331
1 INTRODUCTION
Container terminal performance is linked to the
mechanism used in the organization of processes
within the container terminal ill are among the ship
andthepierordownloadtheinmatesoperations Om
then monitoring operations in the squares and the
extentofmasteryofthoseoperationsandavailability
Simulation of a Container Terminal and it’s Reflect on
Port Economy
A.Elentably
KingAbdul‐AzizUniversity,Jeddah,SaudiArabia
ABSTRACT:Thecombinationbetweenthedesignandprojectofcontainerterminalsandthereflectonport’s
economy may be carried out through two main approaches: optimization or simulation. Although the
approaches based on optimization models allow a more elegant and compact formulation of the problem,
simulationmodels are mainlyba
sedon discrete event simulation(DES) models andhelp to achieve several
aims: then measure this impact on port economy before and after implemented this updating overcome
mathematicallimitationsofoptimizationapproaches,supportandmakecomputer‐generatedstrategies/policies
moreunderstandable,andsupportdecisionmakersindailydecisionprocessesthrougha“whatif”a
pproach.
SeveralapplicationsofDESmodelshavebeenproposedandsimulationresultsconfirmthatsuchanapproach
isquiteeffectiveatsimulatingcontainerterminaloperations.Mostofthecontributionsintheliteraturedevelop
object oriented simulation models and pursue a macroscopic approach which gathers elementary handling
a
ctivities(e.g.usingcranes,reachstackers,shuttles)intoafewmacro‐activities(e.g.unloadingvessels:crane‐
dock‐reachstacker‐shuttle‐yard), simulatethe movementof an“aggregation” ofcontainers andtherefore do
nottake intoaccount the effectsof container types (e.g.20’ vs 40’,full vs empty), theincidence of different
handlinga
ctivitiesthatmayseemsimilarbutshowdifferenttimedurationandvariability/dispersion(e.g.crane
unloading a container to dock or to a shuttle) and the differences within the same handling activity (e.g.
stacking/loading/unloading time with respect to the tier number). Such contributions primarily focus on
modeling architecture, on software implementation issues and on simulating design/real scenarios. Acti
vity
durationisoftenassumedtobedeterministic,andthosefewauthorsthatestimatespecificstochastichandling
equipmentmodelsdonotclearlystatehowtheywerecalibrated,whatdatawereusedandwhattheparameter
Values are. Finally, no one investigates the effect
s of different modeling hypotheses on the simulation of
containerterminalperformances.Thefocusofthispaperisontheeffectsthatdifferenthypothesesonhandling
equipmentmodelscalibrationmayhaveonthesimulation(discreteevent)ofcontainerterminalperformances.
Sucheffects could notbenegligible and should be invest
igatedwith respect to different planning horizons,
suchasstrategicortactical.Theaimistoproposetoanalysts,modelersandpractitionersasortofaguideline
usefulto point outthestrengths or weaknesses ofdifferentapproaches.Drawing on the model architecture
whichwillbeaffectedonporteconomics.
http://www.transnav.eu
the International Journal
on Marine Navigation
and Safety of Sea Transportation
Volume 10
Number 2
June 2016
DOI:10.12716/1001.10.02.16
332
oftimetowaitfortheshipandthenreduceoperating
costs of the vessel and also limit the consumption
increasing cranes and subsequent spare parts and
maintenance operations that the use of any
organization
TheDesignandprojectofcontainerterminalsmay
be carried out through two main approaches:
optimizationorsimulation.Althoughtheapproaches
based on optimization models allow a more elegant
andcompactformulationof theproblem, simulation
modelsaremainlybasedondiscreteeventsimulation
models and help to achieve several aims: overcome
mathematicallimitationsofoptimizationapproaches,
support and make computer‐generated
strategies/policiesmoreunderstandable,
andsupport
decisionmakersindailydecisionprocessesthrougha
“what if” approach. Several applications of models
have been proposed and simulation results confirm
thatsuchanapproachisquiteeffectiveatsimulating
container terminal operations. Most of the
contributionsintheliteraturedevelopobject‐oriented
simulation models and pursue a
macroscopic
approach which gathers elementary handling
activities (e.g. using cranes, reach stackers, shuttles)
into a few macro‐activities (e.g. unloading vessels:
crane‐dock‐reach stacker‐shuttle‐yard), simulate the
movement of an “aggregation” of containers and
therefore do not take into account the effects of
container types (e.g. 20’ vs
40’, full vs empty), the
incidence of different handling activities that may
seem similar but show different time duration and
variability/dispersion (e.g. crane unloading a
containertodockortoashuttle)andthedifferences
within the same handling activity (e.g.
stacking/loading/unloading time with respect to the
tier number). Such contributions
primarily focus on
modeling architecture, on software implementation
issues and on simulating design/real scenarios.
Activity duration is often assumed to be
deterministic, and those few authors that estimate
specificstochastichandlingequipmentmodelsdonot
clearly state how they were calibrated, what data
wereusedandwhattheparametervalues
are.Finally,
no one investigates the effects of different modeling
hypotheses on the simulation of container terminal
performances.Thefocusofthispaperisontheeffects
that different hypotheses on handling equipment
models calibration may have on the simulation
(discrete event) of container terminal performances.
Such effects could not be
negligible and should be
investigated with respect to different planning
horizons, such as strategic or tactical. The aim is to
proposetoanalysts,modelersandpractitionersasort
of a guideline useful to point out the strengths or
weaknessesofdifferent approaches. Drawing on the
model architecture proposed in a
previous
contribution by the same authors(,de Luca, 2005), a
discrete event simulation model is developed and
appliedtotheRedseaContainerTerminalinorderto
dealwiththefollowingissues:
Analysis of the effects of different estimation
approaches (sample mean and random variable
estimations) on estimating whole terminal
performance,
hence on container terminal planning
strategies. In particular, analyses were made for
different time horizons: long‐term planning
interventions/investments, medium/short period,
short‐termorreal‐timeapplications.
‐Analysisoftheeffectsofdifferenthypotheseson
the level of aggregation of elementary activities
(undifferentiatedvs.containertypemodel).
Thepaperis
dividedintofoursections.Inthefirst
section (section 2) an in depth literature survey is
proposed.Theaimistogobackoveraboutthirtyyear
ofcontainerterminalsimulationmodels,tohighlight
weaknesses points of the existing approaches to
handling equipment activities simulation, and to
propose a synthetic
but complete outline of the
modelscalibratedandoftheir
parameters.Insection 3abriefdescriptionof the
discreteeventsimulationmodelisreported.Insection
4resultsfrommodelapplicationareproposedwhile
themainconclusionsaredrawninsection5.
2 LITERATUREREVIEW
The existing literature reports approaches
to either
managingacontainerterminalasasystemandtrying
to simulate all elements or managing a subset of
activities(simultaneouslyorsequentiallyfollowinga
predefinedhierarchy).Themaincontributionsseekto
maximizeoverallterminalefficiencyortheefficiency
ofaspecificsub‐area(oractivity)insidetheterminal.
The most widely followed approaches are based on
deterministic optimization methods, although
recently a stochastic optimization model was
proposed(Murty.2005).Suchapproachesschematize
container terminal activities through single queue
modelsorthroughanetworkofqueues.Followinga
stochastic approach, both modeling solutions may
lead to analytical problems and/or unsatisfactory
results if the probability distribution of activities
involved does not belong to the Erlangen family
(Nilse, 1977; Ramani,1996). Moreover, the resulting
network could be very complicated and theoretical
solution might not be easy to obtain. In such a
context, an effective and challenging alternative
approachforcontainerterminalsystemanalysis
may
berepresentedbydiscretesimulation.
Simulation can help to achieve various aims:
overcome mathematical limitations of optimization
approaches, allow a more detailed and realistic
representation of terminal characteristics, support
decision makers in daily decision processes through
assessment of “what if” scenarios and make
computer‐generated strategies/policies more
understandable.Simulationis
notanewmethodology
inportoperations.Severalworkshavebeenpresented
since the 1980s, most of them concerning port
operations management. Many of the proposed
models do not focus on the details regarding the
modelset‐up,itscalibrationanditsvalidation;buton
the application and/or the simulation
of design
scenarios. Moreover, although the estimation of
handling activity models should be one of the main
issues of all container terminal applications, this
problemdoesnotseemtobetreatedindepthinmost
applications. While many contributions do not
presentanyinformationonhandlingactivitymodels
used, the remaining
contributions carry out very
333
simple approaches (deterministic) and/or give scant
informationontheestimationapproachadopted,the
experimentaldataused,theparametersestimatedand
onparametervalues.
Theaim ofour analysisis twofold topropose an
extensive review of the main contributions in the
literature, to focus on the approaches, models and
parameters
usedtomodelhandlingactivities.
StartingfromthepioneeringworkofCollier(1980)
investigating the role of simulation as an aid to the
study of a port as a system, the 1980s saw several
works implementing the first simulation‐based
models. In Agerschou et al.(1983), Tugcu (1983)
proposedasimulation
modelfortheportofIstanbul,
dealing with berth assignment and unloading
operations. Vessel arrival is simulated through
Poissondistribution, whereas empirical distributions
areusedfortheremainingactivities.ElSheikhetal.
(1987)developedasimulationmodelfortheship‐to‐
berth allocation problem; the phenomenon is
modelled as a
sequence of queues, and vessel
interarrival and service time are modelled through
exponential distribution functions. In the same year,
Park and Noh (1987) used a Monte Carlo type
simulationapproachtoplanportcapacity,Comerand
Taborga (1987) developed one of the first port
simulation softwares (PORTSIM), and Chung et
al.
(1988) proposed a methodology based on a graphic
simulationsystemtosimulatetheuseofbufferspace
toincreasetheuseofhandlingequipmentandreduce
totalcontainerloadingtime.
Inthe1990smucheffortwasspentonsimulating
terminalcontainers: thenumberofapplicationsbased
on simulation increased, terminals
were modelled
morerealisticallythroughdisaggregationofthemain
operationsinseveralelementaryactivities,andmuch
more attention was laid on real case studies. The
focus of most contributions was on developing
practical tools to simulate terminal operations, on
software issues and/or on model validation. Less
attention was focused on
modelling handling
activities and/ormodeldetails. Kondratowicz(1990),
within a general method for modelling seaport and
inlandterminalsinintermodal freighttransportation
systems, proposed an object‐oriented model,
TRANSNODE, to simulate different application
scenarios. Silberholz et al. (1991) described a
simulation program that models the transfer of
containerized cargo to and
from ships, Mosca et al.
(1992)usedsimulationtoascertaintheefficiencyofan
automatic flatar system servicing a rail‐mounted
crane, and Hassan (1993) gave an overview of a
computer simulation program used as a decision
supporttool toevaluate andimproveport activities.
LaiandLam(1994)examined
strategiesforallocation
ofyardequipmentforalargecontaineryardinHong‐
Kong.Inthesameyear,
Hayuth et al. (1994) used a discrete event
simulation to build a port simulator, but the main
emphasis was on software and on hardware
problems.Keyissuesoftheapplicationofmodelling
and
simulationwerediscussedinTolujevetal.(1996)
and Merkutyev et al. (1998), both contributions
proposing an application to the Riga Harbour
Container Terminal. Gambardella et al. (1998)
proposedadiscreteeventsimulationmodel(basedon
process oriented paradigm) to simulate vessel
loading/unloading. The model was applied to the
Italian
container terminal of La Spezia (Italy), with
scant information on the data used and on the
characteristics of the equipment used in the
application. The same case study was analyzed by
Mastrolillietal(1998), usingamodelsimilarto that
proposedinGambardellaetal.(1998)andproposing
acalibration
andavalidationprocedureofsimulator
parameters. Means and standard deviations are
estimated for quay crane, yard crane and straddle
carrier service time, whereas speed of cranes and
travel time of shuttle trailers are assumed
deterministic, as well as vessel arrival and truck
arrival. Nevins et al. (1998) developed PORTSIM,
a
seaport simulation model able to animate and
visualize seaport processes and in the same year
Signorile(1998)developedasoftwaretooltosupport
terminaloperatorsinmakingstrategicdecisions.The
main emphasis was on optimizing container
placementinaterminal;ageneticalgorithmapproach
was adopted, a simple application proposed,
yet no
details can be found on the performance functions
used. The same authors (Bruzzone et al.1999)
investigated the effectiveness and benefits of a
simulationapproachasadecisionsupportsystemfor
complex container terminals. Interesting modeling
details were proposed by Koh et al. (1994), Walton
(1996)andRamani(1996).
Kohetal.(1994)developed
an object‐oriented approach using MODSIM
simulation software. The proposed model relies on
experimental data, average values are used for
handling equipment, whereas Weibull distribution
seemstofitcranecycletimebetter.Holguìn‐Veraand
Walton proposed a simulation model based on the
next event approach.
The model is calibrated on
experimental data and two approaches are carried
out: a deterministic one based on empirical
distributionandastochasticone.Gantrycrane,yard
craneandcranemovementsaresimulatedthrougha
random variable made up by systematic and a
random component. While the systematic
componentsare
estimatedusingmultipleregression,
the corresponding random parts are not clearly
introduced. Ramani (1996) designed and developed
aninteractivecomputersimulationmodeltosupport
the logistics planning of container operations. The
model provides estimates for port performance
indicators.
Since the end of the 1990s, the most important
ports in the world
have been modeled through
discreteeventsimulationmodels,andgreaterinterest
is shown in the calibration of handling activities
models. Choi (1999) develop an object‐oriented
simulation model using SIMPLElanguage and
applyittoanalyzethecontainerterminalsystemused
inPusan. The system is analyzed as a whole
(gates,
yards and berths), deterministic and stochastic
distribution functions are considered: deterministic
fortrailerspeed and forinter arrival time oftrailers
and tractors; uniform for service time at the gates;
exponential for inter arrival time of trailers, vessels
andservicetimeofcranes.
The same case study proposed by Yun
(2000)
follows an object oriented approach, developing a
modeltosimulatetwo different terminals locatedin
Pusan.Thesimulationtoolisgenericandtransferable
334
toany otherterminal;itis based onVisual C++and
gives accurate results once validated on historical
data. As regards equipment characteristics, averages
are used for cranes and trailer speed, whereas
distribution functions are used for crane operation
time (Normal distribution). It is not clear whether
performance characteristics were
estimated. Hussain
(2000)dealwithberthoperationandcraneallocation
problems. Their discrete event simulation model is
based on data collected at the port of Kelang and
specific analyses are carried out to identify the
distribution functions for inter‐arrival time of ships
(Weibull distribution) and for service time at
berths
(distribution not mentioned). The model is
implementedinARENAsoftwareandisvalidatedon
historical data. Mazza (2001) examine the vessel
arrival‐departure process, developing a queuing
network model through an object‐oriented approach
implemented in VISUAL SLAM language. Since no
detaileddisaggregatedataareavailable,afirstorder
Erlangen
distribution is applied for those services
withthesupposed larger variance, a higher order is
adopted for more regular services and, finally, a
triangulardistributionisusedtoassignthenumberof
containerstocranes.
Angelides(2002)developadiscreteeventmodelto
simulate the inbound container handling problem.
The
model is implemented in an EXTEND software
package and applied to the port of Thessaloniki.
Truck inter‐arrival times follow an Erlangen
distribution, whereas maximum, minimum or most
probablevaluesareestimatedforspeedandactivity
time of equipment involved. Developing a
microscopic simulation model, Chin et al. (2002)
evaluate the
effectiveness of automated guidance
vehicles. The focus is on the application and no
detailsaregiveneitheronthemodelsordataused.
Yeung(2002) proposea discrete eventsimulation
modelemployingtheWitnessprogramtoanalyzethe
performanceofHongKong’sKwaiChungcontainer.
Although the model encompasses all
the operations
that may occur in a terminal, the focus is on vessel
arrivals and their distribution among the existing
buffers and operators. While arrivals are simulated
throughadistributionfunction(k‐stageErlangen),the
remaining operations are analyzed in a very
aggregate way and average values are considered
(averagehandling
capacity).
Kiaetal.(2002)useaportsimulatordevelopedin
TAYLORIIsoftwaretoinvestigatetheeffectivenessof
two different operational systems applied to the
terminal of Melbourne. With the emphasis on
terminal capacity, all the activities that occur inside
the terminal are not explicitly simulated but
aggregatedin
onevariablerepresentedbythevessel’s
servicetime.Althoughnodetailsarereportedonthe
model structure, interesting statistical analyses are
presented on vessel arrival patterns (exponential
distribution for inter arrival times) and on vessel
servicetime(k‐stageErlangdistribution).
Parola and Sciomachen (2005) present a discrete
eventmodelto
simulatethelogisticchainofasystem
made by two ports, three possible destinations and
connections between them (by road and/or by rail).
The simulation is undertaken through WITNESS
simulation software and the main emphasis is on
vessel berthing, vessel loading/unloading and gate
operations. Vessel inter arrival is represented by
an
exponential distribution function (estimated), crane
workingtimeandtruckwaitingtime byatruncated
normal distribution. It is not clear whether the
probability distributions were estimated or simply
takenfromtheliterature.Biellietal.(2006)developa
simulation tool in JAVA programming language to
simulatetheport of Casablanca.
Thefocusison the
architecture and on software issues; handling
activitiesarehypothesizedasdeterministic.
Petrovic(2007)simulateunloadingservicesofbulk
cargovessels.Theystresstherelevanceofastochastic
approachandschematizethesystemasathree‐phase
queuingsystemwithdifferentnumbersofserversin
each
phase.A simulationtoolis created inPASCAL
programming language, and all variables are
generated using the Monte‐Carlo method according
to distribution functions obtained from an existing
river terminal: normal for anchorage operations and
for crane unloading times, exponential for inter‐
arrivalofvessels.
Cortès et al. (2007) set out
to simulate the whole
freight transport process in the Guadalquivir river
estuary. Despite a detailed description of operations
and the software modules implemented, little
information on equipment characteristics and time
duration is reported. Deterministic functions appear
tohavebeen usedfor gantrycranes,exponential for
thetransfertimeindock
assignmentwhileforvessel
arrivaltimeanempiricaldistributionfunctionisused.
Cho (2007) propose a model to simulate the
effectivenessof adynamicplanning systemfor yard
tractors utilizing real‐time location systems
technology. AutoMod 11.1 software is used and
statistical modelsareproposed. Of the contributions
introduced so far,
as already pointed out, only ten
papers give information on the handling equipment
models used. Half of them adopt a stochastic
approachandshowestimatedparametervalues.Most
of the contributions deal with vessel
loading/unloading operations. There is substantial
heterogeneity regarding the level of aggregation of
activities involved and how
such activities are
aggregated in a single macro‐activity: El Sheikh
(1987),Choi(2000),Kiaetal.(2002)andYeung(2002)
analyse the entire time to load (unload) a vessel
(vesselcycletime);Kohetal.(1994)andBugavicand
Petrovic(2007)investigatethecranecycletime(time
neededto:lock
ontothecontainer,hoistandtraverse,
lower and locate, unlock and return); crane loading
time to/from a vessel is analysed by Tugcu (1983),
Thiers(1998),YunandChoi(1999),Merkuryevaetal.
(2000), KMI (2000), Sciomachen (2005), Bielli et al.
(2006),andCho(2007).Asregardsvesselcycletime,a
stochastic approach is unanimously proposed.In
particular, El Sheikh (1987), Kia et al. (2002) and
Yeung(2002)suggestusingErlangrandomvariables
whereas Choi (2000) proposes normal random
variablesfortwocranetypes(quay,yard).Asregards
cranecycletime,Kohetal.(1994)advisetheuseofa
Weibullrandomvariable;
Bugavic and Petrovic (2007), for a bulk cargo
terminal, propose normal random variables and
reporttheestimatedparameters.Withregardtocrane
loading/unloading time, Tugcu (1983), Thiers (1998),
335
KMI (2000) and Bielli et al. (2006) follow a
deterministic approach, contrasting with the
stochasticapproachadoptedbyYunandChoi(1999),
Merkuryevaetal.(2000),LeeandCho(2007),Parola
(2005). Choi (1999) propose the exponential
distribution function both for quay crane and yard
crane;Merkuryevaetal. (2000)
propose the uniform
distributionfunctionforquaycraneandatriangular
distributionfunctionforyardgantrycrane;
Cho (2007) suggest the exponential distribution
functionforquaycraneandatriangulardistribution
functionforyardgantrycraneoperationtime.
ParolaandSciomachen(2005)estimatedanormal
random variable but do not
report parameter
values.With respect to crane speed, all propose
deterministicandaggregatemodels whileonlyChoi
(1999),Choi(2000),KMI(2000)andLegatoetal.(2008)
reporttheestimatedmeanvalues.
With respect to other handling equipment, not
muchcanbefoundintheliterature:Angelides(2002)
use deterministic values for a
straddle carrier,
whereasMerkuryevaetal.(2000)proposeatriangular
distribution function for the forklift. As regards
shuttleperformances(speed,traveltime,waitingtime
…), the few models existing are hard to transfer to
different case studies (due to the influence of path
length,pathwinding,trafficvehiclecongestioninside
the terminal and so on). Hence they are omitted in
thissurvey.Foreachtypeofhandlingequipmentand
for each activity simulated, probability distribution
andcorrespondingparametersarereported.
3 MODEL
The proposed approach schematizes a container
terminal(CT)asadiscreteeventsystemandmodels
itsfunctioningthrougha
simulator.Adiscreteevent
system can be defined as an interacting set of
entities/objectsthatevolvesthroughdifferentstatesas
internal or external events happen. Entities/objects
may be physical, conceptual (information flows) or
mathematical, and can be resident or transient.
Resident entities remain part of the system for long
intervals
of time; transient entities enter into and
departfromthesystemseveraltimes.Entitiescanbe
characterized by parameters and/or variables.
Parameters define static (stationary) characteristics
thatneverchange,variablesdefinethestate(dynamic
characteristics) of each entity and may change over
timeandcanfurtherbeclassifiedasdeterministicor
stochastic. In a CT entities represent the handling
equipment, the containers and all those physical
locationsrelevanttoCToperations(dock,yard,gates,
etc..).
Handlingequipmentisa residentandactiveentity
andmaybecharacterizedbyparameters,variables
andanactivity.
Containersaretransientandpassiveentities.
Physicallocationsareresidentandpassive entities.
As for containers, they may be characterized by
parameters and variables. Apart from the above‐
describedentitiesotherentitiescanbeconsidered.
Such entities do not usually move containers but
cancontrol/manageentitiesthathandlecontainers
andcanthuschangetheirattributes.The
changein
suchattributesmaybedrivenbysimpleheuristic
rules (e.g. if there are more than four trucks
waiting for a reach stacker, use one more reach
stacker) or by sub‐models that change entity
attributes, trying to optimize overall terminal
performanceinrealtime.
In discrete event modelling
the model is defined
once the case study is defined and three main tasks
shouldbecarriedout.
Identification of the terminal’s logical and
functionalarchitecture.
Demandcharacterizationandestimation.
Supplycharacterizationandcalibration.
3.1 Casestudy
In this paper the red sea Container Terminalis
analyzed. is
a major private container terminal
operatorinsouthernItaly,andisbothsmallandvery
efficient: it handles close to 0.45 MTEUs per year in
less than 10ha (100,000 m2), which amounts to 45
TEUs/ha. The red sea Container Terminalcan be
divided into three subsystems: enter/exit port gates
(land‐
side), container yards, and berths (sea‐side).
Container handling equipment comprises storage
cranes, loading/unloading cranes, yard tractors,
trailersand reach stackers.The basic activities occur
simultaneouslyandinteractively,andcanbegrouped
into four main operations: receiving (gate – yard),
delivery (yard – gate), loading (yard – berth) and
unloading(berth
–yard).
3.2 Modelarchitecture
Three different macro‐activities were taken into
account:import,exportandtranshipment.Apartfrom
vessel arrival and berthing (not relevant to our case
study) and apart from truck arrival, all the typical
activities of a container terminal were explicitly
simulated.
3.3 Demandcharacterization
Demandis
representedbysinglecontainers.Foreach
macro‐operation (import, export, transhipment), the
demand flows were characterized over space, time
and type. As regards spatial characterization,
container flows were subdivided by origin and
destination zone and were arranged in origin
destinationmatrices.Inparticular,foreachoperation
macro‐origin and macro destination
zones were
identified, usually corresponding to quays, yards,
gates. Different matrices were estimated for each
containertype(20feetvs.40feet,fullvs.empty,…),
each demand flow was characterized by its
distributionovertime.(details;deLuca,2009).
3.4 Supplycharacterization
Asintroducedintheprevioussections,in
acontainer
terminal macro operations, operations and handling
activitiesmaybedistinguished.Macrooperationsare
set up by operations; operations are set up by
elementaryhandlingactivities.Insuchaclassification
336
the different entities involved must be characterized
bytheirgeometricalcharacteristics(ifphysicalpoints)
andbythecorrespondingperformancesupplied(time
durationand/ortransportcapacity).Storage capacity
was estimated for quays and yards; averages and
probabilitydistribution functions were estimated for
handlingequipmentstimeduration.Inthefollowing
tables, results
of estimation (sample means and
probabilityfunctionparameters)arereportedforeach
handlingequipmentandforeachactivity.Detailson
the pursued estimation methodologies and/or
comments on estimation and calibration results may
befoundindeLuca,2009).
Handling equipments involved were: mobile
harbour crane (MHC), gantry crane (GC), reach
stacker(RS).MHCsoperatingintheredseaContainer
Terminal are three Gottwald HMK 260 mounted on
rubber‐types and are mainly devoted to
loading/unloadingcontainersto/fromberthedvessels.
The results, reported in table 6, concern loading
activitiesfromshuttletovesselorfromdocktovessel,
and unloading activities from
vessel to dock. The
following container types were considered:
undifferentiatedcontainers, 20’,40’and20’x20’.Since
most red sea Container Terminal loading/unloading
activities concern full containers, the analysis is
mainly focused on full containers. Some results on
emptycontainersareproposedonlyforactivitiesthat
systematically involve empty containers. Statistical
analysis for undifferentiated containers shows that
the distribution function is always statistically
significant.Thesamerandomvariableseemstobethe
best approximation for loading and unloading
activities that involve 20’ and 40’ (full or empty)
containersmeansandstandard deviations related to
distribution are reported. operating in the Red ea
Container Terminal are four rubber‐tyred gantry
cranes used both for ovement/storage of containers
and for loading of shuttles/trucks. This crane type
usually consists of three separate movements for
container transportation. The first movement is
performedby thehoist,which raisesand lowersthe
container.Thesecondisthetrolley
gear,whichallows
thehoisttobepositioneddirectlyabovethecontainer
forplacement.
The third is the gantry, which allows the entire
cranetobemovedalongtheworkingarea.
The analyses carried out concern loading and
unloading to the shuttle/truck, and loading and
unloading to the stack (sometimes called
pile). Each
activitywasanalyzeddistinguishingundifferentiated
containers from 20’ and 40’ containers. Moreover,
loading time from stack is reported, further
distinguishingthetier.Theanalysisisfocusedonfull
containers,sincetheseactivitiesarethemostfrequent
in the red sea Container Terminal. Finally, averages
and standard deviations were
estimated for trolley
speed and crane speed. As regards undifferentiated
containers, the Gamma distribution function proved
the best solution for all analysed activities. Similar
resultswereachievedonanalysingactivitiesforeach
container type and each tier number. means and
standarddeviationsarereportedforeachactivity.
The RSs operating in
the red sea Container
Terminalareelevenandareequippedwithatwin‐lift
spreader able to move two full 20’ containers. They
are used both to transport containers in short
distances very quickly and to pile/storage them in
variousrows.
The analyses carried out concern: loading to
shuttle/truck, unloading
from shuttle/truck and
stacking. Each activity was analyzed distinguishing
undifferentiated containers from 20’ and 40’
containers. Moreover, stacking was analyzed
distinguishing the tier number. The analysis is
focusedonfullcontainers sinceinredseaContainer
Terminal the main activities are related to full
containers. For the stacking time, the time
duration
foreachtier,uptofive,wascomputed,butitwasnot
possible to distinguish containers typology. For the
mentionedactivitiesGammarandomvariablefitsthe
data better due to best values of the validation test
regardsRSsspeed,theauthorssuggesttoestimatethe
timedurationofthese
activitiesdirectly.
4 SIMULATION
Toplan investmentsfor a containerterminal several
project scenarios need to be compared through
performance indicator estimation. These indicators
couldbeglobal, ifreferringtothecontainerterminal
asawhole(aggregateindicators),orlocalifreferring
toasinglecontainer(disaggregateindicators).
Global indicators are
generally used to evaluate
the benefits of long‐term investments; while local
indicators are used to evaluate the benefits of
medium/short‐term investment and for real time
applications. To test the applicability of the model
architecture proposed for all the cited kinds of
application, the implemented model was validated
with
respecttoperformanceindicatorscoherentwith
thosemeasuredbytheterminalmonitoringofficeand
summarizedabove:
globalperformanceindicators
Terminal operation time: daily time required to
bringallterminalactivitiestoaclose;
localperformanceindicators
handlingequipmentindicators;
vesselloadingand/orunloadingtime;
quay/yardcraneidle
time;
shuttlewaitingtime;
shuttletransfertime;
reachstackerstackingtime;
reachstackeridletime;
gatein/outwaitingtime;
containerindicator;
containeroperationtime:timerequiredtomovea
container with handling equipment (e.g., time
spent moving a container from quay to vessel or
from shuttle to stack). Starting from the model
architectureproposedintheprevioussection, four
differentmodelsbasedonfourdifferenthandling
equipmentmodels,wereimplemented:
SampleMeanUndifferentiatedmodel.1
Sample mean values are used to estimate
handlingequipmenttimedurationandthereis
nodistinctionbetweencontainerstype.
SampleMeanContainerTypemodels.2
Sample mean values are used to estimate
handling equipment time duration and
337
containers type are explicitly taken into
account: 20’ full and/or empty; 40’ full and/or
empty;2x20’full.
RandomVariableUndifferentiatedmodel.3
Thetimeassociatedtoeachsingleactivityisthe
realization of a random variable, handling
equipments time duration is modeled as a
random variable and there
is no distinction
betweencontainerstype.
Random Variable Container Type models.
Handlingequipmenttimedurationismodeled
as a random varia ble and containers type are
explicitly taken into account: 20’ full and/or
empty;40’fulland/orempty;2x20’full.
The results interms of simulationtime point out
thatrandomvariablemodelsrequireacomputational
time much greater than sample mean ones. The
formerrequireabout20minutes,thelatterarebelow
one minute. Results in terms of global indicators
show an average absolute percentage error of more
than 10% for the handling model, whereas in using
thehandling
modelthepercentageestimationerroris
lower than 5%. Using the Container Type models,
resultsintermsofglobalindicatorsshowanaverage
absolutepercentageerrorofabout9%forthesample
mean model, whereas in using the random varia ble
modelthepercentageestimationerrorisabout3%.
Theuse
ofsamplemeanhandlingmodelsdoesnot
produceverygoodresultsintermsoflocalindicators;
averagepercentageestimationerrors exceed13% for
handlingequipmentindicatorsandareabout30%for
containerindicators.
Resultsobtainedusingrandom variablehandling
models are significant: average absolute percentage
errors for handling equipment indicators are
more
than6%withthehandlingmodel,andabout3%with
handlingmodels.Withrespecttocontainerindicators,
when only using the handling models the absolute
percentageestimation erroris acceptableinallother
casestheestimationerrorsareabout30%.
5 CONCLUSIONS
In literature numerous efforts may be found
in the
fieldofsimulationofacontainerterminal,mostofthe
existing papers are only focused on the application
and/oronthecomparisonofdesignscenariosanddo
not pay great attention on the model set‐up, its
calibration and its validation. If on the one hand,
many contributions do
not present any information
on equipment handling models used, the remaining
contributions carry out very simple approaches
(deterministic)and/orgivescantyinformation:onthe
estimation approach pursued, on experimental data
used, on parameters estimated and on parameters
value. Moreover, no one investigates the effects that
different hypotheses on handling equipment
models
calibration may have on the simulation of container
terminal performances. Such effects could not be
negligibleandshouldbeinvestigatedwithrespectto
different planning horizons, such as strategic or
tactical. In this paper a discrete event simulation
model was proposed and applied to the red sea
container terminal in
order to address some of the
openissuesintroducedabove.Theaimwastosuggest
to analysts, modellers and practitioners a sort of a
guidelines useful to point out the strengths or
weaknessesofdifferentapproaches.Guidelineswere
presentedthrough:
apreliminaryindepthliteraturesurvey;
the description of the developed discrete event
models, with particular attention to estimation
results of handling activity models for three
handling equipment (mobile harbour cranes,
gantry cranes, reach stackers) and for different
container type (undifferentiated, 20 feet, 40 feet,
empty,full….);
the
simulation of the effects of different
hypotheses regardingthe approach to estimate
handlingactivitiestimeduration(samplemeanvs
random varia ble estimation),the level of
aggregation of handling activities (e.g. vessel
loading vs explicit simulation of elementary
activities sequence),the segmentation of
containertype.
REFERENCES
[1]Cartenì, A., Cantarella, G.E., de Luca, S. (2005) A
simulationmodel foracontainerterminal,Proceedings
of the European Transport Annual Meeting:
TransportationPlanningMethods,PTRC,London.
[2]Cartenì,A.,deLuca,S.(2009)SimulationoftheRedea
Container Terminal, technical paper, Dept. Of Civil
Engineering,UniversityofRed
ea.
[3]de Luca, S., Cantarella, G.E., Cartenì, A. (2008) A
macroscopic model of a container terminal based on
diachronic networks, Schedule‐Based Modeling of
Transportation Networks: Theory and Applications
(Nuzzolo,A.,Wilson,N.H.M.,eds),Springer.
[4]ElSheikh,A. R.;Paul,R.J., Harding,A.S., Balmer,D.W.
(1987) A Microcomputer‐Based Simulation Study of a
Port,TheJournaloftheOperationalResearchSociety,38
(8)673‐681.
[5]Tuǧcu, S. (1983) A Simulation Study on the
Determinationofthe Best InvestmentPlanfor Istanbul
SeaportTheJournaloftheOperationalResearchSociety,
34(6)479‐487.
[6]
Park, C. S., Noh, Y. D. (1987) A port simulation model
forbulkcargo
operations,Simulation,48(6)236‐246.
[7]Silberholz,M.B.,Golden,B.L.,Baker,E.K.(1991)Using
simulation to study the impact of work rules on
productivity at marine container terminals, Computers
andOperationsResearch,18(5), 433–452.
[8]Collier,P.I.(1980)Simulationasanaid
tothestudyofa
portasasystem,ShipOperationAutomationIII,51‐6.
[9]Gambardella, L. M., Rizzoli, A. E., Zaffalon M. (1998),
Simulation and planning of an intermodal container
terminal,Simulation,71(2)107‐116.
[10]Nevins,M.R.,Macal, C. M., Love,R. J., Bragen,M. J.
(1998)
Simulation, animation, and visualization of
seaportoperations,Simulation,71(2)96‐106.
