The Simulation and Study of the Operating System Efficiency Improvement of a Container Terminal

. With the rapid expansion of Chinese container terminal throughput, each container terminal increasingly pays attention to the operational efficiency of storage yard. In this paper, aim at the problem of excessive containers relocation assignments to establish the slotting optimization allocation model, and verify the validity of the model by using Matlab; Using tandem queuing network modeling to build model of the container terminal operating system,and simulate the container terminal logistics system by Witness. It can provide a reference for efficiency improvement of the container terminal operating system.


introduction
The large scale ship of container terminal brings all kinds of pressure to the container terminals. How to solve the problem of the construction and planning of container berths has become the scholars' key focus. Kim and Hong [1] used the heuristic methods and branch and bound method simultaneously to study containers relocation problems. K.H. Kim [2] considered some of the nature of export containers, he mainly used a dynamic programming method to determined the result. Yuewen Gao and Yousan Ji [3] analyzed the weight factor, and used the search technology to show that this method of container terminals had reference value to reduce the number of containers relocation. Jumin Hao [4] builded an optimization model about an inner shell of the container terminal yard, it improved the utilization of terminal yard effectively. Jianfeng Shen [5] proposed palletizing models, including matching rules, most preferred, and the region partition.
The main object of the study is a container terminal logistics system, for containers relocation assignment, mechanical resource configuration issues, a terminal operation system is optimized. Research ideas and methods can provide reference for related wharf optimized operating system.

Model assumptions
It assumed that the zone and the bay of zone had been allocated in container terminal yard; According to container-delivering information, basic information of all containers have been identified, including the box weight, port of destination, etc; The containers relocation assignments just were operated during the bay. It assumed that all prepared assigned container are the same type.

Model parameters
J expressed the number of stacked inside each Bay; K represented the number of Bay inside per segment; I expressed the number of stack height inside each Bay. M is represented as a large positive values; N represented the total number of planned container terminal yard; d n expressed as a destination of the n-th container; w n was expressed as an n-th grade container carrier; c n expressed as the n-th container property value; v k expressed as k-th Bay carrying capacity. S nkj noted that the i-layer box position of j of k-th Bay, which can occupied by the n-th container, With 1 said it had been allocated, with 0 expressing others: This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits distribution, and reproduction in any medium, provided the original work is properly cited.

XQUHVWULFWHG XVH
Article available at http://www.matec-conferences.org or http://dx.doi.org/10.1051/matecconf/20164402021 Minimizing the number of containers relocation during the Bay is the goal of container allocation problem.
The objective function: Restrictions: (2) ensured that the container can not be more than the amount of energy they carry large; The constraints (3) expressed that if it is on the second layer container, others should be placed beneath the corresponding container. Constraints (4) ensured that a entering the sequence number of second layer containers was greater than its corresponding approach below the serial number.
This paper used a heuristic to calculate. It was thought as a two-dimensional coordinate system, longitudinal explained to give the container two integer , lengthways explained achieving sequential, the so-called container properties.
The destination A and the weight grade B could be determined site adopted container . Clear the level six weight classes, with a 1-6 six digital representation,

Fig1. Rough classification
According to the poly-line to know the initial state of container , it is shown in Figure 2. As can be seen from this figure, the second and third columns belonged to the long column; the first column, the fourth column and the fifth column were moderate; the sixth column belonged short columns.Over four boxes under ultrahigh column forms obtained were stacked in order with short column by column, the first step was amended, and formed

Fig2. Rough classification
Fig3. The first correction According to the above method, the Pareto chart was corrected once again , the results was shown in Figure 4.

Determine the container terminal tandem queuing network model
The relationship between several logistics nodes configuration of the container terminal was shown in Figure 5: Node configure of container terminals system

Container Terminal series line network multinode configuration analysis
The job level of container terminal logistics network was made up of a plurality of unipolar queue subsystem. Container flows between each node formed a queuing network. According to production of operating line, a series queuing network analytic model was established. Container series line network structure of operating line production was shown in Figure 6   In this paper, simulate the existing container terminal processes. The main application was the open-loop series multistage queuing network. In the process of import container loading and unloading, First Container loaded onto the truck by crane, Transport the containers shipped under the gantry by the truck, Then place container onto the yard prearranged location by gantry crane. In this paper, the based on container terminal network queuing structure system, Create a system simulation model, and use witness system make the practical operation, summary statistical indicators.

Set basic parameters
The cranes node was the import containers beginning serving node, the gantry crane node was the ending service node, while exports container was opposite. Network can be considered as only one type of customer (container). It was shown in Figure 8:

Fig8. Container Terminals System of Tandem Open
Queuing Network The truck arrived the quay crane service center, quay crane was idle, but it might not be able to receive services directly. It was carried out equipping in accordance with the relevant information received in advance of the container ship dock. If one of trucks was matching, the truck was responsible for the work of this operating line . If the job line was idle, the containers could be set loaded onto the truck. If there were other trucks to went on the loading operation, the truck was set to enter the system waits.

The simulation process of witness
During operation of the model system simulation, container automatically entered the system, when it arrived the bridge node, which obeyed Poisson distribution, it was shown in Figure 9. 1.7 represented that a boat was removed every 1.7 minutes. The containers were transported to the yar to stacking by using the truck. In the simulation, the truck obeyed second-Ireland distribution (2.0,2,2). The simulation time was set at 2 months (60 days).

Fig9. Simulation model of mechanism configuration
of container terminal In this paper, choose four ratio to analyze the various indicators, as shown in Table 1

Analysis of test results
It could be observed from simulation, when different ratios, through objective analysis option 1: 5: 5 was the best configuration, the equipment utilization was higher, queue length was relatively short.

Conclusion
First, the optimization stockpiling mathematical model was established. Using MATLAB to select a randomly data, Use optimized model to stacking containers, Simulate the process of shipment suitcase, and get the most optimal stockpiling mathematical model.Second, the paper created a container terminal simulation model based on series line network. Based on the historical operating data of a container terminal, the model was established according to the reasonable options of this paper, and use WITNESS to make simulation experiments. The test effectively demonstrated the effectiveness of the proposed methods, there had a good reference for the scheduling of container handling equipment.