279
1 INTRODUCTION
Analysesofshiptrafficareimportant,e.g.toestimate
emissionofgreenhousegases,monitorfleetefficiency
andforconductingstudiesonshipsafety. Automatic
Identification System (AIS) data has become an
integral part of these studies, as they provide
positionalandoperationalinformationforalargepart
oftheshippingfleet.
AIS is a communication system tha
t uses the
maritime Very High Frequency (VHF) bands to
transmit ship movement and technical data at
specified intervals. This includes static data, such as
the shipʹs name, draught, destination and Estimated
TimeofArrival(ETA),aswellasdynamicdatafrom
the ships sensors, such as speed and position (ITU
2014).Atypi
caluseofAISistoexchangeinformation
between vessels that are in the same area, to
automaticallyidentifyothershipsandavoidhighrisk
situations. It is also used in traffic monitoring, to
provideguidancebyvesseltrafficservices(VTS)and
byma
nyothershoresideusers.Thedevelopmentof
AIS was a joint project between the International
Maritime Organization (IMO) and the International
Association of Marine Aids to Navigation and
Lighthouse Authorities (IALA). The International
Convention for Safety of Life at Sea (SOLAS) states
that all ships of 300 gross tonnage and upwards
engagedinint
ernationalvoyages, cargoships of500
gross tonnage and upwards not engaged on
internationalvoyages,aswell asall passenger ships
built after 2002, or operated after 2008, should have
an AIS (IMO 2002). This essentially means that all
largershipsengagedinglobalshippingshouldhave
AISequipment.Nationalrequirementswillnormally
alsorequireshipsnotcoveredbyIMOregulationsto
carryAIStra
nsmitters.Thismeansthatmorethan85
000shipsworld‐widewilltransmitAISdata(Mantell
2014).
AISdataisgathered by AIS receivers, which can
be found on board ships, on buoys, on land (IALA
2011)andmorerecentlyonsat
ellites(hereafterS‐AIS).
Expanding the Possibilities of AIS Data with Heuristics
B.B.Smestad&B.E.Asbjørnslett
NorwegianUniversityofScienceandTechnology,Trondheim,Norway
Ø.J.Rødseth
SintefOcean,Trondheim,Norway
ABSTRACT:AutomaticIdentificationSystem(AIS)isprimarilyusedasatrackingsystemforships,butwith
thelaunchofsatellitestocollectthesedata,newandpreviouslyuntestedpossibilitiesareemerging.Thispaper
presentsthedevelopmentofheuristics forestablishingthespecificshiptypeusinginformationretrievedfrom
AISdataalone.Theseheuristicsexpandthepossibilit
iesofAISdata,asthespecificshiptypeisvitalforseveral
transportation research cases, such as emission analyses of ship traffic and studies on slow steaming. The
presentedmethodfordevelopingheuristicscanbeusedforawiderrangeofvessels.Theseheuristicsma
yform
the basis of large‐scale studies on ship traffic using AIS data when it is not feasible or desirable to use
commercialshipdataregisters.
http://www.transnav.eu
the International Journal
on Marine Navigation
and Safety of Sea Transportation
Volume 11
Number 2
June 2017
DOI:10.12716/1001.11.02.10
280
Land based AIS receivers can detect AIS messages
normallyupto40‐50nauticalmilesoffshore(Skauen
2013),shipsfurtheroff‐shorewillremainundetected
bylandbasedAISreceivers.In2005,researchersfrom
the Norwegian Defence Research Establishment
published the first study investigating whether
satellites could be used to
gather AIS signals (Wahl
2005).In2008,afollowupstudybyHøyeetal.(2008),
foundthatAIS signalscouldbe detectedbysatellite
based AIS receivers positioned in altitudes of up to
1000 km. However, since the AIS system was not
initiallydesignedforspacebasedreceivers,but
rather
to be a ship‐to‐ship communication system, there
were some problems. A satellite will have a much
largercoverageareathanAISreceiversweredesigned
for, which could lead to interference problems
betweenthedifferentships’AISsignals.Accordingto
thestudy,theresultcouldbethatsomeAIS
messages
wouldnotbedetectedbythesatellite.Inpracticethis
leads to a more reliable satellite coverage in areas
withlesstraffic,whilehightraffickedareascanhave
interference problems. In 2010, the Norwegian AIS
satelliteAISSat‐1waslaunched.Thissatelliteisina
sun‐synchronous polar orbit
at 630 km altitude
(Eriksen 2010). The satellite transmits the AIS
messages it receives to Svalbard Ground Station at
each passing. Eriksen et al. (2010) states that over a
time span of 24 hours, areas along the equator is
coveredtwotothreetimes,whiletheHighNorthand
Southis
coveredupto15times.In2013,AISSat‐2was
launched to give extended coverage. This gave a
higherupdateratetotheSvalbardGroundStation,as
wellasahigherglobaldetectionrate.
The use of AIS data in studies on maritime
transportation has become increasingly prevalent.
Smithet
al.(2014)preparedareportasapartofthe
WorldShippingEfficiencyIndicesprojectfundedby
the International Council on Clean Transportation.
ThestudycombinedglobalS‐AISdatafrom2011with
technicalshipdatafromsourceslikeClarksonsWorld
FleetRegister, andtheSecondIMOGreenhouseGas
Study (Buhaug
2009). The S‐AIS data provided
operationalcharacteristics,suchasspeedandloading
condition. In addition, estimates on the distance
travelled were derived from the S‐AIS data. Data
from Clarksons World Fleet Register provided
technical specifications, such as the ship type (for
instance LNG tanker or crude oil tanker)
for each
individualship.
TheThirdGreenhouse Gas(GHG)studybySmith
et al. (2014) had an advantage over the preceding
studies,asitcouldutilizeS‐AISdata.Thesedatawere
usedtogetmorepreciseactivitymeasuresandbetter
emissions estimates for each ship. This was
aggregatedtothe
totalemissionsforeachship type.
In the previous study, emissions were estimated by
using the annual average activity for the different
shiptypes.
Categorizingshipsintoshiptypeandsizecategory
is vital to perform studies on operational efficiency
and greenhouse gas emissions. Knowing the design
speedisnecessary
fordevelopingspeed‐relativefuel
consumption models for ships‐where the design
speed is the speed giving theʺoptimalʺ trade‐off
between speed and fuel consumption. The design
speed is amongst others a factor of the block
coefficientoftheship,whichinturnislargelygiven
bytheshiptype.
Previousstudies,suchasSmithetal.
(2014), have used commercial vessel databases to
retrieve the ship type for each specific ship in the
study.However,usingexternaldatabasestoretrieve
theshiptypecanbecostlyasthesedatabasesrequire
asubscription.Ontheotherhand,manualretrievalof
the ship type from open databases can be time
consuming.Thecombinationofthesetwofactorsmay
inhibit studies on maritime transportations using
estimationbasedonAISdata.
IntheSESAMEStraitsproject(SESAME2017),the
challenge was to give guidance to ships headed for
and in the Straits of Malacca
and Singapore and to
estimate possible fuel savings by suggesting more
efficientspeedstotheships.Aproblem, however,is
to find enough information about the ships to do a
reasonableestimationoffueluseandfuelsavingsfor
differentspeeds.Thisinformationcanbebought,but
injustfive
days,morethan3000differentshipswere
recorded by the AIS‐stations in the area.As the
market for such services are limited and quite cost
sensitive, it was not very attractive to buy the
information.
Theresearchquestionsthatemerged,whenfaced
with these challenges was: How well can AIS
data
aloneidentifytheshiptype andsize?Canheuristics
for identifying the ship type for any ship be
constructed? The objective of this study was to
establishheuristicsforidentifyingtheshiptypefora
largeproportionoftheworldfleet,usingS‐AISdata.
The method for constructing
the heuristics is
outlinedinSection2,whiletheheuristicsparameters
canbefoundinSection3.Theperformanceforthese
heuristics is provided in Section 4, while the results
and the validity of the heuristics are discussed in
Section5.Aconclusionisgiveninthefinalsection.
2 METHOD
SatelliteAISdataspanningthetimeperiodofMay1st
2014toSeptember15th2014wasretrieved.TheS‐AIS
data had been collected using the two satellites
AISSat‐1 and AISSat‐2, and was provided by the
Norwegian Coastal Administration for use in the
SESAME Straits research project. AISSat‐
2 data was
onlyavailableafteritslaunchinJuly2014.
These S‐AIS data included static and/or dynamic
AIS messages for 85,108 ships, identified by unique
MMSI numbers. 43,671 of these ships had both
dynamic and static data. Mantell et al. (2014) stated
thatthetotalworldfleetconsistedof
88,483shipsas
ofMay2014.Approximately95%oftheworldfleetis
presentinourdata,andabouthalfoftheworldfleet
is represented with both dynamic and static data.
TheseS‐AISdataisshownasgroupAinFigure1.
We developed heuristics for a selection
of ship
types with high relevance to international shipping
(Table1).Thisselectionisinlinewiththeselectionin
otherstudiessuchasSmithetal.(2014).
281
Table1. AIS vessel groups, ship types and sizes in this
study
_______________________________________________
AISvessel ShiptypeShipsize
group
_______________________________________________
Tankers LNGandLPGCarriers General,Q‐Flexand
Q‐max
OilTankersUL&VLCC
Cargoships Containervessels Panamax
BulkcarriersPanamax
_______________________________________________
TheClarksonsGroupprovidesadatabasewherea
selectionofvesselsofeachshiptypeandsizecategory
are listed by the ship’s name (Clarksons 2015). This
dataisshown inFigure1 asgroupB.Theshipsare
only identified by their name, and not by a more
uniqueidentifier
suchastheirIMOorMMSInumber.
VesselcharacteristicswerealsoretrievedfromtheS‐
AISdatabymatchingthenameoftheshipfromthe
vessel database to the name registered in the S‐AIS
data. The ships that were present in both the S‐AIS
data and the
vessel database are a candidate group,
formedbyasubsetofthetwogroups,andisshown
asgroupCinFigure1.
Figure1.Theprocessusedforconstructingtheheuristics.
ThevesselsarematchedbetweenClarksonsvessel
sheets andthe S‐AIS data based on their name, and
nottheiruniqueIMOnumber,sothereisapossibility
thatshipsfromothershipclasses,withthesameship
name, are included in the candidate group. To
mitigatethissourceoferrors,
adatacleaningprocess
was required. In the data cleaning, ships with
dimensionsoutsidetheexpectedintervalfortheship
type in question were removed. For instance, cargo
shipsandtankersaretypicallyclassifiedintodifferent
size categories, which often correspond to the
maximum dimensions of important seaways and
ports,
suchasthePanamaCanalandtheSuezCanal.
If a ship was categorized as a Panamax ship in the
Clarksonsvessel database,buthadreportedawidth
ordraughtexceeding thesetofmaximumdimensions
in the Panama Canal in the S‐AIS data, it was not
includedinthe
traininggroup.
Theinitialversionofthemethod usedmaximum
observed speed as one of the parameters for
classification. Early testing of this heuristic showed
thatsomeshipsweremisidentified.Asanexample,a
274mlongand48mbroadoiltankerhadamaximum
observedspeedof20knots.
Becauseoftherelatively
high speed, this vessel was classified as an LNG
carrier.However,speed recordingsfromAIS data is
most commonly speed over ground, and not speed
relativetothewater.Theserecordingsmaythusbea
result of particularly favorable wind and current
conditions,andnotnecessarilyerrors
in speed data.
To find the frequency of the different speed
recordings,allreportedspeedswerebucketedinone
knot intervals. Out of 165 speed recordings for this
vessel, there was only one record of the maximum
recorded speed of 20 knots. The highest speed,
amongst those with the highest
frequency, was 14
knots.Thedatashowedthatthevesselhadthisspeed
at ten occasions. To avoid these rare occurrences of
highspeed,anewconstraintwasputintheheuristics;
foramaximumspeed to be valid, the vessel should
havetenormoreAISrecordsofhavingthat
speed.
After the data had been cleaned, we used the
resultingshipsasatraininggroupfortheheuristics,
shown by group T in Figure 1. Using this training
group, common dimensional traits and operational
characteristics for each ship type was derived by
inspection,andultimatelyusedtoformthe
heuristics.
ThiswasrepeatedforeveryshiptypeinTable1.The
process of making heuristics for panamax bulk
carriersisusedasanexampleandoutlinedbelow.
The heuristic, which consists of constraints on
dimensions, draught, speed and AIS vessel group,
wereappliedonthefullsetofS‐AIS
data(groupA) as
aperformancetest.Theperformanceofeachheuristic
wascheckedbymanuallyconfirmingthespecificship
type of all ships classified by the heuristic, using
onlineshipdatabases.Thesearedatabaseswherethe
shiptypeofasingleshipcanbefoundusingtheIMO
or
MMSI number. The accuracy of a heuristic was
definedasthenumberofshipscorrectlyidentifiedby
ship type, divided by the total number of ships
identified.
2.1 DevelopingheuristicsforPanamaxBulkCarriers
Thecandidategroupfortheheuristictraininggroup
was made by identifying all panamax bulk carriers
present
inboththeClarksonsvesseldatabaseandthe
S‐AIS data. Out of the 2459 panamax sized bulk
carriers in Clarksons vessel database at the time of
retrieval(spring2015),2200shipswerealsopresentin
theS‐AISdata.
2.1.1 DataCleaning
2.1.1.1 Erroneousshipdimensions
Thebreadthwasrequiredtobelessthan34m,as
themaximum widthofthePanama Canalis33.5m.
Theextra0.5mwasallowed,assomepanamaxbulk
carriersseemedtoberegisteredwithawidthof34m
intheS‐AISdata,probably
duetoaroundingerror.
Thisconstraintisillustratedbythetophorizontalline
in Figure 2. There were a lot of vessels in the
candidategroupexceedingthisbreadth.Thefactthat
seemingly panamax vessels could exceed this
constraint can be attributed to the lack of a unique
identifier
inthe vesselsheetsasearlierdescribed.In
other words, these may have been non‐panamax
vesselshavingthesamenameasthepanamaxvessels
in the Clarksons vessel sheets. To ensure that only
Panamaxvesselswerepresentin thetraining group,
an additional breadth constraint of minimum 30 m
was
added.Vesselsbelowthisbreadthwouldfallinto
othershipcategories.Thisconstraintisillustratedby
thebottomhorizontallineinFigure2.Thesebreadth
requirements reduced the candidate group to 1668
ships.
282
Figure2. Length and breadth for the candidate group of
panamax sized bulk carriers. The horizontal lines indicate
themaximumandminimumallowedbreadth.
After the breadth constraints was enforced, three
vesselsintheresultinggrouphad alengthover250
m. The dimensions of these three vessels were
manuallyinspectedinanopenshipdatabase,tocheck
for any errors. The longest ship, Vishva Anand,
actuallyhadalengthof229m,notthe
332mitwas
recordedwithin theS‐AISdata.The second longest
shipwasacontainervesselmisidentifiedasthebulk
carrierSantaRegina,astheysharedtheirname.The
last vessel was the 259 m long bulk carrier Orissa.
This is an exceptionally long bulk carrier, with
a
breadthofonly32m.Sincethesethreevesselseither
werewronglyregisteredor exceptionally large, they
were excluded. After these exclusions, 1665 vessels
remained. The rest of the vessels in the candidate
grouphadreasonablesizes,andwehadnoreasonto
suspectthattheir dimensions wereerroneous.These
shipscouldnowbeusedasa traininggroupforthe
heuristics.
2.1.2 Heuristictraining
2.1.2.1 Maximumspeedconstraint
Becauseofthehighutilizationoftheship’svolume
in bulk carriers, it was expected that the maximum
speed as registered by AIS‐S is lower compared to
otherdimensionallysimilarvessels,suchascontainer
vessels.Asmanyas92%ofthecontainervesselshad
anobservedmaximum
speedof15.9knotsormore,
while 92% of the Panamax bulk carriers had an
observedmaximumspeedof 15 knotsorless.There
wasagroupofshipsreportingspeedsupto18knots,
which can be seen in Figure 3. This can be due to
especially favorable wind
and current conditions. It
canalsobeduetoothershipsbeingmisidentifiedas
bulkcarriers.Becauseofthesefindings,themaximum
recordedspeedallowedintheheuristicwassetto15
knots.
Figure3.Maximumspeedandlengthofthetraininggroup
Panamaxsizedbulkcarriers.
2.1.2.2
Draughtconstraint
Bulkcarrierstypicallycarriesunpackeddrycargo.
Themaincargotypesarecoal,ironore,cereals,sugar
or cement. They have a high utilization of their
volume, as the cargo is held in several transverse
cargoholdsoverthefullshipbreadth.Becauseofthe
high utilization of the
ship’s volume, a high
maximum draught and large differences between
maximum (when the ship is fully loaded) and
minimum (when the ship sails without cargo)
draughtareexpected.
Figure4showsthedraught,lengthandbreadthof
the ships in the training group. There was no
apparent correlation between these
variables.
However,alloftheshipsinthetraininggrouphada
maximumdraughtabove5m,sothisconstraintwas
includedintheheuristic.
The scatterplot in Figure 5 shows the lack of
apparentcorrelationbetweenthemaximumchangein
draughtovertherecordingperiodversusbreadthor
length.
283
Figure4. Draught, breadth and length for the training group of panamax sized bulk carriers. The left plot shows the
maximumdraughtandbreadth,whiletherightplotshowsmaximumdraughtandlength.
Figure5.Changeindraught,relativetobreadthandlengthforthetraininggroupofpanamaxsizedbulkcarriers.Theleft
plotshowschangeindraughtandbreadth,whiletherightplotshowschangeindraughtandlength.
The change in draught ranged from zero to just
below19m.99%ofthevesselsinthetraininggroup
had a change of draught less than 9.5 meters.
Containervesselsaremostlikelytheshiptypetobe
misidentifiedasbulkcarriers,because theyhavethe
same AIS ship type
(cargo ship), but unlike bulk
carriers, container vessels have a lower difference
betweenthedraughtinafully‐orlessloadedstate.
Thus,alowerlimitformaximumchangeofdraught
wasareasonableboundarytoset.Whentheheuristic
for the container vessels was developed, we found
that 97%
of the container vessels had a change of
draught of less than 5.5 m. The lower limit for the
maximum change of draught was set to 5.5 m, to
exclude these container vessels, and the training
groupwasreducedto1,210vessels.
2.1.3 Heuristictesting
Whentestingthe heuristicforthe
panamaxbulk
carriers(Table2) on theS‐AIS data,atotal of 6,024
vessels matched the dimensional data in the
heuristic. After applying the minimum change in
draught and the maximum speed constraint, 2,346
vessels remained. 1,210 of these ships were in the
training group. This means that the
heuristic
identifiedanadditional1,136shipsaspanamaxbulk
carriers. Manualinspectionshowedthat43ofthese
vessels were misidentified. Out of the misidentified
ships, 38 were general cargo ships and three were
offshore support vessels. In addition, a container
vesselandavehiclescarrierweremisidentified.With
43vesselsmisidentified
outof2,346identifiedvessels
theaccuracyoftheheuristicwas98%.
Asimilarexercise was doneforeach of the ship
typeslistedinTable1andcorrespondingparameters
determined.
3 HEURISTICS
Theparametersetsfortheheuristicsforthedifferent
ship types can be found in Table 2
to Table 5. The
parameters were developed using the method
outlined in the previous section. First, a candidate
group out of ship registry data was formed, then
erroneous data were removed. This subset of the
candidate group formed a training group, where
common traits was derived. These common traits
284
wereusedtoestablishparameterstoidentifyaship
typeandshipsize.Theheuristicsreflectthedifferent
ship characteristics, spanning from minimum and
maximum draught to maximum and minimum
speed,dependingon thetraits oftheshiptype and
size.
3.1 Bulkcarriers
Heuristics were developed for the subgroup
of
panamax bulk carriers (Table 2). The maximum
recorded speed was set to less than or equal to 15
knots,andthelength,breadthandminimumchange
ofdraughtwassetaccordingtotheparametersofthe
traininggroup.
3.2 Containervessels
HeuristicsforcontainervesselsareprovidedinTable
3.Inthedevelopmentofthisheuristic,thevesselsin
the panamax training group were split into two
groupsdependentonlengthandmaximumdraught.
To accommodate for this, one group is termed
Panamax while the longer group with a higher
maximum draught is termed Post‐panamax. This is
not fully
consistent with general terminology as
Panamaxincludeslengthsupto280m.
3.3 Gascarriers
Based on the training group for gas carriers, the
characteristic parameters were divided into three
maingroups:Ageneralgroupaswellasagroupfor
Q‐FlexvesselsandagroupforQ‐Max
vessels(Table
4).TheseheuristicssharestheAISvesselgroup,the
maximum recorded speed, the maximum draught
and maximum change of draught, while only the
breadthandlengthdiffer.
3.4 Oiltankers
TheparametersetinTable5isfortwogroupsofoil
tankers: ultra large crude carriers (ULCC) and
very
largecrudecarriers(VLCC).
4 RESULTS
The satellite AIS data was stored on a database.
Usingtheheuristicsparametersastheretrievalquery
parameters, we retrieved all the ships that matched
the different categories. For each ship that was
identified as one of the aforementioned ship types,
wedid
amanualcheckagainstpublicshipdatabases
tocheckifitwascorrectlyidentified.Thiswasdone
by using the ship’s IMO number. In Table 6, the
accuracyofeachheuristic,aswellasthenumberof
vesselsin theworldfleet forthatship typeandthe
numberofvessels
inthetraininggroupcanbefound.
Theaccuracywasquantifiedasthepercentageofthe
numberofvesselscorrectlyidentified.
Table2.Heuristicsforpanamaxbulkcarriers.
__________________________________________________________________________________________________
Shipsize Length Breadth Min.changeofdraught Min.draught Max.recordedspeedAISvesselgroup
[m] [m][m][m][kn]
__________________________________________________________________________________________________
Panamax* 180‐250 30‐305.55<=15Cargoship
__________________________________________________________________________________________________
*Subgroup
Table3.Heuristicsforpanamaxcontainervessels.
__________________________________________________________________________________________________
Shipsize Length Breadth Min.changeofdraught Max.draught Max.recordedspeedAISvesselgroup
[m] [m][m][m][kn]
__________________________________________________________________________________________________
Panamax 210‐269.9 31‐335.513>=15.9Cargoship
Post‐panamax 270‐300 31‐335.514>=15.9Cargoship
__________________________________________________________________________________________________
Table4.Heuristicsforgascarriersofdifferentshipsizes.
__________________________________________________________________________________________________
Shipsize Length Breadth Max.changeofdraught Max.draught Max.recordedspeedAISvesselgroup
[m] [m][m][m][kn]
__________________________________________________________________________________________________
Generalgroup 270‐300 40‐523.513>=16Tanker
Q‐Flex314‐316 48‐503.513>=16Tanker
Q‐Max344‐345 46‐543.513>=16Tanker
__________________________________________________________________________________________________
Table5.HeuristicsforULCCandVLCCoil tankers.
__________________________________________________________________________________________________
Shipsize Length Breadth Min.change Min.draughtMin.draughtMax.recordedspeedAISvesselgroup
[m] [m] ofdraught[m] [m][m][kn]
__________________________________________________________________________________________________
ULCC&VLCC 320‐400 50‐7081025<=16Tanker
__________________________________________________________________________________________________
285
5 DISCUSSION
Simple heuristics to identify the ship type from
satellite AIS data was developed, utilizing a
comprehensivesetofS‐AISdata.Theheuristicswere
developed by combining data from a commercial
vesseldatabasewithS‐AISdatatoformacandidate
group. The candidate groups were inspected for
erroneous data and used as training groups for the
heuristics.
Theheuristicsdeveloped isfullybasedon S‐AIS
data from static and dynamic messages, containing
information such as AIS‐ship type, general
dimensions,draughtandspeed.
It should be noted that the maximum and
minimum draught, as well as
the maximum speed,
are products of the operating conditions of the
vessels. These operating conditions can be affected
byfactorssuchasseasonalmicro‐variations,aswell
as yearly macro‐varia tions. Corbett et al. (2009) has
shown that the average speed of the shipping fleet
canbeinfluencedbyfuelcost.
Maximumdraughtis
influenced by the loading condition of the ship,
which again is influenced by the market the ship
operates in. A strong market means that a large
quantityofgoodsistransported, andthe maximum
draught recorded can therefore be expected to be
higher, compared to in a
weak market. In a strong
market, ships can be expected to have a different
operatingspeedthaninaweakmarket.AstheS‐AIS
data spans a relatively short time‐period, the
heuristicscouldturnouttohaveloweraccuracywith
S‐AISdatafromanothertime‐period.
However, the
heuristics in this paper are partly
basedonstaticinformationsuchasshipdimensions.
This combined with a large number of ships can
negatesomeoftheexpectedvariation.Withdatafor
a longer time‐period, the number of vessels
identified,comparedtotheworldfleet,isexpectedto
rise.
Alongerperiodoftimeandmoredatameansa
higher probability for ships to exceed constraints,
suchasmax/minspeedandmax/mindraught.
Anotherlimitationisthatwhenvesselsheetswere
usedtomake training groupstodevelopheuristics,
the only way to match information for each vessel
from the
S‐AIS data was through the ship’s name.
This was the case because the ships were only
identifiedbytheirnameinthevesselsheetsthatwere
used to develop the template groups. As a ship’s
nameisnotauniqueproperty,suchastheMMSIor
IMOnumber,it
isexpectedthatsomeoftheshipsin
thetemplategroupmaybelongtoanothershiptype,
class or size. To avoid this, the candidate groups
themselves were refined through a manualprocess,
where ships not abiding the expected dimensions
were sorted out. The development of the heuristic
was in all
an iterative process, where experience
basedconstraintswereputinplacetoensureabest
possibletemplategroup.
Nonetheless, the results shown in Table 6 show
theproficiencyofthismethod.
Forthegas tankers, especially theQ‐flexand Q‐
maxgroups,fewshipsoutsideofthetraininggroup
werecorrectlyidentifiedasgascarriers,andthusthe
accuracy is somewhat misleading. This could either
becausedbytoostrictrestrictions,meaningthatthe
traininggroupessentiallywasnotrepresentativefor
therestofthegastankers,ormoreprobable,byalow
numberof vesselsofthistypein
theworldfleet.In
otherwords,itislikelythatmostoftheworldfleet
was the training group. In a typical statistical, or
machine learning, model this could be seen as
overfitting, however these heuristics are more
descriptive than predictive in nature, and thus this
become a lesser issue.
However, future studies
shouldbeconductedtoconfirmtheaccuracyofthese
heuristics.
Table6.Accuracyofthedifferentheuristicsmeasuredbythenumberofcorrectlyidentifiedvesselsoutofthoseidentifiedas
acertainshiptypeandsize.ThenumberofvesselsintheworldfleetisaccordingtoMantelletal.(2014)
__________________________________________________________________________________________________
ShipTypeAISvessel Vesselsinthe Vesselsinthe Vesselsidentified Vesselscorrectly Accuracy
group worldfleet traininggroup inS‐AISdataidentified
__________________________________________________________________________________________________
GasCarriers Tankers
Generalgroup24925124999%
Q‐Flex262828100%
Q‐Max101010100%
OilTankers Tankers
UL&VLCC62430937437299.4%
ContainerVessels Cargoships
Panamax87566580772990.3%
BulkCarriers Cargoships
Panamax2,4051,2102,3462,30398%
__________________________________________________________________________________________________
286
6 CONCLUSION
Based on global S‐AIS data from a period of four
months, we developed heuristics to determine the
typeofaship from AISdataalone.Theseheuristics
gives high confidence identificationof ship types of
importance to the international shipping industry.
Thisispresentedasaproof
ofconcepttoshowthata
simple method of identification can give very good
resultsinthespecificcases.Wealsobelievethatthis
method can be extended to many other ship types,
possibly by extending the number of identification
parameters.OurstudyshowedthatS‐AISdatacanbe
erroneous, and require cleaning before reliable
identificationcanbemade.However,thiscleaningis
based on simple and easy to implement criteria. To
furtherimprovetheaccuracyforallshipclasses,data
from a longer time‐period should be used. More
refined heuristics can possibly be made using
techniques such as
advanced cluster analysis.
However,aswehaveshown,acceptableaccuracywas
reachedbyusing themethodoutlined inthispaper.
These heuristics and method are unprecedented in
literatureandenablestudiesonemissionswhereship
type is a factor to be conducted without the use of
commercialvesseldatabases.
REFERENCES
Buhaug,Ø.,Corbett, J.,Endresen,Ø., Eyring,V., Faber,J.,
Hanayama, S., Lee, D., Lee, D., Lindstad, H. &
Markowska,A.2009.SecondIMOGHGStudy.London:
InternationalMaritimeOrganization(IMO).
Clarksons. World fleet register.Retrieved from
http://www.crsl.com/.[Accessed2015].
Corbett, J.J., Wang, H. & Winebrake, J.J. 2009. The
effectiveness
andcostsofspeedreductionsonemissions
frominternationalshipping.TransportationResearchPart
D:TransportandEnvironment14(8):593‐598.
Eriksen,T.,Skauen, A.N.,Narheim,B. &Olsen,R.B. 2010.
Trackingshiptrafficwithspace‐basedAIS:experiencegained
infirstmonthsofoperations.2010InternationalWaterside
SecurityConference(WSS).
Høye,
G.K.,Eriksen,T.,Meland,B.J.&Narheim,B.T.2008.
Space‐basedAISforglobalmaritimetrafficmonitoring.
ActaAstronautica62(2):240‐245.
IALA 2011. Guideline 1082‐An Overview of AIS.Saint
Germain en Laye: International Association of Marine
AidstoNavigationandLighthouseAuthorities(IALA).
IMO 2002. Resolution A.917(22) GUIDELINES
FOR THE
ONBOARD OPERATIONAL USE OF SHIPBORNE
AUTOMATIC IDENTIFICATION SYSTEMS (AIS).
London:InternationalMaritimeOrganization(IMO).
ITU 2014. Recommendation ITU‐R M.1371‐5, Technical
characteristics for an automatic identification system using
time division multiple access in the VHF maritime mobile
frequency band. Geneva: International
TelecommunicationUnion(ITU).
Mantell,C.,
Benson,R.,Stopford,M.,Crowe,T.&Gordon,
S. 2014. Shipping Intelligence Weekly 02.05.2014.
ClarksonResearchServicesLimited.
SESAME. SESAME Straits Web Pages. Retrieved from
http://www.straits‐stms.com/.[Accessed2017].
Skauen, A.N., Helleren, Ø., Olsen, Ø. & Olsen, R. 2013.
Operator andUser Perspectiveof FractionatedAIS Satellite
Systems. Proceedings of
the AIAA/USU Conference on
SmallSatellites,AroundtheCorner,SSC13‐XI‐5.
Smith, T., Jalkanen, J., Anderson, B., Corbett, J., Faber, J.,
Hanayama,S., O’Keeffe,E., Parker,S., Johansson,L. &
Aldous, L. 2014. Third IMO GHG study. London:
InternationalMaritimeOrganization(IMO).
Wahl,T.,Høye,G.K.,Lyngvi,A.&
Narheim,B.T.2005.New
possiblerolesofsmallsatellitesinmaritimesurveillance.
ActaAstronautica56(1):273‐277.
