KR100411583B1 - method for optimizing container berth schedule and computer-readable medium - Google Patents

Abstract

The present invention relates to a method for optimizing a berth schedule of a container terminal and a recording medium, and more particularly, to plan an appropriate berth schedule for each vessel in a container terminal, to provide a berth schedule that minimizes the cost of the container terminal and each vessel. A berth schedule optimization method and a recording medium are provided, and the berth schedule optimization method according to an aspect of the present invention inputs information on each vessel to be docked to the berth and a preferred berth position for each vessel according to the situation of the device head. A first step of receiving, a second step of determining an eyepiece time and an eyepiece position at which an individual minimum cost is incurred that is not considered for each vessel, a third step of calculating the cost of using a berth by time using a sub-gradient optimization technique, Whether or not the cost of berths obtained in the third step is optimized In order to determine whether the calculation result does not converge based on the fourth step of determining whether the calculation result of the third step has converged according to a predetermined condition and the result determined in the fourth step, And a fifth step of updating a variable value and repeating the process of the third step from the second step.

Description

Method for optimizing container berth schedule and computer-readable medium

The present invention relates to a method for optimizing a berth schedule of a container terminal and a recording medium, and more particularly, to plan an appropriate berth schedule for each vessel in a container terminal, to provide a berth schedule that minimizes the cost of the container terminal and each vessel. An object of the present invention is to provide a method for optimizing a berth schedule and recording medium.

The areas required for planning and operation in the container terminal are as follows.

1) Plan: berth plan, unloading plan, yard plan, resource plan

2) Operation: Operation of import and export work, operation of ship operation

The present invention, which relates to the berth plan, is a part of the plan of the container terminal, and the related fields are described as follows.

A site plan to plan where to place the containers coming into the container terminal, a loading / unload plan to plan loading and unloading the container when the vessel is docked, and a resource plan to schedule the equipment to handle the container. have. In these plans, the berth plan is also very important because the loading / unloading plan is established in accordance with the berth plan.

In the prior art, planners of container terminals are notified of the arrival time and departure time required for each ship when preparing the berth plan, so that the berth plan is established so that several vessels do not overlap in space in time. At home and abroad, the academy of berth planning has been approached in various ways to solve this problem.

1 Lai and Shih (1992): They presented a simulation policy on how to assign berths by dividing the types of ships.

2) Brown et al. (1994): In this study, the berth plan was drawn up by considering various realistic constraints.

Lim (1998): In this study, we introduced a model for preparing a berth plan in which the berth berths are fixed for each ship and only the eyepiece time is determined.

The prior art study lacked the perception of the berth plan as a problem of determining the eyepiece time and eyepiece position at the same time. The reason is that the berths are often shared and used by many ship owners. However, the berths are often used separately for each ship, so only the docking time of each ship is determined within one berth. In the case of joint use of berths, the berth planning problem was studied in the direction of finishing all the vessels from the container terminal's point of view. In practical terms, however, this approach is problematic because the berth plan is to determine the docking position and the docking time of the vessel where the cost of the container terminal and each vessel is minimized.

The present invention is to solve such a problem, in consideration of the position of the container to be loaded in the container in the berth plan can shorten the berthing time for each vessel, the overall view of the container terminal The present invention aims to provide a method and recording medium for berth scheduling to provide an optimized algorithm for presenting and solving a new model for determining eyepiece time and position.

1 shows a flowchart of the method for optimizing the berth schedule of the present invention.

Figure 2 shows the vessel-specific input data based on the data of the week of the vessels for performing the calculation by the method of the present invention.

Figure 3 is based on the model presented in the present invention, comparing the results obtained by solving the three-day berth plan by the prior art solutions and the results obtained by the sub-gradient optimization technique of the present invention.

4 shows an example in which the calculation is performed by the method of the present invention when the solution for a long period of 7 days is to be performed as in the actual berth plan.

5 shows an example of an output screen providing an optimized berth schedule calculated by the berth schedule optimization method of the present invention.

In order to achieve the above object, the berth schedule optimization method according to a feature of the present invention comprises the first step of receiving the information about each ship to be docked to the berth and the preferred berth position for each vessel according to the situation of the device head, A second step of determining an eyepiece time and an eyepiece position at which an individual minimum cost occurs that does not take into consideration other vessels for each ship, based on the data input in the first step, and the eyepiece time for each of the ships determined in the second step; And a third step of calculating the use cost of timeline berths by a sub-gradient optimization technique based on the eyepiece position, and to determine whether the cost of using the berths obtained in the third step is optimized. The fourth step of determining whether the calculation result has converged by a predetermined condition and the result determined in the fourth step. If it is determined that the resulting acid is not converged, and updates the value of the variables required for the iteration, and from the second step to a fifth step for repeating the process of the third step.

Preferably, the berth schedule optimization method is a case where the eyepiece position and eyepiece time of each ship overlap for a specific position and a specific time when it is determined that the calculation result has converged based on the result determined in the fourth step. It further comprises a sixth step for resolving the collision between the ship by correcting in the calculation result.

Preferably, the berth schedule optimization method is characterized in that in the first step, the input information for each ship includes the length of each ship, work required time, arrival time, departure request time.

Preferably, the berth schedule optimization method includes an eyepiece time that satisfies the following equation in order to determine the eyepiece time and eyepiece position at which the individual minimum cost is incurred in which the other vessel is not considered for each vessel in the second step. Find the eyepiece position and make it upper bound.

,

,

(for i = 1 ... m , j = a _k ... n , k = 1,2 ... l ),

(for i = 1 ... m , j = a _k ... n , k = 1,2 ... l ),

(for k = 1,2 ... l ),

x _ijk , z _ijk = 0 or 1 (for all i, j and k).

here, ,

L = berths,

l = number of ships,

m = time during the planning period,

n = eyepiece location when dividing the berth into small spaces,

p _k = ship's preferred location,

a _k = berthing time of the ship,

d _k = time required for the ship to depart,

b _k = working time of the ship,

_k = length of ship,

z _ijk = 1 at eyepiece position j at eyepiece time i of ship k

Decision Tree,

x _ijk = position and time values occupied by ship k have 1

Decision Tree,

c _1k = additional occurrence if the ship is not in the docking position

Port cost,

c _2k = Occurs when the ship cannot meet the required time of departure

Defined as penalty cost.

Preferably, in the third step, the berth schedule optimization method includes: in step 3-1, using the eyepiece time and eyepiece position value obtained in the second step, determining a value of the following equation as lower bound;

,

,

,

(for i = 1 ... m , j = a _k ... n , k = 1,2 ... l ),

(for i = 1 ... m , j = a _k ... n , k = 1,2 ... l ),

(for k = 1,2 ... l ),

x _ijk , z _ijk = 0 or 1 (for all i, j and k),

In addition, step 3-2, which finds the eyepiece time and eyepiece position satisfying the following equations for all the ships by lower bounding the value obtained in step 3-1.

,

,

,

(for i = 1 ... m , j = a _k ... n , k = 1,2 ... l ),

(for i = 1 ... m , j = a _k ... n , k = 1,2 ... l ),

(for k = 1,2 ... l ),

x _ijk , z _ijk = 0 or 1 (for all i, j and k) and

Step 3-3 to obtain each subgradient value G _ij by the following equation

(for i = 1 ... m, j = 1 ... n)

Characterized in that consists of.

Preferably, in the fourth step, in the fourth step, in order to determine whether the value obtained in the third step is a converged value, the variable value is determined by the following equation,

,

Where Z _MAX is the fourth lower value as the highest lowerbound value found while the calculation in the third step is repeated;

If the Z _MAX obtained in the step 4-1 is smaller than the predetermined set value, characterized in that it comprises a step 4-2 of stopping the operation to find the solution, storing the Z _MAX found up to this time and stops the calculation.

Preferably, in the fifth step, in the fifth step, when the Z _MAX is not smaller than a predetermined setting value, steps 5-1 and 5-1 of updating a variable value according to the following equation: Step 5-2,

,

And

And a fifth step of repeating the calculation from the second step by the variables updated in steps 5-1 and 5-2.

In order to establish an optimized berth plan of the container terminal according to another aspect of the present invention, the recording medium recording the berth schedule optimization method for determining the berth position and berthing time of each optimized vessel incurred a minimum cost, the berth On the basis of the first step of receiving the information of each vessel to be docked to the vessel and the preferred eyepiece position for each vessel according to the situation of the device head, the other vessel is not considered for each vessel based on the data input in the first stage Based on the eyepiece time and eyepiece position determined in the second step of determining the eyepiece time and eyepiece position at which the minimum cost occurs, the use cost of the berths for each time zone is calculated by the sub-gradient optimization technique. In the third step, in order to determine whether the cost of using the berths obtained in the third step is optimized, If it is determined that the calculation result does not converge based on the fourth step of determining whether the calculation result of the third step has converged according to a predetermined condition and the result determined in the fourth step, each variable value required for the iterative calculation is determined. And a computer readable recording medium having recorded thereon a program for executing the fifth step of updating and repeating the process of the third step from the second step.

Preferably, the computer-readable recording medium overlaps the eyepiece position and eyepiece time of each ship for a specific position and a specific time when it is determined that the calculation result has converged based on the result determined in the fourth step. And a computer-readable recording medium having recorded thereon a program for executing the berth schedule optimization method further comprising a sixth step for resolving the collision between ships by correcting the case in the calculation result.

Preferably, the computer-readable recording medium in the first step, the input information for each ship is the berth schedule optimization method including the length of each ship, work required time, arrival time, departure request time And a computer readable recording medium having recorded thereon a program to be executed by the computer.

Preferably, in the second step, the computer-readable recording medium satisfies Equation 1 in order to determine the eyepiece time and eyepiece position at which individual minimum costs incurred that do not consider other vessels for each vessel occur. It is characterized in that the computer-readable recording medium recording a program for executing the berth schedule optimization method to find the eyepiece time and the eyepiece position and upper bound it.

Preferably, in the third step, the computer-readable recording medium uses the eyepiece time and the eyepiece position value obtained in the second step to determine the value of Equation 2 as lower bound. Step -1 and Step 3-2 for finding the eyepiece time and eyepiece position satisfying Equation 3 for all the ships by lower bounding the value obtained in Step 3-1, by Equation 4 And a computer-readable recording medium having recorded thereon a program for executing the ore schedule optimization method comprising the third to third steps of obtaining each sub-gradient value G _ij .

Preferably, in the fourth step, the computer-readable recording medium determines the variable value according to Equation 5 to determine whether the value obtained in the third step is a converged value. here, Z _MAX is the sun Z _MAX obtained in the step 4-1 and the step 4-1, the lower bound for the value of the maximum found during the computation of the third step iteration is smaller than a predetermined value stop the operation to obtain characterized in that a computer-readable recording medium storing a program to store the Z _MAX sought by this time and run the berth predetermined optimization method comprising the steps 4-2 to stop the calculation from the computer.

Preferably, in the fifth step, in the fifth step, when the Z _MAX is not smaller than a predetermined value, the computer-readable recording medium updates the variable value according to Equation 6 below. According to the steps updated in step 5-2, step 5-1 and step 5-2, step 5-3, wherein the calculation from the second step is repeated. And a computer readable recording medium having recorded thereon a program to be executed by the computer.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

In the present invention, an optimization technique and a subgradient optimization technique are proposed as a method of making a mathematical model and solving a solution for preparing a berth plan. Among the two techniques proposed by the present invention, the optimization technique has been proved through experiments to be effective only when the planning period of the berth plan is less than 3 days. The reason for this is that the optimization technique rapidly increases the calculation time as the number of ships increases, and in the case of plans exceeding 3 days, the calculation time often exceeds 48 hours. In contrast, in the case of the subgradient optimization technique, the results of experiments in the 7-day berth plan were found in 5 minutes, which proves that the subgradient optimization technique can be achieved well. there was.

Container terminals should have a berth plan to reduce the operating costs of the terminal while accommodating individual vessel requirements. We have developed an optimized berth planning technique that can meet the various requirements of ships and minimize the operating cost of the terminal. The optimal berth planning technique uses Lagrange mitigation, which is a method of management science, to develop berth planning.

1 shows a flowchart of the method for optimizing the berth schedule of the present invention.

As shown in FIG. 1, the method for optimizing a berth schedule according to a preferred embodiment of the present invention includes the first step of receiving information about each ship to be docked at the berth and the preferred berth position for each ship according to the situation of the device head ( S1), based on the data input in the first step, a second step (S2) to determine the eyepiece time and eyepiece position at which the individual minimum cost incurred that does not consider other vessels for each vessel (S2), determined in the second step On the basis of the eyepiece time and eyepiece position of each ship, whether or not the cost of berths obtained in the third step is optimized (S3), wherein the cost of berths for each time zone is calculated by a sub-gradient optimization technique. In order to judge, the fourth step S4 of determining whether or not the calculation result of the third step has converged according to a predetermined condition and the result determined in the fourth step If it is determined that W is not a converging said calculation result, and updates the value of the variables required for the iterative calculation, and from said second step includes the step of claim 5 (S5) to repeat the process of the third step.

Symbols used for the detailed description of the present invention are as follows.

L = length of berths

l = number of ships

m = time during the planning period

n = Possible docking position when dividing the berth into small spaces

p _k = ship's preferred location

a _k = berthing time of the ship

d _k = time required to depart the ship

b _k = working time of the ship

l _k = length of ship

z _ijk = Decision variable with 1 if it is located at eyepiece position j at eyepiece time i of ship k

x _ijk = position and time values occupied by ship k are decision variables with 1

c _1k = additional transport costs incurred if the ship is not in the docking position

c _2k = penalties incurred when the vessel is unable to meet the required time of departure

The model for the berth planning is as follows.

,

(for i = 1 ... m , j = 1 ... n ),

(for i = 1 ... m , j = a _k ... n , k = 1,2 ... l ),

(for i = 1 ... m , j = a _k ... n , k = 1,2 ... l ),

(for k = 1,2 ... l ),

x _ijk , z _ijk = 0 or 1 (for all i, j and k),

here .

The first equation of Equation 7 shows that the service cost for the ship docked in the berth is minimized. The problem is modified to solve the problem effectively. Thus, when the second equation of Equation 7 is combined with the first equation, the problem becomes as in Equation 1 above. here Is a dual variable of the first equation of Equation 7.

The dual problem of Equation 1 is the same as Equation 2.

In addition, Equation 1 may be divided by each vessel and expressed as Equation 3 above.

In summary, the process of solving the berth assignment problem is as follows:

Step 1: One solution is obtained from Equation 1 above, and the upper bound is obtained.

Step 2: Optimally solve the R _k problem of Equation 3 above for all ships. At this time, the value of Equation 2 is lower bound (Z _LB ).

Step 3: The subgradient value G _ij is obtained by the method of Equation 4 above.

Step 4: Update the step size and the current highest lower bound by equations (5) and (6) above.

Where Z _MAX is the highest lower bound value found so far. In this example, lambda was initially set to 2 and was reduced to 1/2 if Z _MAX did not increase during 30 iterations. Z _MAX will stop working to obtain it is smaller than 0.005 stores the Z _MAX sought by this time and finish.

Step 5: The dual variable is updated as in Equation 6 above.

Then go to Step 2.

After solving and completing this procedure, the results of berth assignments may overlap between ships. At this time, if the ships are overlapped in the direction of the smallest increase in total cost, the result can be very close to the optimum solution.

Figure 2 shows the ship-specific input data based on the data of the week of the ships for performing the calculation by the method.

FIG. 3 compares the results of solving the three-day berth plan by the prior art solution and the results of the subgradient optimization technique of the present invention, based on the model presented in the present invention. Here, it can be confirmed that the subgradient optimization technique of the present invention has obtained an optimal solution except two cases. And in terms of calculation time it can be seen that reduced to 7% with respect to the prior art technique.

However, in the case of actual berth planning, the period to be planned is seven days or more. In such a case, it is impossible to solve by the conventional technique. The reason is that in the case of the optimization technique, the calculation time increases exponentially when the number of ships increases, so that the solution cannot be obtained.

4 shows an example in which the calculation is performed by the method of the present invention when the solution for a long period of 7 days is to be performed as in the actual berth plan.

The subgradient optimization technique of the present invention has obtained an accurate solution as shown in FIG. 4 even in a long period as in the case of the actual berth plan.

The term "evaluation criteria" shown in FIG. 3 is a value used to measure the performance of the solution presented by the subgradient technique when the optimal solution cannot be obtained as an estimate for the optimal solution provided by the subgradient optimization technique of the present invention. And performance is the value of the results / evaluation criteria. Experiments verified that the subgradient optimization algorithm presented in the present invention can solve the one-week berth plan result within 3 minutes.

5 is a diagram illustrating a result of performing a preferred embodiment of the present invention using a GUI program.

The method and recording medium for berth schedule optimization according to the present invention can be modified and applied in various forms within the scope of the technical idea of the present invention and are not limited to the above preferred embodiment.

The embodiments and the drawings are only for the purpose of describing the contents of the invention in detail, and are not intended to limit the scope of the technical idea of the invention, the present invention described above has a general knowledge in the technical field to which the present invention belongs It is apparent that the present invention is not limited to the above embodiments and the accompanying drawings, since various substitutions, modifications, and changes are possible without departing from the technical spirit of the present invention.

By using the method and the recording medium of the berth schedule optimization according to the present invention, it is possible to effectively operate the berths, which are important resources in the container terminal, and improve the quality of service for the ship. The present invention has a significant significance in overcoming the limitations of the mathematical approach while satisfying the actual needs of the field. In order to verify the effect of the present invention, experiments were carried out by receiving actual data from the eastern container terminal, and the method of the present invention provides almost the same results for the problems solved by the method provided by the prior art, while calculating the computation time of the prior art. Compared to 10% of the time required to solve. In addition, a one-week berth plan, which is a realistic berth plan that is too time-consuming and hardly solved by the conventional method, was solved to a level close to an optimal solution within three minutes. Therefore, by applying the method proposed in the present invention to the actual container terminal can not only significantly improve the quality of service for the vessel can also provide an important effect that can significantly reduce the logistics cost in the container terminal.

Claims (6)

In the berth schedule optimization method for determining the docking position and the docking time of each optimized vessel incurred the minimum cost for the optimal berth planning of the container terminal, A first step of receiving information about each ship to be docked at the berth and a preferred docking position for each ship according to a situation of a device head; A second step of determining an eyepiece time and an eyepiece position at which an individual minimum cost incurred that does not consider other vessels for each vessel is based on the data input in the first step; A third step of calculating the use cost of the berths according to time slots based on the eyepiece time and eyepiece position of each ship determined in the second step by a sub-gradient optimization technique; A fourth step of determining, based on a predetermined condition, whether or not the calculation result of the third step has converged in order to determine whether the cost of using the berth obtained in the third step is optimized; If it is determined that the calculation result does not converge according to the result determined in the fourth step, a fifth step of updating each variable value necessary for the iteration calculation and repeating the process of the third step from the second step ; And If it is determined that the calculation result is converged by the result determined in the fourth step, the collision between ships is eliminated by correcting the case where the eyepiece position and eyepiece time of each ship overlap with each other for a specific position and a specific time. And a sixth step for the sequencing schedule optimization method. The method of claim 1, In the first step The input information for each ship is the berth schedule optimization method comprising the length of each ship, work required time, arrival time, departure time required. The method of claim 1, In the second step In order to determine the eyepiece time and eyepiece position at which an individual minimum cost occurs that does not consider other vessels for each vessel, the berth schedule optimization method characterized by finding the eyepiece time and eyepiece position satisfying the following equation and making it upper bound. . , , (for i = 1 ... m , j = a _k ... n , k = 1,2 ... l ), (for i = 1 ... m , j = a _k ... n , k = 1,2 ... l ), (for k = 1,2 ... l ), x _ijk , z _ijk = 0 or 1 (for all i, j and k). here, , L = berths, l = number of ships, m = time during the planning period, n = eyepiece location when dividing the berth into small spaces, p _k = ship's preferred location, a _k = berthing time of the ship, d _k = time required for the ship to depart, b _k = working time of the ship, _k = length of ship, z _ijk = 1 at eyepiece position j at eyepiece time i of ship k Decision Tree, x _ijk = position and time values occupied by ship k have 1 Decision Tree, c _1k = additional occurrence if the ship is not in the docking position Port cost, c _2k = Occurs when the ship cannot meet the required time of departure Defined as penalty cost. The method of claim 1, The third step is Step 3-1 to determine the value of the following equation lower bound using the eyepiece time and eyepiece position value obtained in the second step. , , , (for i = 1 ... m , j = a _k ... n , k = 1,2 ... l ), (for i = 1 ... m , j = a _k ... n , k = 1,2 ... l ), (for k = 1,2 ... l ), x _ijk , z _ijk = 0 or 1 (for all i, j and k); Step 3-2, which finds the eyepiece time and eyepiece position satisfying the following equation for all the ships by setting the value obtained in the above step 3-1 as the lower bound: (R _k ) , , , (for i = 1 ... m , j = a _k ... n , k = 1,2 ... l ), (for i = 1 ... m , j = a _k ... n , k = 1,2 ... l ), (for k = 1,2 ... l ), x _ijk , z _ijk = 0 or 1 (for all i, j and k); And Steps 3-3 to find each subgradient value G _ij by the following equation (for i = 1 ... m, j = 1 ... n) Berth schedule optimization method comprising the. The method of claim 1, The fourth step is In order to determine whether the value obtained in the third step is a converged value, the variable value is determined by the following equation, , Where Z _MAX is a value of the highest lower bound found while the calculation in the third step is repeated; If the Z _MAX obtained in the step 4-1 is smaller than the predetermined set value, the step of optimizing the berth schedule, characterized in that the step of stopping the work to solve the solution, and to remember the Z _MAX found up to this time and stop the calculation Way. The method of claim 1, The fifth step is If the Z _MAX is not smaller than a predetermined set value, A step 5-1 of updating the step size value by; Step 5-2 of updating the dual variable value by; And And a step 5-3, wherein the calculation from the second step is repeated by the variables updated in steps 5-1 and 5-2.