Le e aging ad anced ma ine design and decision-making me hods o
add ess he ma i ime deca boniza ion challenge
Joseph Bu goyne1,* and Thomas A. McKenney1
1 Uni e si y o Michigan, Ann A bo , MI, USA
Abs ac . Global clima e a ge s and s ic e ma i ime egula ions ha e accele a ed he u gency o shipping o
deca bonize. The consequen challenge lies no only in he imma u i y o possible solu ions bu in he cascading
unce ain ies ha shape in es men , design, and ope a ional choices. This pape examines how unce ain y has
been add essed in ma i ime deca boniza ion esea ch and wha lessons can be d awn om indus y. This e iew
maps comme cial, egula o y, echnological, and p io i y unce ain y agains he analy ical me hods mos o en
applied, om scena io analysis, mul i-c i e ia decision analysis(MCDA), o op imiza ion and simula ion. The
e iew shows a hea y eliance on scena io-based explo a ion, equen use o MCDA o na iga e compe ing
p io i ies, and egula o y unce ain ies o en being amed in ine ec i e s a ic iews. I also highligh s he
implici hyb idiza ion o me hods, which is o en unde epo ed and limi s anspa ency. To mo e om
desc ip ion o ac ion, his s udy in oduces a s uc u ed decision amewo k ha adap s exis ing unce ain y
pos u es in o a decision ee o ma . An illus a i e case demons a es how he amewo k wo ks o align
o ganiza ional con ex – esou ces, isk ole ance, imelines – wi h app op ia e me hods o guide o e all
s a egy. The con ibu ion lies in linking analy ical igo wi h decision pos u es, o e ing p ac ical suppo o
di e en s akeholde s na iga ing he unce ain y o ma i ime deca boniza ion.
Keywo ds: Ma i ime deca boniza ion; decision-making unde unce ain y; s a egic amewo k; scena io
anlaysis, s akeholde con ex ; egula o y unce iany; comme cial unce ain y; echnolgoical unce ain y
1 In oduc ion
The ma i ime indus y aces an u gen need o deca bonize in esponse o global clima e a ge s, including he
Pa is Ag eemen and he IMO's ne -ze o emissions goal by 2050. This challenge equi es a subs an ial shi in
app oach, d i en by he eme gence o al e na i e uels, new echnologies, and e ol ing egula o y and ma ke
condi ions. Indus y s akeholde s mus enhance hei decision-making p ocesses o ship design and ope a ion,
ocusing on ene gy e iciency, educed emissions, and ca bon p icing. While a ious echnologies a e being
de eloped o achie e ma i ime deca boniza ion, hey in oduce new economic and echnical conside a ions,
including inc eased ship cos s and complexi y, eliabili y conce ns, and g ea e unce ain y o e all.
The signi icance o hese eme ging conside a ions o majo shipping segmen s c ea es a gap ha academic
esea ch, wi h i s expe ience in complex ship design and decision-making unde unce ain y, is well-posi ioned o
add ess. Indus y epo s on ma i ime deca boniza ion p o ide egula ly upda ed o e iews on ene gy e iciency,
al e na i e uels, and gene al pa hways o achie e sec o -wide deca boniza ion by 2050, mainly se ing egula o s
and o e ing help ul bu non-speci ic backg ound o ship designe s and ope a o s . Meanwhile, academic esea ch
inc easingly explo es sus ainabili y, deca boniza ion, and complex ship design, especially he challenges o
de ining equi emen s o echnologically ad anced and ope a ionally di e se essels such as na al comba an s
and, mo e ecen ly, comme cial ships like con aine ships, bulk ca ie s, and anke s, which a e now acing simila
complexi ies due o deca boniza ion demands [4], [5], [6].
Ul ima ely, he e is an impe a i e o immedia e ac ion, sugges ing ha a combina ion o egula o y
compliance, echnical p og ess, and inno a i e comme cial s a egies is essen ial o deca bonize he ma i ime
indus y. Decision-making unde unce ain y is an impo an aspec ha equi es new s a egies, ools, and a
holis ic app oach ha includes design, ope a ion, and lee -le el pe spec i es. Conside ing he complexi y and
dynamic na u e o he deca boniza ion challenge, he ma i ime indus y can employ a blend o s a egic app oaches
o manage unce ain ies.
* Co espondence o: jbu [email protected]
16 h In e na ional Symposium on P ac ical Design o Ships and O he Floa ing S uc u es PRADS 2025
Ann A bo , MI, USA, Oc obe 19 h – 23 d 2025
2
This pape builds upon “The Impac o Ma i ime Deca boniza ion on Ship Design: S a e-o - he-A -Repo ”
p esen ed a he In e na ional Ma ine Design Con e ence in June 2024 [5] by:
• P o iding a amewo k o guide he selec ion o a decision-making app oach,
• Re iewing in de ail exis ing me hods o decision-making unde unce ain y and hei cu en
applica ions o ma i ime deca boniza ion, and
• Iden i ying emaining gaps in add essing he comple e deca boniza ion challenge.
This esea ch aims o s imula e u he discussion and os e collabo a ion be ween academia and indus y in
ackling he u gen ma i ime deca boniza ion challenge and ca e s o bo h esea che s and p ac i ione s, p o iding
ac ionable insigh s o hei daily wo k. Fo esea che s, indus y de elopmen s a e mo ing quickly, and help is
needed o b ing ele an use cases and ensu e hey a e add essing he igh challenges and applying hei esea ch
in he mos e ec i e way possible o maximize impac . This includes clea ly iden i ying he bes applica ions o
gi en me hods a ailable o unde de elopmen . Fo p ac i ione s, he e is a need o b ing ship design me hods and
decision-making p ocesses in o he no mal way o wo king o bes handle he inhe en unce ain ies and dynamics
associa ed wi h he ma i ime deca boniza ion challenge.
2 The Ma i ime Deca boniza ion Challenge
The indus y’s deca boniza ion challenge can be cap u ed in h ee main s a emen s:
• Fi s , we need o ac now in an indus y ha is no easy o change.
• Second, he e a e many op ions, mos o which a e no ully ma u e.
• And hi d, key s akeholde s mus handle unce ain y and dynamic condi ions.
Shipping accoun s o oughly 3% o global emissions and is conside ed a “ha d- o-aba e” sec o [7]. I
ma i ime emissions a e no educed, he sec o may be esponsible o a g ea e sha e o global emissions by 2050,
as o he sec o s such as powe and oad anspo deca bonize a a as e pace [1], [2], [3]. His o ical esis ance o
change, long essel li espans, high capi al in ensi y, and slow-mo ing global go e nance cons ain swi
ans o ma ion. Ye he sec o is now en e ing a pe iod o accele a ing ans o ma ion, spu ed by egula o y shi s
– especially he IMO’s enhanced a ge s o ne -ze o emissions by 2050 and s ic e egional policies like hose
om he EU – ha a e accele a ing change and aising compliance isks.
Responses o his ising p essu e a e mul i ace ed and equi e de elopmen in h ee main a eas: egula ion,
echnical, and comme cial. Regula o y measu es a e becoming s ic e and mo e complex, c ea ing di ec
compliance equi emen s and indi ec ma ke signals. Technical solu ions a e p oli e a ing, om ene gy e iciency
measu es o new uel sys ems, bu emain a a ying le els o ma u i y and in eg a ion complexi y. Comme cial
iabili y unde pins adop ion, o cing owne s and ope a o s o weigh unce ain uel supply chains and ola ile
p icing. E en u he , hese unce ain ies a e no expe ienced in isola ion. The complex ela ionships be ween
ma i ime s akeholde s esul s in decision-making being in e dependen ac oss di e en ac o s.
Figu e 1. Ma i ime deca boniza ion ship design cause-and-e ec chain (McKenney 2024)
3
Figu e 1 summa izes he ma i ime deca boniza ion challenge wi hin he ship design con ex as a cause-and-
e ec chain, whe e egula o y p essu es d i e shi s in design objec i es, which in u n s imula e he adop ion o
echnical solu ions and in oduce new laye s o eme ging conside a ions. A each s age, unce ain ies accumula e,
egula o y shi s al e compliance equi emen s and u u e baselines; design emphasis owa d e iciency and
emissions educ ion o ces ade-o s in pe o mance; echnical solu ions such as al e na i e uels o EE
echnologies ca y isk o imma u i y, in eg a ion complexi y, and unce ain y ega ding long- e m iabili y. These
compounding unce ain ies do no simply complica e he echnical ask o design, bu di ec ly shape he
comme cial and s a egic decisions ha ollow. The chain no only highly he d i e s o deca boniza ion bu also
he expanding decision space in which shipowne s, po s, and designe s mus ope a e unde deep and pe sis en
unce ain y.
2.1 App oaching Decision-Making
Decision-making unde unce ain y is widely s udied, bu wha makes he ma i ime case dis inc is he scale
and speed o change being demanded in a adi ionally conse a i e indus y. The sec o has long elied on
inc emen alism: de e minis ic o ecas s, e iciency- ocused upg ades, and s epwise egula o y compliance. These
app oaches p o ed e ec i e in ela i ely s able con ex s, whe e uel ma ke s we e p edic able, and egula ions
ad anced g adually. Howe e , oday, e ol ing egula ion and consequen comme cial ola ili y combine o
o e whelm inc emen al app oaches, lea ing decision make s exposed.
The deca boniza ion challenge o ces s akeholde s o con on ques ions beyond echnical easibili y o sho -
e m p o i abili y. Shipowne s, po s, and echnology de elope s each ace high-s akes choices, om e o i s and
in as uc u e planning o pilo ing new echnologies, o en unde unclea egula o y o ma ke condi ions. These
decisions no only equi e ho ough analysis bu also depend on o ganiza ional s a egy and a e shaped by
esou ces, capabili ies, imelines, and isk ole ance. The same unce ain y may be ole able o a di e si ied lee
owne , ye exis en ial o a smalle ope a o .
S eng hening hese decisions equi es g ounding s a egy in analy ical suppo . Sophis ica ed me hods
inc easingly explo ed in academia, i.e., scena io analysis, simula ions, and mul i-c i e ia e alua ion, o e igo in
compa ing ade-o s and es ing obus ness. Howe e , hei impac depends on how hey a e aligned wi h he
pos u e o he decision-make . A i s mo e , a cau ious ollowe , and a hedge may ace he same unce ain y ye
equi e di e en analy ical app oaches o ac e ec i ely. This ecogni ion ames he pu pose o he ollowing
li e a u e e iew. Mapping cu en pa e ns in how me hods a e applied o di e en unce ain y ypes is a necessa y
i s s ep owa d in eg a ing analy ical me hods wi h s a egic decision-making. Sec ion 3, he e o e, e iews how
he li e a u e has engaged wi h comme cial, egula o y, and echnological unce ain ies, and e alua es he
analy ical me hods mos o en employed in esponse.
3 Li e a u e Re iew: Me hod Use in Ma i ime Deca boniza ion
This sec ion e iews how a ious decision-making me hods ha e been applied in ma i ime con ex s o da e,
and wha pa e ns eme ge om hei usage. To g ound hese obse a ions in e idence, an ex ensi e li e a u e
e iew was conduc ed, su eying o e 50 pape s ocused on decision-making unde unce ain y wi h an
emphasis on ma i ime deca boniza ion. While in ended o build a ep esen a i e pic u e o how unce ain y is
add essed in ma i ime deca boniza ion and planning, i mus be no ed ha his da a se may no cap u e e e y
indi idual expe ience o app oach – he esul s should be ead as a e lec ion o b oade pa e ns a he han a
de ini i e census.
This body o li e a u e spans om echno-economic models, ship design s udies, uel selec ion amewo ks,
and li e cycle assessmen s, and e lec s a wide ange o academic and applied pe spec i es. The objec i es o he
e iew we e o:
• In es iga e how and when unce ain y is acknowledged in ma i ime deca boniza ion esea ch,
• Map he me hods employed o add ess unce ain y in ma i ime and compa able complex sys ems,
• Iden i y ecu ing pa e ns, eme ging p ac ices, and me hodological gaps ha can in o m u u e ool
selec ion.
4
3.1 Inc easing Academic Engagemen
O e he pas 20 yea s, he e has been a g owing academic and comme cial esponse o he challenges o
ma i ime deca boniza ion and he b oade p oblem o decision-making unde unce ain y. Ea ly con ibu ions in
he mid- o-la e 2000s and ea ly 2010s we e limi ed in numbe and scope, o en ocused on disce ning wha
s akeholde s’ p io i ies should be and om sus ainabili y planning in adjacen ields such as o es y,
anspo a ion, o powe sys ems [8], [9].
Figu e 2. Numbe o publica ions on decision-making unde unce ain y in deca boniza ion and sus ainabili y,
cumula i e ba s composed o ma i ime- ocused publica ions and o he s, as no a ed.
Following 2015, a clea shi occu ed, wi h a mo e ocused wa e o esea ch mos likely ied o emission
egula ion, he eme gence o al e na i e uels, and a clea e de ini ion o policy in he shipping sec o . The adop ion
and enac men o policies and ad isemen s such as he IMO’s ini ial GHG s a egy and he EU’s Fi o 55 packages
may ha e p omp ed he su ge in s udies a ge ing ma i ime-speci ic applica ions[10]. Since 2020, he olume o
published wo k add essing unce ain y in ma i ime deca boniza ion has accele a ed d ama ically. This ise in
engagemen e lec s ecogni ion ha unce ain y is no a pe iphe al conce n in ma i ime deca boniza ion; a he , i
is cen al o he p oblem.
3.2 Unce ain y Themes in Li e a u e
To be e unde s and how he s udies ha e cha ac e ized and add essed unce ain y, his sec ion explo es he
speci ic ypes o unce ain y o in e es wi hin he e iewed li e a u e. Unce ain y in ma i ime deca boniza ion
was mos equen ly amed a ound h ee dominan ca ego ies: comme cial, egula o y, and echnological. One o
he b oades ca ego ies, comme cial, includes ac o s such as uel p ice, li e cycle cos , ca bon ax, ma ke
conside a ions, and in es men iming. Regula o y unce ain y s ems om he ambigui y o he scope, s eng h,
and iming o local and global policy. Technological unce ain y e e s o he s ill-de eloping ma u i y, a ailabili y,
and long- e m iabili y o eme ging uel and p opulsion op ions. P io i y unce ain y cap u es he ambigui y o
s akeholde alues and objec i es. Unlike comme cial o echnological unce ain ies, which can mo e easily be
quan i ied, p io i y unce ain y e lec s shi ing and con es ed iews on which ou comes—cos , compliance,
pe o mance, sa e y— should d i e decisions. In ma i ime deca boniza ion, hese p io i ies di e ac oss ac o s and
e ol e as ma ke s and egula ions change.
No. Publica ions by Yea
2000 2005 2010 2015 2020 2025
Yea
0
1
2
3
4
5
6
7
No. o Publica ions
Ma i ime
Non-Ma i ime
5
Figu e 3. Pe cen acknowledgemen o unce ain y ype seen in he li e a u e.
Comme cial unce ain y was he mos p e alen ca ego y, accoun ing o nea ly 40% o men ions ac oss he
e iewed s udies. This e lec s he capi al-in ensi e, asse -hea y na u e o he ma i ime indus y, whe e long
in es men cycles can magni y exposu e and isk om inancial a iables such as uel p ice ola ili y, ca bon
p icing, and ma ke demand shi s. S udies like Ve ga a Pa edes e al. and Aspen e al. emphasize uel p ice and
li ecycle cos p ojec ions as cen al o de e mining he iabili y o al e na i e uel pa hways o ma i ime
deca boniza ion [11], [12]. In es men iming unce ain y also appea ed equen ly, especially in analyses
weighing e o i op ions e sus newbuilds in an icipa ion o u u e egula o y schemes and consequen ca bon
cos s [11].
The second mos dominan ca ego y was echnological unce ain y, encompassing ques ions o echnology
ma u i y, a ailabili y, and o e all iabili y in he ope a ional li espan o ma ine asse s. This high le el o
acknowledgemen no only e lec s he now apidly e ol ing echnology landscape, bu he app ehension in he
indus y o adop hem, due o unce ain y in uel a ailabili y and compa ibili y locking owne s in o subop imal
pa hways. I is impo an o no e how he e iewed s udies no only ocused on he unce ain y o he pe o mance
o new echnologies, as seen in Chae e al, bu also he lexibili y in e o i ing and he obus ness o hese
echnologies as seen in Fo e ich e al. [13], [14].
Regula o y unce ain y, while showing in 20% o cases, is almos ce ainly unde ep esen ed ela i e o i s
impo ance in indus y discou se. Howe e , many s udies ame egula o y changes indi ec ly h ough hei
economic impac s, i.e., ca bon axing, emissions ading, o compliance cos ing, as opposed o s andalone policy
unce ain y. Wi h he sc eening c i e ia used in he li e a u e e iew, hose ins ances we e classi ied unde
comme cial as opposed o egula o y, shi ing he o e all weigh ing o hese ca ego ies. Secondly, egula o y
unce ain y is o en b oadly men ioned bu di ec ly implemen ed as ixed scena io inpu s a he han as unce ain
a iables o be explo ed hemsel es.
Unce ain y Acknowledged (%)
Comme cial
P io i y
Regula o y
Technological
Unce ain y Type
0
5
10
15
20
25
30
35
Acknowledged (%)
6
3.3 Me hods Employed o Handle Unce ain y
The me hods used ac oss he e iewed pape s o acknowledge unce ain y we e ca ego ized in o six
o e a ching g oups, based on hei unde lying app oach o unce ain y modeling. The g oupings used o he
li e a u e e iew a e as ollows:
• Scena io-Based Me hods
o Explo a o y app oaches o add essing deep unce ain y h ough al e na i e u u es and
pa hways.
• Mul i-C i e ia Decision Analysis (MCDA)
o Tools o s uc u ing ade-o s among al e na i es, o en inco po a ing s akeholde p e e ences
o weigh ed c i e ia.
• Pe o mance-Based Tools
o S uc u ed assessmen amewo ks (e.g., Li e Cycle Assessmen s (LCA), Techno-economic
Analysis (TEA) ha embed unce ain y in pe o mance e alua ion.
• Simula ion-Based Me hods
o P obabilis ic modeling o sys em dynamics, cap u ing a iabili y h ough s ochas ic o agen -
based simula ions.
• Op imiza ion-Based Me hods
o Quan i a i e app oaches o inding op imal solu ions unde cons ain s, including obus and
s ochas ic op imiza ion.
Figu e 4. Pe cen me hod usage by ype in he li e a u e.
Figu e 4 displays he pe cen con ibu ion o each me hod g oup ac oss he e iewed pape s. The dis ibu ion
highligh s mul iple in e es ing pa e ns in how he ield has chosen o been cons ained o handle unce ain y in
ma i ime deca boniza ion decision-making. The i s o which is he domina ion o Scena io-Based me hods,
e lec ing he ield’s ecogni ion ha he deep unce ain ies a ound uel a ailabili y, echnology ma u a ion, and
egula o y e alua ion o en demand explo a o y a he han p edic i e app oaches. The lack o eliable o ecas s
and p ojec ions and he long- e m planning ho izons inhe en o he ma i ime indus y again s ongly suppo he
p e alence o scena io hinking seen wi hin he li e a u e. Complemen a y o his, Op imiza ion-Based me hods
ep esen ed unde 4% o con ibu ions, sugges ing ha adi ional op imiza ion app oaches a e less sui able o he
ague and uns uc u ed challenges seen in ma i ime deca boniza ion.
Me hod Con ibu ions (%)
MCDA
Op imiza ion-Based
Pe o mance-Based
Scena io-Based
Simula ion-Based
Me hod G oup
0
5
10
15
20
25
30
35
Con ibu ions (%)
7
The second mos employed me hod was MCDA, p esen in jus unde a qua e o all he publica ions. The
equency o MCDA speaks o he complex s akeholde en i onmen and he unclea and o en compe ing
objec i es seen in ma i ime deca boniza ion decisions, such as cos , emissions, sa e y, and eliabili y. The s ong
p esence o MCDA sugges s ha esea che s and p ac i ione s in his a ea a e ac i ely g appling wi h how o
balance he echnical e alua ions wi h quali a i e s akeholde inpu and p e e ence-d i en ade-o s. Scena io-
based and MCDA me hods a e also mo e in ui i e and can be explained easily wi hou de ailed knowledge o he
unde lying me hodology, unlike Op imiza ion- and Simula ion-based me hods, o example.
3.4 Mapping Me hod Types o Unce ain y Types
Figu e 5 plo s he numbe o imes me hod ypes ha e been used o add ess he unce ain y ca ego ies iden i ied
in he e iew. The pu pose o his mapping is o e eal which combina ions o unce ain y ype and decision
suppo , o modeling me hod, ha e been s udied and which ha e ecei ed li le o no a en ion.
Figu e 5. Me hod employed o handle ype o unce ain y in li e a u e hea map.
Consis en wi h he dis ibu ion seen in Figu e 4 (me hod con ibu ion cha ), comme cial and echnological
unce ain ies we e he mos widely assessed o e all, wi h esea ch employing me hods ac oss all ca ego ies. The
s onges pai ing in he da ase occu ed be ween comme cial unce ain ies and scena io-based me hods, ollowed
by op imiza ion-based and pe o mance-based ools. Technological unce ain ies we e he second mos common
ca ego y, appea ing 21 imes, mos equen ly app oached wi h MCDA and scena io-based me hods. P io i y
unce ain ies appea ed less o en bu we e domina ed by MCDA, wi h all o he me hods showing minimal
connec ions. MCDA was second only o scena io-based me hods o e all in e ms o equency o use.
Clus e s o low ac i i y we e concen a ed in he o en less di ec ly quan i iable unce ain y ypes o p io i y
and egula o y. Regula o y unce ain ies showed no eco ded pai ings wi h MCDA o pe o mance-based ools,
and only small coun s wi h op imiza ion-based, simula ion-based, and scena io-based me hods. Pe o mance-
based, simula ion-based, and op imiza ion-based me hods also had minimal applica ion o p io i y unce ain ies,
which we e o e whelmingly handled h ough MCDA.
MCDA
Op imiza ion-based
Pe o mance-based
Scena io-based
Simula ion-based
Me hod Type
Comme cial
P io i y
Regula o y
Technological
Unce ain y Type
Unce ain y Type s Me hod Type Co-occu ence
0
5
10
15
8
These pa e ns, pa icula ly he concen a ion o scena io wo k in comme cial con ex s and he sp ead o
echnological unce ain ies ac oss mul iple me hods, a e examined u he in he discussion o unde s and hei
d i e s and implica ions o decision-making unde unce ain y.
4 Discussion
4.1 Inc easing Engagemen : To Wha & Why?
The d ama ic inc ease in publica ions add essing unce ain y in ma i ime deca boniza ion e lec s a mo e han
jus academic momen um. Comme cial in e es de i ed om ising egula o y p essu e has o ced he ealiza ion
o deca boniza ion no longe being a dis an o op ional objec i e. Ea ly li e a u e in his space p ima ily e ol ed
a ound iden i ying wha one should ca e abou and wha migh be possible. In ecen yea s, ha aming has
ans o med o a mo e ope a ionally ocused ques ion o : gi en he unce ain ies ha de ine upcoming decisions,
how can one make sound choices?
This ansi ion shows bo h p essu e wi hin he indus y and he adop ion o app oaches om o he ields ha
ha e long aced planning unde unce ain y. The esul ing decision-making landscape has ma u ed g ea ly. Wha
once e ol ed a ound gene alized scena io mapping o MCDA ankings has e ol ed in o s uc u ed decision
wo k lows, combining scena io de elopmen , pe o mance modeling, simula ions o dynamic beha io s, and
op imiza ion o sizing o con igu a ion. This “chaining” o me hods p o ides clea e links be ween de ined
unce ain ies and esul ing decisions, a ea u e less equen ly seen in ea lie wo k.
Wi hin hese s uc u ed wo k lows, comme cial and echnological unce ain ies con inue o domina e. This is
unsu p ising: ope a ional expenses a e a cen al d i e o shipping in es men , and bo h comme cial a iables ( uel
p ice ajec o ies, ca bon cos s, u iliza ion a es) and echnical ac o s ( echnology ma u i y, in eg a ion isks,
eliabili y) can swing o al cos s a mo e han ma ginal changes in e iciency. These echnological unce ain ies
can also compound po en ial exposu e by delaying adop ion o o cing expensi e edundancies as echnology
de elops.
In his en i onmen , decision-making becomes as much abou managing u u e exposu e as i is abou hi ing
absolu e pe o mance a ge s. This is exac ly whe e scena io-based me hods and mo e obus ness-o ien ed
amewo ks like obus decision making become essen ial, whe e ensu ing op ions emain de ensible ac oss a
spec um o unp edic able a iabili y is pa amoun . Scena io me hods hold he planning s age, c ea ing a space o
plausible u u es whe e di ec ly applied ools such as op imiza ion, simula ion, and pe o mance-based ools can
wo k wi hin o con igu e sys ems, es sensi i i ies, and e en u he s uc u e adeo s h ough MCDA.
4.2 Me hod Unce ain y Clus e ing: Insigh s & Implica ions
The co-occu ence mapping shows a ew clea clus e s, again wi h he s onges concen a ion be ween
comme cial unce ain ies and scena io-based me hods. This pai ing e lec s he complex eali y o he cu en and
u u e ma ke in li e a u e. As Moshiul e al (2023) no e, “Fuel p ice ola ili y and ca bon cos exposu e emain
he mos in luen ial and leas con ollable d i e s in pa hway selec ion”, making explo a o y me hods a mo e
de ensible s a ing poin han de e minis ic and single- alue o ecas s [15]. This endency pa allels p ac ices in
inance, whe e scena io-based app oaches like eal op ions analysis ha e eme ged p ecisely because de e minis ic
o ecas ing me hods ha e consis en ly ailed o cap u e ma ke ola ili y and s uc u al shi s [16]. Real op ions
analysis i sel ep esen s a o m o scena io hinking, whe e adap i e capaci y is alued o e op imiza ion o
singula u u es. This app oach has p o en a mo e obus han adi ional ne p esen alue calcula ion in ola ile
ma ke s [17]. This o e all pai ing is e ec i e because comme cial unce ain ies o en exhibi deep and s uc u al
unce ain y whe e his o ical pa e ns a e no p o en o p edic u u e condi ions.
Howe e , he heigh ened ocus on comme cial unce ain ies has p omp ed a wide ange o me hodological
ea men s beyond scena io hinking. Op imiza ion and simula ion app oaches ha e also been employed
equen ly, bu mos o en wi h scena io me hods and MCDA being used a ea ly s ages o ame u he analysis.
As seen wi h Wang & Teo and Zwaginga & P uyn, lee -le el cos planning equen ly begins wi h s ochas ic uel
p ice dis ibu ions ha a e subsequen ly es ed in op imiza ion amewo ks o lee enewal and deploymen [18],
[19]. In addi ion, he li e a u e has iden i ied hese dis ibu ions as “o en subjec i e cons uc s and a e ounded on
9
explo a o y and scena io hinking [20]. This hyb idiza ion is no an excep ion bu has inc easingly become he
no m.
4.2.1 Hyb idiza ion: F om Implici o Explici
The clus e ing pa e ns p esen ed e eal se e al consis en ends ha poin owa d bo h he p e alence and
limi a ions o cu en me hodological app oaches in ma i ime deca boniza ion. The pa e n o scena io me hods
shaping ini ial explo a o y s ages while o he me hods d i e applied analysis sugges s a b oade eali y no ully
cap u ed in he li e a u e e iew. Mos me hods and models a e, o some ex en , implici ly hyb id. In his con ex ,
hyb id e e s o combining mul iple me hod ypes in he pu sui o educing and es ing unce ain y pa ame e s.
While a mo e limi ed subse o publica ions explici ly desc ibes hei app oaches as hyb id (e.g., combining
scena io explo a ion wi h simula ion), in p ac ice, mos e o s ely on ups eam assump ions and inpu s d awn
om explo a o y wo k. As Kwakkel & Haasnoo (2019) poin ou , “simula ion models o ope a ional eliabili y
equi e demand, uel p ice, and echnology pe o mance dis ibu ions – all o which a e scena io a i ac s in hei
own igh [21].” The dis inc ion be ween pu e and hyb id me hods is o en mo e seman ic han p ac ical and is a
mo e p e alen han clus e ing analysis alone migh sugges .
4.2.2 P io i y Unce ain y and MCDA
The p io i y unce ain y clus e shows he mos in ui i e me hod usage, wi h MCDA app oaches ha ing a hea y
concen a ion. P io i y unce ain y s ems om con lic ing s akeholde alua ions, whe e a ying s akeholde
p io i ies and isk ole ances can clash. This dilemma equi es s uc u ed and ai me hods o p e e ence elici a ion
and accoun ing, which is he p ima y s eng h o MCDA me hods such as Analy ical Hie a chy P ocess (AHP),
Technique o O de P e e ence by Simila i y (TOPSIS), Mul i-c i e ia Op imiza ion and Comp omise Solu ion
(VIKOR), and P e e ence Ranking O ganiza ion Me hod o En ichmen E alua ions (PROMETHEE). Howe e ,
he isola ion o p io i y unce ain y om o he s udied me hod ypes e eals a po en ial blind spo : s akeholde
p e e ences a e a ely independen o de eloping echnical and comme cial eali ies, ye ew s udies combine
p io i y elici a ion wi h simula ion o scena io-based explo a ion.
One s udy ha does is Baud y e al.’s (2018) ange-based Mul i-Ac o Mul i-C i e ia Analysis (MAMCA)
amewo k, whe e s akeholde p e e ence analysis is combined wi h Mon e Ca lo simula ions o explo e how
s akeholde p e e ence p opaga es ac oss scena ios [22]. Howe e , hese in eg a ed app oaches emain
unde ep esen ed in ma i ime deca boniza ion li e a u e o e all. The seemingly s a ic ea men o s akeholde
p io i ies misses he dynamic eali y whe e s akeholde a i udes can shi d as ically as ma ke and egula o y
condi ions e ol e. Fo example, a po au ho i y’s weighing o cos e iciency and en i onmen al pe o mance
migh change en i ely i ca bon p icing eaches $200/ onne e sus $50/ onne, o i low- o ze o-emission uel
a ailabili y is abundan o sca ce. Regula o y o comme cial shi s like hese can undamen ally al e s akeholde
p io i ies and downs eam decisions. Ad ancing me hods o cap u e hese dynamics is a c i ical a ea o u he
esea ch in ma i ime deca boniza ion decision-making.
4.2.3 Regula o y Unce ain ies’ S a ic Acknowledgmen
A mo e ja ing pa e n in he clus e ing is he o e all weak engagemen wi h egula o y unce ain y. This is
coun e in ui i e gi en he undamen al impo ance o egula ion as a cen al d i e o deca boniza ion o e all.
Howe e , as p e iously men ioned, egula o y unce ain y is o en oiced h ough comme cial measu es, whe e
policy ou comes a e ea ed as s a ic condi ions, i.e., ca bon p icing o X amoun in yea Y. Doing so can acili a e
decision making whe e wicked and in ac able unce ain ies can be ea ed in much mo e analy ically manageable
o ms.
Howe e , his end denies he dynamic and poli ical p ocesses in which hese egula ions ake place; hey
shi based on indus y beha io and emission ou comes. Policy can adap o he e y beha io s being modeled,
and he eedback loops be ween indus y decisions and egula o y esponse can be easily igno ed. Some s udies
call ou his limi a ion, as in Psa a is e al. (2018), no ing ha when assessing egula o y unce ain y, many
app oaches “do no include he dynamic eedback ha migh exis be ween measu es due o hei in e ac ions [23]”
. Dynamic policy modeling app oaches like agen -based adap a ion o in eg a ed assessmen models om clima e,
economy, and ene gy ma ke s p esen s ong a eas o possible me hod ans e o ma i ime deca boniza ion
con ex s.
While hese clus e s illus a e how di e en me hods align wi h speci ic unce ain ies, hey also highligh he
limi a ions o cu en academic wo k. Hyb idiza ion is inc easingly p esen , bu o en amed implici ly, lea ing
16
[37] K. Ande sson, S. B ynol , J. Hansson, and M. G ahn, “C i e ia and Decision Suppo
o A Sus ainable Choice o Al e na i e Ma ine Fuels,” Sus ainabili y, ol. 12, no. 9, p.
3623, Ap . 2020, doi: 10.3390/su12093623.
[38] L. Be ge and M. Ma inacci, “Model Unce ain y in Clima e Change Economics: A
Re iew and P oposed F amewo k o Fu u e Resea ch,” En i on. Resou . Econ., ol.
77, no. 3, pp. 475–501, No . 2020, doi: 10.1007/s10640-020-00503-3.
[39] A. Shah, S. Hallega e, R. Lempe , C. B own, and S. Gill, In es men Decision Making
unde Deep Unce ain y - Applica ion o Clima e Change. Wo ld Bank, Washing on,
DC, 2012. doi: 10.1596/1813-9450-6193.
[40] Q. Sun, L. Chen, M. C. Chou, and Q. Meng, “Mi iga ing he inancial isk behind
emission cap compliance: A case in ma i ime anspo a ion,” P od. Ope . Manag., ol.
32, no. 1, pp. 283–300, Jan. 2023, doi: 10.1111/poms.13837.
[41] O. Cabezas-Basu ko, E. Mesbahi, and S. R. Moloney, “Me hodology o sus ainabili y
analysis o ships,” Ships O sho e S uc ., ol. 3, no. 1, pp. 1–11, Feb. 2008, doi:
10.1080/17445300701673841.
[42] S. O. E iks ad and S. Ehle s, “Decision Suppo F amewo k o Exploi ing No he n Sea
Rou e T anspo Oppo uni ies,” Ship Technol. Res., ol. 59, no. 2, pp. 34–42, Ap .
2012, doi: 10.1179/s .2012.59.2.003.
[43] R. Heijungs and M. A. J. Huijb eg s, “A Re iew o App oaches o T ea Unce ain y in
LCA,” July 2004.
[44] L. C e , M. Baud y, and F. Lan z, “How o implemen he 2023 IMO GHG s a egy?
Insigh s on he impo ance o combining policy ins umen s and on he ole o
unce ain y,” Ma . Policy, ol. 169, p. 106332, No . 2024, doi:
10.1016/j.ma pol.2024.106332.
[45] P. Ko kmaz, D. Schmid, and U. Fahl, “Inco po a ing unce ain ies owa ds a sus ainable
Eu opean ene gy sys em: A s ochas ic app oach o deca boniza ion pa hs ocusing on
he anspo sec o ,” Ene gy S a egy Re ., ol. 38, p. 100707, No . 2021, doi:
10.1016/j.es .2021.100707.
[46] W. Sch eude , J. C. Sloo weg, and B. Van De Zwaan, “Techno-economic assessmen
o low-ca bon ammonia as uel o he ma i ime sec o ,” Appl. Ene gy Combus . Sci.,
ol. 22, p. 100330, June 2025, doi: 10.1016/j.jaecs.2025.100330.
[47] F. Haag, P. Reiche , M. Mau e , and J. Liene , “In eg a ing unce ain y o p e e ences
and p edic ions in decision models: An applica ion o egional was ewa e planning,” J.
En i on. Manage., ol. 252, p. 109652, Dec. 2019, doi:
10.1016/j.jen man.2019.109652.
[48] I. N. Du bach and T. J. S ewa , “Modeling unce ain y in mul i-c i e ia decision
analysis,” Eu . J. Ope . Res., ol. 223, no. 1, pp. 1–14, No . 2012, doi:
10.1016/j.ejo .2012.04.038.
[49] K. Q. Bui, L. P. Pe e a, and J. Emblems åg, “Li e-cycle cos analysis o an inno a i e
ma ine dual- uel engine unde unce ain ies,” J. Clean. P od., ol. 380, p. 134847, Dec.
2022, doi: 10.1016/j.jclep o.2022.134847.
[50] J. Ren and M. Lü zen, “Fuzzy mul i-c i e ia decision-making me hod o echnology
selec ion o emissions educ ion om shipping unde unce ain ies,” T ansp. Res. Pa
T ansp. En i on., ol. 40, pp. 43–60, Oc . 2015, doi: 10.1016/j. d.2015.07.012.
[51] M. Zhu, K. F. Yuen, J. W. Ge, and K. X. Li, “Impac o ma i ime emissions ading
sys em on lee deploymen and mi iga ion o CO2 emission,” T ansp. Res. Pa T ansp.
En i on., ol. 62, pp. 474–488, July 2018, doi: 10.1016/j. d.2018.03.016.
17
[52] J. Zou and B. Yang, “E alua ion o al e na i e ma ine uels om dual pe spec i es
conside ing mul iple essel sizes,” T ansp. Res. Pa T ansp. En i on., ol. 115, p.
103583, Feb. 2023, doi: 10.1016/j. d.2022.103583.
[53] T.-Y. Chou and G.-S. Liang, “Applica ion o a uzzy mul i-c i e ia decision-making
model o shipping company pe o mance e alua ion,” Ma i . Policy Manag., ol. 28,
no. 4, pp. 375–392, Oc . 2001, doi: 10.1080/03088830110049951.
[54] D. McCollum, G. Gould, and D. G eene, “G eenhouse Gas emissions om a ia ion and
ma ine anspo a ion: mi iGa ion po en ial and policies,” 2009.
[55] J. Vaca-Cab e o, N. González-Cancelas, A. Cama e o-O i e, and J. Quijada-Ala cón,
“Bayesian Ne wo ks Applied o he Ma i ime Emissions T ading Sys em: A Tool o
Decision-Making in Eu opean Po s,” In en ions, ol. 10, no. 2, p. 28, Ma . 2025, doi:
10.3390/in en ions10020028.
[56] S. Chen, S. Zheng, and Q. Zhang, “In es men decisions unde unce ain y on LNG-
powe ed essels o en i onmen al compliance,” J. Shipp. T ade, ol. 3, no. 1, p. 5, Dec.
2018, doi: 10.1186/s41072-018-0031-4.
[57] S. Baş uğ, E. F. Akgül, H. Ha alambides, and T. No eboom, “A decision-making
amewo k o he unding o shipping deca boniza ion ini ia i es in non-EU coun ies:
insigh s om Tü kiye,” J. Shipp. T ade, ol. 9, no. 1, p. 12, Ap . 2024, doi:
10.1186/s41072-024-00172-1.
[58] S. Badakhshan, H. D. Kaushik, and J. Zhang, “S ochas ic Op imiza ion o Small
Modula Reac o and Ba e y Sizing o Ma i ime Deca boniza ion Unde Voyage
Unce ain ies,” IEEE T ans. T ansp. Elec i ica ion, pp. 1–1, 2025, doi:
10.1109/TTE.2025.3577938.
[59] R. Halim, L. Ki s ein, O. Me k, and L. Ma inez, “Deca boniza ion Pa hways o
In e na ional Ma i ime T anspo : A Model-Based Policy Impac Assessmen ,”
Sus ainabili y, ol. 10, no. 7, p. 2243, June 2018, doi: 10.3390/su10072243.
[60] J.-D. Cap ace, C. H. Ma ques, L. F. Assis, A. Lucchesi, and P. C. Pe eda, “Sus ainable
Shipping: Modeling Technological Pa hways Towa d Ne -Ze o Emissions in Ma i ime
T anspo (Pa I),” Sus ainabili y, ol. 17, no. 8, p. 3733, Ap . 2025, doi:
10.3390/su17083733.
[61] F. Zanobe i, G. Pio, S. Ja a zadeh, M. Muñoz O iz, and V. Cozzani, “Deca boniza ion
o ma i ime anspo : Sus ainabili y assessmen o al e na i e powe sys ems,” J.
Clean. P od., ol. 417, p. 137989, Sep . 2023, doi: 10.1016/j.jclep o.2023.137989.
