Wireless Sensor Network, 2010, 2, 629-638
doi:10.4236/wsn.2010.28074 Published Online August 2010 (http://www.SciRP.org/journal/wsn)
A Study on Effective System for Harbor Container
Delivery & Cargo Work Automatuion
Dong-Hoon Kim, Jun-Yeob Song1, Seung-Ho Lee, Il-Yong Kang2, Suk-Keun Cha3
1Intelligent Manufacturing Systems Research Division, Korea Institute of Machinery & Materials, Daejeo n, Korea
2R&D Center, Promecs Co., Ltd., Jeonju, Korea
3ACS Co., Ltd., Seoul, Korea
Received June 2, 2010; revised June 10, 2010; accepted June 17, 2010
Abstract
In this article, we have attempted to analyze current situation and the problem of domestic and overseas har-
bor container delivery & cargo work automation centered on major harbors and to suggest effective way to
deal with the issue in order to improve the productivity of container cargo work per crane, the major index of
productivity of high value-added shipbuilding industry. In particular, we have suggested the way to realize
effective automation system that can improve the efficiency of harbor container delivery & cargo work
through the development of high-tech measuring automation technology using microwave radar and applied
design that have broken away from traditional automation system and traditional problems such as depend-
ence on manual work and the problem of laser method in which workers cannot identify laser beam under
sunlight and workers’ eyesight can be weakened by being exposed to laser beam.
Keywords: Harbor, Container, Delivery, Cargo Work, Automation
1. Introduction
Shipbuilding industry is undeniably high value-added
industry with domestic, overseas competitiveness and
will be developed continuously in the future. In ship-
building industry, harbor is a work place as well as an
important factor determining productivity as shown in
Figure 1. Harbor has close relation with crane and the
productivity of container cargo work per crane is a rep-
resentative index by which we can measure the service
level of a harbor and the productivity per shipment can
be considered more reasonable index recently. As shown
Table 1 and Figure 2, Busan harbor, the representative
port of South Korea, is ranked high among Asian ports in
terms of capacity in container transportation. However,
its container productivity per crane is about 25.1 con-
tainers per hour, which is the lowest level except Kaoh-
siung port in Taiwan compared with Hong Kong (40)
and Singapore (39.4) etc. Thus, effective management of
yard cargo work is necessary to improve port productiv-
ity including Busan harbor. It is absolutely necessary to
improve container cargo work by focusing on the system
automation within the area of apron (the part attached to
the inner wall of port dock and a place where container
cargo work is conducted by container crane) in container
cargo work automation for this purpose. In this article,
we have suggested the effective way to overcome the
problem and developed a prototype device by analyzing
current situation and the problems of related method.
2. Current Situation of Domestic and
Overseas Harbor Container
In case of domestic existing ports, most of the work is
done manually due to outdated infrastructure and the
opposition of port trade union and Hanjin shipping co.
applied laser beam method to the new Gamman dock in
Busan harbor in terms of automation but it is being
managed along with manual work including dispatching
guide staff in a rainy or foggy day. In most cases, proper
number of staff to guide containers to arrive at cargo
work area of crane is estimated to be abou t 7 but 48 peo-
ple are being dispatched per container crane in three
8-hour shifts in reality to avoid danger or overwork,
which is creating excessive fixed costs and weakening
business competitiveness.

D. H. KIM ET AL. 630
Table 1. Crane Productivity of major ports.
Division Pusan Singapore Hongkong Kaohsiung Tokyo Shanghai Unit
Crane
Productivity 25.1 39.4 40 15.6 30.4 32.6
Number of
containers/
Hour
Figure 1. Container loading dock.
Figure 2. Comparison of volume of cargo between major
ports.
Crane maintenance rate of operation is about 60%
considering proper rate of operation to cope with the
change of volume of cargo and actual working hour is
estimated to be about 4.8 hours and workers spend 40%
of 8 hour daily working time (3.2 hours) for standby time
since workers are dispatched by crane unit.
But U-Port project was initiated by the former Minis-
try of Maritime Affairs & Fisheries in Korea and is under
way currently and Gwang Yang port, as an example,
began the construction of Automated Terminal in 2003
and is expected to start test-operation in 2009, which
seems to create high expected effect for competitiveness
and high value-added industry in the future.
As shown in Figure 3, the domestic and overseas re-
search trend of related technology is as follows. In Korea,
a method that finds the route by receiving information
about transport vehicle’s stop location manually through
assistant worker positioned outside has been used re-
cently in most case and the research about technology
Figure 3. Introduction of traditional system.
using optical sensor, optical fiber and camera are being
conducted partly but its application on the field is diffi-
cult due to noise etc. In case of foreign countries, there
are HHLA port, Hamburg, Rashid port, Dubai, Mel-
bourne port, Australia, Katka port, Finland, and Salerno
port, Italia etc as examples of using laser and there are
Vancom terminal and Delta terminal of Vancouver port,
Canada etc as examples of using RF Tag. But there are
problems in these cases in which the angel of laser beam
changes due to crane vibration because the device needs
to be installed on the crane itself in case of laser beam;
the credibility and recognition rate drop due to environ-
mental factors occurring frequently such as direct sunlight,
lighting, rainy weather, heavy rainfall, fog, and dust; the
risk of safety accident, such as weakening of eyesight or
losing eyesight when the eyes of worker are exposed to
laser beam, is high; it is difficult to process various in-
formation; and the application of camera is not effective
either because there are many days with bad weather
such as fog etc.
Other than these cases, ultrasound can be used addi-
tionally for collision prevention due to the limitation of
measuring range and RF Tag is being used for limited ID
monitoring to establish management information system
concerning electronic document exchange information
and loading control such as yard schedule planning and
ship transport planning as part of information system.
3. Effective Way of Container Delivery &
Cargo Work Automation
The development of a device that provides the stop loca-
tion of Yard Trailer is required in case of cargo work of
loading and unloading container by crane between Ship
and Yard Trailer (Y/T) as an effective way of container

D. H. KIM ET AL.631
delivery & cargo work automation. For this goal, the
development of hardware and software meeting the re-
quirement of the field characteristics of port without be-
ing interrupted by various external environments is nec-
essary. It is required to develop position measure system
based on RF that consists of Transponder and Base Sta-
tion, control system that transmits or processes the posi-
tion information received from the system to the driver
of transport vehicle (refers to control device in case of
Automatic Guided vehicle), and controller that transmits
and processes the stop position information of vehicle
calculated in one of the base stations installed on multi-
ple crane supporters to transport vehicle wirelessly in
case of transport vehicle and base station.
In other words, it is necessary to research the way to
make precise tracking of position possible, processing
technology of various sensor data, position measure
method based on RF, and application of system. It is
considered as an effective way for container delivery &
cargo work automation to develop high-tech measure-
ment automation technology using microwave radar sys-
tem that overcame traditional system with problems such
as dependence on manual work and identification prob-
lem in case of rainy or bad weather etc, to improve the
performance port automation system through applied
design and to achieve localization of the system. It will
become a solution meeting the requirement of field char-
acteristics of port while minimizing interruption from
various external environments. As shown in Figure 4, the
detailed specification of port automation system to
improve container delivery & cargo work can be defined
as follows. Also, Figure 5 shows the overview of man-
agement solution and func tion for suggested system.
1) Develop Automatic Recognition Function by each
Container Size.
2) Develop Position & Tracking Solution.
3) Ensure necessary performance under bad condition
such as temperature (–20 to 70℃℃), humidity(snow,
rain, fog), and wind etc.
4) Enable recognition and cargo work for the next
generation Twin Lift.
5) Decide the position of Yard Crane (RTTC), Track-
ing Solu t ion, and Embedded T ype Device.
4. The Development of Prototype Device
The purpose of developing prototype device in this re-
search is to enable precise tracking of position and to re-
search processing technology of various sensor data, posi-
tion measurement method based on RF and application of
system. The purpose is to develop high-tech measurement
automation technology using microwave radar system that
overcame traditional system with problems such as de-
pendence on manual work and identification problem in
case of rainy or bad weather etc, to improve the perform-
ance of port automation system through applied design
and to achieve localization of the system. For this purpose,
we plan to develop hardware and software meeting the
requirement of field characteristics of port without being
interrupted by various ext ernal envi ronm ent.
Figure 4. The goal and performance of automation system for delivery & cargo work.

D. H. KIM ET AL. 632
Figure 5. Overview of management solution and function.
In other words, we have suggested development sys-
tem and implemented basic design to satisfy the specifi-
cation (transmission range: more than 100 m, margin of
error: less than ± 0.3 m) for the final goal of developing
position measurement system based on RF suitable for
port environment that consists of Transponder and Base
station, control system that transmits or processes posi-
tion information received from the system to the driver
of transport vehicle (refers to control device in case of
AGV), and controller that transmits and processes stop
position information calculated in one of the base sta-
tions installed on multiple crane supporters in case of
transport vehicle and base station.
In development prototype system, TP (Transponder),
which receives signal from BS (Base station) located at 4
legs of container crane as shown in Figure 6, receives
information from the distance of R1, R2, R3, R4 and the
information gets converted to position information by
controller. The internal system configuration and opera-
tion diagram of overall system are shown in Figure 7
and Figure 8.
Figure 9 shows monitoring device and the design con-
tents of monitoring screen to manage development pro-
totype system effectively. BS (b ase station) and TP (Tran-
sponder) send and receive information from each other in
5.8 GHz frequency band between Yard Trailer and Crane
and monitoring device for vehicle provides the informa-
tion to driver through user interface and the information
gets transmitted to operation control room (situation
room) in 2.4 GHz frequency band so that monitoring is
made possible in control room.
As shown Figure 10, in terms of calculating position,
the position of moving unit (MU) is determined by meas-
uring the distance to 3 fixed units (RU) (minimum theo-
retical number). The followings are the calculation meth-
ods and we proceeded with research using RTOF
method.
1) TDOA (Time-difference of arrival)
2) RTOF (Round-trip time-of-flight)
3) FMCW(Linear frequency modulation)
Measurement mechanism can be explained simply as
follows. The distance (D) between TP and BS is calcu-
lated by TOA method (Time of arrival) using RTT (round
trip time). TOA method using RTT determines the time
in which a signal gets transmitted between BS and TP
from the time in which a signal travels back and forth
between BS and TP. Coordinates of each TP(x1, y1, H),
(x2, y2, H), (x3, y3, H), and the height of BS(H’) are
known values and D1, D2, D3 are calculated by TOA
method. D1’, D2’, D3’ can be calculated by the Pythago-
rean formula because the values of D1, D2, D3 and H-H'
are already known. Thus, if you substitute all values into
the following equ ation, you will get 3 Equ ations about x,
y, by which you can calculate the coordinates( x, y).

 

 

 
2
2
11
22
11
2
2
22
22
22
2
2
33
22
33
DDHH
xx yy
DDHH
xx yy
DDHH
xx yy
 

 

 

(1)
In order to conduct a mock test, we formulated Auto
routing and Firmware module as shown in Figure 11 and
plan to monitor moving position between Container crane

D. H. KIM ET AL.633
Figure 6. Overview of development prototype system.
Figure 7. Internal system configuration.

D. H. KIM ET AL. 634
Figure 8. System operation diagram.
Figure 9. Monitoring device and monitoring screen d es ign.

D. H. KIM ET AL.635
Figure 10. Measurement mech anism.
Figure 11. Design of auto routing and firmware module.

D. H. KIM ET AL. 636
and Yard trailer on this basis. We also developed em-
bedded prototype device for Yard Trailer (Y/T) as shown
in Figure 12 on the basis.
5. Operation Result
As shown in Figure 13 and Figure 14, we confirmed
that the gauge of Embedded KIT installed in Demo Ve-
hicle reached red line and that the time indicating the
distance is zero and the time indicating container of con-
tainer monitoring program is at the middle point were
meeting at the same point.
Figure 13. Embedded KIT monitoring scr ee n.
For error test at randomly fixed position, it is an ex-
periment to find out hunting level of data by collecting
raw data when vehicle is stopped at a random position.
As shown in Table 2, we confirmed that the average
error was less than 30 cm in all cases as the result of
measuring more than 2 positions (coordinates) 5 times
for 30 seconds.
For error test in case of returning to starting point, we
measured the first position we measured and position
accuracy when Demo Vehicle returned to starting point.
It is a test to find out whether the system can always
identify the same position information. As the result of
measurement shown in Table 3, the position accuracy
when Demo vehicle returned to starting point was less
Figure 14. Main monitoring screen.
Table 2. The result of error test at randomly fixed position.
than 7 cm in maximum, which is very good result. We
also confirmed that the system can identify the same po-
sition information in any situ ation. Measured coordinate s(11.4, –8) (20.4, –6.1)
Average measured
coordinates 11.36, –7.91 20.36, –6.02
Average Error x: 4 cm, y: 9 cm x: 4 cm, y: 8 cm
For error test according to the number of TP, we con-
ducted a test by turning off the power of TP with Demo
Table 3. The result of Error Test in case of returning to
starting point.
Measured
coordinates First measurement (10.7, –8.1) measurement after return (10.7, –8.1)
Average measured
coordinates 10.73, –8.15 10.73, –8.15
Average Error x: 2 cm, y: 7 cm x: 3 cm, y: 5 cm
Table 4. The result of Error Test according to the number
of TP.
Number of TP Average Error
3 x: 14.3 cm, y: 8.3 cm
4 x: 11.3 cm, y: 3 cm
5 x: 12.5 cm, y: 14 cm
6 x: 5 cm, y: 8.5 cm
Vehicle located at a random position. We used 3-6 TPs
and it is a test about TP that cannot secure LOS and this
test was conducted to prepare against poles or cargo in
case of moving. As the result of measurement shown in Figure 12. Design of monitoring system in control room.

D. H. KIM ET AL.637
Table 4, there was a little difference of error according
to the number of TP but the margin of error in all cases
was less than 30 cm.
Table 5 shows the summary of the contents and result
of these experiments. The excellence of development
system compared with traditional system can be expected
as the result of this research as shown in Tab le 6.
6. Conclusions
Automation system between crane and yard trailer to
improve container delivery & cargo work is an original
technology that can prepare for the centralization of port
container volume, which will have ripple effect on the
economy. In terms of the effectiveness of container crane
used for port cargo work currently, 48 workers are being
dispatched per crane in 38-hour shifts; Crane mainte-
nance rate of operation is about 60% considering proper
rate of operation to cope with the change of volume of
cargo; workers spend 40% of 8 hour daily working time
(3.2 hours) for standby time since workers are dispatched
by crane unit, which has been causing excessive fixed
costs and weakening business competitiveness. It is con-
cluded that we can be well prepared for the centralization
of port container volume of cargo due to the pursuit of
building bigger and faster ships through the development
Table 5. Summary of test result.
Test Evaluation Contents Level of achievement
Error Test at random fixed position Confirm the credibility of data by collecting raw data when demo vehicle is stopped at a random
location
Confirmd that average error was less than 30 cm
in allcases after measuring more than 2 positions
(coordinates) 5 times for 30 se c o nds.
Error Test in case of returning to start-
ing point
Measure the position accuracy when demo vehi-
cle returned to starting point based on the first
measured position to confirmthe reliability of
measurement
Position accuracy after returning to starting point
was less than 7 cm in maximum, which is very
good result. Confirmed that the system can iden-
tify the same position information in any situa-
tion.
Error Test according to the number of
TP
Test by turning off power supply of TP with
demo vehicle located at random position, Test
using 3-6 TPs to prepare for a situation in which
LOS cannot be secured such as poles or cargo
etc.
There was a little difference of error according to
the number of TP but the margin of error was
less than 30 cm in all cases.
Operation Test Test to confirm vehicle stop through Embedded
Kit in demo vehicle a n d monitoring program. Confirmed that the system operated normalloy as the result of mock test using demo vehicle.
Table 6. Comparison between development system and traditional system.
Item Traditional system (Laser scanner, beam,
optical sensor) Microwave Type
Visual 1 dimensional simple sensor detection 1 2 dimensional Visual Position Tracking
Multiple recognition vehicle recognition system need to be in-
stalled separately. recognize all vehicle’ ID in the Area (unlim- ited no of vehicle)
Environm ental influence Vulnerable against port environment such as
snow, rain, fog etc No Environmental influence
Control System Impossible Control of Container Control Vehicle is Pos-
sible
Multiple Control Impossible Recognize the number of vehicles and lane in
case of multiple vehicle operation.
Loading & unloading recognition Impossible Recognize loading & unloading of container
Moving Direction Recognition Additional function to recognize opposite
direction is needed. Recognize Moving Direction regardless of the direction.
Anti-vibration Recognition err or depending on the vibration
of container Safe against the vibration of container
Unmanned Operation Impossible Completely Unmanned Y/T Design is Possi-
ble.
Speed Recognition not provided Recognize Vehicle Speed
Maintenance Cost High Maintenance Cost Very little Maintenance Cost
Position Recognition not provided Trance Vehicle Moving Route

D. H. KIM ET AL. 638
of this technology. We think that the commercialization
of this technolog y in the future will make great contribu-
tion to improving nation al and corporate competitiven ess
because of less financial loss due to the postponement of
shipping and prompt processing of export & import pro-
cedure by making it possible to work 24 hour a day
throughout the year by unmanned cargo work regardless
of external environment.
7. References
[1] D. H. Kim and J. Y. Song, “Ubiquitous-Based Mobile
Control and Monitoring of CNC Machines for Develop-
ment of u-Machine,” Journal of Mechanical Science and
Technology, Vol. 20, No. 4, 2006, pp. 455-466.
[2] D. H. Kim and J. Y. Song, “Knowledge-Evolutionary
Intelligent Machine-Tool-Part 1: Design of Dialogue
Agent Based on Standard Platform,” Journal of Mecha-
nical Science and Technology, Vol. 20, No. 11, 2006, pp.
1863-1872.
[3] R. Cheah, “Design and Implementation of an MMS
Environment on ISODE,” Computer Communications,
Vol. 20, No. 15, 1997, pp. 1354-1364.
[4] Zigbee Alliance. http://www.zigbee.org
[5] P. Bahl and V. N. Padmanabhan, “RADAR: An In-
Building RF-Based User Location and Tracking System,”
IEEE International Conference on Computer Communi-
cations, New Orleans, 2000, pp. 775-784.
[6] “Current Situation of Container Dock Operation,”
Ministry of Maritime Affair, South Korea, 2007.
[7] “Analysis Report of World’s Major Terminals,” Cargo
System, 2005.