Project MARTHA: The Final Report

Full text

The Final Report
Project
MARTHA

Contents
2 | Project MARTHA Project MARTHA | 3
I. Introduction 4
II. Methodology and Research Design 6
III. The Results 8
III.1 Questionnaires and Interviews 8
III.2 Findings from the questionnaire data 12
III.3 Field Study 15
IV. Dissemination activities 23
V. Fatigue Risk Management: A Case Study 25
VI. Main Findings and Conclusions 29
VII. Future Research 31
Acknowledgements
The partners wish
to acknowledge the
following, without
whom, this project
could not have been
a success:
First and foremost, the
volunteer participants
who gave up their time to
complete questionnaires,
undertake weeks of
diary keeping and wore
actiwatches for long periods
in the service of the project
and for the benet of their
fellow seafarers.
Secondly, the companies
who participated. We
are indebted to the
two Chinese state-run
companies who operate
bulk carriers and tankers
for their kind help and
support. In Europe, we
were privileged to work
with the MF Shipping
Group (MFG) of Delfzijl
(‘Company A’) and the
Zodiac Shipmanagement
Co was (‘Company B’).
We appreciated the
friendly help and support
they gave us during the
project and we hope
that the experience has
helped them in their
understanding of fatigue
management of the
fatigue risk.
The many individuals,
fellow researchers and
maritime organisations
who have supported us
over the last three years.
Finally, our sponsors – the
TK Foundation who have
supported the project
thoughout and without
whom, we could not have
achieved so much.
The partners met many times
during the project and several
sta from each organisation made
contributions. The main researchers
were:
The Centre for Maritime Health
and Society, University of Southern
Denmark:
Dr Jorgen Riis Jepsen
Dalian Maritime University:
Dr Zhiwei Zhao
InterManager:
Capt Kuba Szymanski
Capt Paddy McKnight
Stress Research Institute, University
of Stockholm:
Dr Goran Kecklund
Wessel van Leeuwen
University of Southampton:
Dr Dominic Taunton
Dr Anne Hillstrom
Dr Tammy Menneer
Gemma Hanson (PhD student)
Stuart Pugh (PhD student)
Warsash Maritime Academy,
Southampton Solent University:
Professor Mike Barnett
Professor Claire Pekcan
Annette Dymond
Branimir Pantaleev
Capt Ivor Salter

4 | Project MARTHA Project MARTHA | 5
Fatigue at sea and related issues, such as stress and workload, are
highly topical and important areas of research as the problems
of mental health and wellbeing are being increasingly recognized
by society. Ships’ crews are under increasing pressure from
competitive voyage schedules and have to handle their tasks with
fewer crew members. Evidence from accident records and research
literature both point to the serious impact that sleepiness and
fatigue may have on the safety and welfare of seafarers.
The shipping industry has recognized the need for research in this area
and the ndings from the predecessor, project HORIZON, have had
a signicant impact already on the understanding within the shipping
industry of the importance of managing fatigue, both in terms of
sleepiness and also in its longer term psycho-social eects. The latter is
much less well-researched, and the results from the MARTHA project have
indicated that fatigue and stress increase for most crew as the voyage
length increases, and motivation decreases. Captains suer more than
their colleagues from both fatigue and stress. Port work is particularly
demanding: the results also show that no one onboard gets adequate
sleep, with the night watch keepers being particularly at risk of falling
asleep. High sleepiness levels can occur at any stage of the voyage but the
quantity and quality of sleep deteriorates over long voyages. The results
from the use of actigraphy have also conrmed many of the perceptions of
seafarers from their interviews and weekly diaries.
MARTHA was conducted by an international partnership of researchers
and industry. The $3 million project was sponsored by the TK Foundation
over a three year period from 2013 to 2016.
Introduction
The Partners
I.
The Stress
Research
Institute

6 | Project MARTHA Project MARTHA | 7
Methodology
and Research
Design
II.
The aim of the study was to explore the levels of sleepiness and the psycho-
social issues associated with long term fatigue and motivation, using a
sample of volunteer seafarers in the naturalistic setting of work onboard
their vessels.
Four shipping companies were selected to assist in the collection of data. Two
companies are managed in Europe: ‘Company A’ operates 43 small product tankers
In North West Europe. The vessels have an Intensive trading pattern with port calls
every three days. The eet is manned primarily with European ocers and Filipino
crew. ‘Company B’ manages 341 large container ships on liner routes which included
the Far East to Europe, and Asia to South America. The number of port calls in each
continent was interspersed with periods on oceanic passage. The ships were manned
with European ocers and mixed nationality crew. Two state-owned Chinese shipping
companies also took part in the project. ‘Company D’ operated 400 bulk carriers
trading worldwide, and ‘Company T’ operated 40 tankers in Far Eastern waters. Both
Chinese managed companies employed all Chinese ocers and crew.
•Questionnaires and interviews
with managers and seafarers in
the four shipping companies.
(Nearly 1,000 questionnaires
were completed).
•Onboard diaries of volunteer
seafarers from the four shipping
companies over a tour of
duty. (The highest number of
continuous weekly diaries was
17 weeks, but depending when
the diary started, they covered as
long as up to 6 months of a tour
of duty).
•Actigraphy data from selected
volunteers. The wearers were
requested to wear the watches
continuously for two weeks at the
start of their tour and two weeks
before signing o at the end of
their tour.
The project employed three main methods for data collection:
Before considering the ndings of the project in more detail, it is important
to distinguish between two separate but related phenomena: sleepiness
and fatigue. The research literature tends to blur the denitons, but the
following diagram provides an explanation of the major dierences:
Questionnaires
distributed to
four compaines.
Total: nearly 1000
questionnaires.
Interviews with
seafarers and
managers in both
China and Europe.
*Interviews, questionnaires and Study 1 conducted in both Europe
and China, but Study 2 only with European companies
1Numbers for both companies were
accurate at the time of the study
Analysis of results
from Study 1 and 2.
Development and
dissamination of
ndings.
Final reports and
publictaions.
Observation study of Masters and
selected crew for a tour of duty
(3-4 months).
• KSS and MPI readings, sleep
and stress levels through weekly
diaries
• Actiwatches
• Hours of work and voyage data
Repeat of study 1 on European
vessels.
Analysis of results of 110 seafarers
dierent ranks.
FRMS workshops, and developing
fatigue training, prediction and
reporting interventions.
MARTHA: The Research Plan – Europe*
Distinguishing Sleepiness
and Fatigue
2014
Surveys SurveysStudy 1 (2014) Study 2 (2015)
2015-2016
Short-term
Sleepiness Long-term
Fatigue
Healthy individuals
Rapid onset
Short duration
Single cause
Short-term eect on daily activities
May cause health disorders
(physical and mental)
Insidious onset
Persists over time
Multi-factor causes
Signicantly aects behaviour
and wellbeing

The Results
III.
8 | Project MARTHA Project MARTHA | 9
The project comprised of three studies carried out simultaneously. Each study
was led by a dierent partner and progress was discussed at regular partner
meetings. The results presented here are given in three separate sections
TOTAL = 937 questionnaires and 51 in-depth interviews
The questionnaires and interviews
with both European and Chinese
seafarers and managers had two
main objectives:
•To examine cultural dierences
in the interpretation of regulatory
frameworks on hours of work
and rest through interviews with
managers and seafarers:
• How do organisational practices
aect seafarer fatigue?
• What are the dierences
between Europe and China?
•To consider the ways in which
the incidence and eects of
fatigue, and the risk of injuries
and accidents at sea, can be
reduced.
Interviews and questionnaires
were conducted with seafarers and
managers as shown in the table to
the right:
III.1 Questionnaires and Interviews
The potential consequences
of fatigue
“If l stand in the same
posture for a long time,
l feel slightly that l am
unable to walk any
further. Although I am
young, I have had such
problems. Sometimes
I felt particularly
uncomfortable and
one of my feet can not
touch the oor. Once it
touched the ground, the
other foot would quickly
take over. This is caused
by long-time standing,
I guess.”
A Chinese AB from
Company D
“When I am tired, I feel fretful, taking extreme
views. I think even a good-tempered guy is easy to
have conict with others when he is fatigued”
A Third Engineer from Company B
Europe
China
‘Company A’ ‘Company B’ TOTALS
Questionnaires 314 140 454
Interviews 5 9 14
‘Company D’ ‘Company T’ TOTALS
Questionnaires 230 253 483
Interviews 20 17 37
Fatigue
Accidents
injuries
Disability
Ill health
Ship
casualties
Power
performance
Sick leave
•The health eects of fatigue
can be physical: for example,
muscle fatigue, body aches,
feeling uncomfortable or weak
in extremities following intensive
work or sustained posture.
Engineering seafarers also mentioned the problem of heat related
disorders in hot weather under intensive physical work.
The health eects of fatigue also cover mental fatigue, and there can be at
least four recognisable symptoms:
•Being fretful, irritable, unhappy and nding it easy to get into conict
with others;

10 | Project MARTHA Project MARTHA | 11
“The symptom of fatigue, for me, is tiredness
and slow response. When somebody told me
something, I nodded. One minute later, I forgot.
When I was reminded, oh, yes, I forgot …”
A Second Ocer from Company A
“Due to the typhoon on that day, other ships all
stopped. But our captain insisted on sailing. We
were very tired at that time actually. Sea water
poured into the engine room, but nobody noticed
that. When it was noticed, it was too late to take any
actions. The ship sank. Fortunately nobody died.”
An AB from Company T
•Slow responses, poor concentration and sleepiness
•Infections
•Psychological distress
•Genetic make-up
•Environmental inuences
•Gene-environment interactions
•Altered perceptions
• fatigue
• pain
•Neurocognitive changes
• concentration
• memory
•Mood alterations
• depression
• anxiety
•Sleep disturbances
•Incidents of insomnia and homesickness are more serious when
seafarers are fatigued.
The eects of sleepiness and fatigue can also be a signifcant and
contributory factor in accident causation, which can result in
environmental pollution, machinery damage and re:
•Altered cortisol regulation
(relative hypocortisolaemia)
•Altered vascomotor
regulation
• Abnormal blood
pressure responses
to postural change
• Dizziness
• Palpitations
•Immunological changes
• Cutaneous anergy
• Markers of immune activation
•Lymphoid organs
• Lymph node tenderness
• Sore throat
• Altered bowel habits
• Abdominal pain and
bloating
• Myalgia and arthraigia
T cells
Cytokines
“I heard that while
the ship was sailing
ahead, the ocer felt
so tired that although
he did not fall asleep,
he had already lost his
consciousness and
forgot to alter course.
The shallow water area
was just around the
corner and the ship was
stranded…”
A Third Ocer from
Company D
Immune
mediators
Hypothalamic-pituitary-
adrenal axis
Pituitary Adrenal
Heart and blood vessels
Neuroendocrine pathway
Sympathetic nervous
system pathway
Immune pathway
Sensory pathway
Immune system
Gastrointestinal tract
Musculoskeletal systekm
Risk factors
CNS symptoms
Neuro-
transmitters
Neuroendocrine
factors
1Source: Australian Clinical Practice
Guidelines (2002) http://www.tnq-support-
group.net/chronic-fatigue-syndrome-
australia-guidelines-2.html

12 | Project MARTHA Project MARTHA | 13
The Chronic Health Eects of Fatigue Dierences between ocers and ratings in their
perceptions of sleepiness, quality of sleep and stress
A comparison of age and
experience levels
between European and
Chinese seafarers
A comparison of dierent
sleep requirements
between European and
Chinese seafarers
Dierences
between ocers
and ratings in
actual versus ideal4
sleep length
III.2 Findings from the questionnaire data
Competing factors
onboard that contribute
to the adverse health
eects of fatigue
•Nutrition
·Quality
·Quantity
·Soft drinks
·Sweets
•Energy expenditure
·Sedentary work
·Exercise
•Smoking
•Alcohol
•Drugs
•Sedentary Work
•Sleeping disorders
·Insomnia
·Sleepapnoea
·hypopnoea
·Delayed/advanced
sleep phase syndrome
•Gastrointestinal
disorders
·Peptic ulcer
·Irritable bowel
syndrome
•Cardiovascular
disorders
·Myocardial infarction
·Stroke
·Hypertension
•Common infections
•Cancer
•Multiple sclerosis
•Mental disorders
·Depression
•Metabolic disorders
·Metabolic syndrome
·Hyperlipemia
·Diabetes
The results from the questionnaires showed that there were some
dierences between the ideal and actual sleep lengths for ocers
and ratings; their perceptions of sleepiness, quality of sleep and
levels of stress were quite similar to each other.
The perceptions are on a scale of 0 to 4, with the
higher numbers indicating worse sleep or stress. The
results suggest that ocers in general are a little more
tired at work, suer a lower quality of sleep and a
higher level of stress than ratings. These ndings are
supported by the results of the eld study, described in
the next section.
The main dierences were found to exist between
European and Chinese seafarers, indicating some
interesting dierences between the cultures of the
two dierent management styles. All companies
employed Asian crew, suggesting that these
dierences are a result of organisational culture
than national culture.
The main dierences were:
•Age and years at sea between crews in European and
Chinese companies;
•Sleep requirements between crews in European and
Chinese companies; and
•Perceptions of sleepiness, quality of sleep and
stress levels between crews in European and
Chinese companies.
Note the dierence between the experience of the
European crews compared to the Chinese crews, given
their similar ages.
Note the dierence between the European crews
compared to the Chinese crews, with regard to the
amount of sleep actually got as opposed to their ideal.
Note that this result suggests that ocers get a little less sleep than
they wish, and a little less than ratings do.
Av. sleep in a 24
hour period
Ideal sleep length
in a 24 hour period
Ocers 7.8 8.1
Ratings 8.4 8.4
Sleepiness at work Quality of sleep Stress at work
Ocers 1.7 1.2 1.2
Ratings 1.4 0.9 0.9
Companies Age Av years at sea
Europe “B” 39 14.6
“A” 38 13.1
China “D” 33 7.5
“T” 36 9.3
Companies
Av.
Sleep in
24 hour
period
Ideal5
sleep
length in
hours
Europe “B” 8.3 7.6
“A” 7.9 7.8
China “D” 7.7 8.6
“T” 8.4 8.8
Increased prevalence of several of these conditions has been demonstrated in seafarers3.
3Source: Poulsen, T.R. (2014) “Health of Danish seafarers and shermen
1970-2010: What have register-based studies found?” Scandinavian
Journal of public Health DOI: 10.1177/1403494814534538
4&5“Ideal” in this context refers to the personal preference for length of
sleep as expressed by the individuals in the questionnaire.

14 | Project MARTHA Project MARTHA | 15
A comparison of perceptions of sleepiness,
quality of sleep and stress between European
and Chinese seafarers
The perceptions are on a scale of 0 to 4, with the higher numbers
indicating worse sleep or stress. Note the dierence in stress between
Europeans and Chinese.
Chinese and European seafarers were also
asked about the factors which contributed to
their fatigue and sleepiness levels. Although
dierent priorites were given depending
on their nationality, the following factors all
featured highly:
•Job security
•Environmental
issues
•Job demands
•Sleep quality
•Irregular working
hours
•Rest hours
•New regulations
and more
requirements
placed on seafarers;
•Increased
inspections and
more paperwork;
•The bad
condition of ships’
accommodation
•The lack of proper
maintenance
•Work in port;
•Working onboard a
new ship;
•The quality and
professionalism of
work colleagues.
Issues which were repeatedly mentioned by
seafarers as contributing to their fatigue and
sleepiness levels were (in no particular order
of priority):
Companies Sleepiness at work Quality of sleep Stress at work
Europe “B” 1.5 0.94 0.66
“A” 1.3 0.92 0.75
China “D” 1.9 1.3 1.5
“T” 1.6 1.2 1.3
III.3 Field Study
The research design of the eld study is shown below. After
completing a background questionnaire, covering various aspects
of health and normal sleeping patterns, each volunteer was asked
to complete a diary on a weekly basis. All diaries were emailed
once a week for the duration of the seafarer’s tour of duty.
Selected volunteers were asked to wear an Actiwatch for two periods
during a tour of duty – the rst two weeks after starting the diry, and the
nal two weeks before departing the vessel at the end of the tour of duty.
European companies only
Study 1: April to July 2014
Study 1: April to July 2014
Study 2: May to August 2015
Study 2: May to August 2015
The data collected from each of the
110 particiapants in the eld study
included:
•Background questionnaire for
each volunteer;
•Weekly diaries (including KSS and
MFI6) for each volunteer;
•Actiwatch data for two weeks at
the beginning and end of tour for
selected volunteers;
•Voyage reports of vessel to cover
period of study; and
•Ocial hours of work/rest for
each individual volunteer.
Background
questionnaire
•BMI
•Working conditions
•Working hours
•Health behaviour
•Sleep habits
•Fatigue
•Sleep problems
•Stress
Weekly diaries
•Sleep (problems)
•Sleep disturbing
conditions
•Stress
•Sleepiness
•Fatigue
•Feelings of
depression
•General health
•Working conditions
Weekly questionnaires
Time at sea (≥ 8 weeks)
Background
questionnaire
Actigraphy
(2 w)
n=4/ship
Actigraphy
(2 w)
n=4/ship
Who took part? What Questions
were asked?
‘Company A’ – small product tankers
‘Company B’ – large container ‘vessels
Ship Capt C/0 2/0 3/0 C/E 2/E 3/E AB Cook Total
477245217439
Ship Capt C/0 2/0 3/0 1/E 2/E Elect Bosun MM AB Cook Total
44112422342126
Ship Capt C/0 2/0 3/0 C/E 2/E 3/E AB Cook Total
243431307126
Ship Capt 2/0 3/0 C/E 1/E 2/E Bosun MM AB Total
223221112519
6KSS = Karolinska Sleepiness Scale
MFI = Multidimensional Fatigue Inventory

16 | Project MARTHA Project MARTHA | 17
Results from the Background questionnaires
The background questionnaires sought a number of factors which could aect
fatigue levels. The graphs below show some of the signicant factors:
The average number of hours per week which
were recorded by all crew onboard is 67 hours per
week. There is not a big dierence between the two
trading companies. It should be noted, however,
these gues do not include overtime hours and
include all ranks onboard.
20
1950 1960 1970 1980 1990 2000
15
10
5
0Year of birth
Number of participants
Mean = 1974-sep-23
Std. Dev. = 3,710E3 days
N = 109
Age (birth date) of the participants:
Body Mass Index (BMI)
Category
from to
Very severely underweight 15.0
Severely underweight 15 16
Underweight 16 18.5
Normal (healthy weight) 18.5 25
Overwieght 25 30
Obese Class I (Moderately
obese) 30 35
Obese Class II (Severely obese) 35 40
Obese Class III (Very Severely
obese) 40 +
BMI (kg/m2)
15 20 25 30 35 40
0
0
4
4
8
8
12
12
BMI
BMI
Company ACompany B
Number of participantsNumber of participants
Normal working hours per week
24 2420 2016 1612 128 84 40
10
20
30
40
50
60
70
80
90
100
110
120
130
Number of participants
Company B Company A
Fatigue: is it higher or lower at the end of a voyage?
With regard to all 110 seafarers in the sample, across both shipping
companies and including seafarers of all ranks, the majority of seafarers
(61%) consider that they are more fatigued at the end of a voyage than at
the beginning, irrespective of the actual length of the voyage.
If this result is studied more closely, we nd that this perception varies
considerably depending on what rank is considered and the types of task
carried out onboard.
The study considered three groups of crew on the vessel: the Captain;
Watch Keeping Ocers; and day workers. Each group contained at least
10 members.
The three pie charts (opposite) illustrate the ndings.
The group representing day work crews, cooks and engineers experience
less fatigue at the end of the voyage than other groups, with a majority of
the group being less fatigued or the same by the end of their tour.
The watch keeping group was represented by Second and Third Deck
Ocers. A small majority of ocers reported they felt more fatigued at the
end of their tour than at the beginning.
The group representing the Masters presented a very dierent picture,
with a very large majority of the group reporting that their fatigue levels
were higher at the end of their tour of duty than at the beginning.
The sample of 110 seafarers asked if crew experienced more stress at
the end of the voyage than at the beginning. Taking the group as a whole,
nealry half the seafarers did experience higher levels of stress by the end
of their tour of duty, as shown in the diagram to the left.
Lower at the end
Higher at the end
Same
8
61
36
Able Seaman
Lower at the end
Higher at the end
Same
Third ocer
Lower at the end
Higher at the end
Same
Captain
Higher at the end
Same
10.28%
48.60%
41.12%
Lower at the end
Higher at the end
Same
_

18 | Project MARTHA Project MARTHA | 19
Results from the Weekly Diaries
The Circadian type, sometimes also
referred to as the ‘diurnal type’ or
‘chronotype’, is the propensity of
individuals to prefer to sleep in
the morning or evening. Although
most people are neither extreme
morning or evening types, at
least 50% of people will recognise
themselves as a “lark” or an “owl”.
In the sample, a majority of
seafarers identied themselves
as morning types, which this is
unusual among shift workers
where most night shift workers are
evening types. The reason for this
dierence is not known, but may be
signicant if individuals are extreme
and are put on watches opposite
to their natural inclinations. For
example, an extreme morning
person on a late evening watch.
Participants were asked to complete a weekly diary for
each week during their tour of duty. In some cases,
the number of diaries was in excess of 10 weeks.
The maximum number of returned diaries was for a
continuous 17 week period.
The daries were returned by email on a weekly basis to
the Project Manager who could liaise with the Captains
about any additional relevant voyage information. In this
respect, each Captain was a “champion” for the project,
motivating other crew to participate by their own
example. This strategy worked well, resulting in a very
high return rate of continuous weekly data.
In some cases, the diaries were not started at the
commencement of an individual’s tour of duty and
dierent periods of time remained after the diaries
were stopped until the end of the contract. The
periods for data collection were determined by the
Captain’s time onboard. It was therefore important for
“anchoring” the results to the same portion of a voyage,
to also know the dates of the start of contract and the
end of contract date for each individual.
In addition to general questions about the individual’s
sleep, two specic and validated measures of
sleepiness and long term fatigue were employed:
the Karolinska Sleepiness Scale (KSS) and the
Multidimensional Fatigue Index (MFI).
The KSS explores subjective feelngs of stress on a
scale of 1 to 9. It has been validated against EEG and
other variables7. Individuals were asked to measure
their sleepiness levels for the 24 hour period before
completion of the weekly diary.
Circadian type
25 2520 2015 1510 105 5
0
a denitie
evening person
a denitie morning
person
more evening than
morning person
more morning than
evening person
neither a morning nor
an evening person
Number of participants
Company B Company A
1. Extremely alert
2. Very alert
3. Alert
4. Quite alert
5. Neither alert nor
sleepy
6. Some signs of
sleepiness
7. Sleepy, no eort
to stay awake
8. Sleepy, some
eort to stay
awake
9. Very sleepy, great
eort to keep
awake, ghting
sleep.
KSS – The Karolinska
Sleepiness Scale
The diagram below shows the
KSS scores over 7 for all 110
seafarers who completed them
at dierent stages of the voyage.
The scores were calculated using
an anchoring mechanism, so
they show comparable scores at
various weeks into a voyage for
each individual. Consequently, the
KSS scores cover from week 1 to
over 6 months.
From previous research, a KSS
score over 7 indicates a high risk of
falling asleep. The most signicant
results, which this diagram
illustrates, are:
•The risk of falling asleep through
tiredness is present at ALL stages
of the voyage, making it a safety
risk at all stages of the voyage;
and
•Very high levels of sleepiness
(KSS of 8 or 9) are apparent
and increasing after 6 months
onboard.
The Multidimensional
Fatigue Inventory
(MFI)8 measures
fatigue using four
statements for each
of the following
dimensions:
•General fatigue
•Physical fatigue
•Mental fatigue
•Reduced activity
•Reduced motivation
From the dierent dimensions of
fatigue, the results in this study
indicate that it is motivation that
decreases with time at sea.
This is a signicant nding because
it oers an explanation for recent
reports of casualties occuring on
vessels where the crew, including
the Captain, have been onboard for
longer than 6 months.
Reduced motivation may lead to
complacency, individuals taking
short cuts and “work-arounds”
and not following the correct
procedures.
0.0
0.5
1.0
1.5
2.0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31
Weeks at Sea
Mean number of ratings per 24h period
KSS9
KSS7
KSS8
Fatigue
dimension
Statements
like (positively
and negatively
formulated)...
General
fatigue
I feel t
I feel tired
Physical
fatigue
Physically I can take
on a lot
Physically I feel I am
in a bad condition
Mental
fatigue
I can concentrate well
My thoughts easily
wander
Reduced
activity
I feel very active
I get little done
Reduced
motivation
I have a lot of plans
I dread having to do
things
7Kaida et al (2006) “Validation of the Karolinska
sleepiness scale against performance and EEG
variables”. Clinical Neurophysiology Vol117,
Issue 7, July 2006, pp 1574–1581
8Smets, E.A. et al (1995) “The
Multidimensional Fatigue
Inventory (MFI) psychometric
qualities of an instrument
to assess fatigue” Journal of
Psychosomatic Research Vol
39, Issue 3, April 1995, pp
315–325

20 | Project MARTHA Project MARTHA | 21
Results from the Actigraphy
Examples of Actiwatch data:
A Bosun on daywork above and below a Captain
A total of 70 European seafarers took part in the Actiwatch study over
the two years of the eld study. These included 15 Captains, 19 Watch
Keeping Deck Ocers, 13 day-working Engineers and 23 Ratings. Although
some Chinese ocers also took part, logistical problems with transporting
the valuable scientic instruments meant that insucient numbers were
collected to make the results statistically signicant and so detailed
analysis of their data is not included in this report.
32 Actiwatches were deployed on the 12 participating vessels and
volunteers were asked to wear them continuously for two weeks at both
the beginning and the end of their tours of duty. Watches were rotated
between crew members to optimise data collection.
The watches are accelerometers so they record movement and can
indicate the dierence between wake periods and sleep perods. To help
the analyst, volunteers were requested to press a marker button each
time they woke up and before falling asleep to produce a blue line on the
graphs to make clearer the dierent periods of sleep and wakefulness.
Software is able to analyse various parameters such as total amount
of sleep in a 24 hour period and wake bouts during a period of sleep.
This latter measure is useful to provide some indication of the quality of
sleep obtained.
The graphs illustrated below and
opposite show the kind of data
which is produced. The rst shows
a typical day-worker, who is able to
get regular and adequate periods
of sleep at night. The second graph
shows a Captain, with evidence of
more disturbed sleep and irregular
patterns of sleep.
It is important to remember that
Actiwatches measure the amount
of sleep obtained, and not how
tired someone may feel, nor long
term fatigue and stress levels.
The analysis of the actigraphy provides objective
data and is a valuable conrmation of the
surveys and questionnaires conducted in the
eld study.
Analysis is ongoing, but some important ndings
have already been conrmed:
Both the amount of sleep and the quality of sleep
(the latter as measured by wake bouts) decreases
over time for all crew. This nding supports
the KSS numbers showing increasing levels of
sleepiness over time.
The Master pressed the event
marker before going to sleep
Period of sleep
Period of work
The actiwatch was set up to start recording on 04 May at midnight BST.
The Bosun started wearing it at midnight local time in Manzanillo (Mexico)
The Bosun pressed the event marker button
The Bosun switched o the light and went to sleep
Period of sleep Period of work
The Bosun remembered to press the
event marker an hour after waking up
0 20 40 60 80 100
0.0
5.0
10.0
15.0
20.0
25.0
Day of study
Linear Regression
R2 Linear = 0.014
All sleep today

22 | Project MARTHA Project MARTHA | 23
A study of dierent ranks shows that it is the night
time Watch Keeper (typically the Second Ocer) who
obtains less sleep than colleagues. This is illustrated in
the graph below:
In addition to the funding of the project, the TK
Foundation also contributed to the funding of two PhD
students, registered at the University of Southampton.
Their studies are continuing beyond the completion
of the MARTHA project, but their studies will provide a
further legacy of the project.
One PhD student is investigating fatigue and
distractions in maritime control rooms. The study will
explore the relationship between fatigue, attention
and safety and investigate the eects of dierent
kinds of distractions on attention, and how fatigue
impacts attention. Experimental laboratory work will be
conducted with sleep restricted groups.
The second PhD student is examining the eects
of stress on attention and visual cognition. Existing
research indicates that stress and fatigue can inuence
each other. This study will examine stress in its ability
to both improve and impair attention through the
experiments using an Attentional Network Test.9
The eect of wake bouts is shown in the graph below. It indicates a signicant increase,
as time (measured in weeks) passes. Increased wake bouts may be seen as an
indication of more disturbed sleep as the voyage progresses. The data is “anchored”, so
it shows the increase from the start of the contract to at least 24 weeks.
0 4 8 12 16 20 24 28
15
20
25
30
35
Weeks since start of contract (1 = rst week)
Number of wake bouts
5.60
5.80
6.00
6.20
6.40
6.60
6.80
7.00
Captain CO 2O 3O CE 1E 2E 3E Crew
Rank
Mean of all sleep today
Dissemination
activities
IV.
The MARTHA partners agreed a publication policy at the start of the project
and the following post-project dissemination activities were agreed at the
nal partner meeting in June 2016. They provide a variety of formats for
publishing the results and leaving a legacy for MARTHA in the future.
Publications
•During the course of the project, a number of
academic papers, conference presentations and
articles for the nautical media were produced.
•To date, these outputs have mostly focussed on the
questionnaire and interview data, undertaken with
the employees of Chinese and European seafarers.
Some of these papers have already been translated
into Mandarin and vice-versa. Various papers from
the eld study and the results of the actigraphy
analysis will follow in due course.
•Presentations on the MARTHA project have also
been given at seminars with both European shipping
companies, as well as at several conferences and
seminars during the project.
•In addition to this industry report, it is also planned
to provide a written submission on long term fatigue
to the IMO. Finally, it is hoped that one legacy of
MARTHA will be the provision of a public website for
fatigue knowledge transfer.
II-2 Events and Workshops
•A number of workshops were
held to obtain feedback on the
ndings of the project and also
to explore the fatigue issues
of most concern to seafarers
and shore-based managers.
Workshops were held in:
•Athens, June 2016
•Warsash, June 2016
•Singapore, October 2016
•Manila, November 2016
The workshops all proved useful
in identifying issues which
participants thought might
cause fatigue onboard as well as
providing some ideas to mitigate
the risks of fatigue onboard
ships. The format was exible
and interactive and the dialogue
focussed on the following two
themes:
•What are the most signicant
causes of lack of sleep and/or
fatigue onboard the ship?
•What are the recommendations
for mitigating lack of sleep and
fatigue and their consequences
in the future?
The responses from the workshop
delegates can be grouped into three
main categories:
•Vessel design and living
environment
•Working conditions; and
•Operational issues.
9Fan, J., McCandliss, B. D., Sommer, T., Raz, A.
& Posner, M. I. (2002). Testing the eciency and
independence of attentional networks. Journal of
Cognitive Neuroscience, 14, 340-347.

Vessel design and living environment
Participants sought improvements in:
•Noise and Vibration levels
•Temperature
•Quality of Accommodation spaces
•Bedding (eg change of mattresses)
•Exercise facilities onboard
Some of these issues are covered in the Maritime Labour
Convention (MLC 2006) to apply to new vessels and measures
will need to be evaluated further, as the requirements come
into force.
Working conditions
Participants sought improvements in:
•Safe manning levels
•Nutrition and good food aboard
•Hours of work and rest
•Stress onboard through harassment and bullying
Operational issues
Participants sought improvements in:
•Being relieved on time and having a KPI to measure it
•Revision of company reporting requirements in order
to reduce bureaucracy
•Communication between ship and shore
•Logistics-port calls to be better organised and
discussed with sea sta
•Timings of inspections onboard by external parties
•Time management. For example: the timing of
Notices of Readiness
•Recovery time during the voyage. For example,
going to anchor.
Participants also recognised that there needs to be
a cultural change in the industry’s attitude towards
fatigue by both seafarers and shore management. The
response: “but it’s always been like this” was no longer
seen as acceptable.
Awareness and cultural change also apply to the
agencies ashore who interact with ships and personnel:
charterers, agents and port state ocials.
24 | Project MARTHA
Fatigue Risk
Management:
A Case Study
V.
Recent studies10 in several safety-critical industries reveal a conceptual
move away from prescriptive regulations - which seek to mitigate the risk of
fatigue through limiting the hours of work, to a more goal-based system - that
involves the employment of Fatigue Risk Management Systems (FRMS). The
systems approach oers an integrated management of fatigue which covers
policies, operational aspects and quality assurance.
FRMS are being adopted in other transport sectors - such as: aviation, road and rail
systems - but evidence from recent marine accident investigations indicates that
the use of FRMS in the shipping industry is less mature than in other safety-critical
transport industries, and less advanced in exploring such concepts in practical
operational settings.
10Gander, P.H.(2015) Reviews of Human
Factors and Ergonomics,
Vol. 10, 2015, pp. 253–271
Project MARTHA | 25

The KPIs on ‘Lost Time Injury Frequency’ (LTIF) and
‘Work and Rest hours Non-conformities’ are illustrated
below by kind permission of the company. The
numbers in the graph below indicate the number of
LTIs per quarter, averaged over the year.
Target for 01f. Lost Time Frequency:1.60
2011 2012 2013 2014 2015 2016
5
4
3
2
1
Calendar Year
KPI
4.46
3.31
0.78
1.57
1.55
0.00
After discussions with the management team, a fatigue
incident report form was introduced, and changes made
to the eet orders on fatigue. One of the key changes was
the introduction of a new Key Performance Indicator (KPI)
for work and rest hours.
Although reporting of fatigue is still an issue to be
tackled, the following quote from the ship’s managers
illustrate the engagement with the materials and
general success of the intiative.
“The publications are very clear and
are in use by our seafarers. We have
not received any fatigue incident report
yet. We are quite happy with our KPI’s
over 2016 with no incidents and LTIF
of zero (which is the rst time since
2011). Also recorded observations with
regard to work and rest hours shows a
considerable decrease since 2013.”
12/12/2016
Project MARTHA | 27
The core elements of a FRMS are:
1. Fatigue Awareness training and cultural change
programmes;
2. A fatigue reporting system within a just culture; and
3. Data-driven analysis for operational fatigue risk
assessment, workload management and monitoring of
adequate sleep for those onboard.11
11Data collection tools may include sleep
diaries and surveys, scientic data, for
example, actigraphy, and fatigue
prediction models
Evaluation of the eectiveness of such systems in
other industries has highlighted a number of issues to
be resolved in the successful implementation, one of
the major challenges being the acceptance of these
systems by the work force.
Changing the culture in shipping, represents a major
challenge, both for individual seafarers and for shore
management. The rst important step is raising
awareness and this can be achieved by simple yet
eective messages about managing fatigue through
guides and booklets.
Fatigue incident reporting is another important element
of the systems approach, and needs to be part of a
transparent and “blame-free” culture. Employees will be
reluctant to report incidents which may be caused by
sleepiness or general fatigue if they think that there will
be recriminations.
The third core element – the use of data in assessing the
sleepiness and fatigue levels of employees is probably the
part most open to misunderstanding. Although in other
industries, like aviation, sytems have been developed to
a high degree. One of the features of FRMS is that it can
be part of the continuous improvement cycle of a Safety
Management System (SMS).
An example of this approach is the ‘maturity model’
concept, which allows for dierent levels of engagement
as the system grows and is accepted by the workforce.
The concepts here are similar to that employed in the
Tanker Management Self Assessment (TMSA) Code, and
may t well within this kind of approach. Starting with
simple messages and guidance can build to a more
interactive approach where schedules can be predicted
using biomathematical prediction tools.
At a higher level of maturity, seafarers can take more
ownership of the system themselves by reporting incidents,
and keeping a check on their own and colleagues fatigue
levels. The company may also consider periodic reviews of
crew fatigue through sleep diaries and other techniques,
similar to those used in the MARTHA project. The technology
exists now to combine the power of big data and predictive
analysis with the science underpinning fatigue, stress, health
and wellbeing to provide better health and welfare services
to seafarers wherever they may be.
The MARTHA project engaged with both European
management companies to explore some of the issues
involved in setting up a FRMS. In one of the companies,
raising awareness was achieved with presentations on
the project at the annual Captain’s seminars, where
considerable interest in the topic was shown by
the participants.
This interest was echoed throughout the project in which
seafarers seemed keen to pariticipate and contribute
their views on this subject.
26 | Project MARTHA

Main
Findings and
Conclusions
VI.
Project MARTHA | 29
The number shown in the graph below is the average
number of received non-conformity reports per quarter
for a given year.
The experience of being involved in a fatigue research
project was generally very positive as the quote below
from one of the companies illustrates:
“We had several objectives for
our participation in the MARTHA
project, including wanting to learn
more about fatigue and proactive
fatigue prevention, to contribute
to a scientic project on this
important topic, and to engage our
crew on a subject that has direct
relevance for them. We believe that
the project has successfully met
our objectives, and we have learnt
more about fatigue, its causes and
preventative measures. We have
shared this knowledge with our
crews through a fatigue awareness
and prevention initiative as part of
our Occupational Health and Safety
Management System (OHSAS 18001).
This included sharing some of the
materials we received during the
MARTHA Project, through onboard
training sessions and visits,
seminars and onboard discussions.
We have also incorporated more
benecial guidance into our
management system to ensure that
the information remains readily
available. We gained a lot from
participating in the project and
believe the knowledge gained will be
benecial to the shipping industry”.
Target for 17. Work / Rest hours 2.00
2011 2012 2013 2014 2015 2016
2.5
2
1.5
1
0.5
Calendar Year
KPI
0.00
1.25
1.75
2.00
1.00
28 | Project MARTHA
The main ndings from the three studies are:
•From a search of current research literature, the
causes and consequences of both sleepiness and
long-term fatigue are now well-established (pages
10-11).
•Although European and Chinese seafarers may
allocate dierent priorites to fatigue factors, they do
share the same perceptions about the major factors
that inuence fatigue onboard.
•There is evidence of higher levels of fatigue and stress
in seafarers from Chinese-managed companies
than European managed ones. This suggests that
dierences in organisational factors are signicant in
aecting fatigue mitigation onboard.
•Long-term fatigue levels are perceived dierently
depending on the nature of the work onboard.
Captains and Watch Keepers appear to fare worse
than day workers.
•Both fatigue and stress levels are perceived as higher
at the end of a voyage than the beginning
by most crew, and port work is seen as more
demanding than work at sea.
•Sleepiness levels vary a little during the voyage,
suggesting there are opportunities for recovery. Overall,
there is a small but signicant decrease in the amount
of sleep in a 24 hour period over the course of time.
•Captains and day workers get more sleep than watch
keepers, but Captains are more at risk of fatigue than
other ranks. Night watch keepers (Second Ocers)
get signicantly less sleep than others.
•Motivation decreases the longer seafarers are away
from home. This has important implications for safety,
as it may be a signicant underpinning root casue of
phenomena such as complacency, short cuts, and not
following procedures.
•Although the amount of sleep experienced by
seafarers during a voyage may stay the same or
decrease slightly, there is evidence that sleep quality
becomes more disturbed over time – there are more
wake bouts and fragmented sleep.

Project MARTHA | 3130 | Project MARTHA
In conclusion, the evidence from three dierent sources of data: 937
questionnaires, 110 weekly diaries and 70 Actiwatch graphs point in the
same direction:
1. Both sleepiness and fatigue are important issues for seafarers and
managers: they both have safety and long-term physical and mental
health implications;
2. Long tours of duty (over 6 months) may lead to increased sleepiness,
loss of sleep quality and reduced motivation. Any of these outcomes
could result in ‘near-misses’ and accidents onboard;
3. Night watch keepers are most at risk from falling asleep on duty;
4. Captains feel stressed and fatigued at the end of their tours of duty and
need recovery time.
5. There are simple operational solutions which can ensure sleep is easier
for those onboard through fatigue risk management. These solutions
should involve seafarers and agencies ashore which impact on shipboard
operations.
6. The introduction of Fatigue Risk Management Systems, as already
used in other safety-critical transport systems, presents an integrated
systems approach to managing the risk of fatigue. It requires ownership
by all in the company, changes in culture and can be introduced
in a gradual process as the company develops its own approach.
The development of new data collection, transmission and analysis
techniques will accelerate the process.
7. In the longer term, improved vessel design will make a signicant
impact in reducing the eects of sleepiness and fatigue.
Future
Research
VII.
The diagram below illustrates some of the main ndings of the MARTHA
project, and also shows some of the new questions raised by this research.
Sleepiness
and
Fatigue
Levels
Months at Sea
3 6
What is the optimum
voyage length?
Sleepiness recovery periods
through time at anchor etc
How long should
recovery be
between tours
of duty?
What is the real shape
of this fatigue curve?
•What is the optimum tour of
duty length? Should there be a
maximum shorter than the
MLC requirement?
•How long should recovery time
between voyages be?
•How does cognitive
performance deteriorate over
time due to fatigue and stress?
•How does “mood” change
over time? Does this have
a signicant eect on the
psychological wellbeing
of seafarers?
Other areas of research include
the further development of
FRMS concepts for the shipping
industry.
Specic goals are:
•The development of improved
fatigue prediction models.
•The development of instruments
to survey psychological wellbeing
over the long term
•The development of models
of how long term fatigue and
recovery may be predicted
Some of the questions to answer include:

For more information contact
secretary@intermanager.org