493
1 INTRODUCTION
Over the past years, container ships have steadily
been growing in size cf. [3]. This development is
mainly caused by a higher demand, which requires
these container vessels to transport more and more
goods. But the constant lengthening, widening, and
deepeningoftheseshipsisbecomingahuge
problem
whenitcomestosafelymaneuveringinharborareas.
”Mostshipcollisions,allisions,andgroundingsoccur
inharbors,becausethatiswherenavigationbecomes
restricted byland, shallow water, other vessels, and
man‐made structureslike jetties,bridges, and piers”
(cf. [5]). To ensure safety in harbor maneuvers, all
relevantcountriesintheworldareobligatedtohave
pilots(e.g.,inGermanycf.[7]).Pilotsareexpertsfor
specific areas and support the ship crew with
important information about water depths, currents,
andmaneuvers.Eveniftheresponsibilityfortheship
remains with the captain, pilots often take over
control
and give steering commands directly to the
helmsman.Therefore,thepilothastobeawareofthe
currentsituation,includingallenvironmentalfactors
to avoid collisions or grounding. Understanding the
situationisofhighimportancebecauseeacherrorhas
the potential to cause a collision or grounding [11].
However, accident
reports reveal that pilots have
problemsmonitoringtheirenvironmentssuccessfully.
As a result, accidents happen on a regular basis as
showninthiswork.
Inthiswork,wewanttoinvestigatecontainership
berthing from the pilot’s perspective to identify
situationsthatarecriticalandlacktechnical support.
Therefore, we ask
the following research questions:
(RQ1) What were themost common causesfor container
shipaccidentsinharborareasinthepast?(RQ2)Whatare
theproblemspilotshavetofaceduringtheberthingprocess
ofcontainershipsnowadays?
Investigations on Container Ship Berthing from the
Pilot’s Perspective: Accident Analysis, Ethnographic
Study, and Online Survey
U.Gruenefeld,T.C.Stratmann,Y.Brueck,A.Hahn&S.Boll
UniversityofOldenburg,Oldenburg,Germany
W.Heuten
OFFIS‐InstituteforIT,Oldenburg,Germany
ABSTRACT:Inrecentyears,containershipshave had to transportmoreand more goods duetoconstantly
growingdemand.Therefore,thecontainershipsforcarryingthesegoodsaregrowinginsize,whiletheharbors
fallshortinadaptingtothesechanges.Asaresult,the
berthingofthesecontainershipsinharborshasbecome
more challenging for harbor pilots. In this work, we identify problems and risks with which pilots are
confronted during the berthing process. First, we analyzed approximately 1500 accident reports from six
different transportation safety authorities and identified their major causes. Second,
we conducted an
ethnographicstudywith harbor pilots in Hamburgtoobserve their actions. Third, wegainedmore specific
insights on pilots environments and communications through an online survey of 30 harbor pilots from
differentEuropeancountries.Weconcludeourworkwithrecommendationsonhowtoreduceproblemsand
risksduring
berthingofcontainervessels.
http://www.transnav.eu
the International Journal
on Marine Navigation
and Safety of Sea Transportation
Volume 12
Number 3
September 2018
DOI:10.12716/1001.12.03.07
494
Table1.Classificationofaccidentreports(percentagereportstotalpercentageofallaccidentsforgivenauthority).
__________________________________________________________________________________________________
AuthorityReports
__________________________________________________________________________________________________
Name CountryPeriodTotal Containerships inharbors inrivers inopensea
__________________________________________________________________________________________________
MAIB GreatBritain 1989‐2015 519 94(18.1%)28(5.4%) 22(4.2%) 44(8.5%)
TSBC Canada1990‐2015 395 18(4.6%)4(1.0%) 10(2.5%) 4(1.0%)
USCG USA1956‐2015 193 13(6.7%)3(1.6%) 8(4.1%) 2(1.0%)
BSU Germany 2008‐2015 177 59(33.3%)21(11.9%) 24(13.6%)
14(7.9%)
NTSBUSA1994‐2015 140 1812.9%)7(5.0%) 9(6.4%) 2(1.4%)
ATSBAustralia 1987‐2015 68 21(30.9%)13(19.1%) 1(1.5%) 7(10.3%)
__________________________________________________________________________________________________
1492 223(14.9%)76(5.1%) 74(5.0%) 73(4.9%)
__________________________________________________________________________________________________
To answer our first research question (RQ1), we
identified problems and risks with which harbor
pilots were confronted during the berthing of
container ships in the past. We identified these
problemsandrisksbyanalyzingapproximately1500
accident reports from six different transportation
safety authorities. Here, we identified the accident
reports
involvingcontainershipsinharborareasand
categorized them with respect to the most common
causes of these accidents. To answer our second
research question (RQ2), we conducted an
ethnographicstudyintheharborofHamburg.Using
theresultsfromouraccidentanalysis,wewereableto
identify critical aspects
of the berthing process that
areproblematictothesameextenttodayastheywere
inthepast.However,wealsoidentifiedproblemsthat
becamemorerelevantinrecentyears.Toensurethat
theresultsfromourethnographicstudyaregenerally
valid,weconductedanonlinesurveywith30harbor
pilots from different European countries. This gave
furtherinsightsintothesecriticalaspects.
2 ACCIDENTREPORTANALYSIS
Knowing the past is the first step to understanding
the present and taking actions for the future.
Therefore,ourfirstresearchquestion(RQ1)addresses
thepastbylookingintothemostcommoncausesfor
container ship accidents in harbor areas. To
understandthesemostcommoncausesforcontainer
ship accidents in harbor maneuvers we looked into
accident reports of different authorities. However,
analyzing accident reports to gather insights is an
usualmethodinthemaritimeresearch.[12]analyzed
100 accidents and found that the
major types of
humanerrorthatcontributetoaccidentsaretheresult
of wrong habits, wrong diagnoses, lack of attention,
lack of training, and unsuitable personality. In
general,about75‐96%ofmarinecasualtiesarecaused,
atleastinpart,bysomeformofhumanerror[9].This
is understandable, since
a maritime system is a
system of people. To systematically investigate
accident reports, several different methodical
approaches were presentedin related work (e.g., [8]
or[4]).Morerecentresearchshowsthat,overthelast
decade, the human error could still not be reduced
andthata lackofsituationawareness
stilldominates
most accidents caused by human error [1]. The
importance of situation awareness was already
identified in 2002 by [6]. Recently, Stratmann et al.
investigated what negatively influences situation
awareness, recognizing that most errors occur at
situation awareness level 1, the assessment of the
current situation [10]. Compared to previous
work,
we concentrate on a subset of accident reports that
involve container ships in harbor areas. Here, we
assessedwhatkindsofaccidentshappenedandwhat
causedtheseaccidents.
2.1 Corpus
Weanalyzed1.492maritimeaccidentreportswritten
in English from six transportation safety authorities
betweentheyears1956and
2015.There‐portswere
gathered from the British Marine Accident
Investigation Branch (MAIB)
(https://www.gov.uk/government/organisations/mari
ne‐accident‐investigation‐branch, last retrieved
September16,2018),theTransportationSafetyBoard
of Canada (TSBC)
(http://www.tsb.gc.ca/eng/rapports‐
reports/marine/index.asp,lastretrievedSeptember16,
2018), the United States Coast Guard (USCG)
(https://www.uscg.mil, last retrieved September 16,
2018), the Bundesstelle
für Seeunfall Untersuchung
(BSU) (https://www.bsu‐bund.de/EN, last retrieved
September 16, 2018),the American National
Transportation Safety Board (NTSB)
(http://www.ntsb.gov/investigations/accidentreports/
pages/accidentreports.aspx, last retrieved September
16, 2018), and the Australian Transportation safety
Board (ATSB) (https://www.atsb.gov.au/marine, last
retrieved September 16, 2018).We crawled the
accidentreportswithscriptsfrom[10].
2.2 Statistics
Depending on the national
traffic of the
transportation safety authorities, acci‐ dents with
different types of vessels are reported (e.g., fishing
vessels, passenger ships, bulk carriers, tankers, etc.).
However, all authorities that we included in our
accident analysis are investigating accidents with
containerships.There‐fore,weidentifiedallaccident
reportsthatinvolvedcontainerships,
whichwere223
outof1.492accidentreports(14.9%)(seeTable1).We
crawled all published accident reports from all
transportation safety authorities, deciding not touse
predefined time windows. Therefore, the periods of
these reports vary among the different authorities
(e.g., the accident reports from USCG start in 1956,
although container ships were not existing at that
time).
Sinceourfocusisinvestigatingharbormaneuvers,
wedividedthecontainershipaccidentsintodifferent
categories (in harbors, in rivers, in open sea). These
495
categoriesrevealthatcontainershipaccidentshappen
moreofteninharborareasthaninriverorsea areas,
evidenced by were 76 out of 223 reports (34.1%).
However, that is only two more accidents than in
rivers (74/223, 33.3%), and three more than in open
sea (73/223, 32.7%). Still, it
means that a third of all
containershipaccidentshappeninharborareas.
Asanext step,wetookall76accident reports of
container ships in harbors as our subset to further
analyze.Weidentifiedseveraldifferentaccidents:in‐
juries (34/76, 44.7%), collisions (29/76, 38.2%),
groundings (5/76, 6.6%), fires
(3/76, 3.9%) and other
(5/76,6.6%).Here,injuryisthemostcommontypeof
accident.Thisincludes,forexample,pilotsfallingoff
oftheirladders,orcrewmemberslosingbalanceand
fallingintothewaterorfromaplatformontheship.
Many of those accidents are fatal. We manually
identified
whatcausedtheseaccidents:humanerrors
(41/76, 53.9%), technical issues (26/76, 34.2%) and
external factors (9/76, 11.8%). Human errors are
mistakesmadebypi‐lotorcrewmemberthatleadto
incidents such as falling off a ladder, whereas
technicalissuesrefertoequipmentproblems,suchas
a ladder falling
apart even though all routines and
checks were successfully implemented. An external
factorinthisscenariocouldbeanunpredictablyhigh
wave hitting the pilot on the ladder. However,
unpredictably high waves are not a phenomena of
harbor areas. Here, we refer to bad weather
conditionssuchasfogorheavy
rain.
2.3 Externalfactors
In previous work, the causes stated by accident
reports are mostly human error or technical issues.
However,webelievethatexternalfactors,especially
weatherconditions,alsoplayanimportantrole.Inthe
following section, we listed some of those accidents
that were to some extent caused
by unforeseen
weatherconditionsorrestrictedvisibility.
MAIB, Report No 22/2013. The vessel’s bridge
team lost situational aware‐ ness in dense fog as
the vessel maneuvered from berth on the north
shore,beforegroundingontheoppositesideofthe
river.
BSU, Investigation Report 507/11. There was a
strong
risingtideandvisibilitywasveryimpaired
duetofog.
MAIB, Report No 7/2011. The vessel’s design
restrictedtheabilityofpersonnelonthebridgeto
seeobjectsnearthevessel.
BSU,InvestigationReport264/10.Theviewastern
onthetugwasseverelyrestrictedbythefunnel.
MAIB,ReportNo17/2008.Therewerenodefined
operational limits or procedures for the tug
operators when assisting / towing in restricted
visibility.
MAIB,ReportNo1/2008.Theactionsofthepilotto
reducethespeedofapproachofthevessel’sbow
wereunsuccessfulbecauseMarinecoToomai was
positionedjust
aftofamidshipsofLogosIIandnot
ontheportquarterasthepilothadassumed;The
pilot did not accurately monitor the position of
Marineco Toomai; There was a lack of co‐
ordination and communication between the pilot
andthetug’sskipper.
MAIB, Report No 8/1998. The
pilotcould not see
thetugfromthebridgeandassumedthatshehad
been running with the ship stern‐to‐stern, from
whichpositionitwouldhavebeenrelativelyeasy
for the tug to position herself on the ship’s port
quarter.
2.4 Discussion
To summarize our results from the accident
report
analysis, we can say that container ship accidents
oftenhappeninharborareas(34.1%).Toanswerour
researchquestion(RQ1),themostcommoncausesare
human errors (53.9%) and technical issues (34.2%).
[10] give detailed insights on human errors in
accidentreports.However,wealsoidentifiedexternal
factors as
a common cause for maritime accidents
(11.8%).Whenanexternalfactoristhecauselistedin
an accident report, it typically refers to restricted
visibility. The recommendation given by many
accident reports is knowing the operational limits
(e.g., for operations in dense fog). However, even
withagoodweatherforecast,it
isnotuncommonthat
aweatherchangesurprisesoperatorsandmaneuvers
cannot be stopped. For the future, it would make
sense to develop technical solutions to solve the
problemofrestrictedvisibility.
3 ETHNOGRAPHICSTUDY
After investigating the most common causes for
accidents in the past, we want to understand
which
problemspilotsarefacingduringtheberthingprocess
nowadays (R2). Furthermore, while accident reports
describe accidents fairly well, they do not give us a
good understanding of the procedure and relevant
informationforpilotsduringthedockingprocess.To
address both points, we conducted an ethnographic
study on a container
vessel. The vessel is called
’Choapa Trader’ (IMO number 9407885) and it is a
containervessel.Thedimensionsoftheshipare294.1
meters long and 32.29 meters wide. The ship has a
gross tonnage of 52.726 tons and was built in 2009.
Thevesselwas goingfromBremerhavento
Hamburg
(bothcitiesinGermany).
3.1
Observations
Duringourobservations,wetookpicturesandwrote
aprotocolofallthepilots’actionsthatweobserved.
These pictures are available for free on Github
(https://github.com/UweGruenefeld/EthnographicStu
dy). We observed that, along the way from
Bremerhaven to Hamburg, at every point in time
therewasonepilotonboard.The
onlyexceptionwas
withintheharbors,wheretwopilotsweresupporting
the put out to sea and the ship berthing. The pilot
changealwaysstartedwiththe old pilot leaving the
ship bridge. The new pilot then boarded on the lee
sideoftheship.Beforetheoldpilotleftthe
ship,both
shortly exchanged important information about the
ship maneuverability, unusual events etc. After that
thenewpilotgoestotheshipbridgeandtheoldpilot
leavestheshipviaapilotladderontothepilotship.
To summarize our observations on the ship bridge,
496
we classified our gathered data into three main
categories.
Environment We observed that many different
objects in the environment are relevant almost
simultaneously (tugboats, buoys, moored ships,
bypassing ships, quay walls) all while the tugboats
weremoving quickly around the ship and changing
theirpositions.However,inmanypointsof
timethe
tugboatsorotherobjectswerenotvisibleforthepilot
becauseofocclusion(e.g.,bycontainer)ordaytime/
weatherconditions(e.g.,upcomingfog).Interestingly,
pilotspreferredtomakeeyecontactwithstaticobjects
tounderstandtheirship’smovement(e.g.,inthelast
part of thedocking process
they looked atthe quay
wallandevaluatedhowfasttheirvesselapproached
it).
Behavior Interestingly, harbor pilots prefer to be
on the outside of the ship bridge. This is mostly
because they want to make eye contact with the
tugboatsorotherimportantobjectsaround(e.g.quay
walls, buoys). The
closer an object is, the more
importantisitforthepilots.Butbecausetheirvision
is occluded by containers or other parts of the ship,
pilotshavetomovearoundontheshipbridgetofind
the best position for seeing the important objects
around. However, in many situations
they cannot
observe all objects at the same time, and therefore
havetomakeassumptionsabouttheirpositions.
Communication Since two pilots assists in the
docking process, one of them takes over the
communicationwithpersonsthat are not located on
the ship bridge (e.g., tugboat drivers, on‐shore
workers). This
communication is done via UKW or
VHF. Commands on the ship bridge are given
verbally to the people. During our observation, it
became clear thatpilots often hadto repeat partsof
their commands because they were hard to
understand, due to environmental conditions (e.g.,
backgroundnoise,wind).
3.2 Interviews
During the trip from Bremerhaven to Hamburg we
conducted semi‐structured interviews with the
captain of the ship and boarding pilots. In the first
partoftheinterview,wetalkedaboutdemographics
andexperienceoncontainervessels.Allparticipants
ofourinterviewsweremale.Thecaptainwas45years
oldand
had28yearsofexperienceasamariner.The
pilots were between 34 and 42 years old and had
between 5 and 14 years of experience as mariners.
Overall,weaskedfourparticipants(onecaptainand
threepilots).
In the second part of the interview, we asked
questions about the
ship berthing process. All
participantsagreedthatberthingtheshipisbyfarthe
most difficult part of the job. We asked the
participantsiftheycanstatewhytheythinkberthing
acontainervesselissodifficult.Intheiranswers,our
participants came up with three main contributing
factors explaining
why ship berthing is difficult: (1)
large container vessels move into small ports
involving many other entities such as tugboats or
buoys,(2)manypeopleassistinthedockingprocess
andthecommunicationamongthemisdifficult,and
(3) a stream of information during each docking
process has to be
processed, and important
informationinthisstreamhastobeidentifiedquickly.
Furthermore, participants stated that a lot of
informationisnotavailableatanymomentbecauseit
either takes too long to get the information, or the
informationissimplynotavailable.Thecombination
ofallthesefactorsmakesthe
berthingprocesshighly
mentallydemandingforthepilot.
In the third and last part of the interview, we
askedtheparticipantsifthey hadanyideasfornew
technologies that might help making harbor
maneuversmoresafe.Thefollowingisalistofthese
ideas:
Augmenting the ship bridge
windows with
information during river maneuvers. Thereby,
marinerscanobservetheenvironment,andatthe
sametimekeeptrackoftheinstruments.
Thefieldofviewonshipbridgesisquitelimited,
whichisnotaprobleminmostcases.However,in
dockingscenarios,itisverydifficulttokeep
track
ofobjectsthatareclosetotheship.
Because the pilots like to have eye contact with
their environments, it is important to be
unrestrictedinmovingaroundontheshipbridge.
Forexample,notebooksthatsomepilotsusepose
a problem, because they do not allow pilots
to
moveontheshipbridge.
Onepilotsaidthatinsomescenariositwouldbe
helpfultobeabletointeractwiththeportablepilot
unit(PPU)inahands‐freeoperationmode.
Some of the pilots mentioned the idea of virtual
objects replacing real objects. Using this approach,
theywouldbeabletoseevirtua l buoys,whichwould
notonlybemucheasiertokeepup‐to‐date,butalso
wouldbevisibleevenifoccludedbysomethingelse.
Most interestingly, all pilots stated that they like
newtechnologies ingeneral.One pilotmentionedin
particular the experience
he had when PPU’s were
introduced. He said his first impression was that it
would not be useful. However, it became a helpful
tool,onewithoutwhichhewouldnotwanttowork.
3.3 Discussion
During the ethnographic study, we observed that
pilots wanted to have eye con‐ tact with objects
aroundtheminordertobeaware oftheirpositions.
This was especially true for objects that were very
close.Withregardtoouraccidentanalysis,itbecame
clear why many accidents happen in restricted
visibility. Keeping eye contact with surrounding
objects is not possible in that situation, and current
technologiesarenotsuitedtoassistbecauseeitheritis
technicallychallengingtomonitorcloseobjectsorthe
virtually perceived information does not feel well‐
integrated into the real environment. A cognitive
process is necessary to integrate the virtual
information into the real world. Additionally, we
observed that the communication with
tugboat
drivers and other workers is quite error‐prone. The
insightsgained fromourethnographicstudyalready
showpossibleanswerstoourresearchquestion(R2).
However, to ensure that the results from our
ethnographicstudyaregenerallyvalid,weconducted
anonlinesurveywithharborpilots.
497
4 ONLINESURVEYWITHHARBORPILOTS
During our ethnographic study we identified
problematicfactorsthatmightleadto newaccidents
inthefuture(e.g.,communicationbetweenpilotsand
tugboat driver seems to be rather error prone).
However,thestudywaslimitedtoexperiencesmade
in one harbor. To identify the
most important
problems, we needed to further understand the
environments in which harbor pilots work and to
identify which problems persist across harbors. To
investigate these aspects, we designed an online
survey that we conducted with harbor pilots from
differentcountries.
4.1 Design
In our survey, we first wanted to
understand more
abouttheenvironment in which harbor pilotswork.
Therefore, we asked questions about the working
environment.Asanextstep,weaskedmorespecific
questions regarding the communication between
pilots and tugboat drivers during the docking
process, because we had identified this as a critical
aspectduringourethnographic
study.Weclosedthe
surveywithquestionsaskingaboutthedemographics
oftheharborpilots.
4.2 Pretest
Beforeaskingpilotsto participate in our survey, we
did a pretest with an experienced pilot from
Germany. Mainly, we focused the pretest on
understand‐abilityandacceptability.First,weasked
thepilot
tofilloutthequestionnaireandmakenotes
duringthatprocess.Apartfromminorcorrections,the
survey was understood well. However, some
questionswerereadwithconcernbyourexperienced
tester. It was said that a survey that raises the
impression that technology will improve jobs in the
future also creates
the fear of jobs becoming
redundantandreplacedbytechnology.Therefore,we
hadtomakesurethatthiswasnottheimpressionone
wouldgetfromreadingthequestionsinoursurvey.
To further ensure this, we added a detailed
introductionexplainingthereasonforthesurvey.
4.3 Participants
Overall,
30harborpilots(nofemale),agedbetween35
and63(M=50.27,SD=7.22)participatedinouronline
survey. Sixteen pilots were German‐ speaking, and
mainlyfromtheportsinHamburgandBremerhaven.
14 pilots were English‐speaking, from ports
distributed over Europe (e.g., Rotterdam, Antwerp).
Theexperienceofparticipantsrangedfromone
to26
years with an average of 15 years (M=14.97).
Moreover, none of the harbor pilots in our survey
workedasapilotinadifferentareabefore(e.g.,river
pilot). We recruited the participants by manually
crawling harbor pilots homepages in Europe and
contactingthemviae‐mail.
4.4
Results
In this section, we report the results of our online
survey in the same order the questions appeared
withinthesurvey.
Q1: Who do you stay in contact with during
harbor maneuvers? The answers were (in order):
tugboat (26/30, 86.7%), mooring hawser (22/30,
73.3%), harbor control (14/30, 46.7%), captain
(12/30, 40.0%), other pilots (10/30, 33.3%), bridge
team (8/30, 26.7%), other ships (6/30, 20.0%) and
rescuestaff(2/30,6.7%).
Q2: How many persons do you communicate
simultaneously with during a harbor maneuver?
The number varies between three and ten. On
average the pilots have to talk to six different
persons
atthesametime(M=5.90,SD=1.74).
Q3: What is the maximum number of tugboats
involved in one of your harbor maneuvers? The
number varies between two and six, but the
average number of tugboats is four (M=4.40,
SD=0.99).
Q4: What information do you communicate with
the tugboats? Mainly,
pilots communicate about
applied forces and their directions (30/30, 100%),
butalsothepositioningoftugboat(12/30,40.0%),
the maneuver plan (7/30, 23,3%), distances (6/30,
20.0%),andmooringlinelength(1/30,3,3%).
Q5:Whataresomesampleinstructionsyougiveto
atugboat?Analyzingtheanswersofthisquestion,
it
becomeclearthatallinstructionswereshortand
concisebutdonotshareauniformstructure.One
pilot said that each harbor uses different
instructions.
Q6: Can you describe any safety‐critical
experiences due to miscommunication with a
tugboat? The missing uniform structure for
exchanging instructions can lead to
miscommunication. One pilot mentioned that he
wasfacingacriticalencounterbecausea tugwas
onthewrongsideoftheship,duetohisnotbeing
experiencedinthelocalinstructions.Overall,25of
30 pilots reported having experienced a critical
situation (e.g., due to language barriers, not
paying attention,
not experience with local
instructionsbutalsolowUKWquality).
Q7:Hasthecommunicationbetweenyouandthe
tugboats changedover thelast few years?If yes,
how? Despite the experience with critical
encounters due to miscommunication, 23 pilots
statedthatnothinghaschangedinthelastdecade
with
regardtocommunicationbetweenpilotsand
tugboat drivers. Theother 7pilots can only state
minimal changes. One pilot mentioned that the
instructionsdidnotchangemuchbecausetheyare
treatedasverytraditionalandlocal.
4.5 Discussion
Theresultsofouronlinesurveyshowthatmostofthe
problems identified
in the ethnographic study in
HamburgalsoapplytootherEuropeanharbors.The
communicationwithsixdifferentpartiesatthesame
time is especially mentally demanding. Given the
error‐pronetechnology,itbecameclearwhyaccidents
with wrongly positioned tugboats happen quiet
frequently. Furthermore, it explains why almost all
498
pilots (83.3%) recalled having experienced safety‐
critical situations due to miscommunication. With
container ships increasing in size, this is a serious
problemthatneedstobedealtwithinthefuture.
5 RECOMMENDATIONS
Basedonourfindingsfromtheaccidentanalysis,the
ethnographic study and the online survey, we have
come up with the following recommendations to
reduce problems and risks during the berthing
processofcontainervessels:
Humanerror.Supportthehumaninthedecision‐
makingprocesstoavoidhumanerror.Asmentioned
earlier,shipbridgesaretechnicalsystemsthatmainly
consistof humans. Developing systemsthat support
humans
mayreducethenumberofhumanerrorsin
thosesystems.
SituationAwareness.Inmanyaccidentreports,we
found that a lack of situation awareness leads to
wrong decisions, causing most collisions. Therefore,
we want to highlight the importance of Bridge
Resource Management for situations of navigation
withapilotonboard,
andtoeffectivelycommunicate
decisions,assuggestedby[2].
Near‐distance Radar. During our studies, we
observedthatinformationaboutthepositionofnear‐
by objects is required to successfully berth a large
containervessel.Therefore,wesuggesttoassistpilots
inharbormaneuvers withanovelradarsystemthat
allows pilots to effectively monitor near‐by objects.
Existingradarsystemscannotsolvethisproblemfor
tworeasons:(1)objectsthatarecomingtooclosewill
not be visible in the radar, and (2) the mapping of
these objects onto a 2D screen makes it harder for
pilots and seafarers to
understand the positions of
theseobjectsintherealworld.Therefore,wesuggest
theuseofhead‐mountedAugmentedRealityglasses
tooverlaytherealworldwithinformationaboutthe
positionsofoccludedobjects.
Communication. During the ethnographic study
and online survey, we identified another important
issue: the communication of
the pilots with the
tugboat driver. In our accident report analysis, we
saw a couple of accidents that were caused by
misunderstandings between the pilot and tugboat
driver. Here, it is contradictory that each harbor
implementsitsownspeechcommands.Furthermore,
it is problematicthat pilots do nothave an effective
way to monitor whether the tugboat driver
implementsthegiveninstructions.
6 CONCLUSION
In our work, we investigated the berth of container
ships from the pilot’s perspective. We revealed that
more than one third of all container ship accidents
happen in harbor environments. Furthermore, our
ethnographicstudyshowedthatpilots
faceproblems
when monitoring their environments because many
objects are hidden behind containers or not visible
duetobadweatherconditions.Inafollowingonline
surveyofpilots,weidentifiedthecommunicationvia
UKW/VHF as problematic due to missing standards
andtechnicallimitations.Wethinkourworkisafirst
step
towards a better understanding of the ship
berthingprocessand itsassociatedproblems.Future
workcanbebuiltuponthisworktodevelopsolutions
forsolvingtheidentifiedproblems.
ACKNOWLEDGEMENT
We thank the Ministry of Science and Culture of
Lower Saxony for supporting us with the graduate
school ”Safe Automation
of Maritime Systems
(SAMS)”. And we would like to thank the BUSS
DATA GmbH for supporting us with our
ethnographicstudy.
REFERENCES
[1]Baker,C.,andMcCafferty,D. Accidentdatabasereview
ofhumanelement concerns:What do the resultsmean
for classification? In Proc. Int Conf. Human Factors in
ShipDesignandOperation,RINAFeb(2005),Citeseer.
[2]Chauvin,C.,Lardjane,S.,Morel,G.,Clostermann, J.‐P.,
and Langard, B. Human and organisational
factors in
maritime accidents: Analysis of collisions at sea using
thehfacs.AccidentAnalysis&Prevention59(2013),26–
37.
[3]Culliane,K.,andKhanna,M.Economiesofscaleinlarge
containerships: optimal size and geographical
implications.JournalofTransportGeography8,3(2000),
181–195.
[4]DelaCampa
Portela,R.Maritimecasualtiesanalysisasa
tool to improve research about human factors on
maritimeenvironment.JournalofMaritimeResearch2,
2(2005),3–18.
[5]Frittelli, J. Ship navigation in harbors: Safety issues. In
CRSReportforCongress(2008).
[6]GrechM. R.,Horberry,T.,andSmith,A.Human
errorin
maritimeoperations:Analysesofaccidentreportsusing
theleximancertool.InProceedingsofthehumanfactors
and ergonomics societyannualmeeting (2002), vol. 46,
Publications,pp.1718–1721.
[7]Hafenlotsenbrüderschaft.Hafenlotsenordnung.http://ww
w. hamburg‐pilot.de/Hafenlotsordnung_2013_Stand_
07052013.pdf, 2013. Letzter Zugriff am 02. November
2017.§5Absatz1Nummer1HmbGVBl(Hamburgisches
Gesetz‐und
Verordnungsblatt).
[8]Koester,T.Humanerrorinthemaritimeworkdomain.
InProceedingsof20thEuropeanAnnualConferenceon
Human Decision Making and Manual Control (2001),
pp.149–158.
[9]Rothblum, A. M. Human error and marine safety. In
National Safety Council Congress and Expo, Orlando,
FL(2000).
[10]Stratmann,T.C.,
andBoll,S.Demonhunt‐theroleof
endsley’s demons of situation awareness in maritime
accidents. In Human‐Centered and Error‐Resilient
Systems Development (Cham, 2016), C. Bogdan, J.
Gulliksen,S.Sauer,P.Forbrig,M. Winckler,C.Johnson,
P. Palanque, R. Bernhaupt, and F. Kis, Eds., Springer
InternationalPublishing,pp.
203–212.
[11]Veit, S.‐M. Die elbe ist ein enges revier.
http://www.taz.de/!879062/,2015.LetzterZugriffam11.
Juni2017.
[12]Wagenaar,W. A.,andGroeneweg,J.Accidentsatsea:
Multiple causes and impossible consequences.
International Journal of Man‐ Machine Studies 27, 5
(1987),587–5
