315
1 INTRODUCTION
Marinepollutionshavebeenalwaysattractiveforthe
mankindsinceancienttimes.Beingecologicallyvery
importantfocalcenters,coastalzoneshavebeenalso
experienced human induced negative environmental
problems,suchas,coastalerosion,urbansprawland
pollution. In the global world of today, the
importance of connection between countries and
continents is increasing and tra
nsportation by sea is
preferred due to economical reasons, because it is
very cheap with respect to other transportations
systems. However, sea transportation causes some
problems also. Sea pollution caused by the sea
transportationisabout20%ofthetotalseapollution.
Sea pollution are caused by ships daily a
ctivities
and accidents. In daily operations, ships are
unloading their churns without convenient
internationalrudiments.Seapollutionbyaccidentsof
tankers and ships with dangerous loads causes
importantproblems.Seapollutionbyshipsdamages
sea habitat, which effects human health obliquely,
beside it restricts utilization of sea for different
purposes(Yiğit
,2006).
Shipping is principal to our well‐being, with
around 90% of European Union international trade
going by sea and more than 3.7 billion tonnes of
freightayearbeingloadedandunloadedinEUports.
If not correctly controlled, the effect on the
environmentcouldbedestructive, asshipsoftencarry
large volumes of hazardous cargo and generate a
significantam
ountofpollutantsthroughouttheirlife
cycles(EMSA,2008).
Investigation of Marine Pollution Caused by Ship
Operations with DEMATEL Method
Ü.Özdemir,H.Yılmaz&E.Başar
KaradenizTechnicalUniversity,Trabzon,Turkey
ABSTRACT:Shipshaveanimportantroleinamongthefactorscausingmarinepollution.Marinepollutionby
shipsdamagessealife,whicheffectshumanhealthindirectly,inadditionitrestrictsusageofseafordifferent
purposes. Increasing comprehensive and compelling liabilities related with environmental components and
subjectsdaybydayareexpectedresultsforenvironmentalscienceandengineeringapplicat
ionsaccordingto
60% of our responsibilities of European Union Integration development. In today’s’ world, where global
warmingisanissueofhighpriorityandmakesusfeelitsimpactsinourlives,leavingalivableworldtothe
nextgenerationsisnowaprima
rygoalforall.Determinationofmarinepollutioncausedbyshipoperations
issue is a mu1tiple criteria decision‐making (MCDM) problem, and requires MCDM methods to solve it.
Therefore, the role of ship factor in maritime pollution and the possible reasons of this argument can be
q
uantitativelyevaluatedbasedonexpertknowledgeandMCDMmethodology.Toinvestigatewhatmakesto
reduce the first “caused by ship operationsʺin marine pollution, the decision‐making trial and evaluation
laboratory(DEMATEL)methodapproachwasappliedinthisstudy.
http://www.transnav.eu
the International Journal
on Marine Navigation
and Safety of Sea Transportation
Volume 10
Number 2
June 2016
DOI:10.12716/1001.10.02.14
316
Many of pollutants are released by vessels either
operationally or accidentally. The most important
environmental damages caused by discharging
household wastes and bilge water, dumping ballast
water and wash water from tankers, emission of
exhaustfumes,leachingofanti‐foulpaints,pollution
with toxic materials, removal, introduction of
organisms and
acoustic and visual disturbances
(Robert,2001).
Shipping cause problem troubles to the
environment both on inland waterways and on the
ocean. These problems come from six major origins;
routinedischargesofoilybilgeandballastwaterfrom
marine shipping; dumping of non‐biodegradable
solid waste into the ocean; accidental spills
of oil,
toxics or other cargo or fuel at ports and while
underway; air emissions from the vesselsʹ power
supplies; port and inland channel construction and
management;andecological damageinconsequence
oftheintroduction ofoutlandishspeciestransported
onvessels(OECD,1997).
The release of oil and other harmful substances
(includingnoxiousliquid substances,sewage,garbage
and air pollution) into the marine environment is
regulated in great detail in the International
ConventionforthePreventionofPollutionfromShips
(MARPOL). This convention was adopted by the
InternationalMaritimeOrganization(IMO)in1973.It
has been amended a number of times and is
being
continuously complemented and strengthened to
meet the ever‐increasing demands of the world
community (EMSA, 2008). In the pa st few decades,
international, regional and national regulation over
shipping matters such as navigational safety, ship‐
sourcepollutionandmaritimesecurityhavegrownto
suchascopethattheglobalshippingindustry
today
facesalitanyofcostlyregulatoryrules.Accordingly,
theshipowner’sconventionalrightoffreenavigation
ispresentlyqualifiedbyconsiderablerequisitessuch
astheprotectionofthemarineenvironmentand the
promotion of maritime safety. In particular, the
emphasis on marine pollution control by relevant
coastal and port
states has come to fundamentally
erodethetraditionalrightoffreenavigationaccruing
to maritime states and their shipping interests (Tan,
2006).
Shipping also causes more invisible types of
pollution. Recent concerns include the harmful
environmental effects of substances in anti‐fouling
paint used on shipsʹ hulls, and of species which
are
transported from one sea area to another in ballast
water tanks. Also, the bilge water includes a high
amount of dirtiness as well as its toxic, corrosive,
inflammable / explosive characteristics. Discharging
ofthebilgewateroutofthevesselthroughthepumps
andwithoutwaitingforalong
timeisrequired,butit
is pumped directly into the marine environment.
Directly discharge process of bilge water is very
harmfulformarineenvironment.Bothconcernshave
led the IMO to adopt specific conventions on the
topics in 2001 and 2004 respectively. In addition,
pollution is caused when ships are constructed
and
maintained,andwhentheyaredismantledattheend
oftheirlifecycles.Thelatterisofparticularconcernat
thepresenttime,giventhatmuchoftheworldʹsship
breakingisdoneincountries whereneither workers
northeenvironmenthaveadequateprotectionagainst
harmful practices and
substances involved in ship
recycling. A convention on the safe and
environmentallysoundrecyclingofshipsiscurrently
beingdraftedbytheIMOanditissetforadoptionin
May2009(Emsa,2008;Anderson,2009).
According to the IMO (International Maritime
Organization) MARPOL Code, marine pollution
caused by ships are
affected by factors such as
operational pollution and accidental pollution.
Operational pollution means the phenomenon that
ship‐cause marine pollution is not confined to
accidents. In fact, the majority of pollutants are
released while the ship is on voyage rather than
accidentally. In this respect, activities include the
chronic discharge
of sewage, tank residues, bunker
oils and garbage, as well as the exchange of ballast
water,emissionsfromvessels’enginesandpollution
duetoanti‐foulingpaintsonships’hulls,dumpingof
garbage and solid waste, resulting of oil and waste
waterafterdeckwashingoperations,pouringofcargo
intothe
sea,giving directlytotheseaofrawsewages.
In this sense, determination of marine pollution
caused by ships operations are a kind of mu1tiple
criteria decision‐making (MCDM) problem. So,
proposed method is developed for selection with
decision‐making trial and evaluation laboratory
(DEMATEL)as ashiproutine operations
processon
marinepollution.
This research utilizes DEMATEL technique to
explaintherelationshipsbetweenthevariouscriteria.
DEMATEL is a comprehensive method for building
and analyzing a structural model involving causal
relationships between complex perspectives. This
study aims to utilize the a kind of mu1tiple criteria
decision‐making (MCDM) method, named
decision‐
making trial and evaluation laboratory (DEMATEL)
technique approach to recognize the influential
criteria of marine pollution caused by ships routine
operations.
2 METHODOLOGY
Marine pollution caused by ships usually occur due
tocombinationofcoincidentalincidentsorprocesses,
as a general rule by negligence of one or more
independent
components that are required to action
accurately for the successful finalizations of the
system requirements. The process of determination
marine pollution caused by ships is required to
handle several complicated factors in a better
conceivableandlogicalmanner.So,determinationof
marine pollution caused by ships issue is a kind of
mu1tiplecriteriadecision‐making(MCDM)problem.
To solve this problem, we used a MCDM method,
called DEMATEL. DEMATEL developed by the
Science and Human Affairs Program of the Battelle
MemorialInstituteofGenevabetween1972and1976
was utilized to study and resolve the complex and
intertwinedproblemgroup(Tzeng
andChiang,2007).
In this study, decision‐making trial and evaluation
laboratory (DEMATEL) method is applied because
thismethodgeneratescausaldiagramstodescribe the
basic concept of contextual relationships and the
strengths of influence among the criteria (Wu and
Lee,2011).
317
In this study firstly, marine pollution caused by
routineshipoperationscriteriawasdetermined.The
criteria involved in the marine pollution selection
havebeenchosenaccording totheIMO(International
Maritime Organization) MARPOL Convention. In
next step, the DEMATEL analysis was applied in
order to determine the criteria as follow
section 2.1.
Toestablishthenetworkrelationshipsamongcriteria
in influence each other for the marine pollution
selection involves a decision making team which
includes 2 academics personal, 4 experienced
captains.Then,decisionmakingteamE1,E2,...E6is
constituted to determine the network relationships.
And give
the performance scores for each expert in
terms of all criteria in the evaluation hierarchical
structure respectively. A questionnaire was used to
find out influential relations from each expert for
ranking each criterion on the appropriate marine
pollution caused by ships routine operations with a
four‐point scale ranging from 0
to 4, representing
from ‘No influence (0),’ to ‘Very high influence (4),’
respectively. For each pairwise comparison, the
decisionmakingteamhavetodeterminetheintensity
of the relative importance between two criteria. The
computation of using DEMATEL technique is based
upon these six experts’ opinions. So there are 6
dimensions, the six 6 X 6 matrices. We utilized the
DEMATEL to construct the influence map in
accordance with the real situation in which criteria
should be interdependent and determine the
importanceofcriteria.
2.1 DEMATELMethod
In generally, the DEMATEL method is used to
illustrate the relations between criteria and
to reach
the main factor/criteria to symbolize the impact of
factor (Tzeng et. al, 2007; Wu and Lee, 2011). The
DEMATELmethodisestablishedondigraphswhich
can discrete involved factors into cause and effect
groups(Yanget.al,2008).Thismethodhasalsobeen
individually used in many activities
such as safety
problems(Liouet.al,2007),transportation(Chenand
Yu, 2008), supply chain management (Chiu et. al,
2007)andautomotiveindustry(WuandLee,2011). 
The DEMATEL method is briefly described as
follow(WuandLee,2011;Liouet.al,2007;Chenand
Yu,2008).
Step 1:
Calculating the direct‐relation matrix.
SupposewehaveLexpertsinthisstudyandncriteria
to consider. Firstly, the relationship between criteria
requires that the comparison scale be designed as
with five levels, where scores ranging from 0 to 4
represent ‘‘no influence’’ to ‘‘very high influence’’,
respectively. Experts
are answered the direct
influencedegreebetweencriterion‘’u’’ andcriterion
‘’v’’, as indicated by Zuv. The initial direct‐relation
matrix Z= [Z
uv]|Y| is determined owing to pairwise
comparisons in terms of influences and directions
between criteria. Then, as the result of these
evaluations, the initial data can be obtained as the
direct‐relationmatrixthatisamatrixZ,inwhichaZ
uv
is denoted as the degree to which the criterion u
affectsthecriterionv.Thescoresbyeachexpertwill
give us a ‘’n x n’’ non‐negative answer matrix X
k
=
[X
k
uv]nxn, with 1≤k≤L. Thus, X
1
, X
2
, . . . , X
L
are the
answer matrices for each of the L experts, and each
elementofisanintegerdenotedbyX
k
uv.Thediagonal
elementsofeachanswermatrixX
k
areallsettozero.
Wecanthencomputethe‘’nxn’’averagematrixZfor
allexpertopinionsbyaveragingtheLexperts’scores
asfollows:

1
1
k
k
uv
nn
nn
k
Zuv x X
L

(1)
Step2:Normalizingthedirect‐relationmatrix.
The normalized direct‐relation matrix M can be
obtainedbyformulas:
M
ZL
 (2)
1
1
, ,v 1,2,...,n
max1
n
v
Lu
un Zuv


 (3)
Step 3: Calculating the total‐relation matrix. After
the normalized direct‐relation matrix M is obtained,
thetotalrelationmatrixKcanbeacquiredbyformula
(4), in which the H is represented as the identity
matrix.

1
KM HM
 (4)
Step4:Thesumofrowsandcolumns areseparately
denotedasDandRwithinthetotal‐relationmatrixK
throughequations(5)to(7).
, 1,2,3,...,
uv
Kkuv n  (5)

1
n
uuv
v
DD k




 (6)

1
n
vuv
u
Rk




 (7)
The DEMATEL method analysis was used to
obtain the initial direct‐relation matrix with using
pairwise comparison with the total relation matrix
with D+R, D‐R values and build a critical relative
graph of criteria in the cluster effect. ‘’Du’’ denotes
therowsumofithrowofmatrix
K.Then,Dudenotes
thesumofinfluencedispatchingfromfactorvtothe
other factors both directly and indirectly. Rv shows
thecolumnsumofvthcolumnofmatrixK.Rvshows
thesumofinfluencethatfactoruisreceivingfromthe
otherfactors.Thesumofrowsum
andcolumnsum
(D+R)showstheindexofrepresentingthestrength
of influence both dispatching and receiving.
Furthermore,if(D‐R)ispositive,thenthefactor‘’u’’
isratherdispatchingtheinfluencetotheotherfactors,
318
andif(D‐R)isnegative,thenthefactor‘’u’’israther
receivingtheinfluencefromtheotherfactors.
Step5:Determiningathresholdvaluetoobtainthe
digraph.SincematrixKprovidesinformationonhow
onefactoraffectsanother,itisnecessaryforadecision
maker to set up
a threshold value to reduce some
negligible effects. For these reason, only the effects
greaterthanthethresholdvalueischosenandshown
indigraph.Inthisstudy,thethresholdvalueissetup
bycomputingthe average oftheelements inMatrix
K. The digraph can be acquired by
mapping the
datasetof(D+R,D‐R).
3 EMPIRICALSTUDY
Anempiricalexampleforthemostimportantcriteria
selection for the marine pollution caused by ship
routine operations is illustrated to demonstrate the
proposedmethodtobemorerationalandsuitablein
thissection.
Six criteria involved in the marine
pollution
caused by ship operations selection are used in this
empirical study. The criteria have been chosen
according to the IMO (International Maritime
Organization) MARPOL Convention. A decision
making team were invitedto answer the
questionnaire. The computation of using DEMATEL
methodisbasedsixdecisionmakingteam’sopinions.
Using
the6X6pairwisecomparisons,theaveragesof
their opinions were calculated . DEMATEL method
was used to evaluate the criteria which potentially
effectiveindeterminingmarinepollution.Thecritical
impactofeachnodeonthemarinepollutionselection
andthenetworkeffectweredetermined.Initialdirect
relation matrix was created based
on pair‐wise
comparisonoftotal relation matrix values (D+Rand
D‐R) and the criteria were clustered in the form of
critical relative graph by applying DEMATEL
technique. The relative graph is necessary for
explaining the network structure of the clusters in
determiningthemost influentialcriteria on
recreational
beach decision. In order to generate the
objective super‐matrix of ANP, the network effect
should be first built using DEMATEL. Then, the
average initial direct matrix Z is obtained based on
formula(1)asTable1.
Normalized initial direct‐relation matrix M is
calculatedthroughformulas(2)and(3).Sequentially,
the total relation matrix K is also derived utilizing
referto(4)showninTable2.
Total sum of effects given and received by each
criterion is seen in Table 3 using formulas (6) and
(7).Table3providesthedirectandindirecteffectsof
six dimensions. Finally, the threshold value
(0.7218)
used in Step 5 is to compute the average of the
elements in Matrix T. The digraph of these six
dimensions is demonstrated and the network
relationshipmapofDEMATELmethodwasobtained
andshowninFig.1.
Table1.InitialdirectmatrixZ.
Table2.TotalinfluentialrelationmatrixK.
Table 3 shows that a “bunker oils and bilge
waters” criterion is the most important dimension
withthelargest(D+R)valueof8,624whereas“Anti‐
fouling paints” criterion is the least important
dimension with the smallest value of 6,697. The
importanceofdimensions canbe determined by the
(D
+R)values.Tofurtherinvestigatethecause‐effect
relationship of dimensions, ballast waters, garbage
and solid wastes and deck, hold, vs. washing
operations are net causes based on positive (D‐R)
values.
319
Table3. Sum of influences given and received on each
criterion
Figure1. Network relationship map of impacts for the
marinepollutioncausedbyshipoperationscriteria
Bunkeroilsandbilgewaters,andsewageswaters
are net receivers due to negative (D‐R) values.
Furthermore, deck, hold, vs. washing operations
(Dailyopr.),ballastwatersandanti‐foulingpaintsare
the three most essential dimensions to improve the
marinepollutioncausedshipsbyfurtherconsidering
the causal relationships. To
further investigate the
cause‐effect relationship of dimensions, “deck, hold,
vs.washing operations,(Daily opr.)”, ballastwaters,
anti‐foulingpaintsandgarbageandsolidwastesare
net causes based on positive (D‐ R) values. Bunker
oils and bilge waters, and sewages waters are net
receivers due to negative (D‐R)
values. From Fig. 1,
bunker oils and bilge waters, garbage and solid
wastes, and “deck, hold, vs. washing operations
(Daily opr.)” are the three most critical dimensions.
Specifically, “deck, hold, vs. washing operations
(Daily opr.)” directly affects bunker oils and bilge
waters and sewages waters. Six dimensions are
influenced or mutually
influenced by any pa ir of
dimensions except “deck, hold, vs. washing
operations(Dailyopr.)”.Insummary,“deck,hold,vs.
washing operations (Daily opr.)” is the most
importantdimensionfollowedbyballastwatersanti‐
fouling paints then garbage and solid wastes
criterion. Therefore, “deck, hold, vs. washing
operations (Daily opr.)”, ballast
waters and anti‐
foulingpaintsarethethreemostessentialdimensions
to improve the minimizing marine pollution caused
byshipoperations byfurtherconsidering thecausal
relationships.
4 CONCLUSION
In this paper, we would like to new framework for
determineandevaluatedthemarinepollutionrelated
to ship operations with above
mentioned these six
criteria. This study presents DEMATEL method for
analyzingtheimportanceofcriteriacorrelationsand
also to describe the contextual relationships among
these criteria. Based to the results, six criteria have
someinterrelationswitheachother.Accordingtothe
impact‐direction map, we can make accurate
decisions. Also,
the results showed that bunker oils
andbilgewaters is a key value factor and powerful
influential criteria. However, by further considering
causalrelationships,improvingbothbunkeroilsand
sewages waters factor dimensions cannot be
effectively strengthen minimizing marine pollution
caused by ship operations because these two
dimensions are influenced by
the other dimensions.
In contrast, “deck, hold, vs. washing operations
(Daily opr.)”, ballast waters, anti‐fouling paints and
garbage and solid wastes factor are all causal
dimensionsandhavepositiveimpactsonbunkeroils
andbilgewaters,andsewageswatersfactorsuchthat
any improvement on these dimensions would also
improve bunker oils and bilge waters, and sewages
waters. The proposed framework brings several
contributions to marine pollution evaluation and
selection. At first, a new model for selecting marine
pollutionwithemphasisonshipoperationsissueshas
been developed. At second, the DEMATEL method
wasappliedinselectingmarinepollutioncause
with
respecttoshiproutineoperations.Thisfeatureisalso
uniquewithregard to previous studies. In addition,
results of these modeling can be used with other
decision making method. Application of other
decision support approaches would help to extend
the analysis such as analytical network process
(ANP).
REFERENCES
Anderson,
D. 1998. The Roles of Flag States, Port
States, Coastal States and International
OrganizationsintheEnforcementofInternational
Rules and Standards Governing the Safety of
Navigation and the Prevention of Pollution from
ShipsundertheUNConventionontheLawofthe
SeaandotherInternationalAgreements.Singapore
JournalofInternationalandComparativeLaw.2,557‐
578.
Chen, C.H. & Yu, W.Y. 2008. Using a Strategies
Approach to Analysis the Location Selection for
High‐TechFirms in Taiwan. Management Research
News,
31(4),228‐244.
Chiu,Y.J.,Chen,H.C.,Tzeng,G.H.&Shyu,J.Z.2006.
Marketing strategy based on customer behavior
for
theLCD‐TV.InternationalJournalofManagement
andDecisionMaking,7(2),143–165.
320
EMSA,2008.EuropeanMaritimeSafetyAgency
Work
Programme2008.s.53.
Liou,J.J.H.,Tzeng,G.H.&Chang,H.C.2007.Airline
safety
measurement using a hybrid model. Air
TransportManagement,13(4),243–249.
OECD, 1997. Organisation for Economic Co‐
Operation and Development, The Environmental
Effects Of Freight. Paris: Head of Publications
Service.
Robert, B. 2001.Protecting the Oceans from Land‐
Based Activities. Oxford: GESAMP Report and
StudiesNo.71.
Tan,A.K.J.2006.Vessel‐SourceMarinePollutionThe
Law and Politics of International Regulation.
Cambridge:CambridgeUniversityPress.
Tzeng, G.H., Chiang, C.H. & Li, C.W. 2007.
Evaluating intertwined effects in e‐learning
programs:anovel hybridMCDMmodel basedon
factoranalysisandDEMATEL.Expert Systemswith
Applications,32(4),1028–1044.
Yang, Y.P.O., Shieh, H. M.,Leu, J.D. & Tzeng G.H.
2008. A Novel Hybrid MCDM Model Combined
with DEMATEL and ANP with Applications.
InternationalJournalofOperationResearch , 5(3),160‐
168.
Yiğit,F. 2006.Gemi KaynaklıKirleticilerveTrabzon
Limanına Gelen Bazı Gemilerin Atıksularının
İncelenmesi. Karadeniz Teknik Üniversitesi, Fen
BilimleriEnstitüsü,YüksekLisansTezi.
Wu, W.W. & Lee, Y.T. 2011. Developing global
managers’ competencies using the fuzzy
DEMATEL method. Expert Systems with
Applications,32(2),499–507.