303
1 CHARTERPARTYAGREEMENT
The terms under which vessels are chartered are
containedintheCharterPartyAgreement.Thesetend
tohaveasimilarformatregardlessofwhohasdrafted
them and they contain similar provisions. This
includes information on bunkers, charterers’ rights
and obligations covering speed and performance
warranted by the owner and the Charterer’ rights
shouldtheynotbemet.AssuchCharterersgenerally
pa
y for the fuel consumed over the voyage and the
CharterPartyAgreementsincommonusagescontain
clauses warranting vessel performance in terms of
fuel consumption and speed. Clearly both Speed
MadeGoodandfuel consumedaredependentonthe
weathercondit
ionsexperienced,inparticularcurrents
andwaves.ThereforetheCharterPartyAgreements
also contain definitions of ‘Good Weather Periods’
under which the performance and speed will be
achieved. Usually this is achieved by placing upper
limitsonwinds,wavesandadversecurrents.Ifthe
condit
ions experienced exceedthese, thenthevessel
does not have to perform as per the Charter Party.
However, the performance description applied for
“good weather” and how/when currents should be
taken into account is not always made clear.
Furthermore speed and performance provisions are
notnormallyconsolidatedinoneclauseorcontractual
document,andcanbepoorlyconstructed,leadingto
the potentia
l for costly disputes over speed and
performancewarranty.
2 METHODOFCALCULATION
The Performance Evaluation method is based on a
Good Weather Analysis, a methodology for speed
A Consultative Approach to Charter Party Agreements
Based on Virtual On Time Arrival
H.Davies
PerMarePerTerras(PMPT)Limited,Newtown,UnitedKingdom
S.Bevan
OceanplusLimited,Newtown,UnitedKingdom
ABSTRACT:CharterParty agreements underpin the relationshipbetweenshipowners and charterers.The
agreementguaranteestheperformanceofavesselintermsofspeedandfuelconsumption.Onthisbasisthe
charterersplanthearrivaloftheircargoandtheirprofitmargin.However,shipperformanceisdegradedby
age,periodsbetweenmaintenanceandma
nyvesselsfailtoperformasexpected.Moreovertheperformanceis
onlywarranted during thespecific conditions stated in the charter party which are not always clear. These
usuallyrefertoBeaufortForce(BF)andtheDouglasSeaandSwell(DSS)scalewhichisarchaicintheageof
NumericalWeatherPrediction.Gi
ventheseconditions,thestageissetforconflictandthereareoftendisputes
overtheweatherconditionsexperienced.Moreoverships’oftendonotarriveontimebecausethechartererhas
assumed that the ship will make good its warranted speed and not ta
ken account of the forecast weather
conditions. The authors propose a new way of approaching charter agreements with the emphasis on
consultationratherthanconfrontationfacilitatedbyanewwebbasedsoftwareplatform.
http://www.transnav.eu
the International Journal
on Marine Navigation
and Safety of Sea Transportation
Volume 11
Number 2
June 2017
DOI:10.12716/1001.11.02.13
304
andbunkeranalysiscalculationsingoodweather,as
setoutinthreeEnglishLawprecedents–TheDidymi
Case[1](LloydsReport108,1988),TheGasEnterprise
Case [2] (Lloyds Report 352, 1993) and The Gaz
Energy Case [3] (English Commercial Court: 2011
High Court of England and Wales 3108
and 2012
High Court of England and Wales 1686). The
performanceofa vessel is assessed ingoodweather
period only in order to determine the good weather
performance speed, or her speed capability in good
weather.Thereafter the good weather performance
speedisappliedontheentirevoyage,as
ifthevessel
has performed the entire voyage in good weather
conditions,theunderpinningassumptionbeingthatif
the vessel did not perform to the Charter Party in
good weather then she would not have done so in
worseconditions.
Thekeytothiscalculationisthereforetoestablish
the good
weather periods. These are usually
determinedbyaweatherserviceprovider,appointed
by the charterer and specified in the Charter Party
Agreement.Thegoodweatherperiodsareidentified
afterthevoyagebyreconstructionoftheship’strack
from the noon day reports and/or by AIS reports.
The ship positions are
then matched to the gridded
output from Numerical Weather Prediction (NWP)
modelsandthecorrespondingvaluesforwind,waves
andcurrentextracted.Thegoodweatherperiodsare
identifiedandspeedandconsumptioncalculated.If
theperformanceachievedingoodweatherislessthan
that specified in the Charter Party, then the
good
weatherspeedandconsumptionisextrapolatedtothe
entire voyage to calculate the additional bunkers
consumed and the additional time taken for the
voyage.
Theweatherreportedbythevesselplaysnorolein
thisprocess.Itisthereforenotunusualfordisputes
todevelopregardingtheweatherconditions
reported
by the vessel as compared to those assessed by the
weatherserviceprovider.
Thecurrenttransactionalapproachtocharterssets
upaconflictsituationinwhich:
1 Ship owners cast the vessel consumption and
performanceinthemostfavourablelightinorder
towinbusiness.
2 Masters are under pressure
to perform and
incentivisedto exaggeratethe weather conditions
experienced; this can be exacerbated by poor
qualityreportingoftheactualweatherandpoorly
calibratedweatherinstruments.
3 Chartererscannotrelyonthevesseltodeliverthe
cargototimeandcost.
The authors will propose an alternative
collaborativeapproach
toshipcharters,butfirstitis
worth exploring the inherent problems with the
currentapproach.
3 GOODWEATHERDEFINITION
Typicallygoodweatherisdefinedas“uptoBeaufort
Force4andDouglasSeaState3…noadverseeffectsof
Swell/Currents”.
Theobviousproblemwiththisdefinitionisthatit
isvery
unusualfortheseconditionstobemetonan
ocean voyage.Figure 1 shows the probability of
windsin excess of BeaufortForce4occurringinthe
North Atlantic during themonth of August. During
thewintermonthsBeaufortForce4isexceedednearly
100%ofthetimeacrosshuge
swathsofoceannorthof
35N.Itisunrealisticforchartererstoexpectvessels
to make an ‘on time arrival’ without excessive
consumption.Itisperhapsnotwidelyunderstoodby
shore based staff that despite their size that ships
cannotovercomethelawsofphysics.
Fuel consumption is a function of
the resistance
thatthevesselhastoovercome.Inturnresistanceisa
function of the vessel block coefficient, loading, and
trim, water density, currents, waves, winds and
biofouling.
Figure1. North Atlantic‐Probability of wind speed
exceedingBeaufortForce4inAugust(SourceCOADS)
Speed is a function of power and drive train
efficiency and is impacted by engine racing,
slammingandpropellerracingwhichareaffectedby
relativewavedirectionandmagnitude.Theimpact
ofspeedmadegoodcanbeconsiderableasshownin
Figure2.
Figure2.SpeedMadeGoodbyaTankermakingrevolutions
for12knotsforHead,Beam,Followingseas(Source:NIMA
PubNo92009)
4 DOUGLASSEASCALE(DSS)
The DSS is a two‐digit scale (proposed by Captain
Douglas, Hydrographer to the Royal Navy in the
1920s)forreportingtheheightofwavesasobserved
at sea. It allows for the distinction between the sea
305
and swell.While the generating wind blows, the
resultingwavesarereferredtoas‘sea’or‘seastate’.
When the winds stops or changes direction, waves
that continue on without relation to local winds are
called‘swell’.TheWMOgivethesedefinitionsas:
Windwaveorwindsea:
Wavesraisedbythewind
blowing in the immediate neighbourhood of an
observationsiteatthetimeofobservation.
Swell:Anysystemofwaterwaveswhichhasleft
its generating area (or observed when the wind
fieldthatgeneratedthewavesnolongerexists).
Waves are generated at a
broad spectrum of
frequenciesandseadoesnothaveaperiodassociated
withitonlyaheight.Incontrast,swelliscomposedof
gravity waves that have been generated elsewhere
and have propagated.Gravity waves experience
frequency dispersion, i.e. waves of different
frequencies travel at different speeds and swell
becomessorted
intowavesofdifferentfrequenciesas
it propagates away from the generating area, with
longperiodswellsarrivingfirst.
Theremaybeaheavy swellpresenteventhough
thewindsarelight.TheDSSwascreatedspecifically
to address this by treating sea and swell separately.
Bydefinitionswellis
notrelatedtothelocalwindand
itisthereforequiteincorrecttolinktheDouglas Sea
Scale Sea State 3 with the Douglas Sea Scale swell
description.
The Douglas Sea and Swell scale has 2
components:‘sea’and‘swell’whichbydefinitionare
independent variables. The Douglas Sea Scale is
reported as 2 numbers; the first referring to the
DouglasSeaState;andthesecondreportingtheswell.
ForexampleiftheDouglasSeaStateheightis1.25m
and the swell is described as short and heavy, then
the Douglas Sea Scale is 36. The WMO has adopted
theDouglas
SeaScale[4]forthereportingofseastate
bymarinersandrecommendsthattheterminologyis
usedin forecasts for shipping[5]. Unfortunately this
format is rarely used and typically only Sea State is
provided.
Douglas Sea Scale 3, as commonly referred to in
Charter Party Agreements, only refers to
the wind
generatedwaveheightandfrequentlyomitstheswell
component.Thelackofaquantifiedvalueforswell
then makes comparison with NWP problematic and
createsproblems for chartercompanieswho wishto
query warranty performance.NWP wave models
are spectral models which work by calculating the
total level
of wave energy in the ocean and then
assigningittoatwo‐dimensionalfrequency‐direction
domain(termedthewavespectrum)usedtodescribe
the average motion of the sea‐surface under waves.
Essentiallythespectrumdecomposesagivensea‐state
into a set of constituent sine waves, each with a
differentdirection,period(inverseoffrequency)and
amplitude(energy).Someofthesearedesignatedas
‘swell waves’ whilst others are designated as ‘wind
waves’.
There is one output parameter that combines sea
and swell that can be meaningfully compared to
observed wave heights. This is the Significant Wave
Height which
is a measure of combined ‘sea’ and
‘swell’andisdefinedasfourtimesthesquarerootof
thefirstmomentofthewavespectrum;thisiscloseto
the average height of the highest 1/3 of the waves
(andhasitsoriginsinMunk &Sverdrup1947[6])and
is
what shipborne observers are expected to report.
Measurements of Significant Wave Height are the
mainsourceofdataforwavemodelsandarederived
from remote sensing using space borne Synthetic
ApertureRadar(SAR)andaltimeters.Measurements
of Significant Wave Height from buoys are also
assimilatedandprovidegroundtruth.
Significant Wave Height therefore represents a
‘commoncurrency’ between observations and NWP.
Weather service providers commonly apply a
SignificantWaveHeightof2.0metresasequivalentto
aCharterPartyentryofDSS3.Howevertherehasnot
yetbeenadefinitiverulingonthisinArbitrationand
theuseof
DSSinCharter Partiescontinueswith the
swellpartomittedinmanycases.
Setting aside issues around height, the period of
theswellcanalsohaveabearingontheseakeepingof
the vessel and hence impact performance – if
successive waves strike the side of a vessel at the
samephaseofsuccessiverolls,relativelysmallwaves
can cause heavy rolling [7]. The IMO has published
algorithms and guidance for Masters for avoiding
dangeroussituations[8]whichcanreadilybeapplied
topredicttheimpactofswellonseakeepingandship
safety (covering reduction of intact stability,
synchronousrollingand
parametricroll,etc.).
5 ACCURACYOFNWP
Theanalysisandforecastsofsurfacewindaremature
andhavea high degree of reliability.For example,
Figure 3 verifies forecast windspeed against
windspeedobservedatmannedobservingstationsin
Europe where the observations are reliable.It
indicates a forecast accuracy of
+/‐ 1 knot at T+72
hours i.e. 3 days.Similarly the most skillful wave
analyseswhencomparedtobuoysareaccurateto+/‐
0.3metre.
Figure3:RMSErrorofECMWFforecastsof10mwindspeed
atT+48and72(Source:ECMWF)
NWPanalysesarebasedonobservationsandare
themostaccurate depiction of the conditions at that
time.Windanalysesarereadilyavailableat6hourly
intervals at 0000/0600/1200/1800UTC and waves at
0000/1200UTCdaily.Eachmodelcreates‘snapshots
ofreality’as3hourlyforecastsfromtheanalysisbase
time. NWP
models therefore have a temporal
granularityof3hoursandiftheshippositionreport
306
doesnotoccur at 00/03/06UTCetc. then this means
that interpolation in time is required between the
forecastvaluesortheshippositionatthetimeofthe
forecast has to be estimated.Similarly the models
outputonagridwithvaluesatdiscretelatitudeand
longitudes. The grids
are currently around 0.25
degrees by 0.25 degrees and so if the ship position
does not fall exactly onto the grid point then
interpolationinspaceisrequiredbetweentheforecast
valuesatadjacentgridpoints.
Meteorologicalandoceanographicphenomenaare
notlinearandarecharacterisedbydiscontinuitiesand
interpolation
ispronetounquantifiableerrors.NWP
should therefore be used with caution to assess the
conditions experienced during a voyage as the
temporal and spatial grid is unlikely to match the
time and location of ship reports and a degree of
interpolationwillberequired.
6 WEATHERROUTEING
Notwithstanding the issues
highlighted in the
previoussectionregardingtheuseofNWPtoassess
theconditionsexperiencedduringavoyage,weather
forecastsare incredibly accurate. Figure 4 shows the
anomaly correlation scores, which are a measure of
accuracy, for the 5 day forecasts produced by the
leadingNWPmodels.Thebestmodels
arenowbetter
than 90% accurate at 5 days and still show skill in
their 14 day forecasts. This accuracy enables
optimised ship routeing to avoid bad weather and
adversecurrentsinordertominimisethefuelburned
overavoyagewhilstensuringOnTimeArrival.The
problem is that this
process is currently conducted
seperately or is not reflected in the Charter Party
Agreement.
Figure4.AnomalyCorrelationof5‐dayforecasts(Northern
andSouthernhemispheres)(SourceNCEP)
The next section proposes a new way of
approaching charter agreements with the emphasis
on consultation rather than confrontation facilitated
by a new web based optimised ship routeing
platform.
7 EFFICIENCY
Aswehaveseen,currents,wavesandwindincrease
the resistance thata vessel must overcome to move.
Clearlythis
resistancevariesintimeandspaceduring
a voyage. Therefore, during the voyage either the
vessel Speed Made Good or the vessel slip can be
expected to fluctuate as a result of the resistance
encountered.
Marine diesel engines operate efficiently in a
narrow power band and in order to minimise
fuel
burn constant power should be maintained within
this power band for as much of the voyage as is
practical. This is clearly demonstrated in Figure 5
which shows the impact of added resistance on the
power required to achieve different speeds as
measuredduringanexperimentonMVRotterdam.A
small difference in current has a large impact on
powerrequired.
Figure5.ImpactofcurrentonMVRotterdampowercurves
(Source:Kessel,Ronald,EmpiricalEstimateoftheInfluence
ofWind,Current&SurfaceWavesonEnergyConsumption
byShips,29April2013,CMRE‐FR‐2013)
It therefore makes sense to assess the total
resistancethatwillhavetobeovercomeontheentire
voyageforanumberofpossibleroutesandtoselect
the least resistance route. Applying basic navigation
principles, resistance can be calculated in terms of
miles. The average speed that will need to
be
achievedovertheentirevoyageinordertomakean
On‐Time‐Arrival can then be calculated, the vessel
canthenmakerevolutionsforthisspeed,andletthe
SpeedMadeGoodfluctuate.
Asimpleexampleillustratestheprinciple.Takea
vessel making revolutions for 10 knots and
encountering
aheadcurrentof1knot.Thevesselwill
onlyachieveaSpeedMadeGoodof9knotsandwill
have to travel 11 miles through the water to make
good10miles.Overadistanceof100miles,thevessel
caneither arrive in 10 hours ma kingrevolutions for
307
11knotsORarrivein11hoursmakingrevolutionsfor
10knots.
Thissimpleexampleshowsthatitisnotsensibleto
estimate arrival timesor fuel consumption basedon
thenominalvalues andunrealisticweatherconditions
currently assumed in Charter Party Agreements.
There is always likely to be a
trade‐off between
operating at the most efficient running speed and
burningleastfuel,andarrivingataspecifictime.This
information can be presented to the charterer and
ship owner to inform their decision and provide
sharedassumptionsregardingtheforecastconditions
over the voyage and their impact on
vessel
performance.
8 TRADINGSPACE
Given some basic information about the vessel and
the voyage, the resistance and distance through the
water can be calculated for the forecast weather
conditions during the voyage.The weather
conditions(andresistance encountered)areafunction
oftimeaswellasspace,theythereforedepend
onthe
vessel’sSpeedMadeGood.Inordertosimplifythe
calculation 4 possible vessel speeds and
correspondingoptimsiedroutesareconsidered;ECO,
Charter, Maximum continuous rating (MCR) and
speed required to achieve On Time Arrival. The
possible combinations of speed, consumption and
arrival time can then be presented along
with the
numberofbadweatherperiodsandimpactofcurrent
thatareexpected.Figure6showsanoptimumroute
summaryfora vesselsailingbetweenPembrokeand
NewYorkinOctober2015whichavoidedex‐Tropical
Storm Joaquin, reduced impact from the adverse
Gulfstream/North Atlantic drift currents and
minimisedthe
distanceintheEmissionControlArea
(ECA) off the US east coast. While Beaufort 4 and
Douglas Sea Scale 3 conditions were expected to
exceed 36 hours due to unsettled weather patterns
acrossthe North Atlantic,anOn‐Time‐Arrivalcould
be achieved at 12.9kts routing south via the Azores
whichadded some 300nmto the shorter great circle
route.
As part of a new consultative approach it is key
thatallpartiesviewthesameinformationthrougha
web browser prior to sailing. The authors have
developed a web interface to enable a consultative
approach,asexemplifiedbytheinformation
revealed
atFigure6whichsummarisestheexpectedperiodsof
badweather,theimpactofcurrent,theearliestarrival
timeandotherrelevantinformationforanoptimised
routewherethepriorityisforanon‐timearrival.
ConstraintssuchaslimitingexposuretoHighRisk
Areas or minimising mileage in
Emission Control
Areas can be added.Algortihms can also be
embedded, for example Seakeeping algorithms or
algorithmstocalculatetheCO
2,SOxandNOx(which
isspeeddependant)foreachofthespeedoptions.
RouteBenefit
Calculation
Total
(nm)
ECA
(nm)
#ofbad
weather
periods
(6hrs)
Total
(nm)
ECA
(nm)
#ofbad
weather
periods
(6hrs)
2908 1244 36 3115 490 23
Distancethrough
thewater(nm)
3314
3115
CPSpeed(kts)
12.5
12.5
Speedrequiredfor
OTA(kts)
13.8
12.9
TotalVoyagetime
(hrs)
256
241
199
6
15
754
241hrs
ReduceddistanceinECA(nm)[DifferenceinECA
mileage]
Earliestarrivaltime[Shortestvoyagetime&
durationatconstantpower]
ShortestRoute(BP
DistanceTable)
RecommendedRoute
Saving(nm)[Additionaldistancethroughthe
water]
AdditionalSlip(%)[Achievedvstheoretical
distanceperproprev]
Saving(hours)[Additionaltimeduetobad
weather/icebergs]
Figure6.Exampleofroutebenefitcalculationsforan
optimisedroutefromPembroke,UKtoNewYork.
9 POSTVOYAGE
Thesameapproachcanbeappliedonceavoyagehas
been completed where an adapted process using
moreaccurateanalysedweatherconditionsinsteadof
forecasts can be used to verify performance.
This
process offers a fast and cost‐effective evaluation of
ship performance related to weather criteria
stipulatedinCharterPartyAgreements. Byexample,
Figure 7 shows an assessment of ship performance
against weather criteria which can easily be viewed
viaawebbrowserbyallparties.Thisexampeshows
that
thespeedmadegoodrarelyachievedtheCharter
Partyspeed.If,inthiseample,theweatherconditions
had beenbenign then this assessment would
provide an early indication that the vessel is
underperforming – something that any
owner/operator would wish to know as soon as
possible.
10 LIMITATIONS
Althoughthe options
presented areonlyasaccurate
as the weather forecasts and the vessel information,
the process and the underpinning assumptions are
explicit and offer a step improvement over current
practice. The assumptions made are shared by all
parties and the resulting decisions form a common
and mutual consultative audit trail for
the voyage.
The use of a web based software tool will allow a
more consultative rather than confrontational
approach to resolving warranty issues that should
308
helpreducetheneedforcostlyarbitrationintheevent
ofadispute.Testingofthisapproachwithinterested
shipping companies and charterers is expected to
commenceshortly.
Figure7.Exampleofapostvoyageperformanceassessment
showingshipspeedmadegoodgenerallybeingbelowthe
CharterPartyspeedforthemajorityofthevoyage.
REFERENCES
[1]The Didymi 1998 2 Lloyd’s Law Report 108 (Didymi
Corporationvs.AtlanticLinesandNavigationInc.)The
“Didymi”case
[2]TheGasEnterprise19932Lloyd’sLawReport352(B.P.
ShippingLtd.vs.ExmarN.V.)The“GasEnterprise”case.
[3]Hyundai Merchant Marine Company Limited v
TrafiguraBeheerB.V.2011.
The“GazEnergy” English
Commercial Court: Flaux J: [2011] High Court of
England and Wales 3108 (Comm) 29 November 2011
and English Commercial Court: Teare J: [2012] High
Court of England and Wales1686 (Comm): 21 June
2012]
[4]CIMO Guide 7th Edition 2014.Guide to Meteorological
Instruments and Methods
of Observation, Chapter 4, para
4.2.2.13.1
[5]https://www.wmo.int/pages/prog/amp/mmop/faq.html
[6]Munk & Sverdrup 1947. Wind, Sea, andSwell: Theory of
Relations for Forecasting, US Hydrographic Office Pub
No.601.
[7]The American Practical Navigator (2002) (Bowditch,
Pub.No.9).Chapter32.
[8]IMO MSC.1/Circ.1228 11 January 2007. Guidance to the
Masteron
avoidingdangeroussituationsinadverseweather
andseaconditions.7‐9
.
