J. Marine Sci. Appl.

(2013) 12: 434-444

DOI: 10.1007/s11804-013-1214-x

Practical Solutions for Reducing Container Ships’

Waiting Times at Ports Using Simulation Model
Abdorreza Sheikholeslami, Gholamreza Ilati* and Yones Eftekhari Yeganeh
Transportation Planning Department, Iran University of Science and Technology, Tehran 16846-13114, Iran

Abstract: The main challenge for container ports is the planning traditional port management organizations for more efficient
required for berthing container ships while docked in port. Growth operations. To attract waterborne commerce and decrease
of containerization is creating problems for ports and container shipping costs, ports and operators are now planning for the
terminals as they reach their capacity limits of various resources
next generation of vessels with increased draft even though
which increasingly leads to traffic and port congestion. Good
planning and management of container terminal operations reduces
these ports are not equipped to accommodate such vessels
waiting time for liner ships. Reducing the waiting time improves through channels leading to them.
the terminal’s productivity and decreases the port difficulties. Two The harbor design often requires a formal and appropriate
important keys to reducing waiting time with berth allocation are assessment of the port, terminals and berths. The design of the
determining suitable access channel depths and increasing the mooring systems for specific berths and ships provides the
number of berths which in this paper are studied and analyzed as data on the loads on the mooring equipment and on the
practical solutions. Simulation based analysis is the only way to motions of the moored ship. This information allows for
understand how various resources interact with each other and how simulating port operations to determine bottle necks for berth
they are affected in the berthing time of ships. We used the
allocation and to reduce the waiting time of ships in port. The
Enterprise Dynamics software to produce simulation models due to
the complexity and nature of the problems. We further present case access channels, turning basins and berthing places have to be
study for berth allocation simulation of the biggest container dimensioned and aligned, the ships to be serviced have to be
terminal in Iran and the optimum access channel depth and the considered, as well as the environmental conditions, the
number of berths are obtained from simulation results. The results available tugs and the aids to navigation, in order to minimize
show a significant reduction in the waiting time for container ships the waiting times.
and can be useful for major functions in operations and Physical limitations such as channel depth, storage yard
development of container ship terminals. space, berthing facilities, and landside productivity determine
Keywords: container ships; waiting time; access channel depth; how much throughput a port can potentially handle in a given
quay length; simulation model; enterprise dynamics; berth
year. The proper planning and management of port operations
allocation
in view of the ever growing demands in global trade
Article ID: 1671-9433(2013)04-0434-11
represents a big challenge because of restrictions such as the
length of the quay and depth of access channels which causes
1 Introduction1 increased difficulties for berthing operations planning and the
Marine transportation is one of the most efficient, loading and unloading of ships.
effective, safe and environmentally sound ways to transport Shippers and carriers are using larger ships in global trade
people and goods. Ports are at the center of global, to gain transportation efficiencies and cost savings, which
intermodal freight systems that connect deep-sea and have enormous importance in this very competitive market.
short-sea shipping routes to remote interior regions through While large ocean-going-vessels, such as those used for
a system of inland waterways, roads, railways and pipelines. waterway barge transport, are substantially less efficient and
As such, ports are an important cornerstone of the cost efective EDRG (2012).
macro-infrastructure in the countries and regions where they An important capacity consideration is the size of the vessel
operate. In the last decade, global shipping has experienced a port can accommodate. Along with other factors, channel
explosive growth. Increasing world trade and globalization, width and depth establish the maximum size vessel that can
developing markets in Asia and the relatively low cost of be call into a port. Access channels are the initial points of
transportation have all contributed to this growth, resulting entry to ports and harbors. The absence of a verified design
in high demand for more terminals, distribution centers and methodology often leads to conservative estimates on channel
production plants. Globalization also has led to increased size, which gives rise to increased initial/maintenance
competition among ports and increased requests from the dredging costs. The depths of most navigation channels
gradually decrease over time due to sedimentation, since these
Received date: 2013-07-07. channels behave as a sediment trap. Maintenance costs are a
Accepted date: 2013-08-28. critical element in the economic feasibility of a port,
*Corresponding author Email: ilati@iust.ac.ir
particularly when a relatively long access channel requires
© Harbin Engineering University and Springer-VerlagBerlin Heidelberg 2013
Journal of Marine Science and Application (2013) 12: 434-444 435

frequent dredging. The present study examines the influences increased demand for the service at container terminals that
of dredging access channels in Iran’s RAJAEE port in now have to serve several vessels per day, load and unload
reducing container ships’ waiting times by using a simulation thousands of containers per day, and have to do so in a timely
model. Simulation can guide terminal managers with manner in order to reduce the time liner ships have to spend at
evaluating all the terminal key resources (cranes, barges, tugs, the terminal and thus gaining a competitive advantage over its
quays, etc.) to understand their interactions with vessel delays neighboring ports in the region. This competitive advantage
and to assess and mitigate the risks arising from them. A list helps the port increase its customers and thus its profits (Ali et
of challenges that a simulation model can help to tackle al., 2011) Port capacity depends upon channel depths, channel
include (Dahal et al., 2007). widths, turning basin size, sufficient bridge heights, and port
• Minimizing infrastructure investment while maintaining support structures with sufficient dock and crane capacity to
a high service level load and unload goods. USACE provides detailed guidance
• Evaluating the peak utilization for different equipment in for the design of inland waterways, the guidance for coastal
the terminal access channels is not comprehensive. The USACE guidance
• Determining the number of cranes defines authorized depths and widths for navigable channels
• Evaluating if the envisioned or current terminal layout is (Demirbilek and Sargent, 1999) .Berth Allocation refers to the
capable of coping with future market demands problem of allocating vessels to berths while minimizing the
• Assessing quay extension investment total service times of vessels. In recent years, simulation has
• Assessing the impact of vessel schedules on the service become an important tool for improving operation and
level performance of ports. According to research by Saanen (2000),
Merkuryev et al.(1998) it can be concluded that the
• Understanding the impact characteristics on the terminal
performances simulation results provide valuable information to support the
decisions made by programmers,operators and terminal
• Assessing the risk of a higher variability of vessel arrivals
managers. Beneficial applications of simulation in support of
• Implementing new strategies to optimize crane allocation
complex management of container ports have been
• Assessing the impact of a new fleet composition
demonstrated by Bruzzone et al.(1999). Highlights of this
• Understanding the correlation between waiting times, study have implications for several applied examples.
quay occupancy and vessel arrival patterns Experimental results have indicated the advantages of the
• Estimating if and how the hatch composition affects the simulation approach in terms of reusability, flexibility, time of
terminal performances modeling and estimation of the operations. Legato et al. (2000)
Many different parameters enter into the planning, design modeled the line of logistics activities related to import, ship
and operation of deep-draft navigation channels. For berthing and departure processes in a container port. They
example, in the planning of a navigation channel, a design could use the process simulation to provide a model of the
ship, typically the maximum size ship from the projected queuing network in a port. The good results and validation of
user fleet is selected on the basis of economic analyses. The output from the model in this study show that the simulation
two main design dimensions of navigation channels are approach can be a viable solution to this problem. Kim and
width and depth, and these must be determined to Moon (2002) presented a mixed-integer-linear-programming
accommodate the design vessel. Likewise, for safe (MIP) model which was formulated for the berth-scheduling
operations within a channel, it is necessary to consider the problem. The simulated annealing algorithm was applied to
effects of winds, waves, tides, currents, visibility, and the berth-scheduling problem to find near-optimal solutions.
navigational aids. The density and type of traffic (one-or Experimental results showed that the simulated annealing
two-way traffic), ship speed, turning basins, and tug algorithm obtains solutions that are similar to the optimal
assistance are other factors that need to be considered with solutions found by the MIP model. A numerical experiment
the operations of channels (Demirbilek and Sargent, 1999). showed that the computational time and quality of solutions
Today with global trade, the density areas of the port and depend on the number of vessels and the ratio of the
container ports, restrictions such as the length of the quay, the overlapped area of rectangles when they are positioned at
depth of the access channel, and the number of tugs and tide their least cost locations to the total area of the rectangles. The
conditions, have made double the difficulties of berthing numbers of optimal berth and quay cranes were studied by
operations planning and the loading and unloading of ships. Nam et al. (2002) at a port in Busan (Korea).
Thus, due to the complexity and nature of the problems, using Several operational models in four different scenarios
the enterprise dynamics simulation approach as a flexible and were presented and their operations were evaluated through
efficient solution is proposed. In this study the objects of the simulation experiments. The results showed that sharing the
simulation model can be explained as follows: quay cranes with the next berths could be effective in
1) Simulated berth allocation process in Rajaee port. improving efficiency. Yun and Choi, (2003) proposed an
2) Calibration and Validation of the simulation model. Object-Oriented Simulation Model for analyzing terminal
3) Reduced ship waiting time by making a change in the containers (including: gateway, container yard, berth and
access channel depth. equipment, such as transfer cranes, gate cranes, trailers and
The growth in container transportation has led to an
436 Abdorreza Sheikholeslami, et al. Practical Solutions for Reducing Container Ships’ Waiting Times at Ports Using Simulation Model

yard tractors). The system's model uses a set of goals in the to achieve the optimized allocation plan based on dynamic
programming language of the Object-Oriented Simulation decision making. Bierwirth and Meisel (2010) presented a
Model, so that the comparative model can control and thorough review of the previous attempts in solving the
manipulate it. Outcomes of statistical analysis of resources berth allocation and quay crane assignment problems.
may be used to assess the capacity and operative efficiency Particular focus in their article is put on integrated solution
of a terminal. Hartmann (2004) has provided the approach approaches which become increasingly important for the
for generating realistic data scenarios in port container terminal managemet. He et al. (2012) imply that the berth
terminals, as inputs to the simulation model and test allocation is an important part of port operations and is also
optimization algorithms. In this study, a scenario includes the key link of logistics. The authors established a 0-1
data about “reaching the ships, trains and tractors during the programming model to minimize both the ship's retention
time” and some information about “the containers which are time in port and the distance between the berth and target
picked up or delivered" and users are able to control various storage yard.
parameters of a kind. Meng et al. (2009) stated that the berth In this paper, incerising of the access channel depth and
allocation is the most important issue for reducing the quay length were studied as two practical solutions to reduce
ship's queuing time. When the port is heavily congested with container ships’ waiting times. Simulation methods and
different types of vessels, effective berth allocation Enterprise Dynamics (ED) software were used for the
techniques could optimize the berth utilization and reduce simulation process. A view of the simulation model is shown
the ship's queuing time. They propose a simulation method in Fig. 1.

Anchorage

Quay

Approach channel

Fig. 1 A view of the simulation model

2 Simulation model failures and weather conditions for navigation and

departures. Other constraints such as access channels, tugs
The simulation model contains the terminal and port entry windows can also be part of the simulation
characteristics including the number and size of the quays model.
and the available loading equipment. It was also taken into Generally, simulation of a system where its variable
account seasonality in demand, and variability in demand. modes occurs only at a point of time discretization
Furthermore, the model takes the infrastructure elements “momentous spot,” is called discrete event simulation. In
such as: loading equipment, conveyors, tugs, and barges fact, in this type of situation, a system in discrete moments
including failure patterns into account. Typical processes of time will be updated. Therefore in this research discrete
that can be part of the model include: vessel arrival (patterns, event simulation software ED is used to be employed in
arrival windows, and queuing rules), quay allocation, and designing random processes, discrete events and dynamic
equipment allocation, berth processes (mooring, paperwork, designs. Applications of this simulation software are used in
loading, offloading etc.), departures, maintenance and the process of model production, transportation (rail, airport
Journal of Marine Science and Application (2013) 12: 434-444 437

and port industries), logistics and warehousing. In this take part in nonphysical characterizations, such as diagrams,
research, the ED logistics package has been put in use to tables or context of models. In the present research 4D
model production. In short, capabilities of this package Script software programming language, a number of key
includes, potential for operation model production of those commands are used for the production of the model (Brito,
like machineries, operators and robots, equipment to transfer 2010; Cakaj, 2010). The conceptual model of the simulation
materials such as cranes, tanks and also in warehousing model shows all processes that inputs do from arrival time
(Banks, 2011; Halverson, 2009). ED software founded on until departure time. The Conceptual Model or Process
the basis of the atoms concept as components of particles of Graph of Simulation Model indicates all processes of a
model production. An atom not only may represent a simulation model quality as from entering up to leaving the
machine, service or a product, but also have the capability to system. (Fig. 2)

Fig.2 Conceptual model

3 Defined features in the simulation model that is in charge of handling the highest volume of container
operations plays a very important and vital role in Iran's
The container sector in Iran is a fast-growing market. The economy and Trade. Terminal containers at RAJAEE port
main container port is located at Bandar Abbas (Iran’s provide services for all linear and feeder ships and 21 linear
RAJAEEport), but growth of the container sector will be shipping lines which are in traffic in this port.
limited by physical constraints in the near future. (Such as Iran’s RAJAEE port is a multipurpose port and it has two
the length of the quay, the depth of the access channel and container terminals. A general view of the quays in this port
the number of tugs and sea tide conditions). is presented in Fig.3. One of the advantages of terminal 2
The distance of Iran’s RAJAEE port to Tehran is 1501 Km, over terminal 1 is the difference between drafts. The drafts
30 Km to the province’s Capital and 40 Km to the nearest in terminal 1 are between 8 and 12 meters but terminal 2 has
airport. The connection of this port to Tehran and other parts drafts between 14 and 16 meters. So it is possible for the
of the country is possible via roads and railways. berthing of the large ships in terminal 2. The quay position
Iran’s RAJAEE port connects to more than 80 ports and capacity in the two terminals are not comparable to each
worldwide and has the highest rate of cargo transit in the other.
country. A large volume of cargo being shipped towards
Central Asia passes through this port.
Iran’s RAJAEE port, as the biggest container port of Iran
438 Abdorreza Sheikholeslami, et al. Practical Solutions for Reducing Container Ships’ Waiting Times at Ports Using Simulation Model

Table 2 Classification of ship types and their characteristics

Type of Average Percentage of
Average
ship/ length ships (with this
draft /m
Percentage /m length & draft)
Termi Termi 173 8.6 35.26 %
nal I nal II Liner 222 12.8 38.28 %
65.80 279 14.5 21.37 %
316 14.5 5.09 %
77 5.1 13.42 %
Feeder 136 6.7 37.34 %
34.20 165 8.2 33.60 %
216 12.5 15.65 %

In continuation by definition of scenarios that are along

with some changes in entrance information, the results
model exit simulations for model calibration will be
considered. For the implementation of the simulation model,
there must be a warming up and observation time for the
model to be made clear. With regards to estimating the
Fig.3 A view of Iran’s Rajaee port arrival time between two ships (about 8 hours) and also the
ship's time in port, the time for warming up and observation
The main objective was to find an efficient solution for time are considered respectively to be 1 week and 3 months.
reducing container ships’ waiting times in Iran’s RAJAEE Also 50 independent observations before the execution are
port. The beginning steps of the simulation model is for recommended.
identifying assumptions, boundaries and determining the Scenario of berthing ships: The purpose of this scenario is
key index system. General assumptions of the simulation to study and evaluate the impacts of the berthing methods
models can be divided into two categories of structural of ships on the indexes such as waiting time while at
assumptions and assumptions about the data. Structural anchorage and operation time. In short, the methods are
assumptions used are about the performance of the system shown in table 4.
and the system boundaries. The simulation model in this Scenario to increase entrance rate of ships anchorage:
study is done based on the modeling of discrete berths and it Average time between arrivals in 2009 to 2011 as compared
is assumed that there are a certain number of berths at each to the average arrival time in 2006 to 2008, have decreased
terminal for ships. (Table 1). Assumptions regarding the data about 25%. So for the scenario of increasing the arrival rate
include the consideration of two types of ships (liner and of ships to anchorage, the 25% average decrease in arrival
feeder) by the specifications of the draft and volume of the time between two ships entering is utilized.?
container (Table 2). Instituting specifications of the The scenario for changing the percentage of entering ship
simulation model including behavioral features of each one types to anchorage: In this scenario, the percentage of the
are outlined in Table 3. change of the impact of various types of ships which enter
the anchorage will be studied. For this purpose, percentage
Table 1 Length and depth of different berths of terminals of entering liner ship types are considered equal to 0%, 50%
and 100% respectively.
Number of Number Berth Initial depth of Queuing discipline policies for ships at anchorage: The
terminal of berth length/m berth draft/m ships which stay at the anchorage areas are to wait in queue
1 4 340 12
and are generally arranged by different regular rules. In a
1 5 300 12.5
1 6 270 12.5 preliminary state the ships in queue are generally arranged
1 7 250 11.7 by entering time to the anchorage area.
2 25 370 16 Simulated models which have been calibrated for 2006 to
2 26 370 16 2008 data are used to predict port situations between 2009 to
2 27 370 16 2011. With different tests to determine the important
parameters of model simulations, preliminary depth of
4 Calibration of the simulation model access channel was found to equal 11 meters and estimated
loading speed of the crane to equal 17.5 containers per hour.
Calibration tests will eliminate doubt about the model
According to the results achieved, the berthing scenario
results from the simulations and that they can improve the
based on strictly random estimates, calculated the average
situation that has been used in the real system. Comparing
waiting time for ships at anchorage better than the other
overall process model and conduct, with the real system is
methods. Also the simulation model result matches up well
generally the purpose of determining the validity of
with the real data (Fig. 4 and Fig. 5).
models.(Perros, 2009).
Journal of Marine Science and Application (2013) 12: 434-444 439

Table 3 Defined subjects and particulars in simulation model

Analysis Quantities total or calculation method Features List Subject

Quantity 1 means liner ship and quantity 2 means
Bernoulli(65.45,1,2) Type
feeder ship.

Ship’s length is obtained on the basis of ship’s type of Liner={173,222,279,316}

Length (Meter)
discrete distribution. Feeder={77,136,165,216}

Number 1 refers to ships which are intended to Movment Course

proceed from anchorage toward the port and number 0 {0,1} (Towards Quay or
for ships which are intended to leave the quay. Departing Quay)
N represents the total number of ships that have been Number of Inward
{1,2,…,N}
produced in simulation. Ships
The color green for a ship indicates suitable
conditions to enter the access channel and the color
red indicates a warning that at the moment, at least
{G, R} Color
one of the necessary accommodating conditions for
the entry of ships to the access channel is not
available.
Full details of classification deep draft ships are Calculated on separating by type and
Draft
summarized in Table 2. length of the ship.
Number of
Is determined by length and GRT of
- Loaded/Unloaded
the ship.
Containers Per Box
Prepration Time
Is assumed not to exceed 30 minutes (1800 seconds) Min{1800,Weibull(3316.23,2.10)} (from Berthing up to
Operation Onset)
While at the beginning moments of loading and Ship
unloading operations there is a sufficient figure of
crane coefficient, loading and unloading time length is
calculated on the basis of the following relationship:
Based on the number of
(Total Operation (Box))/(QCNorm*25)
loaded/unloaded containers,
In this regard QCNormenacts a coefficient of the
loading/unloading speed of each Containers Loading
number of cranes needed for the ship. In cases where
crane, and also the number of cranes and Unloading Time
the number of the cranes are not available, loading
assigned for each ship.
and operations with the number of the existing cranes
has been started and when extra cranes for loading
and unloading are added, the remaining time to
complete loading and unloading operations, is
calculated and will be used in the simulation model.

Is assumed not to exceed 1 hour (3600 seconds) Min{3600,Pearsont5(8771.35,3.30)} Unberthing Time

The number 1 indicates that at the moment, the ship

has all necessary conditions including suitable water
Ship’s Situation to
depth, the availability of channel and suitable location
{0,1} Enter Access
for berthing and by all means is ready to move
Channel
forward. Otherwise, this figure of time equates to
zero.
NT indicates number of terminals (in here NT=2) and
Number of Allocated
the assigned berth is elected according to a berthing {1,2,…,NT}
Berths
method as soon as a ship enters the access channel.
With the assumption of discrete quay length, the
Berthing Location in
entering ship chooses one of the suitable places for {1,2,….,NBi}
Entering Terminal
berthing.
440 Abdorreza Sheikholeslami, et al. Practical Solutions for Reducing Container Ships’ Waiting Times at Ports Using Simulation Model

This time, at the moment the ship

leaves the anchorage is calculated as
Waiting Time at
- the time difference between the ship
Anchorage
leaving the anchorage and the time it
reached the anchorage.
This time when the ship departs from
anchorage is calculated as the time
- Waiting Time in Port
difference between the ship leaving
the terminal and arriving at the port.

The deep draft access channel is upgraded over time Initial depth of draft of access
Draft Depth at Any
based on its initial quantity and changes in sea level channel is considered equal to
Time
due to the tides. 11meters.
Access
The number 1 indicates that a ship in the access Channel
Movment Course of
channel from anchorage is toward the port. Toward
{0,1} Ships in Access
the port on the move. The number 0 indicates that the
Channel
ship is in the access channel and departing the port.
“I” indicator is used for terminals I=1,2 Number of Terminal
This figure is fixed during simulation. Li Terminal Length
While a ship is moving from anchorage to port,
occupied length of the quay that the ship is supposed Occupied Terminal
[0,Li]
to enter will be updated in accordance with the ship’s Length at Any Time
Quay
length.
Number of berths for first and second terminals will Number of Berths at
NBi
be respectively considered to be 3 and 4. Terminal
Depth of any berth, in time, will be Draft Depth of Each
- updated on basis of initial and tidal Berth in Any
effects. Terminal
In the initial state, there are 6 standby
- Number(Quantity)
tugs in the port for shipment.
Time duration to pass through access channel is
calculated by continuous uniform distribution function
Time Length to Pass
with least time limit of 30 minutes and over 60
Uniform[30,60] Access Channel
minutes. Also time for a free tug move without towing
(Minutes)
any ship in access channel considered roughly 20% of
Tug
the time when towing a ship in the access channel.
The number 0 means that a tug is engaged with
serving floating object. Number 1 is equivalent to tug
being free at anchorage. Number 2 indicates that the Availability at Any
tug is idle and stays along the quay. The assumption is {0,1,2} Moment Including
that tugs after completed towage of floating objects, (Location Position)
will stay at their present position (beginning or the
end of access channels).

Average
Time Si Simulation model
Spent
in R Real data
Anchorage
(min)

Sictly random Most appropriate berth length Most appropriate berth draft

Fig.4 Comparison of time at anchorage area in simulation model and real data in different berthing scenarios (2006-2008)
Journal of Marine Science and Application (2013) 12: 434-444 441

Average
Time Si Simulation model
Spent
in R Real data
Port
(min)

Strictly random Most appropriate berth length Most appropriate berth draft

Fig.5 Comparison of average time spent in port in simulation model and real data in different berthing scenarios (2006-2008)

Table 4 Scenarios for selection of ship berthing allocation

Berth Selection Method Analysis

In this case, out of preferred berth location for each ship in terminal, one is selected by random.
Strictly Random
When only one suitable location is available, the ship will be berthed at that location.

In this case, the ship will be selected for the berth in accordance with parity of its length. In other
Most Appropriate Length words, a berth which has the least length difference with the ship, minimum length of the berth
will be not useable.
Most Appropriate Berth In this case, the ship will be selected for the berth which has the slightest difference in draft depth.
Draft In another words, a berth which has the least draft difference with the ship will be selected.

Average
Time
Spent
In Simulation model
Anchorage
(min) Real data

Strictly random Most appropriate berth length Most appropriate berth draft

Fig.6 Comparison of time at anchorage area in simulation model and real data in different berthing scenarios (2009-2011)

Average
Time Simulation model
Spent
In Real data
Port
(min)

Strictly random Most appropriate berth length Most appropriate berth draft

Fig.7 Comparison of average time spent in port in simulation model and real data in different berthing scenarios (2009-2011)
442 Abdorreza Sheikholeslami, et al. Practical Solutions for Reducing Container Ships’ Waiting Times at Ports Using Simulation Model

5 Validation of simulation model allocation process. Any vessel must have appropriate draft
with the access channel. Otherwise the ship will have to wait
In this section, results of model simulations for the for the tide to increase the depth of the access channel.
current status of the port in various scenarios for berthing of The accessibility of a port depends upon several factors,
ships are presented. For comparison simulation model such as the depth of its access channel, which also
results given by the real data and with consideration to the determines the depth of the docks or basins. In this case it is
fact that arrangement of anchorage in reality is a First assumed that one or a few meters of the access channel is
In-First Out (FIFO) scheme, therefore in all scenarios added to the minimum depth. This process further increases
arrangement of queues are considered in the FIFO scheme. the extent of its efficiency, and will continue. Such a
Also the arrival ships rates are in accordance with the real strategy can reduce "average time spent in port" in regards
data between 2009 and 2011 following a negative to two areas, anchorage (applied to the vessels ready to
distribution view with 389.17 minutes. Also percentage of arrive) and at berth (vessels ready for departure). More
the liner type is the primary (65.45%). According to the specifically, the amount of waiting time at the anchorage
results achieved, simulation model has managed, time area after evacuation of the berth which occurs with every
waiting for a ship in anchorage and also the time of the ship vessel, because of needed allocation for the next vessel, such
in port was predicted with proper care. (Fig.6 and Fig.7) conditions is due to the lack of a sufficient depth access
canal with regard to the tidal effects, this is a considerable
6 The impacts of increasing the depth of decrease. Also, some of the ships leave the berth after their
access channels through dredging loading has been completed, so no problems for smooth
departure from the port occurs. Perhaps the amount of
Port locations and sites are preliminarily constrained and
imposed waiting time for the next ships coming into the
quality of maritime access related to the depth of the
anchorage area also (for quicker evacuation of quays)
waterway system and port access channels. As is clear, time
decreases. Tables 5, 6, 7 and 8 describe the effects of
optimization of berth allocation requires the identification of
increasing the depth of the access channel (Depth of
main bottlenecks for the process of providing service to
available access channels equals 11m). In this paper, the
container ships. One of these bottlenecks is the depth of
strictly random method for the berthing scenario is used.
access channels that acts as constraints on the berth

Table 5 Effects of increasing the access channel depth in the rate of 11.5 meters
Standard Low limit High limit
Index Average min max
deviation /95% /95%
Average time in anchorage (minutes) 223.4 92.11 197.85 248.94 104.89 580.59
Average handling time (minutes) 1562.95 49.4 1549.25 1576.65 1470.91 1715.83
Average berthing time (minutes) 1774.63 51.74 1760.28 1788.98 1670.96 1932.36
Average time spent in port(minutes) 2087.98 114.42 2056.25 2119.71 1882.68 2479.58
Terminal 1 utilization rate (%) 50.65 1.77 50.16 51.14 47.33 54.92
Terminal 2 utilization rate (%) 31.64 3.4 30.69 32.58 23.31 39.09

Table 6 Effects of increasing the access channel depth in the rate of 12 meters
Standard Low limit High limit
Index Average min max
deviation /95% /95%
Average time in anchorage (minutes) 180.82 102.84 152.3 209.34 71.59 675.43
Average handling time (minutes) 1557 46.31 1544.15 1569.84 1440.12 1641.71
Average berthing time (minutes) 1766.54 50.27 1752.6 1780.48 1631.89 1852.32
Average time spent in port(minutes) 2037.51 117.83 2004.83 2070.19 1799.9 2497.63
Terminal 1 utilization rate (%) 50.6 1.96 50.05 51.14 44.2 54.32
Terminal 2 utilization rate (%) 31.28 3.63 30.27 32.29 24.81 37.69

Table 7 Effects of increasing the access channel depth in the rate of 13 meters
Standard Low limit High limit
Index Average min max
deviation (95%) (95%)
Average time in anchorage (minutes) 167.36 81.26 144.83 189.9 42.93 436.59
Average handling time (minutes) 1566.23 42.64 1554.4 1578.05 1455.4 1663.97
Average berthing time (minutes) 1771.77 43.71 1759.64 1783.89 1652.98 1862.57
Average time spent in port(minutes) 2029.28 96.87 2002.41 2056.15 1857.23 2302.59
Terminal 1 utilization rate (%) 48.4 2.34 47.75 49.05 43.08 52.83
Terminal 2 utilization rate (%) 33.41 3.33 32.48 34.33 25.5 39.56
Table 8 Effects of increasing the access channel depth in the rate of 14 meters
 Journal of Marine Science and Application (2013) 12: 434-444 443

Standard Low limit High limit

Index Average min max
deviation (95%) (95%)
Average time in anchorage (minutes) 148.9 72.61 128.76 169.04 31.99 423.18
Average handling time (minutes) 1560.46 56.09 1544.91 1576.02 1424.69 1679.28
Average berthing time (minutes) 1767.32 55.23 1752 1782.63 1639.38 1883.59
Average time spent in port(minutes) 2006.69 80.06 1984.48 2028.89 1854.65 2236.93
Terminal 1 utilization rate (%) 48.04 2.13 47.45 48.63 44.36 53.65
Terminal 2 utilization rate (%) 34.03 3.22 33.14 34.93 28.24 43.8

conditions, it will proceed to queue for entering the access

channel. These necessary conditions include, appropriate
Average depth of the access channel and existence of at least one
Time
Spent proper berth at the quays. At any moment, if the vessel lacks
in the above mentioned conditions, then it will be directed to
Anchorage
/min the initial (waiting) queue. Therefore the berth allocation of
simulation models can help operators to identify the precise
11 11.5 12 13 14 restrictions of berthing ships. In order to make better
Access channel depth decisions, the simulation model typically produces results
Fig.8 Effects of increasing the access channel depth in such as, quay utilization, crane utilization, turn-around
reducing average waiting time spent in anchorage times, waiting times per vessel type (waiting for channel,
delays because of weather, waiting for tugs, etc.), tug
utilization, landside infrastructure utilization and the usage
of the access channel. In this paper two practical solutions
Average
Time are analyzed to reduce waiting time in berth allocation.
Spent 1) Determine the access channel depth and 2) Increase the
in
Anchorage length of the quay. Results have shown that increasing
/min access channel depth to 11.5 meters and increasing the
number of berths to 10 will have the affect of reducing
7 8 9 10 11 container ships’waiting times significantly (Fig.8 and Fig.9).
Number of berths With increasing the access channel depth to 11.5 meters,
Fig.9 Effects of increasing the number of berths in reducing waiting time will be reduced by approximately 80% as
average waiting time spent in anchorage compared to the access channel depth at 11 meters. It should
be noted that the average operation time could be increased
7 Effects of increasing the number of berths when the number of berths are increased but the number of
quay (increasing quay length) cranes are fixed. Because more ships simultaneously can
berth with the number of available cranes, loading and
The assumption is that one or several berths which are unloading operations will be slowed down.
exactly the same or similar should be added to the existing
quay in the container terminal, to the sources added. Such
an approach can reduce "Average Time Spent in Port" in the
8 References
anchorage area during peak periods (high utilization). All of Ali I, AbouelseoudY, HamdyElwanyM (2011).Container terminal
the new berths will be constructed in terminal 2 with a depth berth allocation and quay crane assignment using IP and
of 370 meters and 16 meters draft. According to Fig. 9, the simulated annealing. Proceedings of the 41st International
waiting time at anchorage will decrease by increasing the Conference on Computers & Industrial Engineering, USA, Los
Angeles, 31-37.
number of berths.
Banks J, CarsonJS (2010). Discrete event system simulation.5th ed.
Pearson Education, Inc., Upper Saddle River, NJ, 14-21.
8 Conclusion Bierwirth Ch, Meisel F (2010). A survey of berth allocation and
quay crane scheduling problems in container terminals.
As the world’s container fleets get upgraded with larger European Journal of Operational Research, 202(3), 615–627.
ships, major ports are facing the challenge of Bruzzone AG, Giribone P, Revetria R (1999). Operative
accommodating deeper vessel drafts. Nowadays proper requirements and advances forth new generation simulators in
planning for standing container ships in port position and multimodal container terminals. Proceedings of the 31st
the necessity for rapid and orderly loading and unloading Conference on Winter Simulation, 2, 1243–1252. ACM Press,
them are among the main challenges concerning container New York.
ports. Any vessel arriving at anchorage will stay in primary Cakaj Sh (2010). Modeling, simulation and optimization: focus on
applications. Publisher: In Tech, India.
queue where the vessel’s situation will be determined
constantly and when the vessel possesses the necessary
444 Abdorreza Sheikholeslami, et al. Practical Solutions for Reducing Container Ships’ Waiting Times at Ports Using Simulation Model

Dahal K, Galloway S, Hopkins I (2007). Modeling, simulation and operational optimization. Technical Report, TBA
optimization of port system management. International Journal Nederland/Delft University of Technology.
of Agile Systems and Management, 2(1), 92-108. Yun WY, Choi YS (2003). Simulator for port container terminal
Demirbilek Z, Sargent F (1999). Deep-draft coastal navigation using an object oriented approach. Working paper, Pusan
entrance channel practice. Coastal Engineering Technical Note National University.
I-63, U.S. Army Corps of Engineers (USACE).
Economic Development Research Group (2012). Panama canal
Author biographies
expansion study phase 1 Report: Developments in trade and
national and global economies. The United States Department Abdorreza Sheikholeslami was born in 1967. He
of Transportation, Maritime Administration, Washington, DC. received his BSc degree and MSc degree from Iran
University of Science and Technology. Now he is
Halverson R (2009). Simulation with enterprise dynamics. an Associate Professor at the School of Civil
Midwestern State University, Wichita Falls, TX, Journal of Engineering, Iran University of Science and
Computing Sciences in Colleges Archive, 21(4), 246-252 Technology. He received his Ph.D degree in
He L, Liu H, Lv H, Zhao J (2012). Optimization study of berth Transportation Engineering and Planning from Iran
University of Science and Technology in 2006. His
allocation based on genetic algorithm. ICLEM 2012: Logistics current research interests include transportation
for Sustained Economic Development—Technology and problems.
Management for Efficiency, 67-73.
Hartmann S(2004). A general framework for scheduling equipment
and manpower on container terminals. OR Spectrum, 26, Gholamreza Ilati was born in 1980. He is a PhD
51–74. candidate at the School of Civil Engineering, Iran
Kim KH, Moon KCh (2002). Berth scheduling by simulated University of Science and Technology. His current
research interests include main challenges in
annealing. Transportation Research Part B, 37, 541–560. maritime transportation specifically berth allocation
Legato P. Mazza R (2000). Berth planning and resources problems, bunkering, etc.
optimization at a container terminal via discrete event
simulation. University della Calabria, 537-547.
Meng N, Chen Y, He L (2009). Dynamic decision-making for port
berth allocation. Second International Conference of Intelligent
Computation Technology and Automation, 4, 188-191.
Merkuryev Y, Tolujew J, Blümel E, Novitsky L, Ginters E, Yones Eftekhari Yeganeh was born in 1986. He
Viktorova E, Merkuryeva G, Pronins J (1998). A modeling and received his M.S.degree in Transportation
simulation methodology for managing the Riga harbor Engineering and Planning from Iran University of
container terminal. Simulation, 71(2), 84-95. Science and Technology in 2011. His current
Nam KC, Kwak KS, Yu MS (2002). Simulation study of container research interests include dry port, port choice and
terminal performance. Journal of Waterway. Port, Coastal and marine transportation.
Ocean Engineering, 128(3), 126–132.
Perros, H (2009). Computer Simulation Techniques-The Definitive
Introduction. Computer Science Department, NC State
University, Raleigh.
Saanen YA(2000). Examining the potential for adapting simulation
software to enable short-term tactical decision making for