67
1 INTRODUCTION
Navigational safety has traditionally focused on
issuesofsecurityandthelossoflivesandproperty.
Moreover,concernforenvironmentalprotectionand
collisionavoidanceisincreasingsignificantly.Inthe
MalaccaStraits,anexaminationofthecasualtydata
between1975and1995showsthatseriousaccidents
haveoccurredinthi
shigh‐densitytrafficarea.(8)
The Malacca Straits are a high‐risk area for
navigation. The channel of the Malacca Straits,
considered to be the busiest shipping lane in the
world, islocatedbetween the east coast of Sumatra
IslandinIndonesiaandthewestcoastofPeninsular
Malaysia,andislinkedtotheStraitsofSingaporeat
it
ssoutheasternend.Atapproximately500miles(800
kilometers)long, theMalaccaStraits are the longest
straitsintheworldusedforinternationalnavigation.
They form the main seaway connecting the Indian
Oceanwiththe ChinaSea,and providethe shortest
route for tankers tra
ding between the Middle East
andAsiancountries.(2)
Inthi
sstudy,anAISisimplementedtostudyship
collision probabilities under The International
Regulations for Preventing Collisions at Sea 1972
(COLREG)guidelines.Inthiscontext,theAISisused
asthesourceofdataforthehazardidentificationand
riskevaluationstepsoftheFSA.
2 LITERATUREREVIEW
Several aut
hors have explored risk assessments for
shipcollisionsinthechannel.Mou etal(2010)used
AIS data to study collision avoidance in busy
waterwaysbyperformingstatisticalanalysesofships
involvedincollisions,establishing therisk
assessment model via the SAMSON program.For
this model, the authors only took into account the
Formal Safety Assessment (FSA) for Analysis of Ship
Collision Using AIS Data
M.B.Zaman&A.Santoso
DepartmentofMarineEngineering,FacultyofMarineTechnology,ITS‐Surabaya,Indonesia
E.Kobayashi&N.Wakabayashi
FacultyofMaritimeSciences,KobeUniversity,Japan
A.Maimun
FacultyofMechanicalEngineering,UniversitiTeknologiMalaysia(UTM),Malaysia
ABSTRACT: Currently, Maritime safety is the best issue in the world. International Maritime organization
(IMO)haverecommendedFSAmethodologytoenhancemaritimesafety.Inthispaper,theresearchconducted
inthe MalaccaStrait. Malacca Straitisan areathathas ahigh riskfor shipping navigation.Many accidents
occurintheareaarelikecollision,fire,groundingandsoon.Thereforeastudyonimprovingsafetyinthi
sarea
is very important. itis toproduce anoutput that can beused to provide input to the master and multiple
stakeholderstoimprovesafetyonboa
rdatthetimeofsailing.Inthisstudy,AISisusedasadatasource.Sea
conditiondatacollectedactualtrafficthroughtheAutomaticIdentificationSystem(AIS)equipmentinstalledat
KobeUniversity,Japan,andUniversitiTeknologiMalaysia(UTM)inJohor,Malaysia.Thedataisappliedto
defineamethodwiththehelpofGeogra
phicInformationSystems(GIS).
http://www.transnav.eu
the International Journal
on Marine Navigation
and Safety of Sea Transportation
Volume 9
Number 1
March 2015
DOI:10.12716/1001.09.01.08
68
ships(ownships)thatencounteredaTSSintheport
of Rotterdam. J. Wang et al. (2010) explored the
formalsafetyassessment(FSA) ofcontainerships.In
their study, they used fault tree analysis (FTA) for
hazardidentificationandriskevaluation.Kobayashi
etal.(2008)presentedguidelinesforshipevacuation
during
atsunami.Todoso,theyanalyzedAISdata
fromashipthatpassedinOsakaBay,Japan.Pitana
et al. (2010) analyzed the evacuation of a large
passenger vessel in the case of a pending tsunami
using a stochastic approach, a discrete event
simulation(DES).In this study, they
obtained AIS
data for calculating the sea traffic in the area.
Zamanetal.(2013)examinedthemaritimesafetyin
the Malacca Strait using AIS data and an analytic
hierarchy process (AHP).This data enabled the
rankingofsituationsbasedonascorethatmeasured
danger.Zaman et al. (2013) established
the ship
collisionusingFMEAFUZZYbasedonitsAISdata.
The probability calculation took into account the
trafficdensityinthechannel.
3 ANALYSISOFAISDATA
Thestudyareaofthisresearchisshowninfigure1,
andabreakdownofthenumberofshipstrackedin
June
2010isshowninfigures2and3.
Figure1.Studyarea
Themostdatareceivedonasingledaywasonthe
6/2/2010, when the number of ships transmitting
informationwas1372.Thefigurealsoshowsthatthe
number of ships tends to rise on 6/2/2010 before
decliningfrom6/3/2010.Thefewestshipswereinthe
Straits on the 6/6/2010, when only
1249 ships were
recorded.
1180
1200
1220
1240
1260
1280
1300
1320
1340
1360
1380
1400
NumberOfShip
Day
NumberofShip
Figure2.NumberofshipperdayinJune2010
Fig.3showsthenumberofshipsusingtheStraits
perhouron6/2/2010.Asharpriseinthenumberof
ships can be seen around 07.00h and 08.00h, when
the number of ships was 1047. The period with the
fewestshipsoccursintheearlyhoursofthemorning,
until
around04.00h.
940
960
980
1000
1020
1040
1060
123456789101112131415161718192021222324
NumberofShip
Hour
NumberofshipPer Hour
Figure3.Numberofshipperhouron2June2010
Table 1 shows the population of ships passing
throughthe Malacca Straits on 6/02/2010.The types
of ship can be broken down as: tanker ships 37%,
cargo ships 26%, tugs 9%, passenger ships 3%,
towing and fishing ships 1%, other ships 10%, and
unknownvessels14%.
Table1.ShippopulationbytypeinJune2010
_______________________________________________
TypeofShipPercentageofShip`sNumber
_______________________________________________
Tanker37%
Cargo26%
OtherShip10%
Tugs9%
Passenger3%
Towing1%
Unknown14%
_______________________________________________
4 FORMALSAFETYASSESSMENT
According to the IMO, the FSA is a rational and
systematicprocess for assessing therisks associated
with any sphere of activity, and for evaluating the
costs and benefits of different options for reducing
thoserisks.TheFSAisalsoaformalandintegrated
approach to
assessment. The purpose of applying
this methodis to use the five‐step procedure of the
FSA to make an overall analysis and enhance
maritimesafety.
(3)
The five‐step procedures of FSA are: hazard
identification, risk assessment, establishes safety
measure, cost‐benefit assessment and
recommendation for decision making. The FSA is a
tool designed to assist maritime regulators in the
69
process of improving and deriving new rules and
regulations.
4.1 HazardIdentification
HazardidentificationforshipsintheMalaccaStraits
is an essential part of the risk assessment process.
Hence, a list of hazards that have befallen ships in
thisregionisimportantinordertoidentifyrelevant
risks.
FMEA
‐Fuzzy Hazard Identification of Ship
Collisions
ThetraditionalFailureModeandEffectAnalysis
(FMEA) determines the risk priorities of failure
modes through the risk priority number (RPN),
whichis the product of the probability (P), severity
(S),anddetection(D)offailure.Thatis,
RPN=P×S×
D (1)
Intheproposedmethod,thefuzzificationprocess
usesasingletonmethodtomapacrisppointintoa
fuzzy set, and the fuzzy set rule base comprises a
collectionoffuzzyif‐thenrules.Thefuzzyinference
engine performs a mapping from fuzzy sets, based
on these fuzzy if
‐then rules and the compositional
rules of inference. In the defuzzifier, the Weighted
MeanofMaximum(WMoM)isusedforcalculatinga
crisp output from the system. The input linguistic
variablesarecreatedbasedonFMEA.Thesearethe
P,S,andDofEquation(1).Theoutputofthis
fuzzy
decisionistheprioritynumber
To create a FMEA model using a fuzzy method,
ten scenarios were established based on AIS and
other data. Such hypothetical scenarios will be
explored.Thosescenarioscreatedasscopedomainin
thefuzzylogicmethod.Thevalueofthemembership
function ranges from 0
to 1. Defuzzified values for
each of the scenarios were thus obtained, and the
resultsareshowninTable2.
Table2.Defuzzifiedvaluesoftencollisionscenarios
_______________________________________________
ScenariosValue
_______________________________________________
OverTrafficDensity5.23
Distancebetweenvesselsareveryclosed3.56
ProbleminHeadSituationandHumanError2.87
ProbleminOvertakingSituationandHumanError 6.56
ProbleminCrossingSituationandHumanError4.68
ErrorofNavigation5.80
FailureofMachineryandElectricity6.13
BadWeather5.90
Characteristicareais
notgood5.75
HumanFactors1.16
_______________________________________________
4.2 RiskAssessment
The second step of the FSA is the risk assessment.
Theriskanalysisiscomposedoftwomainactivities:
probabilitymodelingandconsequencemodeling.
4.2.1 ProbabilityAnalysis
Inthispaper,probabilitieshavebeenestablished
based on AIS data and hazard analysis. The factors
analyzed in assessing the ship collision probability
are the head‐on situation, crossing situation,
overtaking situation, and traffic density, based on
AISandGIS(GeographicInformation System)data.
Thetrafficdensitycanbedeterminedas:
m
s
cc
N
DW
(2)
whereN
mis the number of ships using the channel,
D
c is the channel length, and
c
W is the channel
width.Fig.1showstheareaselectedtocalculatethe
traffic density and ship collision probability, based
onAISandGISdata.
Fig.1Areaselectedfor shipcollision probability
calculation,basedonAISandGISdata
Theshipcollisionprobabilityperpassagecanbe
expressedas:
Pa=
i
N ×Pc (3)
where
i
N is the probability number of collisions
per passage, and Pc is failures per passage or
encounter.Pccanbeexpressedas:
Pc=μ
c×T (4)
whereμ
cisfailuresperhourandTisthetimetaken
perpassage.
Theprobabilitynumberofcollisionsinthehead‐
on and overtaking condition per passage can be
expressed.Asfollows,assumingthatfourgroupsof
shipshaveidentical characteristicssuchashead on,
overtaking, left and right side crossing.
The
expressionis:
i
N =4xBxDxρs (5)
and according to the reference,
(7)
the number of
collisionsinthecrossingconditionperpassageis:
4
2
m
i
c
N
NLB
D




(6)
In Equations (5)–(6), B is the mean beam of
meeting (m), L is the mean length of meeting (m),
and D
s is the sailing passage distance, and
m
N is
arrivalfrequencyofmeetingships(ship/time).Inthis
paper, equation 5‐6 taken for calculation of ship
collisionprobability intheMalacca Straits based on
AISdataandGIS.Thenumberofcollisionsperyear
canthenbedeterminedas:
Na=Pa×(365×24/T)
(7)
Tables 3–8 show estimations of ship collision
probabilities in the selected area of the Malacca
Straits.Inthetable3‐8alsoshowthat
m
N isarrival
frequency of meeting ships. In this case,
m
N is
determined based on AIS and GIS, and
i
N is
determinedaccordingtoequation5and6.
70
Table3.CollisionprobabilitybasedonAISdataat02.00h
__________________________________________________________________________________________________
NodeNm Ni
c
 Dc(m) L(m)B(m)T PcPan PaClass
__________________________________________________________________________________________________
1 ShipHeadon12 0.0115287 0.0000151 0.000015 1.729E‐07
2 ShipOvertaking 15 0.0144108 0.000015 24688 26432 1 0.000015 2.162E‐070.0159154Occasional
3 ShipCrossing36 0.0951823 0.0000151 0.000015 1.428E‐06
TotalNumberofShip 631.817E‐06
__________________________________________________________________________________________________
Table4.CollisionprobabilitybasedonAISdataat06.00h
__________________________________________________________________________________________________
NodeNm Ni
c
 Dc(m) L(m)B(m)T PcPan PaClass
__________________________________________________________________________________________________
1 ShipHeadon23 0.0230655 0.0000151 0.000015 3.46E‐07
2 ShipOvertaking 35 0.0350996 0.000015 24688 167 31 1 0.000015 5.265E‐07 0.029533 Occasional
3 ShipCrossing65 0.1665912 0.0000151 0.000015 2.499E‐06
TotalNumberofShip 1233.371E‐06
__________________________________________________________________________________________________
Table5.CollisionprobabilitybasedonAISdataat10.00h
__________________________________________________________________________________________________
NodeNm Ni
c
 Dc(m) L(m)B(m)T PcPan PaClass
__________________________________________________________________________________________________
1 ShipHeadon25 0.0465466 0.0000151 0.000015 6.982E‐07
2 ShipOvertaking 35 0.0651652 0.000015 24688 315 50 1 0.000015 9.775E‐07 0.0513325 Probable
3 ShipCrossing68 0.2789462 0.0000151 0.000015 4.184E‐06
TotalNumberofShip 1285.86E‐06
__________________________________________________________________________________________________
Table6.CollisionprobabilitybasedonAISdataat14.00h
__________________________________________________________________________________________________
NodeNm Ni
c
 Dc(m) L(m)B(m)T PcPan PaClass
__________________________________________________________________________________________________
1 ShipHeadon14 0.0039360 0.0000151 0.000015 5.904E‐08
2 ShipOvertaking 23 0.0064663 0.000015 24688 44 8 1 0.000015 9.699E‐08 0.0047686 Remote
3 ShipCrossing37 0.0258884 0.0000151 0.000015 3.883E‐07
TotalNumberofShip 745.444E‐07
__________________________________________________________________________________________________
Table7.CollisionprobabilitybasedonAISdataat18.00h
__________________________________________________________________________________________________
NodeNm Ni
c
 Dc(m) L(m)B(m)T PcPan PaClass
__________________________________________________________________________________________________
1 ShipHeadon16 0.0171589 0.0000151 0.000015 2.574E‐07
2 ShipOvertaking 35 0.0375352 0.000015 24688 232 32 1 0.000015 5.630E‐07 0.029074 Occasional
3 ShipCrossing63 0.166569 0.0000151 0.000015 2.499E‐06
TotalNumberofShip 1143.319E‐06
__________________________________________________________________________________________________
Table8.CollisionprobabilitybasedonAISdataat23.00h
__________________________________________________________________________________________________
NodeNm Ni
c
 Dc(m) L(m)B(m)T PcPan PaClass
__________________________________________________________________________________________________
1 ShipHeadon19 0.0199517 0.0000151 0.000015 2.993E‐07
2 ShipOvertaking 26 0.0273024 0.000015 24688 190 32 1 0.000015 4.095E‐07 0.0228851Occasional
3 ShipCrossing48 0.1269097 0.0000151 0.000015 1.904E‐06
TotalNumberofShip 932.612E‐06
__________________________________________________________________________________________________
4.2.2 Consequenceassessment
The consequence analysis for each scenario was
carriedout.Fivecategories comprise the built risk
level by using a risk matrix. Table 7 shows the
probability index and the consequence categories.
The consequence analysis is classified as the
following:doesnotresultininjuries,minorinjuries,
majorinjuries,
deathortotaldisability,anddeathor
totaldisabilityofseveral people.Theresultsofthe
consequenceanalysisareplottedasariskmatrix.
Table7.Probabilityindexandconsequencecategories
_______________________________________________
ProbabilityDescription
Index
_______________________________________________
1 Veryunlikely LessthanonceP<1/1000
per1000years
2 RemoteOnceper100–1000years P<1/100
3 Occasional Onceper10–100years P<1/10
4 Probable Onceper1–10yearsP<1
5 Frequent Morethanonceperyear P=1
_______________________________________________
ConsequenceDescription
categories
_______________________________________________
ADoesnotresultininjuries
BMinorinjuries
CMajorinjuries
DDeathortotaldisability
EDeathortotaldisabilityforseveralpeople
_______________________________________________
4.2.3 RiskMatrix
Figures 4, 5, and 6 show the risk matrices for
probabilityandconsequenceanalysesintheMalacca
StraitbasedonAISandGISdata.BasedonAISdata,
the scenario of probability assessment was carried
out with different times using actual data.In this
case,thescenarios
takeninthefollowingtimeshave
high traffic areas: 02:00, 10:00, and 22:00.Figure 4
shows the risk matrix at 02:00, based on AIS data.
In this case, the risk matrix is established based on
the results of the probability and consequence
assessments.Inthiscondition,thenumberofships
71
inhead‐onencountersis12,incrossingencountersis
15,andinovertakingencountersis36.
ThenumbersofshipsisdeterminedbasedonAIS
data in the selected area in the Malacca Strait.
Basedontheprobabilityindex,thehead‐on,crossing,
and overtaking encounters are classified,
respectively,
at points 4, 4, and 5. Based on the
consequence analysis, the head‐on, crossing, and
overtakingencounters are classified,respectively,at
pointsC,C,andD.Thetolerableconditionsarefor
thehead‐onandcrossingencounters.Anintolerable
conditionistheovertakingencounter.
Figure 5 shows the risk
matrix at the 10:00
scenario based on AIS for which the risk level was
establishedinthehead‐on,crossing,andovertaking
conditions.BasedonAISdata,thenumberofships
inhead‐onconditionsis25,crossingconditionsis35,
and overtaking conditions is 68.Based on the
probability
index, the head‐on, crossing, and
overtaking conditions are classified, respectively, at
points4,5,and5.Inaddition,basedonconsequence
analysis, head on, crossing and overtaking are
classified,respectively,atpointC,D,andD.Inthis
case, risk level conditions are the following: a
tolerable level for head
‐on, an intolerable level for
crossing,andanintolerablelevelforovertaking.
Figure6showstheriskmatrixat22:00basedon
AISdata.Therisklevelwas establishedinthehead‐
on, crossing, and overtaking conditions.In these
conditions,thenumberofships inhead‐onis19,in
crossingis26,andinovertakingis48.Therisklevel
conditions are the following: a tolerable level for
head‐on, a tolerable level for crossing, and an
intolerable level for overtaking.The results of
navigation safety based on risk assessments using
AIS data for different times are important for
navigators
toobserveifintransitinthisarea.These
resultsarealsousefultoensuresafetymeasuresand
risk mitigation for enhancing safety in the Malacca
Strait.
N=Negligible;T=Tolerable;I=Intolerable
1 2 3 4 5
A N N N N T
B N N N T T
C N N T T
(
H,C
)
I
D N T T I I
(
O
)
E T T I I I
Figure4.RiskmatrixbasedonAISdataat02:00
1 2 3 4 5
A N N N N T
B N N N T T
C N N T T
(
H
)
I
D N T T I I
C,O
E T T I I I
Figure5.RiskmatrixbasedonAISdataat10:00
1 2 3 4 5
A N N N N T
B N N N T T
C N N T T
(
H,C
)
I
D N T T I I
(
O
)
E T T I I I
Figure6.RiskmatrixbasedonAISdataat22:00
4.3 Safetymeasures
Thisstepaimsatproposinganeffectiveandpractical
safetymeasure.High‐riskareasareidentifiedfrom
theinformationobtainedintheriskassessment,and
then, the development of risk control meas ures can
be initiated.Risk control measures can assist in
reducingtheoccurrencelikelihoodoffailuresand/or
mitigating their possible consequences.Structural
review techniques may be used to identify all
possible risk control measures for cost‐effective
decision‐making.Thesafetymeasuresare
generated from the results of the risk assessment,
which is established based on AIS data.. Table 9
shows the safety measures adopted to reduce
risk
duringshipcollision.
4.4 CostBenefitAnalysis
The technique of cost‐benefit assessment consist on
thefollowingthreesteps:
1 Estimationofthebenefit
2 Estimationofthecost
3 Combinationofbenefitandcost
Aftercomplementingthecost‐benefitassessment,
theresultswiththehighestoverallscoresareselected
4.5 Recommendation
Based on analysis of FSA from step 1‐4, then there
aresomerecommendationasfollows:
1 Improved safety at sea transportation is very
important.Itisnecessary cautionwhen
navigatingashipatsea.
2 Allcrewofshipsshouldimprovetheabilityand
experiencetothe
training.
3 Foravoidhumanerroratthetimeofsailingona
ship, it would require good coordination and
communicationbetweenthecrewwhentheship
4 Maintenance great vessels necessary for the
conditionofmachinery,electricaland navigation
systemsoperateproperly.
72
Table9Safetymeasureofshipcollision
__________________________________________________________________________________________________
AccidentHazardProbability/Consequence RiskSafetyMeasuretoreducerisk
__________________________________________________________________________________________________
EventCausesProbabilityConsequence
_________________________________________________________________________________________
HumanFatigue&lackof Frequent Death/disability Intolerable Increaseknowledge&skills&
errororknowledge&skillspromotecultureofsafety
ShipTypeofships, Probable Death/disability Tolerable Replaceoldshipswithnew
Conditions length,speed,stateshipsandconductcareful
ofloadingexaminationsoftheships
conditions
Collision EnvironmentalDistance
between Probable Majorinjury Tolerable Makenavigationalaids
factorsvesselsiscloseavailable
Machinery Failureofmain Probable Majorinjury Tolerable Conductregularmaintenance
factorsengineorelectronics
Navigational InappropriatecrewProbable MajorinjuryTolerableIncreasecrewmanning
factorsmanningcapabilities
__________________________________________________________________________________________________
5 CONCLUSIONS
This paper presented an implementation of the AIS
for a study on FSA. In this context, the AIS was
implemented as a source of data for the hazard
identificationandshipcollisionprobabilityoftherisk
assessmentstepoftheFSA.
BasedontheAISdata,theshippopulation
passing
through the Malacca Straits on 6/2/2010 was
calculated. This was broken down as: tanker ships
37%, cargo ships 26%, other ships 10%, tugs 9%,
passengerships3%,towingandfishingships1%,and
unknownvessels14%.
Inthispaper,theidentificationofhazardsanalysis
has established. The ranking of
hazard analysis are
probleminheadsituation(6.56), Errorof navigation
(6.13), failure of machinery and electricity (5.9),
problem in crossing situation (5.8), bad weather
(5.75), over traffic density (5.23), problem in
overtaking situation (4.68), Distance between vessels
are very close (3.56), Speed of own ship and target
ship(2.87)and
characteristicareaisnotgood(1.16).
The risk analysis comprised two main activities:
probability modeling and consequence modeling. In
this paper, probabilities were established based on
AISdataandhazardanalysis.
A number of situational and traffic density
factors were analyzed in order to calculate ship
collisionprobabilitiesbasedon
AISandGISdata. In
this case, based on hazard analysis, the head on
condition, overtaking condition and crossing
condition taken to make analysis of ship collision
probabilityin the Malacca Straits. Theresult of ship
collision probability calculation taken based on AIS
dataandGISat02.00h,10.00hand
22.00h.
ACKNOWLEDGMENT
Theauthorswishtothank membersof themaritime
safetysystemlaboratoryatKobeUniversityfortheir
assistance and continuous support of analysis and
evaluationoftheAISandotherdata.
REFERENCES
[1]CheeKuang Tam., Bucknall, R. (2010) “Collision risk
assessment for ship,” Journal of Marine Science and
Technology(JMST),15,pp.257‐270.
[2]Chua Thia‐Eng, Inggrid R.L. Gorge, S. Adrian Ross
(2000) ” The Malacca Straits,”Elsevier Science Direct,
Vol.41,pp.160‐178.
[3]IMO (2002): MSC Circ.1023/MEPC 392, GUIDELINES
FOR
FORMALSAFETYASSESSMENT(FSA)FORUSE
INTHEIMORULE‐MAKINGPROCESS.
[4]Kobayashi, E., Koshimura, S., Yoneda, S. (2008)
“Guidelines for ship evacuation from tsunami attack,”
ProceedingsoftheInternationalSocietyofOffshoreand
PolarEngineers(ISOP)pp.68‐70.
[5]Pitana, T., Kobayashi, E. (2010) “Assessment of ship
evacuation response to pending tsunamis,” Journal of
MarineScienceandTechnology15(3)pp.242‐256.
[6]Papanikolaou,P.,Eliopoulou,A.,Alissafaki,A.,Mikelis.
A., Aksu. S., Delautre. S. (2010) “Casualty analysis of
Aframaxtankers,”ElsevierScienceDirectdoi:10.1016
[7]Sulaiman, O.O., Saharuddin, A. H., Kader, A. S. A. ,
Laily, A.
R. N. (2011) “Safety and Environmental Risk
and Reliability Model for Inland Waterway Collision
Accident Frequency,” International Journal of Energy
andEnvironmet4(3)pp.461‐477.
[8]Sharina Gran. (1999) “The Impact of Transportation on
WidlifeintheMalaccaStraits”,TEDCaseStudies,9(3);
No.573
[9]Wang, J., Foinikis,
P. “Formal safety assessment of
containership.” (2010) Elsevier ScienceDirect, pp. 143‐
157.
[10]Zaman,M.B.,Kobayashi,E.,Wakabayashi,N.,Pitana,
T., Maimun, A. (2013): Implementation of Automatic
Identification System (AIS) for Evaluation of Marine
Traffic Safety in Strait of Malacca using Analytic
Hierarchy Process (AHP), Journal of Japan Society of
Naval Architects and Ocean Engineers, Vol.16, pp.141‐
153.
[11]Zaman, M. B., Kobayashi, E., Sahbi, K., Wakabayashi,
N., Maimun, A. (2013): Fuzzy FMEA Model for Risk
Evaluation of Ship Collision in the Malacca Straits:
Based on AIS Data, Journal of Simulation,Vol.8, pp:91‐
104.