Self-rated health and individual differences. In Chronobiology International.
https://doi.org/10.1080/07420520802110704
International Civil Aviation Organization (ICAO). (1998). Human Factors Training Manual
(DOC 9683-A). Montreal: International Civil Aviation Organization (ICAO).
International Civil Aviation Organization (ICAO). (2013). Doc 9995 Manual of Evidence-
based Training (1st. ed.). Montreal: International Civil Aviation Organization (ICAO).
International Civil Aviation Organization (ICAO). (2017). Doc 9966 Manual for the
Oversight of Fatigue Management Approaches (2nd ed). Montreal: International Civil
Aviation Organization (ICAO).
Jackson, C. A., & Earl, L. (2006). Prevalence of fatigue among commercial pilots.
Occupational Medicine. https://doi.org/10.1093/occmed/kql021
Lee, S., & Kim, J. K. (2018). Factors contributing to the risk of airline pilot fatigue. Journal
of Air Transport Management, 67(July 2017), 197–207.
https://doi.org/10.1016/j.jairtraman.2017.12.009
Lerman, S. E., Eskin, E., Flower, D. J., George, E. C., Gerson, B., Hartenbaum, N., …
Moore-Ede, M. (2012). Fatigue Risk Management in the Workplace. Journal of
Occupational and Environmental Medicine.
https://doi.org/10.1097/JOM.0b013e318247a3b0
Levine, D. M., Stephan, D. F., Krehbiel, T. C., & Berenson, M. L. (2008). Statistics for
Managers using Microsoft Excel. (R. Nakosteen, Ed.) (5th ed.). Boston: Prentice Hall.
Marcora, S. M., Staiano, W., & Manning, V. (2008). Mental fatigue impairs physical
performance in humans. Journal of Applied Physiology, 106(3), 857–864.
https://doi.org/10.1152/japplphysiol.91324.2008
National Aeronautics and Space Administration (NASA). (1999). NASA Aviation Safety
Reporting System. An Analysis of Part 121/135 Duty Schedule-related Fatigue Incidents.
NTSB. (2016). NTSB Most Wanted List 2016. REDUCE FATIGUE-RELATED ACCIDENTS.
Petrilli, R. M., Roach, G. D., Dawson, D., & Lamond, N. (2006). The sleep, subjective
fatigue, and sustained attention of commercial airline pilots during an international
pattern. In Chronobiology International. https://doi.org/10.1080/07420520601085925
Phillips, B. A., & Danner, F. J. (1995). Cigarette Smoking and Sleep Disturbance. Addictive
Disorders and Their Treatment, 17(1), 40–48.
https://doi.org/10.1097/ADT.0000000000000123
Powell, D. M. C., Spencer, M. B., Holland, D., Broadbent, E., & Petrie, K. J. (2007). Pilot
fatigue in short-haul operations: Effects of number of sectors, duty length, and time of
day. Aviation Space and Environmental Medicine, 78(7), 698–701.
PT Garuda Indonesia (Persero) Tbk. (2018). Company Profile. Retrieved from
https://www.garuda-indonesia.com/id/en/corporate-partners/company-
profile/index.page?
Reinhart, R. O. (2007). Basic Flight Physiology (3rd ed). New York: McGraw-Hill.
Roach, G D, Reid, K. J., & Dawson, D. (2003). The amount of sleep obtained by locomotive
engineers: effects of break duration and time of break onset. Occupational and
Environmental Medicine. https://doi.org/10.1136/oem.60.12.e17
Roach, Gregory D., Sargent, C., Darwent, D., & Dawson, D. (2012). Duty periods with early
start times restrict the amount of sleep obtained by short-haul airline pilots. Accident
Analysis and Prevention, 45(SUPPL.), 22–26. https://doi.org/10.1016/j.aap.2011.09.020
Rood, G. M., & James, S. H. (2016). Noise. In D. P. Gradwell (Ed.), Ernsting’s Aviation and
Space Medicine (5th ed., pp. 747–767). Boca Raton: CRC Press.
Saddoune, M., Desaulniers, G., & Soumis, F. (2013). Aircrew pairings with possible
repetitions of the same flight number. Computers and Operations Research.
https://doi.org/10.1016/j.cor.2010.11.003
Salazar, G. J., & Caldwell, J. A. (2010). Fatigue in Aviation. Travel Medicine and Infectious
Disease, 3(May), 85–96. https://doi.org/10.1016/j.tmaid.2004.07.008
Sallinen, M., Härmä, M., Mutanen, P., Ranta, R., Virkkala, J., & Müller, K. (2003). Sleep-
wake rhythm in an irregular shift system. Journal of Sleep Research.
https://doi.org/10.1046/j.1365-2869.2003.00346.x
Shappell, S. A., & Wiegmann, D. A. (2000). The Human Factors Analysis and Classification
System – HFACS. DOT/FAA/AM-00/7 Office, DOT/FAA/AM, 15.
https://doi.org/10.1177/1062860613491623
Signal, T. L., Gale, J., & Gander, P. H. (2005). Sleep measurement in flight crew: Comparing
actigraphic and subjective estimates to polysomnography. Aviation Space and
Environmental Medicine. https://doi.org/10.1017/S0029665107005605
Smith, P. B. (1999). Predicting Process Difficulties in Multicultural Teams. In Diversity and
Group Effectiveness (p. 12). Sussex. https://doi.org/10.3102/00346543067001043
Spencer, M. B., Stone, B. M., Rogers, A. S., & Nicholson, A. N. (1991). Circadian
rhythmicity and sleep of aircrew during polar schedules. Aviation Space and
Environmental Medicine.
Stott, J. R. R. (2016). Vibration. In D. P. Gradwell (Ed.), Ernsting’s Aviation and Space
Medicine (5th ed., pp. 213–228). Boca Raton: CRC Press.
Taherdoost, H. (2016). Sampling Methods in Research Methodology; How to Choose a
Sampling Technique for Research. International Journal of Academic Research in
Management (IJARM), 5(2), 18–27. https://doi.org/10.2139/ssrn.3205035
Thomas, L. C., Gast, C., Grube, R., & Craig, K. (2015). Fatigue Detection in Commercial
Flight Operations: Results Using Physiological Measures. Procedia Manufacturing.
https://doi.org/10.1016/j.promfg.2015.07.383
Usman, H., & Sobari, N. (2013). Aplikasi Teknik Multivariate Untuk Riset Pemasaran.
Jakarta: Rajawali Pers.
Van Cauter, E., Holmbäck, U., Knutson, K., Leproult, R., Miller, A., Nedeltcheva, A., …
Spiegel, K. (2007). Impact of sleep and sleep loss on neuroendocrine and metabolic
function. In Hormone Research (p. 9). Chicago: Karger.
https://doi.org/10.1159/000097543
Yuliawati, I., Barus, S. B., Aryani, S., Taryanto, M., Mariska, I., Haryati, E. R., … Hasiholan,
O. (2015). Hasil Penelitian Balai Kesehatan Penerbangan tentang Faktor resiko
kelelahan pada personil penerbangan sipil dalam rangkapenerapan Fatigue Risk
Management System di Indonesia. Jakarta: Balai Kesehatan Penerbangan.
Yuliawati, I., Yuliana, Ningsih, K. U., Wahdah, N., Ariani, S., Khrisnapandit, I., …
Hasiholan, O. (2018). Hasil Kajian Balai Kesehatan Penerbangan tentang Fatigue Risk
Management System: Pengaruh faktor resiko kelelahan terhadap obesitas pada personil
penerbangan di Indonesia. Jakarta: Balai Kesehatan Penerbangan.