Arrangement calculation device, arrangement calculation method, and storage medium

Abstract

Provided are an arrangement calculation device and the like capable of providing an arrangement of containers that reduces the cost of handling the containers. An arrangement calculation device according to one aspect disclosed herein comprises: a generation unit that generates an Ising model for allocating the positions of containers in a temporary storage location, having one or more available positions at which the containers can be arranged, to the available positions; an operation processing unit for executing, in an information processing device, an arrangement determination process for changing the Ising model using quantum annealing; and an output unit. The containers include existing containers for which the positions are set, and target containers for which the positions are to be determined. The generation unit generates the Ising model indicating the positions of the existing containers. The operation processing unit acquires the Ising model and the shipment sequence of the containers. The arrangement determination process is a process for determining the positions of the target containers by changing the Ising model so as to reduce an evaluation value, which is lower when the number of combinations of the container and another container that is shipped out later than and arranged on the container is smaller, while satisfying the constraints that each of the containers are at different positions and any one of the containers is arranged in the available position below the positions of the containers. The output unit outputs the positions of the target containers.

Description

Layout calculation device, layout calculation method, and storage medium

[0001] The present disclosure relates to a layout calculation device, a layout calculation method, and a storage medium for calculating a layout.

[0002] When placing containers in a container yard with limited space, such as a port, it may be necessary to load other containers on top of one another. If containers are loaded into the container yard without considering the order in which they will be unloaded, a container being unloaded earlier may be placed below a container being unloaded later. In such cases, the costs of operations such as loading, unloading, and unloading of containers may increase.

[0003] Patent Document 1 describes a container terminal system that includes an AI (Artificial Intelligence) system that autonomously learns from past container terminal operation records and outputs control values ​​based on data related to container loading and unloading, and a terminal system. The control values ​​include, for example, a container storage plan and instructions for the timing and location of on-site trailers to store containers.

[0004] JP 2019-104578 A

[0005] The operation indicated by the control value in Patent Document 1 is the result of machine learning. Therefore, the operation indicated by the control value in Patent Document 1 will not be an operation different from operations performed in the past. Therefore, the technology in Patent Document 1 cannot obtain an operation with reduced cost compared to operations performed in the past.

[0006] One of the objects of the present disclosure is to provide a location calculation device, a location calculation method, and a storage medium that can provide a container location that reduces the cost of container operation.

[0007] a calculation unit that causes an information processing device to execute a placement determination process that modifies the Ising model using quantum annealing, wherein the containers consist of existing containers whose positions are fixed and a target container whose position is to be determined; the calculation unit that obtains the Ising model and an unloading order of the containers; and the placement determination process that determines the position of the target container by modifying the Ising model so that an evaluation value that decreases as the number of combinations of the container and another container that is unloaded after the container and placed above the container in the unloading order decreases, while satisfying constraints that the positions of the containers are different from each other and that one of the containers is placed at the possible position below the position of the container; and the output unit that outputs the position of the target container.

[0008] A placement calculation method according to one aspect of the present disclosure generates an Ising model for assigning a position of a container in a temporary storage location having one or more possible positions where the container can be placed to the possible positions, and causes an information processing device to execute a placement determination process that modifies the Ising model using quantum annealing, the containers being composed of existing containers whose positions are fixed and a target container whose position is to be determined, generates the Ising model indicating the positions of the existing containers, and obtains the Ising model and a carry-out order of the containers, the placement determination process is a process that determines the position of the target container by modifying the Ising model so that an evaluation value that decreases as the number of combinations of the container and another container that will be unloaded after the container and be placed above the container in the carry-out order decreases, while satisfying constraints that the positions of the containers are different from each other and that one of the containers is to be placed at the possible position below the position of the container, and outputs the position of the target container.

[0009] According to one aspect of the present disclosure, there is provided a storage medium for storing a program that causes a computer to execute a generation process of generating an Ising model for assigning a position of a container in a temporary storage location having one or more possible positions where the container can be placed to the possible positions, an arithmetic process of causing an information processing device to execute a placement determination process of modifying the Ising model using quantum annealing, and an output process, wherein the containers include existing containers whose positions are fixed and a target container whose position is to be determined, the generation process generates the Ising model indicating the positions of the existing containers, the arithmetic process acquires the Ising model and a carry-out order of the containers, and the placement determination process determines the position of the target container by modifying the Ising model so that an evaluation value that decreases as the number of combinations of the container and another container that is carried out after the container and placed above the container in the carry-out order decreases, while satisfying constraints that the positions of the containers are different from each other and that one of the containers is placed at the possible position below the position of the container, and the output process outputs the position of the target container.

[0010] Advantageously, the present disclosure provides a container arrangement that reduces the cost of operating the container.

[0011] FIG. 1 is a block diagram illustrating an example of the configuration of a location calculation device according to the present disclosure. FIG. 2 is a flowchart illustrating an example of the operation of a location calculation device according to the present disclosure. FIG. 3 is a block diagram illustrating an example of the configuration of a location calculation system according to the present disclosure. FIG. 4 is a flowchart illustrating an example of the operation of learning a prediction model of a location calculation system according to the present disclosure. FIG. 5 is a flowchart illustrating an example of the operation of learning a prediction model of a location calculation system according to the present disclosure. FIG. 6 is a flowchart illustrating an example of a process for determining a position of a location calculation system according to the present disclosure. FIG. 7 is a flowchart illustrating an example of a process for determining a position of a location calculation system according to the present disclosure. FIG. 8 is a flowchart illustrating an example of a process for determining a position of a location calculation system according to the present disclosure. FIG. 9 is a flowchart illustrating an example of a process for determining a position of a location calculation system according to the present disclosure. FIG. 10 is a flowchart illustrating an example of a process for determining a position of a location calculation system according to the present disclosure. FIG. 11 is a block diagram illustrating an example of the configuration of a location calculation system according to the present disclosure. FIG. 12 is a flowchart illustrating an example of the operation of a location calculation system according to the present disclosure. FIG. 13 is a flowchart illustrating an example of the operation of a location calculation system according to the present disclosure. FIG. 14 is a block diagram showing an example of the configuration of an allocation calculation system according to the present disclosure. FIG. 15 is a flowchart showing an example of the operation of the allocation calculation system according to the present disclosure. FIG. 16 is a flowchart showing an example of the operation of the allocation calculation system according to the present disclosure. FIG. 17 is a block diagram showing an example of the configuration of an allocation calculation system according to the present disclosure. FIG. 18 is a flowchart showing an example of the operation of receiving a weight in the allocation calculation system according to the present disclosure. FIG. 19 is a diagram schematically showing an example of a possible installation location. FIG. 20 is a diagram showing an example of an array showing the allocation of containers. FIG. 21 is a diagram showing an example of the hardware configuration of a computer capable of realizing the allocation calculation device, operation device, and prediction device according to the present disclosure.

[0012] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings.

[0013] First Embodiment First, a configuration of a first embodiment of the present disclosure will be described.

[0014] <Configuration> FIG. 1 is a block diagram illustrating an example of the configuration of a layout calculation device according to the present disclosure.

[0015] The configuration of a layout calculation device according to a first embodiment of the present disclosure will be described in detail with reference to Fig. 1. In the block diagram of the present disclosure, elements that exchange data are connected by solid lines. However, the combinations of elements that exchange data are not necessarily limited to the combinations of elements connected by solid lines.

[0016] In the example shown in FIG. 1 , the layout calculation device 10 according to this embodiment includes a generation unit 120 , a calculation processing unit 130 , and an output unit 140 .

[0017] The layout calculation device 10 according to the first embodiment of the present disclosure includes a generation unit 120 , a calculation processing unit 130 , and an output unit 140 .

[0018] The generator 120 generates an Ising model for assigning the location of a container in a temporary storage location having one or more possible locations where the container can be placed to the possible locations.

[0019] The calculation unit 130 causes the information processing device to execute a placement determination process that changes the Ising model using quantum annealing.

[0020] The containers include an existing container whose position is fixed and a target container whose position is to be determined.

[0021] The generation unit 120 generates the Ising model indicating the position of the existing container.

[0022] The calculation processing unit 130 acquires the Ising model and the container unloading order.

[0023] The placement determination process is a process of determining the position of the target container by modifying the Ising model so as to lower the evaluation value in the unloading order while satisfying constraint conditions.

[0024] The constraint is that the positions of the containers are different from each other and any of the containers is placed in the possible positions below the position of the container.

[0025] The evaluation value is a value that decreases as the number of combinations of the container and other containers that are carried out after the container and placed on top of the container decreases.

[0026] The output unit 140 outputs the position of the target container.

[0027] The above description of the layout calculation device 10 can also be rephrased as follows.

[0028] The generation unit 120 generates an Ising model that can indicate the position of a container in a temporary storage location for each possible position where the container can be placed, and that indicates the position of the container whose position is fixed.

[0029] The calculation processing unit 130 causes the information processing device to execute the placement determination process.

[0030] The placement determination process is a process using quantum annealing to determine the position of a target container, which is the container whose position is to be determined, from the Ising model and the container unloading order so as to satisfy constraints and optimize an evaluation value, and to modify the Ising model to indicate the position of the target container. The constraints are that the positions of the containers are different and that any of the containers is placed at the possible position below the position of the container. The evaluation value is based on the number of combinations of the container and other containers that are unloaded after the container and placed above the container.

[0031] Determining the position of a target container so as to optimize the evaluation value means, for example, determining the position of the target container so as to optimize the evaluation value represented by an objective function. Determining the position of a target container so as to optimize the evaluation value may also mean determining multiple combinations of target container positions such that the evaluation value represented by the objective function satisfies a goodness condition. The goodness condition may be, for example, that the evaluation value is better than a reference value (e.g., a threshold value that serves as a reference for the goodness of the evaluation value). The goodness condition may be, for example, that the evaluation value determined by the combination of the positions of the target containers is within a predetermined number of positions from the best (in other words, that the ranking of the evaluation value is a predetermined rank or higher). The goodness condition may be, for example, that the evaluation value is better than a reference value and, in addition, that the evaluation value determined by the combination of the positions of the target containers is within a predetermined number of positions from the best (in other words, that the ranking of the evaluation value is a predetermined rank or higher).

[0032] The information processing device is a quantum computer or a pseudo quantum computer. A pseudo quantum computer is a computer that simulates a quantum computer and performs calculations similar to those performed by a quantum computer. Quantum annealing includes simulated annealing. Simulated annealing is a calculation process similar to quantum annealing performed in a pseudo quantum computer. When the information processing device is a pseudo quantum computer, the information processing device performs the placement determination process using simulated annealing as quantum annealing. When the information processing device is a quantum computer, the information processing device performs the placement determination process using quantum annealing that is not simulated annealing.

[0033] The output unit 140 outputs the position of the target container.

[0034] For purposes of this disclosure, containers can be stacked on top of other containers, although the number of containers that can be stacked on top of one another is finite.

[0035] A temporary container storage location is a location where containers are not permanently installed, but where containers are loaded and unloaded. An example of a temporary container storage location is a container yard at a wharf. A temporary container storage location may also be a container storage area at a railway freight station, a container storage area at a truck transfer base, or the like. In a temporary container storage location, the locations where containers can be placed are predetermined.

[0036] A detailed example of this embodiment will be described later.

[0037] <Operation> FIG. 2 is a flowchart illustrating an example of the operation of the layout calculation device according to the present disclosure.

[0038] Hereinafter, the operation of the layout calculation device according to the first embodiment of the present disclosure will be described in detail with reference to FIG.

[0039] 2, the generation unit 120 generates an Ising model for allocating container positions to a temporary storage location having one or more possible positions where the container can be placed (step S11). Next, the calculation processing unit 130 causes the information processing device to execute a placement determination process that modifies the Ising model using quantum annealing so as to lower the evaluation value while satisfying the constraints (step S12). Then, the output unit 140 outputs the position of the target container (step S13).

[0040] <Effects> The present disclosure has the effect of providing a container placement that reduces the cost of container operation. This is because the generation unit 120 generates an Ising model that can indicate the location of a container and indicates the location of a container whose location is fixed. The calculation processing unit 130 then causes an information processing device to execute a placement determination process using quantum annealing, in which the calculation processing unit 130 determines the location of a target container based on the Ising model and the container unloading order so as to satisfy constraints and optimize the evaluation value, and modifies the Ising model to indicate the location of the target container.

[0041] This makes it possible to calculate how to place containers (cargo) in a way that reduces the cost of moving containers placed above the container being carried out when the container is carried out in the order in which it is carried out. In other words, it is possible to solve the combinatorial optimization problem of which containers should be placed in which positions.

[0042] If there is insufficient space for storing containers, containers that are to be removed later must be stacked on top of containers that are to be removed earlier. As a result, when a container is removed, the containers stacked on top of that container must be temporarily moved to another location. This operation is called a reordering operation. In order to minimize the number of container movements during the reordering operation, it is necessary to calculate a placement method that minimizes inconsistencies between the removal order and the stacking order. The objective function in this embodiment calculates the number of inconsistencies between the removal order and the stacking order, and therefore, by minimizing this objective function, a combinatorial optimization problem can be solved.

[0043] By reducing the number of times containers are moved during handling operations, it is possible to reduce the working time required for container operations and to minimize the energy consumption required to operate the cranes.

[0044] In such a combinatorial optimization problem, the number of combinations is enormous. Therefore, it is not easy to solve the combinatorial optimization problem within a practical time using a method that tries all possible combinations. If the calculation time (i.e., the search time) is limited to a realistic time (i.e., a time shorter than the time from when the container unloading order is obtained to when the containers are transported to the temporary storage location), an optimal solution is not likely to be obtained, and an approximate solution with low accuracy is likely to be obtained. If a method that relies on the experience and intuition of an expert is used, that is, if an expert determines the container placement based on experience and intuition, an optimal solution is not necessarily obtained. In this embodiment, it is expected that the cost of container operation will be reduced compared to these cases.

[0045] <Detailed Example of First Embodiment> In the present disclosure, for example, as the installation possible locations where a container can be installed, a plurality of installation possible locations (for example, I ０ ×J ０ The maximum number of containers that can be stacked in one installation location on the installation surface is determined based on the strength of the container, for example (for example, K ０ In other words, the maximum number of containers that can be placed on a container directly in one installation location defined on the installation surface is (K ０ -1) containers can be stacked. (K -1) containers on top of a container placed directly on the installation area defined on the installation surface. ０ -1) containers can be placed in the installation location. ０ ×J ０ ×K ０ containers can be placed in the temporary container storage area. ０ ×J ０ ×K ０ There are locations where the container can be temporarily stored. ０ , J ０ , K. ０ are natural numbers. Hereinafter, the position of a possible installation location will be referred to as a possible position. A possible position can be represented by a three-dimensional vector (i, j, k).

[0046] In the present disclosure, when a container is placed in one possible location, no other containers can be placed in that possible location. That is, the number of containers that can be placed in one possible location (in other words, one possible position) is one. In other words, the positions at which each container is placed (i.e., each position of the container) are different. Also, containers can only be placed on the installation surface of the temporary storage location or on top of other containers. In other words, other containers are placed in all possible locations below the possible location at which a container is placed. In other words, some container (i.e., some other container) is placed in each possible position below the location of the container. In the present disclosure, these (i.e., each position of the container is different and some container is placed in each possible position below the location of the container) are constraints that must be met.

[0047] An installable location below this installable location (referred to as the target location) is an installable location directly below the target location, i.e., an installable location whose horizontal position is the same as the horizontal position of the target location and whose vertical position is lower than the vertical position of the target location. The horizontal position is, for example, the component (e.g., (i, j)) of the position of the installable location in three-dimensional space that is parallel to the installation surface. The vertical position is, for example, the component (e.g., (k)) of the position of the installable location in three-dimensional space that is perpendicular to the installation surface.

[0048] The location of each container in such a temporary storage location can be represented in a three-dimensional voxel format. For example, the location of each container in the temporary storage location can be represented by a three-dimensional variable whose value is an ID (identifier) ​​uniquely assigned to the container. In this case, the value of a component indicating a possible location where a container is not installed is a value that is defined as a value representing that a container is not installed and is not used for the container ID.

[0049] Fig. 19 is a diagram schematically illustrating an example of a possible installation location. In the example shown in Fig. 19, a rectangular parallelepiped area represents a possible installation location. Furthermore, the number shown in the rectangular parallelepiped area represents the ID of the container placed in the possible installation location.

[0050] In the following description, the arrangement of the voxel-formatted containers described above is represented by a four-dimensional array x(i, j, k, p) consisting of three-dimensional positions (i, j, k) and one-dimensional container IDs (p). When the value x(i, j, k, p) of this array is 1, it indicates that a container with ID p is placed at position (i, j, k). When the value x(i, j, k, p) of this array is 0, it indicates that a container with ID p is not placed at position (i, j, k). In this example, the presence or absence of a container placed at each possible position is represented by a binary value of 0 or 1. The binary value representing the presence or absence of a container placed at a possible position is not limited to this example. The binary value representing the presence or absence of a container placed at a possible position may be another different binary combination, such as a combination of +1 and −1.

[0051] FIG. 20 is a diagram showing an example of an array representing the placement of containers. In the example shown in FIG. 20, i, j, and k on the horizontal axis represent the positions of possible installation locations. p on the vertical axis represents the ID of a container. In the example shown in FIG. 20, the position where a container with ID p is placed is represented by "1." In the example of FIG. 20, a cell without a "1" indicates that a container with ID p is not placed at the position indicated by (i, j, k). In the example of FIG. 20, a "0" indicating that no container is placed is omitted in a cell without a "1." The array shown in FIG. 20 represents an example where containers are placed as shown in FIG. 19.

[0052] The uniqueness of the container ID is expressed by the following one-hot constraint formula:

[0053] where i is 0 or greater ０ j is a natural number less than 0 ０ k is a natural number less than 0, and k is a natural number greater than or equal to 0. ０ Among the above constraints, the requirement that the positions of the containers be different (i.e., that only one container can be placed at one possible position) is expressed by the following one-hot constraint formula:

[0054] Here, p is a natural number greater than or equal to 0 and less than P0. ０ is the number of containers. Among the above constraints, the condition that a container is placed at each possible position below the container position (i.e., a container must exist at a possible position below a possible position where a container is placed) is expressed by the following equation.

[0055] In this case, the location of the container (i.e., the possible locations where the container is located) is (i １ , j １ , k １ ) The ID of the container is p １ If x(i １ , j １ , k １ , p １ ) = 1. For each k that is greater than or equal to 0 and less than k1, the sum shown in Equation 3 is 1.

[0056] When the arrangement of containers is represented by a four-dimensional array x(i, j, k, p), the expressions shown in Equations 1 to 3 can be said to be constraints.

[0057] The Ising model is a data array representing a four-dimensional array x(i, j, k, p) that represents the arrangement of containers. The data array representing the array x(i, j, k, p) does not necessarily have to be a four-dimensional array. For example, the data array representing the array x(i, j, k, p) can be expressed as a two-dimensional array x(i, j, k, p) that represents the array x(i, j, k, p). １ (q, p) (where q is 0 or more ０ ×J ０ ×K ０ In this case, the value q may be, for example, a natural number less than i. ０ ×j+I ０ ×J ０ × k. In this case, the Ising model is expressed as such a two-dimensional array x １ (q, p).

[0058] The generation unit 120 generates an array x １ The value of the element of (q, p) is set to 1, and an array x indicates the ID of the existing container and the position where the existing container is not placed. １ The value of the element of (q, p) is set to 0. In addition, the generation unit 120 generates an array x １ The value of the element (q, p) is set to 0. As a result, the generation unit 120 generates an Ising model that can indicate the position of the container in the temporary storage location for each possible position where the container can be placed, and that indicates the position of the existing container whose position is fixed. In this embodiment, a container other than the existing container is the target container.

[0059] For example, when a container is carried into a temporary storage location from a ship or the like, the existing container is a container that has already been placed in the temporary storage location before the container is carried into the temporary storage location from the ship or the like. The container carried into the temporary storage location from the ship or the like is the target container. Note that the target container may be a container selected from the containers placed in the temporary storage location, as in the embodiment described below. In this case, the existing container is a container that has not been selected from the containers placed in the temporary storage location.

[0060] If a container stacked below another container is removed from the temporary storage location before the container below it, the other container stacked above the container being removed must be moved to another available location where no containers are placed. By arranging the containers so that the removal order, which is the order in which the container stacked below the other containers is removed from the temporary storage location, is later than the removal order of the other containers stacked above it, the cost of container operation is reduced.

[0061] The cost of handling a container may be, for example, at least one of the time required to move a container, the number of operations required to move the container, and the cost of such operations when the container is brought in, relocated, and / or taken out. The time required to move a container depends, for example, on the distance traveled by the container and the number of operations required. The number of operations depends, for example, on the number of operations required. The cost of handling a container may be, for example, the energy cost and labor cost required to operate the container. The cost of handling a container depends, for example, on the distance traveled by the container and the number of operations required.

[0062] In this way, the cost of operating a container can be represented, for example, by the number of container movements required to unload the container. In this embodiment, the movement of a container is the movement of the container for unloading the container, i.e., the movement of all containers placed above the container to other possible locations in the temporary storage area in order to unload the container to be unloaded. The cost of moving a container depends on the number of combinations of the container and other containers placed above the container and whose unloading order is later than that of the container.

[0063] The calculation processing unit 130 causes the information processing device to execute a placement determination process that determines, using quantum annealing, the position of the target container such that the above-mentioned constraints are satisfied and the evaluation value based on the number of such combinations is minimized.

[0064] An evaluation value based on the number of such combinations is, for example, the number of combinations of a container and another container that is placed above the container and whose carrying-out order is later than that of the container.

[0065] The number of such combinations is expressed, for example, by the following formula:

[0066] where the variables p1 and p2 represent the IDs of the containers. The variable k1 represents the number of rows of possible positions. The value order(p1) is the order of removal of the container with ID p1, and the value order(p2) is the order of removal of the container with ID p2. In this formula, i is a number between 0 and I. ０ j is a natural number less than 0 ０ k is a natural number less than 0, and k is a natural number greater than or equal to 0. ０ p1 and p2 are natural numbers less than 0. ０ where k1 is a natural number greater than or equal to 0 and less than k. The function f1{diff} is a function that takes a value of 1 when the value of diff is positive, and takes a value of 0 when the value of diff is 0 or less.

[0067] In this embodiment, the container unloading order is known and is provided to the arrangement calculation device 10.

[0068] The evaluation value based on the number of combinations may be the sum of the differences in the carry-out order between a container and another container placed above that container and having a later carry-out order than that of the container. Such an evaluation value is expressed by the above-mentioned formula 4. In this case, however, the function f1{diff} is a function whose value is the value of diff when the value of diff is positive, and whose value is 0 when the value of diff is 0 or less.

[0069] The evaluation value based on the number of combinations may be the sum of values ​​that increase as the difference in the carry-out order between a container and another container placed above that container and whose carry-out order is later than that of the container is increased. Such an evaluation value is also expressed by the above-mentioned formula 4. However, in this case, the function f1{diff} is a function that, when the value of diff is positive, increases as diff increases, and becomes 0 when the value of diff is 0 or less. The function f1{diff} may be expressed, for example, by a constant multiple of the value of diff, a power of the value of diff, or a polynomial with the value of diff as a variable, within the range where the value of diff is positive.

[0070] The evaluation value based on the number of combinations may be expressed as the sum of the evaluation value based on the number of combinations described above and other evaluation values. The evaluation value based on the number of combinations may be expressed as a weighted sum of the evaluation value based on the number of combinations described above and other evaluation values. The weighted sum of multiple evaluation values ​​is the sum of multiple evaluation values ​​and the product of weights and the evaluation values, respectively. The function representing the evaluation value based on the number of combinations is an objective function that represents the degree of quality of the position of the target container in the placement determination process. In the above example, the smaller the value of the objective function, the better the position of the target container. In the description of the present disclosure, when the evaluation value is represented by the sum of multiple evaluation values ​​or the weighted sum of multiple evaluation values, each of the multiple evaluation values ​​is referred to as an element evaluation value. In other words, the evaluation value is represented by the sum of multiple element evaluation values ​​or the weighted sum of multiple element evaluation values. In the description of the present disclosure, each of the multiple equations representing the multiple evaluation values ​​is also referred to as an element objective function. In this case, the objective function is represented by the sum of multiple element objective functions or the weighted sum of multiple element objective functions.

[0071] <First Modification of First Embodiment> The cost of container operation may be associated with the distance between consecutive containers in the unloading order. When unloading containers using a truck, trailer, or the like, the truck, etc., unloading the container may be parked near the container being unloaded. In such a case, the sum of the distances between consecutive containers in the unloading order represents an approximate value of the travel distance of the crane (i.e., the horizontal travel distance of the tip of the crane). If the location where the container to be unloaded is loaded onto a transport device such as a truck is far from the location where the container is placed, the distance between consecutive containers in the unloading order is the travel distance between the positions where the crane starts unloading the container (i.e., the position where the crane lifts the container) between consecutive containers in the unloading order. In this case, too, the smaller the sum of the distances between consecutive containers in the unloading order, the easier the work of the crane operator moving the containers may be.

[0072] The sum of the distances between successive containers in the unloading order is expressed by the following formula:

[0073] In this formula, the function f2(diff) is a function that takes a value of 1 when the value of diff is 1, and takes a value of 0 when the value of diff is other than 1. The values ​​of variables i1 and i2 are 0 or more and ０ The values ​​of variables j1 and j2 are 0 or more and J ０ The values ​​of variables k1 and k2 are 0 to K ０ The values ​​of variables p1 and p2 are 0 or more and P ０ The function distance(i1, j1, k1, i2, j2, k2) represents the distance between the position represented by (i1, j1, k1) and the position represented by (i2, j2, k2).

[0074] In this modification, the objective function is, for example, the sum of an element objective function expressed by Expression 4 and an element objective function expressed by Expression 5. The evaluation value is the sum of the value of the element objective function expressed by Expression 4 (i.e., element evaluation value) and the value of the element objective function expressed by Expression 5 (i.e., element evaluation value).

[0075] The formula shown in Equation 5 may represent a value that depends on the distance between containers other than the sum of the distances between containers whose carrying-out order is consecutive. For example, the function f2(diff) may be a function whose value is the square of the value of diff. The function f2(diff) may be a function whose value is the square of diff when the value of diff is positive, and whose value is 0 when the value of diff is not positive.

[0076] <Second Modification of First Embodiment> The cost of operating a container can also be expressed, for example, by the distance traveled by the container when at least one of the containers is carried in and out. The distance traveled by the container when at least one of the containers is carried in and out is correlated with the sum of the distance between a predetermined position and the position where the container is placed. This predetermined position is, for example, the position of a representative point of a ship or the like that carries the container when the container is carried from a ship or the like to a temporary storage location, or the position of a representative point of a quay where the ship or the like that carries the container to be carried in docks. The representative point may be a point that is determined as appropriate.

[0077] If a location where the container to be removed is placed on a transport device such as a truck or freight train that transports the container from the temporary storage location, the above-mentioned specified location may be a location determined as a representative location where the container to be removed is placed.

[0078] The sum of the distances between the predetermined position and the position where the container is placed is expressed by the following formula:

[0079] In this formula, (i0, j0, k0) indicates a predetermined position (also referred to as a reference position). As described above, this predetermined position may be the position of a representative point of a ship or the like that will carry the container when the container is carried from the ship or the like to the temporary storage location, or the position of a representative point of a quay where the ship or the like that will carry the container to be carried in will dock (hereinafter referred to as a carry-in reference position). This predetermined position may also be a position that is determined as a representative position where a container that will be carried out will be placed (hereinafter referred to as an unloading reference position).

[0080] In this modification, the objective function is, for example, the sum of the element objective function expressed by Expression 4 and the element objective function expressed by Expression 6. The evaluation value is the sum of the value of the element objective function expressed by Expression 4 (i.e., the element evaluation value) and the value of the element objective function expressed by Expression 6 (i.e., the element evaluation value).

[0081] The objective function may be, for example, the sum of an element objective function expressed by Expression 4, an element objective function expressed by Expression 5, and an element objective function expressed by Expression 6. In this case, the evaluation value is the sum of the value of the element objective function expressed by Expression 4 (i.e., element evaluation value), the value of the element objective function expressed by Expression 5 (i.e., element evaluation value), and the value of the element objective function expressed by Expression 6 (i.e., element evaluation value).

[0082] When the value of (i0, j0, k0) is a value indicating the carry-in reference position, it can be said that twice the value of the element objective function expressed by equation 6 is an approximate value of the crane travel distance when the target container is carried in. When the value of (i0, j0, k0) is a value indicating the carry-in reference position, it can be said that twice the value of the element objective function expressed by equation 6 is an approximate value of the crane travel distance when the target container is carried out. However, in this case, the crane travel distance for remand is not reflected in the approximate value of the crane travel distance.

[0083] The element objective function expressed by Equation 6 may be the sum of OF3 when the value of (i0, j0, k0) indicates the carry-in reference position and OF3 when the value of (i0, j0, k0) indicates the carry-out reference position. In this case, the value of the element objective function expressed by Equation 6 (i.e., the element evaluation value) is the sum of the value of OF3 when the value of (i0, j0, k0) indicates the carry-in reference position and the value of OF3 when the value of (i0, j0, k0) indicates the carry-out reference position. In this case, the element evaluation value can be said to be an approximate value of the crane travel distance when carrying in and out the target container. However, in this case, the crane travel distance for handling is not reflected in the approximate value of the crane travel distance.

[0084] Second Embodiment Next, a layout calculation system according to a second embodiment of the present disclosure will be described in detail with reference to the drawings.

[0085] <Configuration> FIG. 3 is a block diagram illustrating an example of the configuration of a layout calculation system according to the present disclosure.

[0086] The configuration of the allocation calculation system according to the second embodiment of the present disclosure will be described in detail below with reference to Fig. 3. In the example shown in Fig. 3, the allocation calculation system includes an allocation calculation device 100, an operation device 200, and a prediction device 300.

[0087] The layout calculation device 100 includes a sequence calculation unit 110 , a generation unit 120 , a calculation processing unit 130 , an output unit 140 , a calculation information storage unit 150 , and a movement planning unit 160 .

[0088] The operation device 200 includes a data accumulation unit 210 , a target container identification unit 220 , a container state identification unit 230 , and a UI (User Interface) unit 240 .

[0089] The prediction device 300 includes a machine learning unit 310 , a model storage unit 320 , an information extraction unit 330 , and a prediction unit 340 .

[0090] <Operation Device 200> The operation device 200 is a device for operating one or more temporary storage locations.

[0091] <Data storage unit 210> The data storage unit 210 stores container transfer information, which is information about containers that have been brought into and taken out of temporary storage locations in the past. In other words, the data storage unit 210 stores container transfer information about containers that have been brought into and taken out of each temporary storage location in the past. The data storage unit 210 stores the container transfer information for each temporary storage location. The container transfer information includes a container profile and container schedule information.

[0092] The container profile is information such as the container number (e.g., ID), the container type, the container's delivery origin information, the shipper, refrigeration information indicating whether the container is refrigerated, and the delivery company that will remove the container from the temporary storage location. The delivery origin information is a pre-specified type of information about the location to which the container is shipped (e.g., at least one of information about the country, region, and port). The delivery origin information may include information about the sender who sent the container. The container profile may also include container destination information. The container destination information is, for example, a pre-specified type of information about the location to which the container is delivered (e.g., the region). The container destination information may also include information about the recipient who will receive the container. The container profile is not limited to the above examples. The container profile may include information other than the information exemplified above. The container profile may not include at least some of the information exemplified above.

[0093] The container schedule information includes information on the delivery date and time, which is the date and time when a container is delivered to a temporary storage location, and information on the delivery date and time, which is the date and time when a container is delivered from a temporary storage location. Date and time information such as the delivery date and time information and the delivery date and time information includes information such as the date, time, day of the week, and whether or not it is a public holiday. The delivery date and time may be the date and time when a transport device that transports a container arrives at a temporary storage location. The delivery date and time may be the date and time when a container is actually delivered from the transport device that transports the container to the temporary storage location. The container schedule information is not limited to the information exemplified above. The container schedule information may include information other than the information exemplified above. The container schedule information may not include at least some of the information exemplified above.

[0094] The container transportation information stored in the data storage unit 210 is read out by the machine learning unit 310 of the prediction device 300.

[0095] <Target Container Identification Unit 220> The target container identification unit 220 identifies a target container. In this embodiment, the target container is a container that is transported from a transport device such as a ship that transports the container to a temporary storage location. The target container identification unit 220 receives, for example, from an information processing device of the transport device such as a ship that transports the target container, information on the temporary storage location to which the target container is transported from the transport device and information on the target container that is transported from the transport device to the temporary storage location. This target container information includes, for example, the ID of the target container, profile information of the target container, and information on the delivery date and time that the target container is delivered to the temporary storage location. The delivery date and time information is, for example, the date and time that the target container is scheduled to be delivered to the temporary storage location. The target container identification unit 220 identifies the target container, for example, by extracting the ID of the target container from the received information on the target container.

[0096] The target container identification unit 220 transmits information about the temporary storage location where the target container will be carried in, the ID of the target container, profile information about the target container, and information about the date and time of carrying in of the target container to the information extraction unit 330 of the prediction device 300. The information about the temporary storage location where the target container will be carried in is, for example, information that identifies the temporary storage location (e.g., ID, etc.).

[0097] The target container identification unit 220 sends the received information on the temporary storage location where the target container will be transported (i.e., the temporary storage location where the target container will be placed) to the container status identification unit 230.

[0098] <Container Status Identification Unit 230> The container status identification unit 230 identifies the location of an existing container, which is a container that has already been placed in the temporary storage location where the target container will be placed and whose position in the temporary storage location has been determined. In this embodiment, the temporary storage location where the target container will be placed is the temporary storage location identified by the target container identification unit 220. The container status identification unit 230 extracts, for example, information about the containers placed in the temporary storage location where the target container will be placed, as information about the existing container, from a database that stores information about containers placed in temporary storage locations. The information about the existing container extracted by the container status identification unit 230 includes the IDs of the containers placed in the temporary storage location where the target container will be placed, profiles of those containers, and information about the locations where those containers are placed.

[0099] The container status identification unit 230 may identify the location of the existing container by, for example, extracting from a database the ID of the existing container that is located in the temporary storage location where the target container is located. The container status identification unit 230 may further read, for example, from a database that stores the profile of the existing container whose ID was extracted and information on the date and time when the existing container was brought into the temporary storage location.

[0100] The container state identification unit 230 transmits the ID of the existing container, the profile of the existing container, and information on the delivery date and time of the existing container to the information extraction unit 330 of the prediction device 300.

[0101] The container state identification unit 230 transmits information about the positions of the existing containers to the generation unit 120 of the arrangement calculation device 100. Note that the information about the positions of containers (e.g., the target container and the existing container) is, for example, a combination of the ID of the container and information indicating the position where the container is placed. The information about the position of the existing container in the temporary storage location is, for example, a combination of the ID of the existing container and information indicating the position where the existing container is placed in the temporary storage location.

[0102] <UI Unit 240> In this embodiment, the UI unit 240 receives the position of the target container (specifically, information on the position of the target container). The UI unit 240 outputs the received position of the target container. The UI unit 240 displays the received position of the target container on the display of the operation device 200, for example.

[0103] <Prediction Device 300> The prediction device 300 is a device that learns a prediction model that predicts the unloading date and time of a container for each temporary storage location, and predicts the unloading date and time of a target container using the prediction model.

[0104] <Machine Learning Unit 310> The machine learning unit 310 uses the container transportation information stored in the data accumulation unit 210 as learning data to train a prediction model that predicts the unloading date and time of a container from the container profile and the loading date and time, using an existing machine learning method. The machine learning unit 310 may train the prediction model for each temporary storage location. The machine learning unit 310 may train the prediction model so that the prediction model predicts the period from the loading date and time of the container to the unloading date and time of the container as the unloading date and time of the container.

[0105] The machine learning unit 310 may configure a prediction model to output, in addition to information indicating the container removal date and time, the prediction accuracy of the information indicating the removal date and time. The prediction accuracy may be expressed, for example, by a probability distribution with the date and time as a variable. The prediction accuracy may be expressed, for example, by a half-width when it is assumed that the probability distribution with the date and time as a variable is represented by a normal distribution.

[0106] The machine learning unit 310 stores the prediction model obtained by learning (specifically, data representing the prediction model) in the model storage unit 320 .

[0107] <Model Storage Unit 320> The model storage unit 320 stores the prediction model (specifically, data representing the prediction model) obtained through learning by the machine learning unit 310. The model storage unit 320 may store a prediction model for each temporary storage location.

[0108] <Information Extraction Unit 330> The information extraction unit 330 receives information on the temporary storage location where the target container is to be carried in, the ID of the target container, profile (also referred to as characteristic information) information on the target container, and information on the date and time of carrying in of the target container from the target container identification unit 220. The information extraction unit 330 receives information on the temporary storage location, the ID of an existing container whose position in the temporary storage location is determined, the profile of the existing container, and information on the date and time of carrying in of the existing container from the container status identification unit 230.

[0109] The information extraction unit 330 extracts information used to predict the container removal date and time from the container (here, the target container and existing container) profile and information on the arrival date and time.

[0110] The information extraction unit 330 sends information about the temporary storage location where the target container will be carried in, and information about the ID, profile, and carry-in date and time of the container (here, the target container and the existing container) to the prediction unit 340. The information about the profile and carry-in date and time sent to the prediction unit 340 may be the information about the profile and carry-in date and time extracted as information used to predict the carry-out date and time of the container.

[0111] <Prediction unit 340> The prediction unit 340 receives information on the temporary storage location where the target container will be carried in, and information on the ID, profile, and delivery date and time of the container (here, the target container and the existing container) from the information extraction unit 330. The information on the profile and delivery date and time received by the prediction unit 340 may be the information on the profile and delivery date and time extracted as information used to predict the delivery date and time of the container.

[0112] The prediction unit 340 predicts the removal date and time of each container (here, the target container and the existing container) using a prediction model trained on the temporary storage location to which the target container is to be delivered. As described above, this prediction model is a prediction model that predicts the removal date and time of the container based on the container profile and the delivery date and time of the container. As described above, the prediction model may predict the time from the delivery date and time of the container to the delivery date and time of the container (hereinafter referred to as elapsed time) based on the container profile and the delivery date and time of the container. In this case, the prediction unit 340 calculates the predicted removal date and time of the container based on the delivery date and time of the container and the predicted elapsed time.

[0113] The prediction unit 340 transmits the predicted removal date and time (specifically, information on the removal date and time) of each container (i.e., the target container and the existing container) to the sequence calculation unit 110 of the layout calculation device 100. The information on the predicted removal date and time of each container is, for example, a combination of the ID and information on the predicted removal date and time of each container.

[0114] <Layout calculation device 100> The generation unit 120, calculation processing unit 130, and output unit 140 of the layout calculation device 100 of this embodiment are the same as the generation unit 120, calculation processing unit 130, and output unit 140 of the first embodiment, respectively, except for the points described below. The generation unit 120, calculation processing unit 130, and output unit 140 of the layout calculation device 100 of this embodiment operate in the same way as the generation unit 120, calculation processing unit 130, and output unit 140 of the first embodiment, respectively, except for the points described below.

[0115] <Sequence Calculation Unit 110> The sequence calculation unit 110 receives the predicted removal date and time (specifically, information on the removal date and time) of each container (i.e., the target container and the existing container) from the prediction unit 340 of the prediction device 300.

[0116] The order calculation unit 110 calculates the order in which the containers (i.e., the target container and the existing containers) will be unloaded from the predicted unloading date and time of each container (i.e., the target container and the existing containers). The order calculation unit 110 sorts the containers (specifically, container IDs) in order of the earliest predicted unloading date and time, for example.

[0117] The sequence calculation unit 110 sends the unloading sequence of the containers (i.e., the target container and the existing containers) to the generation unit 120. The container unloading sequence is information indicating the sequence in which the containers will be unloaded from the temporary storage location, i.e., information indicating the result of sorting the containers in order of earliest predicted unloading date and time. The unloading sequence may be, for example, an array of container IDs sorted in the sequence in which the containers will be unloaded from the temporary storage location. The unloading sequence may be, for example, a combination of the ID of each container and information indicating the sequence in which the containers will be unloaded from the temporary storage location.

[0118] <Generation unit 120> The generation unit 120 receives information on the positions of existing containers from the container state identification unit 230. The generation unit 120 receives the carry-out order of the containers (i.e., the target container and existing containers) from the order calculation unit 110. The generation unit 120 reads out the constraint conditions and the objective function for calculating the above-mentioned evaluation value (specifically, information representing the constraint conditions and information representing the objective function) from the calculation information storage unit 150.

[0119] The generation unit 120 generates an Ising model from the information on the positions of the existing containers as described above.

[0120] The generation unit 120 sends the unloading order of the containers (i.e., the target container and the existing container), the constraint conditions, the objective function, and the Ising model to the calculation processing unit 130. Specifically, the generation unit 120 sends information representing the unloading order of the containers, information representing the constraint conditions, information representing the objective function, and the Ising model to the calculation processing unit 130.

[0121] The objective function in this embodiment is, for example, any of the objective functions described in the detailed example of the first embodiment, the first modified example of the first embodiment, and the second modified example of the first embodiment.

[0122] <Calculation Information Storage Unit 150> The calculation information storage unit 150 stores constraint conditions and objective functions (specifically, information representing the constraint conditions and information representing the objective functions).

[0123] <Calculation processing unit 130> The calculation processing unit 130 receives the unloading order of containers (i.e., the target container and the existing containers), constraint conditions, an objective function, and an Ising model from the generation unit 120. Specifically, the calculation processing unit 130 receives information representing the unloading order of containers, information representing the constraint conditions, information representing the objective function, and the Ising model from the generation unit 120.

[0124] The calculation processing unit 130 causes the information processing device to execute the above-mentioned placement determination process using quantum annealing, using the order of removal of containers (i.e., the target container and existing containers), constraints, an objective function, and an Ising model.

[0125] The calculation processing unit 130 acquires information on the Ising model after the placement determination process has been performed from the information processing device. The calculation processing unit 130 identifies information on the position of the target container in the acquired Ising model.

[0126] The calculation processing unit 130 sends the obtained information on the location of the target container to the output unit 140.

[0127] The calculation processing unit 130 sends information on the positions of the existing containers and the target container to the movement planning unit 160.

[0128] <Movement Planner 160> The movement planner 160 receives information on the positions of existing containers and information on the position of a target container from the processor 130.

[0129] The movement plan unit 160 calculates a movement plan for the target container to be brought into the temporary storage location, in order to place the target container at the location indicated by the location information of the target container. The movement plan for the target container includes combinations of source and destination locations for the target container to be brought into the temporary storage location, and the order in which the combinations will be moved.

[0130] The target container identification unit 220 may transmit information indicating the order in which the target containers will be removed from, for example, a transport device to the movement planner 160. In this case, the movement planner 160 calculates, using an existing calculation method, a movement plan for the target containers when the target containers are brought into the temporary storage location in the order in which they will be removed from, for example, a transport device.

[0131] In the following description, another container that is to be delivered after the container to be delivered and that is to be placed at a position below the position where the container to be delivered is referred to as a "reversed-order container." Note that a "position below the position where the container is to be delivered" refers to a position whose horizontal position is the same as the horizontal position of the position where the container is to be delivered and whose vertical position is lower than the vertical position of the position where the container is to be delivered. If there is no reversed-order container for the container to be delivered, the movement planning unit 160 generates a movement plan, for example, to move the container to be delivered directly from the conveying device to the position where the container to be delivered is to be delivered. Furthermore, if there is a reversed-order container for the container to be delivered, the movement planning unit 160 identifies, for example, a position (referred to as a "temporary storage position") where the container to be delivered can be placed until all the reversed-order containers have been placed. Then, the movement planning unit 160 identifies the temporary storage position that is closest to the position where the container to be delivered is to be placed, among the possible temporary storage positions. The temporary storage position for a container is a position where no other containers will be placed from the time the container to be brought in is brought in until all of the reversal-order containers for the container to be brought in have been placed. In this case, the movement planning unit 160 generates a movement plan such that, for example, the container to be brought in is moved from the conveying device to the specified position, and after all of the reversal-order containers for the container to be brought in have been placed, the container to be brought in is moved from the specified position to the position where the container to be brought in will be placed.

[0132] <Output Unit 140 > The output unit 140 outputs information on the location of the target container to the UI unit 240 of the operation device 200 .

[0133] <Operation> Next, the operation of the layout calculation system according to the second embodiment of the present disclosure will be described in detail with reference to the drawings.

[0134] FIG. 4 is a flowchart illustrating an example of an operation for learning a prediction model of the layout calculation system according to the present disclosure.

[0135] First, an operation of learning a prediction model of the allocation calculation system according to the second embodiment of the present disclosure (specifically, an operation of learning a prediction model of the prediction device 300) will be described in detail with reference to FIG. 4 .

[0136] 4 , first, the machine learning unit 310 reads from the data accumulation unit 210 of the operation device 200 the records of the profile and the date and time of delivery of a container that has been previously delivered to and delivered from a temporary storage location (step S101). The machine learning unit 310 uses the records of the container profile and the date and time of delivery of the container to learn a prediction model that predicts the date and time of delivery of a container from the container profile and the date and time of delivery (step S102). The machine learning unit 310 then stores the prediction model obtained by learning in the model storage unit 320. The model storage unit 320 stores the prediction model (step S103).

[0137] FIG. 5 is a flowchart illustrating an example of an operation for learning a prediction model of the layout calculation system according to the present disclosure.

[0138] Hereinafter, the operation of predicting the container removal date and time of the arrangement calculation system (specifically, the prediction device 300) according to the second embodiment of the present disclosure will be described in detail with reference to FIG. 5 .

[0139] In the example shown in FIG. 5 , the information extraction unit 330 acquires the profile of an existing container and information on the delivery date and time from the container status identification unit 230 of the operation device 200 (step S111). The information extraction unit 330 acquires the profile of a target container and information on the delivery date and time from the target container identification unit 220 (step S112). The information extraction unit 330 extracts the profile and information on the delivery date and time to be used for prediction from the acquired information on the profiles and delivery dates of the containers (i.e., the existing container and the target container) (step S113). In other words, for each container, the information extraction unit 330 extracts the portion of the acquired profile to be used for prediction and the portion of the acquired information on the delivery date and time to be used for prediction. When all portions of the acquired profile are to be used for prediction, the information extraction unit 330 extracts all of the acquired profile as the portion to be used for prediction (i.e., the profile to be used for prediction). If all parts of the acquired information on the delivery date and time are to be used for prediction, the information extraction unit 330 extracts all of the acquired information on the delivery date and time as the part to be used for prediction (i.e., the information on the delivery date and time to be used for prediction).

[0140] Next, the prediction unit 340 uses the prediction model to predict the container removal date and time from the container profile and the arrival date and time (step S114). The prediction unit 340 predicts the removal date and time for each container (i.e., the existing container and the target container). As described above, the prediction unit 340 (in other words, the prediction model) outputs the elapsed time from the arrival date and time to the removal date and time as information indicating the predicted removal date and time. The prediction unit 340 may calculate the removal date and time from the arrival date and time and the predicted elapsed time from the arrival date and time to the removal date and time. The prediction unit 340 outputs the container removal date and time obtained by prediction using the prediction model (step S115).

[0141] FIG. 6 is a flowchart illustrating an example of a process for determining a position in the location calculation system according to the present disclosure.

[0142] Hereinafter, a process of determining a position in the location calculation system (specifically, the location calculation device 100) according to the second embodiment of the present disclosure will be described in detail with reference to FIG. 6 .

[0143] 6, first, the sequence calculation unit 110 receives information on the unloading date and time of the containers (step S121). Specifically, the sequence calculation unit 110 receives information on the unloading date and time of each container (i.e., the existing container and the target container). The sequence calculation unit 110 calculates the unloading sequence of the containers (step S122).

[0144] The generation unit 120 receives information about the positions of the existing containers (step S123), and then generates an Ising model as described above (step S124).

[0145] The calculation processing unit 130 causes the information processing device to execute a placement determination process using quantum annealing to determine the position of the target container so as to satisfy the constraints and optimize the evaluation value (step S125). As a result, the Ising model indicates the determined position of the target container.

[0146] Then, the output unit 140 outputs the position of the target container (step S126).

[0147] <Effects> This embodiment has the same effects as the first embodiment, for the same reasons as those for the effects of the first embodiment.

[0148] <First Modification of Second Embodiment> A first modification of the second embodiment of the present disclosure is the same as the second embodiment except for the differences described below.

[0149] <Configuration> An information processing device that executes a placement determination process using quantum annealing can calculate, through the placement determination process, not only the position of a target container with the best evaluation value (specifically, a combination of the positions of multiple target containers), but also the position of a target container with a lower evaluation value. Hereinafter, the position of a target container (specifically, a combination of the positions of multiple target containers) will be referred to as a placement plan for the target container.

[0150] In this modification, the calculation processing unit 130 causes the information processing device to execute a placement determination process using quantum annealing to determine a placement plan that is a position of a target container that satisfies the constraints and whose evaluation value satisfies a criterion (hereinafter also referred to as the evaluation criterion). The number of placement plans to be determined is one or more. The placement plans that are positions of target containers whose evaluation value satisfies the criterion (i.e., placement plans whose evaluation value satisfies the criterion) are, for example, a predetermined number of placement plans in descending order of the evaluation value. The placement plan whose evaluation value satisfies the criterion may be a placement plan whose evaluation value is better than a predetermined evaluation value. The placement plan whose evaluation value satisfies the criterion may be a placement plan that satisfies at least one of the following: the ranking of the evaluation value is within a predetermined rank and the evaluation value is better than a predetermined evaluation value.

[0151] <Operation> FIG. 7 is a flowchart illustrating an example of a process for determining a position in the layout calculation system according to the present disclosure.

[0152] Hereinafter, a process of determining a position in the location calculation system (specifically, the location calculation device 100) according to the first modified example of the second embodiment of the present disclosure will be described in detail with reference to FIG.

[0153] In the example shown in FIG. 7, the operations from step S131 to step S134 are the same as the operations from step S121 to step S124 shown in FIG.

[0154] The calculation processing unit 130 causes the information processing device to execute a placement determination process using quantum annealing to determine a placement plan that is a position of a target container that satisfies the constraint conditions and the evaluation value satisfies the criterion (also referred to as the evaluation criterion) (step S125).

[0155] Then, the output unit 140 outputs the determined placement plan (step S126).

[0156] <Second Modification of Second Embodiment> As a second modification of the second embodiment of the present disclosure, an operation when the movement planner 160 calculates the movement plan for the container described above will be described.

[0157] <Operation> FIGS. 8 and 9 are flowcharts illustrating an example of a process for determining a position in the location calculation system according to the present disclosure.

[0158] Hereinafter, a process of determining a position in a location calculation system (specifically, location calculation device 100) according to a second modified example of the second embodiment of the present disclosure will be described in detail with reference to FIGS. 8 and 9 .

[0159] In the example shown in FIGS. 8 and 9, the operations from step S141 to step S146 are the same as the operations from step S121 to step S126 shown in FIG.

[0160] 9, after the operation of step S146, the movement planner 160 calculates a movement plan for the container (step S147), and the output unit 140 outputs the movement plan for the container (step S148).

[0161] The layout calculation apparatus 100 may perform the operation of step S146 and the operations of steps S147 and S148 in an order different from that shown in the example of Fig. 9. For example, the layout calculation apparatus 100 may perform the operation of step S146 after the operation of step S147. The layout calculation apparatus 100 may perform the operation of step S146 in parallel with the operation of step S147.

[0162] Third Embodiment A third embodiment of the present disclosure will be described in detail below with reference to the drawings. This embodiment is an embodiment that further takes into consideration the loading state of a ship.

[0163] <Configuration> A location calculation system according to a third embodiment of the present disclosure will be described in detail using Fig. 2. The location calculation system according to the third embodiment of the present disclosure is the same as the location calculation system according to the second embodiment, except for the differences described below. The components of the location calculation system according to the third embodiment of the present disclosure are the same as the components of the location calculation system according to the second embodiment that have the same names and reference numerals, except for the differences described below.

[0164] <Target container identification unit 220> The target container identification unit 220 receives information on the location of the target container on the transport device on which the target container is loaded (in other words, the loading location of the target container on the transport device) from, for example, an information processing device of a transport device such as a ship that transports the target container.

[0165] The target container identification unit 220 sends information on the loading position of the target container on the transport device to the container status identification unit 230. As will be described later, the container status identification unit 230 transmits information on the loading position of the target container on the transport device to the generation unit 120. In other words, the target container identification unit 220 transmits information on the loading position of the target container on the transport device to the generation unit 120 via the container status identification unit 230.

[0166] <Container status identification unit 230> The container status identification unit 230 identifies information on the loading position of the target container on a transport device (e.g., a ship). Specifically, the container status identification unit 230 receives information on the loading position of the target container on a transport device via the target container identification unit 220 from an information processing device of the transport device, such as a ship, that transports the target container.

[0167] The container state identification unit 230 transmits the received information on the loading position of the target container on the transport device to the generation unit 120 of the arrangement calculation device 100.

[0168] <Calculation Information Storage Unit 150> The calculation information storage unit 150 stores an objective function based on the loading position of the target container on the transport device. The objective function based on the loading position of the target container on the transport device is, for example, a function that represents the number of combinations of a container and another container that is placed below the target container in the temporary storage location and is placed on a lower level than the target container on the transport device. In other words, such combinations are combinations of two target containers that have the same horizontal loading position on the ship and the same horizontal position in the temporary storage location, and in which the vertical relationship of the loading positions and the vertical relationship of the positions in the temporary storage location are the same. Such combinations are referred to as same-order loading combinations. In other words, this objective function represents the number of same-order loading combinations. The objective function based on the loading position of the target container on the transporting device is the sum of, for example, the objective functions described in the explanations of the detailed example of the first embodiment, the first modified example of the first embodiment, and the second modified example of the first embodiment, and the objective function expressed by the following equation (in other words, the element objective function). The following equation represents the number of combinations of a container and another container that is placed below that container in the temporary storage location and on a lower level than that container in the transporting device.

[0169] In the formula shown in Expression 7, ship(p) indicates the number of steps from the lowest position at the position where the container with ID p is placed in a transport device, for example, a ship, where the container with ID p is placed. The number of steps from the lowest position is, for example, the number of steps when the number of steps at the lowest position is 0. Note that the number of steps at the lowest position may also be 1.

[0170] <Generation Unit 120> The generation unit 120 of this embodiment is the same as the generation unit 120 of the second embodiment, except for the differences described below and the fact that the objective function is the above-mentioned objective function.

[0171] The generation unit 120 receives information on the loading position of the target container on the transport device from the container state identification unit 230. The generation unit 120 also sends the information on the loading position of the target container on the transport device to the calculation processing unit 130.

[0172] <Calculation Processing Unit 130> The calculation processing unit 130 receives information on the loading position of the target container on the transport device from the generation unit 120.

[0173] The calculation processing unit 130 is the same as the generation unit 120 of the second embodiment, except for the differences described above, the fact that the objective function is the above-mentioned objective function, and the fact that the evaluation value is calculated further using information on the loading position of the target container on the transport device.

[0174] <Operation> FIG. 10 is a flowchart illustrating an example of a process for determining a position in the location calculation system according to the present disclosure.

[0175] Hereinafter, a process of determining a position in the location calculation system (specifically, the location calculation device 100) according to the third embodiment of the present disclosure will be described in detail with reference to FIG.

[0176] In the example shown in FIG. 10, the operations in steps S201 and S202 are the same as the operations in steps S121 and S122 shown in FIG.

[0177] In step S203, the generation unit 120 receives information on the loading position of the target container on the ship and information on the positions of the existing containers (step S203).

[0178] The operations from step S204 to step S206 in Fig. 10 are the same as the operations from step S124 to step S126 in Fig. 6. However, in this embodiment, the calculation processing unit 130 further uses information on the loading position of the target container on the ship in the operation of step S205.

[0179] <Effects> This embodiment has the same effects as the first embodiment, for the same reasons as those for the effects of the first embodiment.

[0180] Fourth Embodiment A fourth embodiment of the present disclosure will be described in detail below with reference to the drawings.

[0181] <Configuration> FIG. 11 is a block diagram illustrating an example of the configuration of a layout calculation system according to the present disclosure.

[0182] An allocation calculation system according to the fourth embodiment of the present disclosure will be described in detail below with reference to FIG. 11 . In the example shown in FIG. 11 , the allocation calculation system includes an allocation calculation device 101, an operation device 201, and a prediction device 300. In the example shown in FIG. 11 , the allocation calculation device 101 includes the components of the allocation calculation device 100 according to the second and third embodiments, and a cost calculation unit 170. The operation device 201 according to this embodiment includes the same components as the components of the operation device 200 according to the second and third embodiments. However, in the example shown in FIG. 11 , a target container identification unit 220 and a UI unit 240 are connected. The prediction device 300 according to this embodiment is the same as the prediction device 300 according to the second and third embodiments.

[0183] In this embodiment, the target container may be a container that is newly carried into a temporary storage location from a transportation device such as a ship. In this case, the allocation calculation device 101, the operation device 200, and the prediction device 300 of this embodiment can perform the same operations as the allocation calculation device 100, the operation device 200, and the prediction device 300 of the second embodiment, respectively. In this case, the allocation calculation device 101, the operation device 200, and the prediction device 300 of this embodiment can also perform the same operations as the allocation calculation device 100, the operation device 200, and the prediction device 300 of the third embodiment, respectively.

[0184] In this embodiment, the target container may be a container designated by a container installed at a temporary storage location. The following describes a case where the target container is a container designated by a container installed at a temporary storage location. The following describes differences between this embodiment, in which the target container is a container designated by a container installed at a temporary storage location, and the second embodiment. This embodiment is the same as the second embodiment except for the differences described below. Note that this embodiment may be the same as the third embodiment except for the differences described below.

[0185] <Operation device 201> As described above, the operation device 201 of this embodiment is the same as the operation device 200 of the second embodiment, except for the differences described below. Below, differences between the target container identification unit 220 and the container status identification unit 230 of this embodiment and the target container identification unit 220 and the container status identification unit 230 of the second embodiment will be described.

[0186] <Target container identification unit 220> The target container identification unit 220 receives designated container information, which is information indicating a finite number of containers designated from among the containers placed in the temporary storage location, via the UI unit 240. The target container identification unit 220 sets the container indicated by the designated container information as the target container.

[0187] The target container identification unit 220 sends information indicating the target container to the container status identification unit 230.

[0188] <Container status identification unit 230> The container status identification unit 230 receives information indicating the target container from the target container identification unit 220. The container status identification unit 230 determines that containers other than the target container, among the containers placed in the temporary storage location, are existing containers.

[0189] The container status identification unit 230 transmits information indicating the location of the target container to the generation unit 120, in addition to information indicating the location of the existing container. Specifically, the container status identification unit 230 may transmit information on the locations of the containers placed in the temporary storage location and information indicating the specified container (i.e., information indicating the target container) to the generation unit 120. Of the containers placed in the temporary storage location, containers other than the container identified by the information indicating the specified container are existing containers.

[0190] The information indicating the designated container (i.e., the target container) is, for example, the ID of the target container. Note that the container status identification unit 230 may transmit information indicating an existing container (e.g., the ID of an existing container) as the information indicating the designated container (i.e., the target container). Of the containers whose location information is included in the location information of containers placed in the temporary storage location, the container that is not indicated by any of the information indicating existing containers is the target container.

[0191] <Layout calculation device 101> As described above, the layout calculation device 101 of this embodiment is the same as the layout calculation device 100 of the second embodiment, except for the differences described below. The following describes the differences between the layout calculation device 101 of this embodiment and the layout calculation device 100 of the second embodiment. As described above, when the configuration of the layout calculation device 101 of this embodiment is compared with the configuration of the layout calculation device 100 of the second embodiment, the layout calculation device 101 of this embodiment further includes a cost calculation unit 170.

[0192] <Generation Unit 120> The generation unit 120 receives information indicating the location of the target container in addition to information indicating the location of the existing container from the container status identification unit 230. Specifically, the generation unit 120 receives, for example, information on the location of a container placed at a temporary storage location and information indicating a designated container (i.e., information indicating the target container) as information indicating the target container from the container status identification unit 230. In this case, the generation unit 120 identifies the location of the existing container from the information on the location of the container placed at the temporary storage location and the information indicating the designated container (i.e., the target container). Specifically, the generation unit 120 identifies, for example, containers other than the target container among the containers placed at the temporary storage location. Then, the generation unit 120 identifies the locations of containers other than the target container among the containers placed at the temporary storage location as the locations of the existing containers.

[0193] The generator 120 generates an Ising model in the same manner as the generator 120 of the second embodiment.

[0194] The generation unit 120 sends information representing the position of the target container to the calculation processing unit 130, in addition to information representing the container unloading order, information representing the constraint conditions, information representing the objective function, and the Ising model. The generation unit 120 may send information on the position of the container placed in the temporary storage location to the calculation processing unit 130 as the information representing the position of the target container.

[0195] <Calculation processing unit 130> The calculation processing unit 130 of this embodiment receives, from the generation unit 120, information representing the container unloading order, information representing the constraint conditions, information representing the objective function, and the Ising model, as well as information representing the position of the target container.

[0196] The calculation processing unit 130 of this embodiment causes the information processing device to execute a placement determination process using quantum annealing, which determines placement plans that are positions of target containers that satisfy constraints and have evaluation values ​​that meet the criterion. The number of placement plans to be determined is one or more. The placement plans that are positions of target containers whose evaluation values ​​meet the criterion (i.e., placement plans whose evaluation values ​​meet the criterion) are, for example, a predetermined number of placement plans in descending order of the evaluation values. The placement plans whose evaluation values ​​meet the criterion may be placement plans whose evaluation values ​​are better than a predetermined evaluation value. The placement plans whose evaluation values ​​meet the criterion may be placement plans that satisfy at least one of the following: the ranking of the evaluation value is within a predetermined ranking; and the evaluation value is better than a predetermined evaluation value.

[0197] In the disclosure, the determined placement plans are represented by an Ising model modified by an information processing device. The Ising model indicates the positions of the determined target containers and the positions of the existing containers. Each of the determined placement plans indicates the positions of the existing containers in addition to the positions of the determined target containers.

[0198] The calculation processing unit 130 of this embodiment transmits the determined allocation plan, the positions of the target container and the existing container, along with information indicating the container carry-out sequence, to the cost calculation unit 170. The positions of the target container and the existing container are the positions of containers placed in the temporary storage location.

[0199] <Cost Calculation Unit 170> The cost calculation unit 170 receives the determined allocation plan, the positions of the target container and the existing container, and information indicating the container removal order from the calculation processing unit 130. As described above, the positions of the target container and the existing container indicate the positions of containers placed in the temporary storage location. The positions of the containers placed in the temporary storage location indicated by the positions of the target container and the existing container are referred to as the initial allocation.

[0200] The cost calculation unit 170 sorts the placement plans in ascending order of the difference between the movement cost of each of the determined placement plans and the movement cost of the initial placement (i.e., the increase in movement cost). A negative increase in movement cost indicates a decrease in movement cost. In this embodiment, the movement cost of the initial placement is the number of times a container is reordered when the container is unloaded according to the unloading order when the container's position in the temporary storage location is the initial placement. The movement cost of the placement plan is the sum of the number of times a container is moved when changing from its initial placement to the placement plan and the number of times a container is reordered when the container is unloaded according to the unloading order when the container's position in the temporary storage location is the placement plan. The calculation of the increase in movement cost will be described in detail below.

[0201] The cost calculation unit 170 calculates the number of container rearrangements (also referred to as the number of unchanged moves) when a container is removed in accordance with the removal order when the container's position in the temporary storage location is the initial placement. In other words, the cost calculation unit 170 calculates the number of rearrangements when a container placed in the temporary storage location as initially placed is removed in accordance with the removal order. Rearrangements are the movement of a container within the temporary storage location. The number of rearrangements is the number of times a container is moved within the temporary storage location when a container is removed from the temporary storage location in accordance with the removal order. Rearrangements include, for example, the movement of other containers placed above the container to be removed next to another location within the temporary storage location.

[0202] The cost calculation unit 170 calculates the number of rearrangements (also referred to as the number of change moves) for each placement plan. The number of rearrangements for a placement plan is the number of rearrangements required when containers placed in temporary storage locations according to the placement plan are transported in the transport order.

[0203] The cost calculation unit 170 calculates the number of container movements (also referred to as the number of rearrangement movements) required when rearranging the containers that are placed in the temporary storage location according to the initial arrangement so that the containers are placed in the positions indicated by each of the arrangement plans. In other words, the number of container movements is the number of container movements required to change the initial arrangement to the arrangement plan.

[0204] The cost calculation unit 170 calculates the increase in the movement cost of each placement plan from the number of rearrangements in the initial placement, the number of rearrangements in each placement plan, and the number of container movements required to change from the initial placement to the placement plan.The cost calculation unit 170 calculates the increase in the movement cost of the placement plan by subtracting the number of rearrangements in the initial placement from the sum of the number of rearrangements in that placement plan and the number of container movements required to change from the initial placement to that placement plan.In this case, the increase in the movement cost of the placement plan corresponds to the increase in the number of container movements at the temporary storage location.

[0205] The cost calculation unit 170 sorts the placement plans in ascending order of the increase in movement cost.

[0206] The cost calculation unit 170 sends information about the placement plans sorted in ascending order of the increase in movement cost to the output unit 140. The cost calculation unit 170 may send information about the placement plans sorted in ascending order of the increase in movement cost and the increase in movement cost of the placement plans to the output unit 140.

[0207] <Output unit 140> The output unit 140 receives information about placement plans sorted in ascending order of the increase in movement cost from the cost calculation unit 170. The output unit 140 outputs the information about placement plans sorted in ascending order of the increase in movement cost to the UI unit 240 of the operation device 201. In other words, the output unit 140 outputs the information about placement plans to the UI unit 240 of the operation device 201 in ascending order of the increase in movement cost.

[0208] The output unit 140 may be configured to receive information about placement plans sorted in ascending order of the increase in travel cost and the increase in travel cost of the placement plans from the cost calculation unit 170. In this case, the output unit 140 outputs a combination of information about placement plans and the increase in travel cost sorted in ascending order of the increase in travel cost.

[0209] The output unit 140 may output placement plans in which the increase in travel cost is a negative value, and may not output placement plans in which the increase in travel cost is a non-negative value.

[0210] <Operation> FIGS. 12 and 13 are flowcharts showing an example of the operation of the layout calculation system according to the present disclosure.

[0211] Hereinafter, the operation of the layout calculation system according to the fourth embodiment of the present disclosure will be described in detail with reference to FIGS. 12 and 13. FIG.

[0212] In the example shown in FIG. 12, the operations in steps S301 and S302 are the same as the operations in steps S121 and S122 shown in FIG. 6 performed by the generation unit 120 according to the second embodiment of the present disclosure.

[0213] Next, the generation unit 120 receives information about the specified container and information about the container's location (step S303). As described above, in this embodiment, the specified container is the target container. Furthermore, the information about the container's location is, for example, information about the location of a container placed in a temporary storage location. That is, the information about the container's location includes information about the location of an existing container, which is a container placed in a temporary storage location and has a fixed location, and information about the location of the target container, which is a container specified by a container placed in the temporary storage location. As described above, an existing container is a container placed in a temporary storage location other than the specified container (in this case, the target container).

[0214] Next, the generator 120 generates an Ising model (step S304). The operation of step S304 is the same as the operation of step S124 shown in FIG. 6 performed by the generator 120 of the second embodiment of the present disclosure.

[0215] 13 , the calculation processing unit 130 causes the information processing device to execute a placement determination process using quantum annealing to determine a placement plan that is a position of a target container that satisfies the constraint conditions and the evaluation value satisfies the standard (step S305). The operation of step S305 is the same as the operation of step S135 shown in FIG. 7 performed by the calculation processing unit 130 of the first modified example of the second embodiment of the present disclosure.

[0216] The cost calculation unit 170 calculates the increase in the movement cost of the placement plan (step S306). The cost calculation unit 170 further sorts the placement plans in ascending order of the increase in the movement cost.

[0217] Then, the output unit 140 outputs the placement plans in ascending order of the increase in movement cost (step S307).

[0218] <Effects> The present embodiment described above has the same effects as the first embodiment. The first reason is the same as the reason why the effects of the first embodiment are obtained. The second reason is that the generation unit 120 generates an Ising model indicating the positions of containers other than the specified container. The calculation processing unit 130 then causes the information processing device to execute a placement determination process that determines the position of the specified container so that the constraints are satisfied and the evaluation value meets the criteria. The cost calculation unit 170 then calculates the increase in the movement cost for each placement plan. The output unit 140 then outputs the placement plans in ascending order of the increase in movement cost. This increase in movement cost is the increase in movement cost when the initial placement is changed to the placement plan and the containers placed in the temporary storage location according to the placement plan are unloaded according to the unloading order, compared to the movement cost when the containers placed in the temporary storage location according to the initial placement are unloaded according to the unloading order. If this increase in movement cost is a negative value, the movement cost, i.e., the cost of operating the container, is reduced.

[0219] Fifth Embodiment Next, a fifth embodiment of the present disclosure will be described in detail with reference to the drawings.

[0220] FIG. 14 is a block diagram illustrating an example of the configuration of a layout calculation system according to the present disclosure.

[0221] Hereinafter, the configuration of the layout calculation system according to the fifth embodiment of the present disclosure will be described in detail with reference to FIG.

[0222] 14 , the allocation calculation system according to the fifth embodiment of the present disclosure includes an allocation calculation device 102, an operation device 200, and a prediction device 300. The operation device 200 of this embodiment is the same as the operation device 200 of the second embodiment of the present disclosure. The allocation calculation device 102 and the prediction device 300 of this embodiment are the same as the allocation calculation device 100 and the prediction device 300 of the second embodiment of the present disclosure, except for the differences described below.

[0223] <Prediction Device 300> The prediction device 300 is the same as the prediction device of the second embodiment except for the differences described below.

[0224] <Machine Learning Unit 310> The machine learning unit 310 uses the container transportation information stored in the data accumulation unit 210 as learning data, and uses an existing machine learning method to train a prediction model that predicts the unloading date and time of a container from the container profile and the loading date and time, and estimates the accuracy of the prediction of the unloading date and time. As described above, the predicted unloading date and time may be expressed as the time elapsed from the loading date and time. In this case, the machine learning unit 310 trains a prediction model that predicts the elapsed time from the loading date and time of the container, estimates the prediction accuracy of the elapsed time (i.e., prediction accuracy), and outputs the predicted elapsed time and the prediction accuracy of the elapsed time.

[0225] As described above, the accuracy of the prediction of the removal date and time may be represented, for example, by a probability distribution with the date and time as a variable. When the prediction model predicts the elapsed time from the carry-in date and time to the removal date and time as information representing the removal date and time, the prediction accuracy may be represented by a probability distribution with the time elapsed from the carry-in date and time as a variable.

[0226] <Prediction Unit 340> The prediction unit 340 predicts the removal date and time of each container (here, the target container and the existing container) using a prediction model trained on the temporary storage location to which the target container is to be delivered, and estimates the accuracy of the prediction of the removal date and time. As described above, if the prediction model outputs the elapsed time from the delivery date and time to the removal date and time as information representing the removal date and time, the prediction unit 340 calculates the removal date and time from the delivery date and time of the container and the elapsed time from the delivery date and time to the removal date and time. If the prediction accuracy is represented by a probability distribution with the time elapsed from the delivery date and time as a variable, the prediction unit 340 converts the prediction accuracy into a probability distribution with the date and time as a variable.

[0227] The prediction unit 340 transmits the predicted removal date and time of each container and the prediction accuracy of the removal date and time to the sequence calculation unit 110 of the allocation calculation device 102. Specifically, the prediction unit 340 transmits information indicating the predicted removal date and time of each container and information indicating the prediction accuracy of the removal date and time to the sequence calculation unit 110 of the allocation calculation device 102.

[0228] <Layout calculation device 102> As described above, the layout calculation device 102 of this embodiment is the same as the layout calculation device 100 of the second embodiment, except for the differences described below. The following describes the differences between the layout calculation device 102 of this embodiment and the layout calculation device 100 of the second embodiment. As described above, when the configuration of the layout calculation device 102 of this embodiment is compared with the configuration of the layout calculation device 100 of the second embodiment, the layout calculation device 102 of this embodiment further includes a grouping unit 180.

[0229] <Sequence calculation unit 110> The sequence calculation unit 110 receives the predicted removal date and time for each container and the prediction accuracy of the removal date and time from the prediction unit 340. Specifically, the sequence calculation unit 110 receives information indicating the predicted removal date and time for each container and information indicating the prediction accuracy of the removal date and time from the prediction unit 340.

[0230] The sequence calculation unit 110 determines the container removal order based on the predicted removal dates and times of each container, for example, by sorting the container IDs in ascending order of the container removal dates and times. The sequence calculation unit 110 further calculates the prediction accuracy of the proximity order relationship for each combination of two containers whose removal orders are adjacent in the determined container removal order using the prediction accuracy of the removal dates and times. The prediction accuracy of the proximity order relationship is expressed, for example, by the probability that the removal order of two containers whose removal orders are adjacent will be correct. For example, if the prediction accuracy of the removal dates and times is expressed by a probability distribution using date and time as a variable, the sequence calculation unit 110 calculates the probability that the two containers will be removed in the order according to the determined removal order using the probability distribution of the removal dates and times of the two containers. Such a probability can be obtained, for example, by calculating the integral of the product of the integral up to a certain time of the probability distribution of the container whose removal order is earlier for two adjacent containers whose removal orders are adjacent, and the value at that time of the probability distribution of the container whose removal order is later. When the probability distribution is given by the distribution of probability values ​​after each predetermined time, the above-mentioned probability can be obtained by calculating the value corresponding to the above-mentioned integral.

[0231] The sequence calculation unit 110 sends the container unloading sequence and the prediction accuracy of the adjacent sequence relationship for each combination of two containers whose unloading sequence is adjacent to each other in the unloading sequence to the grouping unit 180. Specifically, the sequence calculation unit 110 sends information representing the container unloading sequence and information representing the prediction accuracy of the adjacent sequence relationship for each combination of two containers whose unloading sequence is adjacent to each other in the unloading sequence to the grouping unit 180.

[0232] <Grouping unit 180> The grouping unit 180 receives the container unloading order and the prediction accuracy of the proximity order relationship for each combination of two containers whose unloading order is adjacent in the unloading order from the order calculation unit 110. Specifically, the grouping unit 180 receives from the order calculation unit 110 information representing the container unloading order and information representing the prediction accuracy of the proximity order relationship for each combination of two containers whose unloading order is adjacent in the unloading order.

[0233] The grouping unit 180 groups the containers using the prediction accuracy of the proximity order relationship. In other words, the grouping unit 180 groups the containers such that, if the prediction accuracy of the proximity order relationship between two containers whose unloading orders are adjacent satisfies an accuracy standard, the two containers are grouped into the same group (also referred to as an unloading group). In further other words, if the prediction accuracy of the proximity order relationship between two containers whose unloading orders are adjacent satisfies an accuracy standard, the grouping unit 180 classifies the two containers into the same group (an unloading group). Grouping of containers involves generating groups of containers using, for example, any of the first to fourth grouping methods described below. The grouping unit 180 may use, for example, one of the methods shown below as the grouping method.

[0234] <First Grouping Method> If the prediction accuracy of the proximity order relationship between two containers whose unloading orders are adjacent in the container unloading order is less than a first predetermined threshold (hereinafter also referred to as an indefinite threshold), the grouping unit 180 places the two containers in the same group (in other words, classifies the two containers). As described above, the prediction accuracy of the proximity order relationship between two containers whose unloading orders are adjacent in the container unloading order is the probability that the unloading order of the two containers is correct. The indefinite threshold corresponds to a threshold for determining whether the unloading order of two adjacent containers can be determined. The accuracy criterion in this case is that the prediction accuracy of the proximity order relationship between two containers whose unloading orders are adjacent in the container unloading order is less than the first predetermined threshold.

[0235] If one of two containers whose proximity order relationship prediction accuracy is less than the arbitrary threshold is already included in a group, the grouping unit 180 incorporates those two containers into that group.If two containers whose proximity order relationship prediction accuracy is less than the arbitrary threshold are included in two different groups, the grouping unit 180 merges those two groups.

[0236] For example, if two containers belong to the same shipper and have a profile of always being removed at the same time, the order in which the two containers are removed may not necessarily be determined. In such cases, the prediction accuracy of the proximity order relationship between the two containers (i.e., the probability that the order of the two containers in the predicted removal order is correct) is low. In other words, two containers whose prediction accuracy of the proximity order relationship is less than an indefinite threshold can be considered to be two containers whose order in which they will be removed first cannot be predicted.

[0237] <Second Grouping Method> If the prediction accuracy of the proximity order relationship between two containers whose unloading orders are adjacent in the container unloading order is greater than a second predetermined threshold (hereinafter also referred to as the deterministic threshold), the grouping unit 180 places those two containers in the same group. The deterministic threshold corresponds to a threshold for determining whether the unloading order of two containers whose unloading orders are adjacent is deterministic. The deterministic threshold is greater than the above-mentioned indeterministic threshold. The deterministic order of the unloading order of two containers means that the unloading order of those two containers can be considered to be fixed. The accuracy standard in this case is that the prediction accuracy of the proximity order relationship between two containers whose unloading orders are adjacent in the container unloading order is greater than the second predetermined threshold.

[0238] If one of two containers whose proximity order relationship has a predicted accuracy greater than the confirmed accuracy is already included in a group, the grouping unit 180 incorporates those two containers into that group.If two containers whose proximity order relationship has a predicted accuracy greater than the confirmed accuracy are included in two different groups, the grouping unit 180 merges those two groups.

[0239] For example, if two containers are transported by the same delivery company and the delivery company always delivers the containers in the order number order, the delivery order of the two containers is determined by the order numbers of the two containers. If the order number is included in the container profile, the prediction accuracy of the proximity order relationship between two containers transported by such a delivery company (i.e., the probability that the order of the two containers in the predicted delivery order is correct) will be high. In other words, two containers whose prediction accuracy of the proximity order relationship is greater than a certainty threshold can be considered to be two containers whose delivery order is deterministic.

[0240] <Third Grouping Method> The grouping unit 180 may use both the first grouping method and the second grouping method to determine whether or not two containers that are adjacent in the unloading order of the containers should be included in the same group.

[0241] <Fourth Grouping Method> In the first and third grouping methods, when one of two containers whose proximity order relationship has a predicted accuracy greater than the confirmed accuracy is already included in a group, the grouping unit 180 may generate a group of those two containers. When two containers whose proximity order relationship has a predicted accuracy greater than the confirmed accuracy are included in two different groups, the grouping unit 180 may generate a group of those two containers.

[0242] In the second and third grouping methods, if one of two containers whose proximity order relationship has a predicted accuracy greater than the confirmed accuracy is already included in a group, the grouping unit 180 may generate a group of those two containers.If two containers whose proximity order relationship has a predicted accuracy greater than the confirmed accuracy are included in two different groups, the grouping unit 180 may generate a group of those two containers.

[0243] <Calculation processing unit 130> The calculation processing unit 130 (specifically, the placement determination process that the calculation processing unit 130 causes the information processing device to execute) calculates an evaluation value without using the combination of two containers included in the same group (specifically, the positions and removal order of the combination of two containers).

[0244] <Operation> Next, the operation of the layout calculation system according to the fifth embodiment of the present disclosure will be described.

[0245] 15 and 16 are flowcharts illustrating an example of the operation of the layout calculation system according to the present disclosure.

[0246] Hereinafter, the operation of the layout calculation system (specifically, the layout calculation device 102) according to the fifth embodiment of the present disclosure will be described in detail with reference to FIGS. 15 and 16. FIG.

[0247] 15, the sequence calculation unit 110 receives information on the unloading date and time of the container (step S401). In this embodiment, in step S401, the sequence calculation unit 110 receives information on the accuracy of the unloading date and time in addition to the information on the unloading date and time of the container.

[0248] Next, the sequence calculation unit 110 calculates the container unloading sequence (step S402). In this embodiment, in step S402, the sequence calculation unit 110 further calculates the prediction accuracy of the proximity sequence relationship using the accuracy of the container unloading date and time.

[0249] Next, the grouping unit 180 groups the containers (step S403).

[0250] In this embodiment, the operations of steps S404 and S405 shown in Fig. 15 and steps S406 and S407 shown in Fig. 16 are similar to the operations of steps S123 to S126 shown in Fig. 6. However, in step S406, the placement determination process calculates the evaluation value without using the positions and carry-out order of the combination of two containers included in the group.

[0251] <Effects> The present embodiment described above has the same effects as the first embodiment, for the same reasons as those for the effects of the first embodiment.

[0252] Sixth Embodiment Next, a layout calculation system according to a sixth embodiment of the present disclosure will be described in detail with reference to the drawings.

[0253] <Configuration> FIG. 17 is a block diagram illustrating an example of the configuration of a layout calculation system according to the present disclosure.

[0254] The configuration of the layout calculation system according to the sixth embodiment of the present disclosure will be described in detail below with reference to FIG.

[0255] In the example shown in FIG. 17 , the allocation calculation system according to this embodiment includes an allocation calculation device 103, an operation device 203, and a prediction device 300. The allocation calculation system according to this embodiment is the same as the allocation calculation system according to the second embodiment, except for the differences described below. The allocation calculation device 103 includes a weight receiving unit 190 in addition to the same components as those of the allocation calculation device 100 according to the second embodiment. The operation device 203 includes the same components as those of the operation device 200 according to the second embodiment. The prediction device 300 is the same as the prediction device 300 according to the second embodiment. Note that in this embodiment, the objective function used to calculate the evaluation value is expressed as the sum of the products of the above-described multiple element objective functions and their respective weights (i.e., the weighted sum of the element objective functions), as expressed by the following equation, for example:

[0256] In the above equation, OF is the objective function, and OF ｉ is the element objective function, and w ｉ is the element objective function OF ｉ The weight of the element objective function is the weight of the element evaluation value calculated by the element objective function. The evaluation value is expressed as the sum of the products of multiple element evaluation values ​​and their respective weights. In the above formula, the value of OF is the evaluation value, and OF ｉ The value is the element evaluation value.

[0257] <UI Unit 240> In this embodiment, the UI unit 240 receives weight values ​​for each of the component objective functions representing the objective functions used to calculate the evaluation value. The UI unit 240 receives weight values ​​input by an operator of the operation device 203 using, for example, an input device. The UI unit 240 may be configured to receive at least one weight value for the component objective functions representing the objective functions used to calculate the evaluation value.

[0258] The UI unit 240 transmits the received weight values ​​to the weight receiving unit 190 of the placement calculation device 103 .

[0259] <Weight Receiving Unit 190> The weight receiving unit 190 receives weights of multiple element objective functions input from the UI unit 240. Specifically, the weight receiving unit 190 receives weights for each of multiple element evaluation values ​​(i.e., multiple element objective functions) from the UI unit 240. The weight receiving unit 190 may be configured to receive input weights, for example, weights for at least some of the multiple element evaluation values ​​(i.e., multiple element objective functions), from the UI unit 240. The weight receiving unit 190 uses the received weights to update information on the objective functions for calculating evaluation values, which is stored in the calculation information storage unit 150, so that the objective function is represented by the sum of products of the received weights and the element objective functions (i.e., the weighted sum of the element objective functions).

[0260] The weight receiving unit 190 may generate a user interface screen for inputting weights. The weight receiving unit 190 may transmit the generated screen to the UI unit 240. The UI unit 240 may receive the screen from the weight receiving unit 190. The UI unit 240 may display the received screen.

[0261] <Generation Unit 120 > The generation unit 120 receives information representing the latest objective function updated by the weight reception unit 190 from the calculation information storage unit 150 .

[0262] <Operation> The operation of the placement calculation system of this embodiment is the same as the operation of the placement calculation system of the second embodiment, except for the operation of receiving the weights, which will be described below.

[0263] FIG. 18 is a flowchart illustrating an example of a weight receiving operation of the placement calculation system according to the present disclosure.

[0264] An example of the weight receiving operation of the placement calculation system according to the sixth embodiment of the present disclosure will be described in detail below with reference to FIG.

[0265] First, the weight receiving unit 190 receives the weight of the element evaluation value (step S501). The weight of the element evaluation value is the weight of the element objective function for calculating the element evaluation value. The weight receiving unit 190 sets the objective function for calculating the evaluation value to the weighted sum of the element objective functions for calculating the element evaluation value (step S502). Then, the placement calculation system stores the objective function (step S503). In other words, the weight receiving unit 190 stores the objective function (specifically, information representing the objective function) in the calculation information storage unit 150.

[0266] <Effects> This embodiment has the same effects as the second embodiment, for the same reasons as those for the effects of the second embodiment.

[0267] Other Embodiments The embodiments and modifications of the present disclosure can be applied to other embodiments or modifications to the extent that they are not inconsistent.

[0268] The layout calculation device according to the present disclosure can be realized by a computer including a memory into which a program read from a storage medium is loaded and a processor that executes the program. The layout calculation device according to the present disclosure can also be realized by dedicated hardware. The layout calculation device according to the present disclosure can also be realized by a combination of the computer and dedicated hardware.

[0269] FIG. 21 is a diagram illustrating an example of the hardware configuration of a computer 1000 that can realize the allocation calculation device, operation device, and prediction device according to the present disclosure. In the example illustrated in FIG. 21 , the computer 1000 includes a processor 1001, a memory 1002, a storage device 1003, and an I / O (Input / Output) interface 1004. The computer 1000 can also access a storage medium 1005. The memory 1002 and the storage device 1003 are, for example, storage devices such as RAM (Random Access Memory) and a hard disk. The storage medium 1005 is, for example, a storage device such as RAM or a hard disk, a ROM (Read Only Memory), or a portable storage medium. The storage device 1003 may be the storage medium 1005. The processor 1001 can read and write data and programs from and to the memory 1002 and the storage device 1003. The processor 1001 can access, for example, other devices via the I / O interface 1004. The processor 1001 can access a storage medium 1005. The storage medium 1005 stores a program that causes the computer 1000 to operate as a layout calculation device according to the present disclosure.

[0270] The processor 1001 loads a program stored in the storage medium 1005, which causes the computer 1000 to operate as the layout calculation device according to the present disclosure, into the memory 1002. The processor 1001 then executes the program loaded into the memory 1002, causing the computer 1000 to operate as the layout calculation device according to the present disclosure.

[0271] The order calculation unit 110, the generation unit 120, the calculation processing unit 130, the output unit 140, the movement planning unit 160, the cost calculation unit 170, the grouping unit 180, and the weight receiving unit 190 can be realized, for example, by a processor 1001 that executes a program loaded in a memory 1002. The calculation information storage unit 150 can be realized by a memory 1002 or a storage device 1003 such as a hard disk drive included in the computer 1000. Some or all of the order calculation unit 110, the generation unit 120, the calculation processing unit 130, the output unit 140, the calculation information storage unit 150, the movement planning unit 160, the cost calculation unit 170, the grouping unit 180, and the weight receiving unit 190 can be realized by dedicated circuits that realize the respective functions.

[0272] The target container identification unit 220, the container status identification unit 230, and the UI unit 240 can be realized, for example, by a processor 1001 that executes a program loaded in a memory 1002. The data accumulation unit 210 can be realized by a memory 1002 or a storage device 1003 such as a hard disk drive included in the computer 1000. Some or all of the data accumulation unit 210, the target container identification unit 220, the container status identification unit 230, and the UI unit 240 can be realized by dedicated circuits that realize the respective functions.

[0273] The machine learning unit 310, the information extraction unit 330, and the prediction unit 340 can be realized, for example, by a processor 1001 that executes a program loaded in a memory 1002. The model storage unit 320 can be realized by a memory 1002 or a storage device 1003 such as a hard disk drive included in the computer 1000. Some or all of the machine learning unit 310, the model storage unit 320, the information extraction unit 330, and the prediction unit 340 can be realized by dedicated circuits that realize the respective functions.

[0274] Furthermore, some or all of the above-described embodiments can be described as, but are not limited to, the following supplementary notes.

[0275] an evaluation value that decreases as the number of combinations of the container and another container that is unloaded after the container and is to be placed above the container in the unloading order decreases, while satisfying constraints that the positions of the containers are different from each other and that one of the containers is to be placed at the possible position below the position of the container; and an output unit. (Supplementary Note 1) An arrangement calculation device comprising: a generation unit that generates an Ising model for assigning a position of the container in a temporary storage location having one or more possible positions at which the container can be placed, to the possible positions; a calculation processing unit that causes an information processing device to execute an arrangement determination process that modifies the Ising model using quantum annealing;

[0276] (Supplementary Note 2) The arrangement calculation device according to Supplementary Note 1, wherein the calculation processing unit causes the information processing device to execute the arrangement determination process of determining the position of the target container so as to optimize the evaluation value further based on a moving distance that is a sum of distances between the positions of two of the containers that are consecutive in the carrying-out order.

[0277] (Supplementary Note 3) The arrangement calculation device according to Supplementary Note 2, wherein the calculation processing unit causes the information processing device to execute the arrangement determination process of determining the position of the target container so as to optimize the evaluation value further based on a distance sum, which is a sum of distances between the positions of each of the containers and a predetermined position.

[0278] (Supplementary Note 4) The target container is the container that is unloaded from a ship and placed in the temporary storage location, and the calculation processing unit causes the information processing device to execute the placement determination process to determine the position of the target container so as to optimize the evaluation value that is further based on the number of same-order loading combinations that are combinations of two of the target containers in which the horizontal positions of the loading positions on the ship are the same and the horizontal positions of the positions in the temporary storage location are the same, and the vertical relationship of the loading positions and the vertical relationship of the positions in the temporary storage location are the same. This is the placement calculation device described in Supplementary Note 1 or 2.

[0279] (Supplementary Note 5) The arrangement calculation device according to Supplementary Note 1 or 2, wherein the calculation processing unit causes the information processing device to execute the arrangement determination process to determine an arrangement plan that is a proposal for the position of each of the target containers, and the evaluation value satisfies a criterion, and the output unit outputs the position of the target container for each of the arrangement plans.

[0280] (Supplementary Note 6) The arrangement calculation device according to Supplementary Note 1 or 2, wherein the target container is the container that is to be newly arranged in the temporary storage location.

[0281] (Supplementary Note 7) The placement calculation device according to Supplementary Note 1, wherein the target containers are a predetermined number of designated containers from the containers placed at the temporary storage location, the calculation processing unit causes the information processing device to execute the placement determination process to determine placement plans that are proposals for the positions of each of the target containers, and the placement plans have the evaluation values ​​that satisfy a criterion, and further comprises a cost calculation unit that calculates a movement cost by subtracting an unchanged number of moves that is the number of moves of the containers to unload the containers placed at the temporary storage location in the unloading order from the sum of an changed number of moves that is the number of moves of the containers to unload the containers placed at the temporary storage location in the unloading order according to each of the placement plans, and a rearrangement number of moves that is the number of moves of the containers necessary to move the containers placed at the temporary storage location to the positions of the containers indicated by each of the placement plans, and

[0282] (Supplementary Note 8) The arrangement calculation device according to Supplementary Note 7, wherein the generation unit receives information on the predetermined number of containers designated from the containers arranged in the temporary storage location.

[0283] (Supplementary Note 9) The arrangement calculation device according to Supplementary Note 1 or 2, further comprising: a grouping unit that classifies two containers, among the containers, into the same carry-out group when the prediction accuracy of the chronological relationship of the carry-out order between the two containers that are adjacent in the carry-out order satisfies a predetermined accuracy standard; and the arrangement determination process calculates the evaluation value without using a combination of two of the containers included in the same carry-out group.

[0284] (Supplementary Note 10) The placement calculation device according to Supplementary Note 9, wherein the predetermined accuracy standard is that the value of the prediction accuracy is smaller than a predetermined threshold value.

[0285] (Supplementary Note 11) The placement calculation device according to Supplementary Note 9, wherein the predetermined accuracy standard is that the value of the prediction accuracy is greater than a predetermined threshold value.

[0286] (Supplementary Note 12) The placement calculation device according to Supplementary Note 1 or 2, wherein the evaluation value is a weighted sum of a plurality of element evaluation values ​​each including the number of the combinations as one element evaluation value, and the calculation processing unit causes the information processing device to execute the placement determination process that calculates the evaluation value using the weight of each of the plurality of element evaluation values.

[0287] (Supplementary Note 13) The placement calculation device according to Supplementary Note 12, further comprising: a weight receiving unit that receives the weight of any of the plurality of element evaluation values.

[0288] (Supplementary Note 14) An arrangement calculation method comprising: generating an Ising model for assigning a position of a container in a temporary storage location having one or more possible positions where the container can be placed to the possible positions; causing an information processing device to execute an arrangement determination process that modifies the Ising model using quantum annealing; the containers consist of existing containers whose positions are fixed and a target container whose position is to be determined; generating the Ising model that indicates the positions of the existing containers; obtaining the Ising model and a carry-out order of the containers; the arrangement determination process is a process that determines the position of the target container by modifying the Ising model so that an evaluation value that becomes lower as the number of combinations of the container and another container that is carried out after the container and placed above the container in the carry-out order becomes smaller, while satisfying constraints that the positions of the containers are different from each other and that one of the containers is placed in the possible position below the position of the container; and outputting the position of the target container.

[0289] (Supplementary Note 15) The method for calculating an arrangement according to Supplementary Note 14, wherein the information processing device is caused to execute the arrangement determination process for determining the position of the target container so as to optimize the evaluation value further based on a moving distance that is the sum of distances between the positions of two of the containers whose carrying-out order is consecutive.

[0290] (Supplementary Note 16) The placement calculation method according to Supplementary Note 15, wherein the information processing device is caused to execute the placement determination process of determining the position of the target container so as to optimize the evaluation value further based on a distance sum, which is the sum of the distances between the positions of each of the containers and a predetermined position.

[0291] (Supplementary Note 17) The target container is the container that is unloaded from a ship and placed at the temporary storage location, and the method of calculating placement described in Supplementary Note 14 or 15 causes the information processing device to execute the placement determination process to determine the position of the target container so as to optimize the evaluation value that is further based on the number of same-order loading combinations that are combinations of two of the target containers in which the horizontal positions of the loading positions on the ship are the same and the horizontal positions of the positions at the temporary storage location are the same, and the vertex relationship of the loading positions and the vertex relationship of the positions at the temporary storage location are the same.

[0292] (Supplementary Note 18) The placement calculation method described in Supplementary Note 14 or 15, further comprising: causing the information processing device to execute the placement determination process to determine placement plans that are proposals for the positions of the target containers, the placement plans having evaluation values ​​that satisfy a criterion; and outputting the positions of the target containers for each of the placement plans.

[0293] (Supplementary Note 19) The method of calculating placement according to Supplementary Note 14 or 15, wherein the target container is the container that is to be newly placed in the temporary storage location.

[0294] (Supplementary Note 20) The method for calculating placement of the target containers is as follows: the target containers are a predetermined number of designated containers from the containers placed at the temporary storage location; the information processing device executes the placement determination process to determine placement plans that are proposals for the positions of each of the target containers, and the placement plans satisfy a criterion in evaluation value; the method calculates a movement cost by subtracting the unchanged number of moves that is the number of moves of the containers to unload the containers placed at the temporary storage location in the unloading order from the sum of the changed number of moves that is the number of moves of the containers to unload the containers placed at the temporary storage location in the unloading order according to each of the placement plans and the rearrangement number of moves that is the number of moves of the containers necessary to move the containers placed at the temporary storage location to the positions of the containers indicated by each of the placement plans; and the method outputs the placement plans in order of decreasing movement cost.

[0295] (Supplementary Note 21) The arrangement calculation method according to Supplementary Note 20, further comprising receiving information on the predetermined number of containers designated from the containers arranged in the temporary storage location.

[0296] (Supplementary Note 22) The method for calculating placement described in Supplementary Note 14 or 15, wherein, when the prediction accuracy of the chronological relationship of the unloading order between two containers of which the unloading order is adjacent satisfies a predetermined accuracy standard, the two containers are classified into the same unloading group, and the placement determination process calculates the evaluation value without using a combination of two of the containers included in the same unloading group.

[0297] (Supplementary Note 23) The placement calculation method according to Supplementary Note 22, wherein the predetermined accuracy standard is that the value of the prediction accuracy is smaller than a predetermined threshold value.

[0298] (Supplementary Note 24) The placement calculation method according to Supplementary Note 22, wherein the predetermined accuracy standard is that the value of the prediction accuracy is greater than a predetermined threshold.

[0299] (Supplementary Note 25) The placement calculation method according to Supplementary Note 14 or 15, wherein the evaluation value is a weighted sum of a plurality of element evaluation values ​​each including the number of the combinations as one element evaluation value, and the information processing device is caused to execute the placement determination process for calculating the evaluation value using the weight of each of the plurality of element evaluation values.

[0300] (Supplementary Note 26) The placement calculation method according to Supplementary Note 25, further comprising receiving the weight of any one of the plurality of element evaluation values.

[0301] (Supplementary Note 27) A method for determining the position of a container in a temporary storage location having one or more possible positions where the container can be placed, by a computer, includes: a generation process for generating an Ising model for allocating the position of the container to the possible position; an arithmetic process for causing an information processing device to execute an arrangement determination process for modifying the Ising model using quantum annealing; and an output process, wherein the containers consist of an existing container whose position is fixed and a target container whose position is to be determined, the generation process generates the Ising model indicating the position of the existing container, the arithmetic process obtains the Ising model and the carry-out order of the containers, and the arrangement determination process is a process for determining the position of the target container by modifying the Ising model so that an evaluation value that decreases as the number of combinations of the container and another container that is carried out after the container and placed above the container in the carry-out order decreases, while satisfying constraints that the positions of the containers are different from each other and that one of the containers is placed at the possible position below the position of the container, and the output process outputs the position of the target container. A storage medium that stores a program.

[0302] (Supplementary Note 28) The storage medium according to Supplementary Note 27, wherein the calculation process causes the information processing device to execute the placement determination process of determining the position of the target container so as to optimize the evaluation value further based on a moving distance that is the sum of distances between the positions of two of the containers that are consecutive in the carrying-out order.

[0303] (Supplementary Note 29) The storage medium described in Supplementary Note 28, wherein the calculation process causes the information processing device to execute the placement determination process of determining the position of the target container so as to optimize the evaluation value further based on a distance sum, which is the sum of distances between the position of each of the containers and a predetermined position.

[0304] (Supplementary Note 30) The storage medium according to Supplementary Note 27 or 28, wherein the target container is the container that is unloaded from a ship and placed at the temporary storage location, and the calculation process causes the information processing device to execute the placement determination process to determine the position of the target container so as to optimize the evaluation value that is further based on the number of same-order loading combinations, which are combinations of two of the target containers in which the horizontal positions of the loading positions on the ship are the same and the horizontal positions of the positions at the temporary storage location are the same, and the vertical relationship of the loading positions and the vertical relationship of the positions at the temporary storage location are the same.

[0305] (Appendix 31) The storage medium described in Appendix 27 or 28, wherein the calculation process causes the information processing device to execute the placement determination process to determine placement plans that are proposals for the positions of each of the target containers, and the placement plans have an evaluation value that satisfies a criterion, and the output process outputs the positions of the target containers for each of the placement plans.

[0306] (Supplementary Note 32) The storage medium according to Supplementary Note 27 or 28, wherein the target container is the container that is to be newly placed in the temporary storage location.

[0307] (Supplementary Note 33) The storage medium described in Supplementary Note 27, wherein the target containers are a predetermined number of designated containers from the containers placed at the temporary storage location, the calculation process causes the information processing device to execute the placement determination process to determine placement plans that are proposals for the positions of each of the target containers, and the placement plans that satisfy a criterion in evaluation value, the program further causes the computer to execute cost calculation process to calculate a value obtained by subtracting the unchangeable number of moves that is the number of moves of the containers to unload the containers placed at the temporary storage location in the unloading order from the sum of the changed number of moves that is the number of moves of the containers to unload the containers placed at the temporary storage location in the unloading order, and the rearrangement number of moves that is the number of moves of the containers necessary to move the containers placed at the temporary storage location to the positions of the containers indicated by each of the placement plans, and the output process outputs the placement plans in order of decreasing move cost.

[0308] (Supplementary Note 34) The storage medium according to Supplementary Note 33, wherein the generation process receives information on the predetermined number of containers designated from the containers placed in the temporary storage location.

[0309] (Supplementary Note 35) The storage medium described in Supplementary Note 27 or 28, wherein the program further causes the computer to execute a grouping process to classify two containers of adjacent containers into the same unloading group when the prediction accuracy of the unloading order relationship between the two containers satisfies a predetermined accuracy standard, and the placement determination process calculates the evaluation value without using a combination of two of the containers included in the same unloading group.

[0310] (Supplementary Note 36) The storage medium according to Supplementary Note 35, wherein the predetermined accuracy standard is that the value of the prediction accuracy is smaller than a predetermined threshold value.

[0311] (Supplementary Note 37) The storage medium according to Supplementary Note 35, wherein the predetermined accuracy standard is that the value of the prediction accuracy is greater than a predetermined threshold value.

[0312] (Supplementary Note 38) The storage medium described in Supplementary Note 27 or 28, wherein the evaluation value is a weighted sum of multiple element evaluation values ​​including the number of combinations as one element evaluation value, and the calculation process causes the information processing device to execute the placement determination process that calculates the evaluation value using the weight of each of the multiple element evaluation values.

[0313] (Supplementary Note 39) The storage medium according to Supplementary Note 38, wherein the program further causes a computer to execute a weight receiving process for receiving the weight of any of the plurality of element evaluation values.

[0314] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure.

[0315] 10 Placement calculation device 100 Placement calculation device 101 Placement calculation device 102 Placement calculation device 103 Placement calculation device 110 Sequence calculation unit 120 Generation unit 130 Calculation processing unit 140 Output unit 150 Calculation information storage unit 160 Movement planning unit 170 Cost calculation unit 180 Grouping unit 190 Receiving unit 200 Operation device 201 Operation device 203 Operation device 210 Data accumulation unit 220 Target container identification unit 230 Container state identification unit 240 UI unit 300 Prediction device 310 Machine learning unit 320 Model storage unit 330 Information extraction unit 340 Prediction unit 1000 Computer 1001 Processor 1002 Memory 1003 Storage device 1004 I / O interface 1005 Storage medium

Claims

1. A placement calculation device comprising: a generation unit that generates an Ising model for assigning the position of the container in a temporary storage location having one or more possible positions where the container can be placed; an arithmetic processing unit that causes an information processing device to execute a placement determination process for changing the Ising model using quantum annealing; and an output unit. The container includes an existing container whose position is determined and a target container whose position is to be determined. The generation unit generates the Ising model indicating the position of the existing container. The arithmetic processing unit acquires the Ising model and the unloading order of the containers. The placement determination process is a process of determining the position of the target container by changing the Ising model so as to minimize an evaluation value such that, while satisfying the constraint condition that the positions of the containers are different from each other and any of the containers is placed in a possible position below the position of the container, the smaller the number of combinations of the container and another container that is unloaded after the container and placed above the container in the unloading order, the lower the evaluation value. The output unit outputs the position of the target container.

2. The placement calculation device according to claim 1, wherein the arithmetic processing unit causes the information processing device to execute the placement determination process for determining the position of the target container so as to optimize the evaluation value further based on a moving distance that is the sum of distances between the positions of two containers whose unloading orders are consecutive among the containers.

3. The placement calculation device according to claim 2, wherein the arithmetic processing unit causes the information processing device to execute the placement determination process for determining the position of the target container so as to optimize the evaluation value further based on a sum of distances that is the sum of distances between each position of the container and a predetermined position.

4. The target container is the container that is unloaded from the ship and placed in the temporary storage location, and the arithmetic processing unit is a combination of two of the target containers in which the horizontal positions of the loading positions loaded on the ship are the same and the horizontal positions of the positions in the temporary storage location are the same, and the vertical relationship of the loading positions and the vertical relationship of the positions in the temporary storage location are the same. The arrangement determination process for determining the position of the target container is executed by the information processing device so as to optimize the evaluation value further based on the number of the same-order loading combinations. The arrangement calculation device according to claim 1 or 2.

5. The arithmetic processing unit causes the information processing device to execute the arrangement determination process of determining an arrangement plan that is a plan of the positions of each of the target containers and satisfies the reference value of the evaluation value, and the output unit outputs the positions of the target containers for each arrangement plan. The arrangement calculation device according to claim 1 or 2.

6. The target container is the container newly placed in the temporary storage location. The arrangement calculation device according to claim 1 or 2.

7. The target container is a specified number of the containers arranged in the temporary storage location, and the arithmetic processing unit causes the information processing device to execute the arrangement determination process of determining an arrangement plan that is a plan of the positions of each of the target containers and satisfies the reference value of the evaluation value. A cost calculation unit is further provided, which calculates, as a movement cost, a value obtained by subtracting the number of unchanged movements, which is the number of movements of the containers arranged in the temporary storage location for unloading according to the unloading order, from the sum of the number of changed movements, which is the number of movements of the containers arranged in the temporary storage location for unloading according to the unloading order according to each of the arrangement plans, and the number of rearrangement movements, which is the number of movements of the containers arranged in the temporary storage location to the positions of the containers indicated by each of the arrangement plans. The output unit outputs the arrangement plans in ascending order of the magnitude of the movement cost. The arrangement calculation device according to claim 1.

8. The generation unit receives information on the specified predetermined number of containers from the containers arranged in the temporary storage location. The arrangement calculation device according to claim 7.

9. The arrangement calculation device according to claim 1 or 2, further comprising a grouping unit that classifies two containers whose loading order is adjacent among the containers into the same loading group when the prediction accuracy of the front-back relationship of the loading order between the two containers satisfies a predetermined accuracy criterion, and the arrangement determination process calculates the evaluation value without using two combinations of the containers included in the same loading group.

10. The arrangement calculation device according to claim 9, wherein the predetermined accuracy criterion is that the value of the prediction accuracy is smaller than a predetermined threshold.

11. The arrangement calculation device according to claim 9, wherein the predetermined accuracy criterion is that the value of the prediction accuracy is larger than a predetermined threshold.

12. The evaluation value is a weighted sum of a plurality of element evaluation values including the number of combinations as one element evaluation value, and the arithmetic processing unit causes the information processing device to execute the arrangement determination process of calculating the evaluation value using the weights of each of the plurality of element evaluation values. The arrangement calculation device according to claim 1 or 2.

13. The arrangement calculation device according to claim 12, further comprising a weight receiving unit that receives any one of the weights of the plurality of element evaluation values.

14. Generate an indexing model for assigning the position of the container in a temporary storage location having one or more possible positions where the container can be placed, and cause an information processing apparatus to execute an arrangement determination process of changing the indexing model using quantum annealing. The container consists of an existing container whose position is determined and a target container whose position is to be determined. Generate the indexing model indicating the position of the existing container, obtain the indexing model and the unloading order of the container. The arrangement determination process is to satisfy the constraint condition that the positions of the respective containers are different from each other and any of the containers is placed at a possible position below the position of the container, and in the unloading order, the smaller the number of combinations of the container and another container that is unloaded after the container and placed on the container, the lower the evaluation value. By changing the indexing model, it is a process of determining the position of the target container, and outputting the position of the target container. Arrangement calculation method.

15. The arrangement calculation method according to claim 14, wherein the information processing apparatus is caused to execute the arrangement determination process of determining the position of the target container so as to optimize the evaluation value further based on a moving distance that is the sum of distances between the positions of two containers whose unloading orders are consecutive among the containers.

16. The arrangement calculation method according to claim 15, wherein the information processing apparatus is caused to execute the arrangement determination process of determining the position of the target container so as to optimize the evaluation value further based on a sum of distances that is the sum of distances between each position of the container and a predetermined position.

17. The target container is the container that is unloaded from the ship and arranged at the temporary storage location, and is a combination of two of the target containers in which the horizontal positions of the loading positions loaded on the ship are the same and the horizontal positions of the positions at the temporary storage location are the same, and the vertical relationship of the loading positions and the vertical relationship of the positions at the temporary storage location are the same. The arrangement determination process for determining the position of the target container is executed by the information processing device so as to optimize the evaluation value further based on the number of the same-order loading combinations. The arrangement calculation method according to claim 14 or 15.

18. An arrangement plan that is a plan for the position of each of the target containers, the arrangement determination process for determining the arrangement plan in which the evaluation value satisfies a standard is executed by the information processing device, and for each arrangement plan, the position of the target container is output. The arrangement calculation method according to claim 14 or 15.

19. The target container is the container newly arranged at the temporary storage location. The arrangement calculation method according to claim 14 or 15. A generation process for generating an indexing model for assigning the position of the container in a temporary storage location having one or more possible positions where the container can be placed; an arrangement determination process for changing the indexing model using quantum annealing; an arithmetic process for causing an information processing apparatus to execute; an output process; causing a computer to execute, the container comprising an existing container whose position is determined and a target container whose position is to be determined, the generation process generating the indexing model indicating the position of the existing container, the arithmetic process acquiring the indexing model and the unloading order of the container, the arrangement determination process satisfying the constraint condition that each of the positions of the containers is different from each other and any of the containers is placed at a possible position below the position of the container, and making the evaluation value lower as the number of combinations of the container and another container that is unloaded after the container and placed above the container is smaller in the unloading order, a process of determining the position of the target container by changing the indexing model, the output process outputting the position of the target container, a storage medium storing a program.

Images