281
1 INTRODUCTION
Historyofstabilitycriteriaisalongone.Itisassumed
thatfirststabilitycriteriawereproposedbyDennyin
1884 and adopted by British Admiralty after the
CAPTAIN disaster. Few years laterGerman
professional Mariners Association recommended
minimumrightingarmscurveafterfounderingof15
fishingvessels during the stormy night 22 to 23
December 1894in North Sea. Benjamin in 1913
proposedminimumdynamicst
abilityarmscurvethat
was used by British Admiralty. During the period
beforetheFirstWorldWar severalotherproposals
were advanced, e.g. by Anderson, Holt, Bruhn,
NiedermairandPierrotet.[1].
The most important proposal was however
adv
ancedin1939byRahola,[2]whoonthebasisof
comparisonofstabilitycharacteristicsofshipswhich
capsized with those which were operated safely
proposedsetofcriteriathatwereunofficiallyinusein
several countries until international stability criteria
developed by the Intergovernmental Maritime
Consultative Organisation (IMCO)were adop
tedin
1968.
Development of internationally adopted stability
criteriawasoneofthebasictasksofIMCOsincethe
establishment of this organization. At the
InternationalConferenceontheSafety ofLife atSea
(SOLAS)in1960thistaskwasrecommendedforthe
Organization. The work on developmentof stability
crit
eria started at the first meeting of the Sub‐
committee on Stability and Subdivision (STAB) of
IMCOin1962.
2 DEVELOPMENTOFTHEINTACTSTABILITY
CODE
The current stability criteria are included in the
InternationalCodeonIntactStability2008adoptedby
the expanded Marine Safety Committee of the
International Maritime Organisation (IMO) on 4
December2008[3](IMCOchangedit
snametoIMOin
1982).
The short history of the Code is as follows. The
first internationally accepted stability criteria were
statistical criteria developed by the STAB Sub‐
committeeofIMCOintheyears1962‐67andadopted
Stability Criteria - Present Status and Perspectives of
Improvement
L.Kobyliński
FoundationforSafetyofNavigationandEnvironmentProtection,Iława,Poland
ABSTRACT:Shorthistoricaldevelopmentandpresentstatusofstabilitycriteriaispresented.Currentworkof
theInternationalMaritimeOrganisationonsecondgenerationstabilitycriteriaiscriticallyassessedshowingthe
advantagesandweekpointsoftheproject.Perspectivesforimprovementofthesafetyagainstst
abilityfailure
arediscussed,includingriskassessmentmethodsandgoalorientedapproachtostabilityproblems.
http://www.transnav.eu
the International Journal
on Marine Navigation
and Safety of Sea Transportation
Volume 8
Number 2
June 2014
DOI:10.12716/1001.08.02.14
282
in 1968 by IMCO Assembly by resolutions
A.167(AS.IV) and A.168(ES.IV). Then, during the
period 1978‐82STAB Sub‐committee developed so
called “Weather Criterion” for passenger and cargo
vesselsthatwas adoptedultimatelyin1985 by IMO
Assembly by ResolutionA.562(14) and later on, for
fishing vessels in 1991 byResolution
A.685(17).There were also adop
ted in the meantime
severalotherresolutionsrelatedtovariousaspectsof
stability,amongstthemresolutionsrelatedtostability
of vessels of different types. All criteria included in
these resolutions were recommended only, none of
themwasconsideredcompulsory.
The criteria developed were criticised from the
very beginning after they were adop
ted. During the
discussionsatSTABSub‐Committeeandalsoinother
places it was stressed that in development of those
criteriaseveralassumptionsweremademakingthose
criteria non‐rigorous. According to many opinions
thosecriteriahardlycouldbe assessedas “rational”,
because they were not based on probabilist
ic
approach.Inlateeightieswith norealistic proposals
ofhowtobasethecriteriaonprobabilityofcapsizing,
andinordertomakeatleastsomestepforward,IMO
decided to make one comprehensive document that
would include all resolutions and criteria already
developed and split between several different
documents. The idea of development of Intact
St
ability Code was advanced and finally, after
discussion, the Code was developed. It was also
agreed that this code should be based on system
approach[4].
FinallyIntactStabilityCodewas adoptedin 1993
byIMOAssemblybyresolutionA.749(18).TheCode
was subsequently amended in 2002. Since then,
however,discussionstartedagainonthepossibilit
ies
to improve level of safety against capsizing and to
revisethecriteria.ItwasagreedbytheIMOthatthe
most importantmotion would be to make stability
criteria compulsory. This was achieved by drafting
neweditionoftheCode,tha
twasadoptedultimately
in 2008.Both editions of the Code included stability
criteriavirtuallyunchangedfromtheoriginalcriteria
recommendedbytheabovementionedresolutions.
3 PRESENTSTATUSOFSTABILITYCRITERIA
Presentstatusofstabilityrequirementsisrepresented
bythe2008editionoftheInternationalIntactStability
Code [3].This edit
ionof the Code consists of three
parts:PartA,PartBandExplanatoryNotes.
Part A of the Code was made compulsory by
proper reference in the SOLAS Convention. It
includesbasiccriteria:statisticalcriteriaandweather
criterion for both passenger and cargo ships and
fishing vessels. Only minor improvements from the
previousedit
ionswereincluded.
PartBoftheCodeincludesprovisionsforspecific
types of ships and other provisions that are
recommended only. That makes possible to amend
thispartoftheCodemoreoftenasdeemednecessary
andtoincludecriteriaandprovisionsthatmaybenot
entirelysufficient
ly validatedfor a trial period. This
partoftheCodeincludesprovisionsfor:
Fishingvessels
Pontoons
Containershipsgreaterthan100m
Offshoresupplyships
Specialpurposeships
Mobileoffshoredrillingunits(MODUs).
PartBofthecodeincludesalsochapterson:
Guidanceinpreparingst
abilityinformation
Stability calculations performed by stability
instruments
Operationalprovisionsagainstcapsizing
Icingconsiderations
Considerations for watertight and weathertight
integrity
Determinationoflightshipparameters.
Two Annexes include detailed guidance for the
conductofanincliningtestandrecommendationfor
skippers of fishing vessels on ensuring a vessel’s
enduranceincondit
ionsoficeformation.Finallythe
Code includes third part comprising explanatory
notestothestabilitycriteria.
4 IMOCURRENTWORKONIMPROVED
STABILITYCRITERIA
Having prepared the 2008 edition of the Intact
StabilityCode,theSLFSub‐Committeewas,however,
not satisfied with the st
ability criteria in force. The
point was raised by some delegations, that several
situations dangerous from the point of view of
stabilityarenotcoveredbythecriteria.Accordingto
some delegationsthe following situations or
stabilityfailuremodesshouldbeconsidered:
Parametricresonanceinfollowingandheadseas
Lossofst
abilityinthewavecrest
Broachingtoandsurfing
Deadshipcondition,and
Excessiveaccelerationswhenrolling
Actually the proposals were not new. Those
situations were considered by the Sub‐committee
during late seventies and early eighties of the last
century. At that time Polish delegation to IMO
proposedtoconsiderthosesituations[5],ap
artfrom
the last one, but after brief discussion the
Subcommitteedecidedthatitwasunabletoworkout
usable recommendations in that respect. Problem of
excessive accelerations was included recently
followingtheproposaloftheGermandelegation.The
Sub‐committee agreed to consider those situations
under the agenda it
em “Second generation stability
criteria”.
Table1.Threelevelsvulnerabilityapproachwithinsecondgenerationstabilitycriteria
__________________________________________________________________________________________________
Level1Level2Directstabilityassessment operationalguidance
__________________________________________________________________________________________________
Stability Simpleandconservative Lessconservativecriteria, Numericalsimulation Basedonexperience
failure criteriabasedongeometrybasedonsimplifiedphysics ofphysicalphenomena
modeofhullandspeedandinvolvingsimplified basedoncomputercodes
developed
__________________________________________________________________________________________________
283
Work on second generation stability criteria
startedin2008[6].Afterratherlengthydiscussionof
thematterduringseveralsessionstheSubcommittee
agreedthat with regard of those situations safety
assessment based on the three‐levels vulnerability
checkhavetobeapplied.Theideaofthisapproachis
that vulnerability
to those stability failures at three
levelshavetobechecked,asshowninthetable1.[7].
Fig.1 shows schematically process of checking
vulnerabilitytoallfourstabilityfailuremodes.
Figure1.Schematicpresentationoftheprocessofchecking
vulnerabilitytofourstabilityfailuremodes(from[8])
The SLF Subcommitteeassumed that three to
four years maybe sufficient to complete the criteria,
however when the work started the Subcommittee
realizedthatthisisnotaneasytask.Aftersixyearsof
discussionandextensiveworkeventhemostsimple
criteria of the level 1 have not been finalized
and
although preliminary draft of those criteria was
prepared, they are still under discussion. It is
assumednowthatatleastthreetofiveyearsmoreare
needed to complete the second generation stability
criteria. It appears that this approach is charged by
some disadvantages and even some countries
expressed
now the view that the criteria proposed
until now are too complicated and in particular at
level three they require application of the rather
sophisticated computer programmes that are not
commercially available. Because of that in their
opinionthewholetaskhastobeabandoned.
5 CRITICALOBSERVATIONSREGARDING
SECONDGENERATION
STABILITYCRITERIA
Attempttoimprovesafetyagainststabilityfailure
asdeemedtobeaccomplishedbydevelopmentofthe
second generation stability criteria has some
advantages but also some weak points. The second
generation stability criteria as they are considered
now are basically design oriented. Their concept is
basedonthe
assumptionthatimprovementofship
constructionandlayoutwillincreaselevelofsafety
against stability failure.That is so, but only to the
certainextent.
Analysis of causes of casualties reveals without
doubtthatinalmostallcasesthecasualtyscenariois
verycomplexandseveralfactorscontributetothe
end
result. Casualties where one single cause may be
identified are extremely rare. Usually, apart from
design faults, also operational factors, including
human factor, play important part. Therefore
improvementofthedesigncharacteristicsoftheship
or even eliminating all possible causes where faulty
designisthemaincauseof
casualtymayaffectonly
small percentage of casualties. Concentrating main
effort on design characteristics of ships is therefore
notthemostimportanttask.
IMO current work on second generation stability
criteria consisting of taking into account some
stabilityfailurescenariosidentifiedasimportantwill
certainlyincreasethelevelofsafetyfrom
thestability
point of view. However, analysis of stability
casualties revealsthatchosen stability failure modes
are not the most frequent ones. Although current
approach to second generation stability criteria
includes consideration of some important hazards it
doesnottakeintoaccounthazardsprobability.
One of the stability failure modes
chosen is
parametric resonance. Extremely large effort to
investigateparametricresonancewasdevotedduring
recent years and a great number of important
papers were published on this subject (e.g. [7, 9]).
Alsoalargenumberofdocumentscontainingstudies
oftheeffectofparametricresonanceonstabilitywere
submitted to IMO.
This resultedin preparation of
preliminary draft of few options ofvulnerability
criteriatoparametricresonanceofthelevel1and2.
Actually, the main reason of taking parametric
resonance into consideration was one casualty of
large container ship [10] where due to parametric
resonancein head waves serious damage
to the
containerstaplesleadingtolossofseveralcontainers
andsomedamagetotheshiphullhappened.Careful
analysis of a number of casualty reports reveals
however, that cases where the casualty may be
attributed solely to parametric resonance are
extremely rare and usually associated with other
hazards.
If we
consider, for example, proposed first level
criteria on avoiding parametric resonance we may
discover that they are most conservative.However
ifthedesignedshipismeetingthosecriteriaitdoes
not mean that occurring parametric resonance is
impossible. Under certain combination of wave
characteristics, speed and heading parametric
resonance may
occur and with no reaction from the
crewitmaybedangerouscausingexcessiverolling.It
is true, that parametric resonance is rare event, and
thereare extremelyraresituationswhereparametric
resonanceactuallyledtostabilityfailure.
Within second generation criteria hazards were
identified, however probability of hazard is not
considered.Itseems,thatiftheprobabilityofhazard
islowerthancertainassumedvalue,suchhazardmay
not be considered further. How to assess hazard
probability is another matter. Obviously probability
ofhazarddependtosomeextentonshipdesign, but
moreonshipoperationandhumanfactor.Parametric
resonancemaybeeasilyavoidedbyfollowingbythe
master operational guidance. Within second
generationstabilitycriteriaoperationalguidancewas
includedinthescheme,thereforeitwas deemedthat
284
this may be necessary, but at present there was no
attempt made to discuss this problem. Operational
guidelines are proposed however as alternative to
directstabilityassessmentasshowninthetable1.
Another hazard considered, loss of stability in
wavecresthappensmoreoften,howeverbotheffects,
parametric resonance and loss off st
ability in wave
crest may be easily avoided with proper handlingof
the ship and not putting the ship in situation when
those phenomena may occur (e.g respecting
recommendation MSC.1/Circ.1228). The same
reasoningisapplicabletosurfridingandbroaching,
hazardsinfactdangerousforratherfastsmallships.
Differentsituationisregardingthehazarddefined
asdeadshipcondit
ion.Thepresentweathercriterion
asitappearsintheISCodeinfactcoversthesituation
of the dead ship condition. In general opinion
weathercriterionisworkingwellfor themajority of
ships,Howeversomeunconventionalships,e.g.ships
with large windage area and large B/L ratio, as for
example large cruise vessels and similar, often have
difficult
y to meet this criterion. On the other hand
experience shows that large passenger ships are
reasonablysafe.
IMO SLF Subcommittee already took care of this
effect developing guidelinesfor alternative
assessment of the weather crit
erion adopted by
resolutionMSC.1/Circ.1200.Intheviewoftheauthor,
problem of dead ship condition, that is virtually
problemoftheweathercriterioncouldbeconsidered
assolved,atleastforthetimebeing.
6 EFFECTOFHUMANFACTORAND
OPERATIONALFACTORS
Humanfactorisanessentialelementofthesystemof
safety against st
ability failure. Accordingto many
sources human factor constitutes main cause of
casualtyinabout80 to 90per centofcasualties and
even more [12,13]. The system of safety where
humanfactoris included is showninfig 2 [12],In
the system there arenumerous int
errelations
between varioussub‐systems.But restricting only to
ship operation it would be necessary to consider
onlythepartofitasmarkedinfig2bythicklinesthat
encapsulateshumanandorganisationalerrors(HOE).
Humanandorganisationerrormaybetheresult
ofdesignandconstructi
onfaults(badcharacteristics
of ships) and force majeure, that are responsible for
about 20% of all HOE casualties, the rest may be
attributed to operational factors that include the
following:
societyandsafetyculture
organization
system
humanperformance(individual).
Fi
gure2.SafetysystemforCRGcasualties(after[12])
Amongst other factors human performance and
safe operation depends strongly on operator
understanding of physical phenomena governing
vesselmotionsinconfused seas,Thisunderstanding
maybeachievedbypropereducationandtraining.
7 IDEASHOWTOIMPROVESAFETYAGAINST
CAPSIZING
Intheopinion of the authorit is nowtimetoa
pply
different, more universal procedures to stability
criteria.Thesearealreadyconsideredwidelyandin
particularin IMO higher bodies. Two approaches
that are not charged with the above discussed
deficiencies:riskanalysisandgoalorientedapproach
areproposed.
7.1 Riskanalysis
Application of risk analysis is considered for some
time and act
ually recommended by the IMO higher
bodies.Thereisbasicdichotomybetweenprescriptive
regulations and risk analysis, and risk analysis has
some advantages in comparison with prescriptive
regulations [14].It was righty pointed out in the
comprehensive paper on second generation stability
criteria [7] that risk analysis have gained greater
acceptance and become standard tool in other
industries. Why not to use thi
s approach to safety
againststabilityfailure?
Theessentialelementoftheriskanalysisandrisk
based requirements is assessment of risk. Risk,
according to the definition is equal to product of
probabilityoffailure(P)anditsconsequences(C):
R=PxC
IMOrecommendstouseintheriskassessmentthe
logarit
hmicscaleintheform:
LogR=log(P)+log(C)
SHIP
OPERATION
CASUALTY
ENVIRONMENT
SHIP DESIGN
SHIP
CONSTRUCTION
N
Safet
y
culture
Or
g
anization
Sy
st
e
m
Human
performance
SHIP
285
Figure 3. Steps of risk analysis
This formulation is more easy to apply and to
construct a risk matrix where for probabilities
(frequencies)offailurerankingisadoptedfromFI=1
(extremely rare) to FI = 7 (frequent) and for
consequences ranking is adopted from SI = 1
(negligible) to SI = 4
(catastrophic) with associated
probabilities[14].Riskanalysisincludesthefollowing
steps(fig.3):
Risk analysis is at present a well‐established
procedure used as a rule, when planning
sophisticated systems. IMO recognized the
advantages of using risk‐based approach as an
alternative to the prescriptive criteria in different
areasofshipsafety
andultimatelytheMarineSafety
Committee of IMO recommended this approach as
Formal Safety Assessment (FSA), in MSC/Circ.1023.
Since then many papers were published on this
subject, however only few concerning stability. Also
theauthorinseveralpapersdiscussedpossibilitiesof
application of the risk assessment methodology to
intact stability
criteria (e.g. [14, 15]), but in practice
existing IMO rules on stability do not include
possibilitytoapplysuchmethods.
Obviously it would be impractical to apply this
methodtoconventionalshipsthatarereasonablysafe,
but the method could be effectively applied to
important and large ships of non‐
conventional
design.Thisononehandmaybethewaytoimprove
safety level whenapplyingexisting criteriaandon
the other hand the way to assure sufficient level of
safety for non‐conventional ships where clause for
alternative methods already included in the SOLAS
Conventionmaybeused.
7.2 Goal
orientedapproach
The most recent concept of assuring safety is goal‐
based approach. Goal based approach does not
specify the means of achieving safety but sets goals
that allow alternative ways of achieving safety[16].
Goal‐based requirements are for some time
considered at IMO and appraised by some authors
[17,
18], and they were introduced in some areas,
albeit not in the systematic manner. Marine Safety
Committee ofIMOcommenced in 2004 at MSCC 78
its work on goal‐based standards in relation to ship
constructionadoptingfive‐tiersystems showninthe
table 2. Goal oriented approach was recently
suggested
tobeappliedtosubdivisionanddamage
stability[19].
Table2.Five‐tiersystemforgoal‐basedrequirements
_______________________________________________
TierI: Goals
TierII: Functionalrequirements
TierIII: Verificationcriteriaofcompliance
TierIV Technicalproceduresandguidelines,
classificationrulesandindustrystandards
TierV Codesofpracticeandsafetyandqualitysystems
forshipbuilding,shipoperation,maintenance,
trainingetc
_______________________________________________
Theadvantageofgoal‐orientedapproachisthatin
ordertoachievegoalwhichistomakeshippingsafer,
variousmeanshavetobeused,notonlyprescriptive
regulations or even risk analysis. All means that
may contribute to safety have to be used as is
specified in the table 2.
This approach does not
exclude risk analysis, which is one of the most
importantmethodtoachievecompliance.
Conclusion. Perspectives of improvement of the
safetylevel.
Itseemsthatthebestwaytoenhancesafetywould
betoputstressonthegoal‐orientedapproach.Using
this approach with respect
to intact stability two
paradigmshavetobetakenintoconsideration:
In general opinion existing criteria are working
well for conventional ships. Experience of
application of the existing intact stability
requirements as in the IS Code 2008 reveals that
stabilityfailureswithshipsmeetingthecriteriaare
extremelyrare.
Risk control
options
Riskassessment
Hazard
probabilities
Hazard
identification
RISK ANALYSIS
Risknotaccepted
Riskaccepted
Hazard
consequences
286
For non‐conventional ships the existing
requirement of the IS Code 2008 may be
inadequateandalternativewaysofassuringsafety
should be used. Applying alternative means is
allowed under the provisions of SOLAS
Convention. Alternative means may include risk
assessment.
Bearingthisinmindthefuturewaysof
improving
safety with regard to stability failures could be as
proposedintheTable3.
In the table distinction is made between
conventionaland non‐conventional ships. There is
obviously the problem how to define non‐
conventional ships. Traditionally under the term
“non‐conventional ships”, ships revealing novel
designfeaturesare
understood.Howeverthisisvery
vague definition because of the difficulty which
featuresmaybeclassifiedas“novel”.Itseemsthatthe
bestwaytoclassifyshipsasnon‐conventionalwould
be to include in this category all ships to which
requirements of the existing IS Code are considered
nonapplicable
intheviewofdesigners,shipowners
or administrations. Experts opinion may be used
whenassessingthiscategory.
Table3.Methodsofsafetyassurance
_______________________________________________
ShipsMethodofstabilitysafetyassessment
_______________________________________________
Conventional, PrescriptivecriteriaasintheISCode
notsophisticated
Noveltypes,large Goalorientedapproachincludingrisk
sophisticatedshipsanalysisundertheprovisionallowing
applicationofalternativemeansof
assuringsafety
_______________________________________________
Application of existing IS Code criteria to
conventional ships is in the opinion of the author
substantiatedbytheexperiencegainedwiththewide
application of those criteria. Those criteria are well
known world‐wide, are reasonably simple and
working quite well. There are really no reasons for
abandonthem.
REFERENCES
[1]Kobyliński L., Kastner S. (2003): Safety and Stability of
Ships,Vol.1,Elsevier
[2]Rahola J. (1935): The judging of the stability of ships.
TransactionsofTheInstitutionofNavalArchitects
[3]IMO (2008): International Code on Intact Stability, 2008.
ResolutionMSC.267(85)
[4]KobylinskiL.(1989):Code of
stabilityforall types of ships
based on System approach 4
th
Internationa Symposium
PRADSVarna,
[5]IMO(1978):Intactstability.Generalphilosophy for ships of
alltypes.SubmittedbyPoland.DocumentSTABXXII/6
[6]IMO(2008):Reporttothe MaritimeSafetyCommittee.Sub‐
committee on stability, and load lines and on fishing vessel
safety,Doc.SLF51/17
[7]PetersW.,Belenky
V.,BasslerC.,SpyrouK.,UmedaN.,
Bulian G., Altmayer B. (2011). The Society of Naval
ArchitectsandMarineEngineers.AnnualMeetingPaper.
[8]IMO(2012):Secondgenerationstabilitycriteria.Submitted
byPoland.DocumentSLF54/14,Annex1
[9]Belenky V. L., Weems K.M.,Paulling J.R. (2003):
Probabilistic
analysisofrollparametricresonanceinheadses.
8thStabConference,Madrid
[10]France,W.N.,Levadou,M.,Treakle,TW.,Paulling,J.R.,
Michel, R.K., Moore, C., 2001: “An investigation of head
sea parametricrolling and itsinfluence on container lashing
systems”.SNAMEAnnualmeeting.
[11]IMO (2006): Revised Guidance to themasterfo r avoiding
dangerous situations in adverse weather and sea conditions.
MSC.1/Circ.1228.
[12]Payer, H. (1994): Schiffssicherheit und das
menschlicheVersagen. Hansa‐Schiffahrt‐Schiffbau‐Hafen,
131Jahrgang,Nr.10
[13]Kobyliński L.: (2005): Performance oriented criteria. .
Proceedings of theInternational HYDMAN
Conference,Ostróda
[14]
KobylińskiL.(2005):Appraisalofriskassessmentapproach
to stability of ships. International Workshop on Ship
Stability,Istambul
[15]Kobyliński L.: (2005): Performance oriented criteria.
ProceedingsoftheInternationalHYDMANConference,
Ostróda
[16]HoppeH.(2006)Goalbasedstandards–anewapproachto
theinternationalregulationof
shipconstruction.IMONews,
Issue1.
[17]VassalosD.(2002):Totalshipsafety–alife‐cyclerisk‐based
DOR for safety. The Stability Research Centre, NAME,
UniversitiesofGlasgowandStrathclyde,May2002.
[18]Chantelauve, G. (2005): On the use of risk analysis in
maritime certification and classification.
Advances in
SafetyandReliabilityESREL2005,Vol.Ip.329.
[19]Papanicolau A. et al. Goalds‐ Goal based damage
stability of passenger ships. SNAME 2012 Annual
Meeting
