The impacts of time segment modeling in berth allocation and quay crane assignment on terminal efficiency
| DOI | https://doi.org/10.1108/IMDS-08-2018-0335 |
| Pages | 968-992 |
| Published date | 10 June 2019 |
| Date | 10 June 2019 |
| Author | Hoi-Lam Ma,Zhengxu Wang,S.H. Chung,Felix T.S. Chan |
| Subject Matter | Information & knowledge management,Information systems,Data management systems,Knowledge management,Knowledge sharing,Management science & operations,Supply chain management,Supply chain information systems,Logistics,Quality management/systems |
The impacts of time segment
modeling in berth allocation and
quay crane assignment on
terminal efficiency
Hoi-Lam Ma
Department of Supply Chain and Information Management,
The Hang Seng University of Hong Kong, Shatin, Hong Kong
Zhengxu Wang
School of Business Administration,
Dongbei University of Finance and Economics, Dalian, China, and
S.H. Chung and Felix T.S. Chan
Department of Industrial and Systems Engineering,
Hong Kong Polytechnic University, Kowloon, Hong Kong
Abstract
Purpose –The purpose of this paper is to study the impacts of time segment modeling approach for berth
allocation and quay crane (QC) assignment on container terminal operations efficiency.
Design/methodology/approach –The authors model the small time segment modeling approach, based
on minutes, which can be a minute, 15 min, etc. Moreover, the authors divided the problem into three
sub-problems and proposed a novel three-level genetic algorithm (3LGA) with QC shifting heuristics to deal
with the problem. The objective function here is to minimize the total service time by using different time
segments for comparison and analysis.
Findings –First, thestudy shows that by reducing thetime segment, the complexityof the problem increases
dramatically.Traditional meta-heuristic,such as genetic algorithm,simulated annealing, etc., becomesnot very
promising.Second, the proposed3LGA with QC shifting heuristicsoutperforms the traditionalones. In addition,
by using a smaller time segment, the idling time of berth and QC can be reduced significantly. This greatly
benefits the containerterminal operations efficiency, and customer service level.
Practical implications –Nowadays, transshipment becomes the mainbusiness to many container terminals,
especially in Southeast Asia (e.g. Hong Kong and Singapore). In these terminals, vessel arrivals are usually very
frequent with small handling volume and very short staying time, e.g. 1.5 h. Therefore, a traditional hourly based
modeling approach may cause significant berth and QC idling, and consequently cannot meet their practical needs.
In this connection, a small time segment modeling approach is requested by industrial practitioners.
Originality/value –In the existing literature, berth allocation and QC assignment are usually in an hourly
based approach. However, such modeling induces much idling time and consequently causes low utilization
and poor service quality level. Therefore, a novel small time segment modeling approach is proposed with a
novel optimization algorithm.
Keywords Transshipment, Berth allocation, QC shifting heuristics, Time steps,
Variable-in-time quay crane assignment
Paper type Research paper
Nomenclature
Input data
Vset of vessels (V¼1, 2, 3, …,I)
Bset of berths in terminal (B¼1, 2,
3, …,J)
Uset of 15-minute time steps (U¼1, 2,
3, …,T)
a
i
expected arrival time of the vessel
(an expected value) i∈V.
Industrial Management & Data
Systems
Vol. 119 No. 5, 2019
pp. 968-992
© Emerald PublishingLimited
0263-5577
DOI 10.1108/IMDS-08-2018-0335
Received 4 August 2018
Revised 4 December 2018
Accepted 2 February 2019
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0263-5577.htm
The authors thanks the editor and the reviewers for their constructive comments and suggestions.
Beside, this research is supported by China Postdoctoral Science Foundation funded project (Grant No.
2018M631792), The Department of Education of Liaoning Province (Grant No. LN2017QN006).
968
IMDS
119,5
v
i
handling volume of vessel i∈V.
qmax
imaximum number of QCs can be
assigned to vessel i∈V.
qmin
iminimum number of QCs can be
assigned to vessel i∈V.
R
i
range of the assignable number of
QCs for vessel i∈V, where R
i
¼
[qmin
i;qmax
i].
PQC productivity, expressed as the
volume (TEU) handled by a QC at a
time step
Qtotal number of QCs in terminal
Na sufficiently large positive constant
1. Introduction
Transshipment hub is one of the most popular terminal models because of the changes in the
supply chain model. Examples can be found in Hong Kong, Singapore, Shanghai, Taiwan
(Kaohsiung), South Korea (Busan), Malaysia (Tanjung Pelepas), the Netherlands (Rotterdam),
etc. (Zhen et al., 2011; Lee et al., 2012; Liang et al., 2012; Tan and Hilmola, 2012). Transshipment
hub mainly deals with transshipment activities, where the vessel turnaround is usually fast.
Vessel arrival rate is relatively more frequent with smaller handling volume and shorter staying
time comparing with those in traditional gateway terminal. The shortest vessel staying time can
even be as short as only 1.5 h including all the documentations , and loading and unloading
operations. On the other hand, in traditional gateway terminal, vessels are usually with larger
handling volume and longer staying time (can be over days).
However, in the existing literature studying on the traditional gateway terminal, most
papers applied an hourly based time segment approach because the vessel staying time is
usually long. For example, the vessel handling time for a large vessel with 5,000 containers
is about 40 h (Queensland Government, 2014). In such situation, quay crane (QC) idling for
an hour may become relatively insignificant and acceptable (Meisel and Bierwirth, 2009;
Giallombardo et al., 2010). However, in transshipment hub, QC idling for an hour becomes
significant, resulting in poor operation efficiency. For this reason, many terminals’
industrialists (e.g. those in Hong Kong) are already changed to 30-min-based planning
approach. In fact, the industrialists are seeking for 15-min-based planning approach to
further enhance their efficiency by reducing the QC idling. However, to the best of the
authors’knowledge, there are no existing papers working in the area of the integrated berth
allocation problem (BAP) with variable-in-time quay crane assignment (QCA), i.e., using
15-min-based or 30-min-based planning approach. Therefore, the objective of this paper is to
fulfill this research gap raised by the practical industrial needs in the terminal industries.
The experimental results obtained by using the new time segment modeling approach show
that vessel’s turnover can be faster by reducing vessel’s waiting time and handling time.
This implies that the efficiency of transshipment hubs can be increased, similar to the
customer service level as well.
Although one may expect that operation efficiency can be improved because of the
reduction in QC idling, the problem complexity modeling in minutes-based approach is in
fact much higher than the traditional hourly based one, especially for the integrated BAP
with the variable-in-time QCA model. The increasing number of variables related to QCA
increases the computational complexity. Meanwhile, the time unit is another factor that
increases the problem complexity dramatically as well. For example, in a typical hourly
based approach, a day is divided into 24 discrete time segments. However, if we model in a
15 min based approach, the number of the time segments will then increase by fourtimes
to 96. This implies that the number of related variables will also increase exponentially.
Therefore, traditional hourly based solution approaches (such as integer programming)
may not be applicable in this case as the computational time will be too long. In this
connection, a new optimization algorithm is required. Since it is known that solving an
969
The impacts of
time segment
modeling
To continue reading
Request your trial