Information processing method, loading method, program, and information processing device

Abstract

The present invention suitably creates a loading plan in accordance with work details. This information processing method comprises: a step for acquiring work information; a step for calculating an initial solution indicating loading information, which indicates articles allocated to a plurality of pallets and a stacking order of the articles, and a loading order, which indicates an order of the pallets on which loading is performed; a step for calculating an initial evaluation value indicating the degree of workload in the initial solution; a step for calculating a neighborhood solution in which the loading information and / or the loading order has been partly changed with respect to the initial solution; a step for calculating a neighborhood evaluation value indicating the degree of workload in the neighborhood solution; and a step for setting the loading information and the loading order used for the loading plan on the basis of the initial evaluation value and the neighborhood evaluation value. The initial evaluation value and the neighborhood evaluation value are calculated in consideration of at least one of multiple gripping in which a plurality of articles are gripped, diversion in which a supply pallet is diverted as a shipping pallet, and reverse loading in which an unnecessary number of articles are removed to another pallet and a supply pallet is diverted as a shipping pallet.

Description

Information processing method, loading method, program, and information processing apparatus

[0001] The present disclosure relates to an information processing method, a loading method, a program, and an information processing apparatus.

[0002] A method of shipping articles by loading the articles onto a tray such as a pallet is known. For example, Patent Document 1 describes an automatic shipping apparatus that conveys article trays on a belt conveyor, loads articles onto the article trays on the belt conveyor, and ships the loaded articles.

[0003] Japanese Unexamined Patent Publication No. 2020-83486

[0004] Here, there is a demand for appropriately creating a loading plan according to the content of work.

[0005] An object of the present disclosure is to provide an information processing method, a loading method, a program, and an information processing apparatus that can appropriately create a loading plan according to the content of work.

[0006] The information processing method relating to this disclosure includes the steps of: acquiring work information indicating goods to be shipped to a destination; calculating an initial solution based on the work information, which includes goods to be assigned to each of a plurality of pallets, stacking information indicating the stacking order of the goods on the pallets, and stacking order indicating the order in which the pallets are stacked; calculating an initial evaluation value indicating the degree of workload of the shipping work in the initial solution based on the initial solution; calculating a neighboring solution based on the initial solution, which is obtained by changing a part of the stacking information and the stacking order from the initial solution; and calculating the shipping in the neighboring solution based on the neighboring solution. The process includes the steps of calculating a proximity evaluation value indicating the degree of workload, and setting the stacking information and stacking order to be used in the stacking plan based on the initial evaluation value and the proximity evaluation value, wherein in the step of calculating the initial evaluation value and the proximity evaluation value, the initial evaluation value and the proximity evaluation value are calculated by considering at least one of the following: multiple gripping, in which multiple items are gripped and stacked in a single stacking operation; repurposing, in which a supply pallet loaded with inventory items is repurposed as a shipping pallet for shipment; and reverse stacking, in which an unnecessary number of items are removed from the supply pallet and the supply pallet is repurposed as the shipping pallet.

[0007] The stacking method relating to this disclosure involves stacking the items on a plurality of pallets based on the stacking information and stacking order set in the information processing method.

[0008] The program relating to this disclosure includes the steps of: acquiring work information indicating goods to be shipped to a destination; calculating an initial solution based on the work information, which includes items to be assigned to each of a plurality of pallets, stacking information indicating the stacking order of the items on the pallets, and stacking order indicating the order in which the pallets are stacked; calculating an initial evaluation value based on the initial solution, which indicates the degree of workload of the shipping work in the initial solution; calculating a neighboring solution based on the initial solution, which is obtained by changing a part of the stacking information and the stacking order from the initial solution; and calculating the degree of workload of the shipping work in the neighboring solution based on the neighboring solution. The computer is made to perform the steps of calculating a neighboring evaluation value indicating compatibility, and setting the stacking information and stacking order to be used in the stacking plan based on the initial evaluation value and the neighboring evaluation value, and in the step of calculating the initial evaluation value and the neighboring evaluation value, the computer is made to calculate the initial evaluation value and the neighboring evaluation value by considering at least one of the following: multiple gripping, in which multiple items are gripped and stacked in one stacking operation; repurposing, in which a supply pallet loaded with inventory items is repurposed as a shipping pallet for shipment; and reverse stacking, in which an unnecessary number of items are removed from the supply pallet and the supply pallet is repurposed as the shipping pallet.

[0009] The information processing device according to this disclosure includes: a work information acquisition unit that acquires work information indicating goods to be shipped to a destination; an initial solution calculation unit that calculates an initial solution based on the work information, indicating the goods to be assigned to each of a plurality of pallets, stacking information indicating the stacking order of the goods on the pallets, and stacking order indicating the order in which the pallets are stacked; an initial evaluation value calculation unit that calculates an initial evaluation value indicating the degree of workload of the shipping work in the initial solution based on the initial solution; a nearest neighbor solution calculation unit that calculates a nearest neighbor solution based on the initial solution, by changing a part of the stacking information and the stacking order from the initial solution; and a nearest neighbor solution calculation unit that calculates a nearest neighbor solution based on the nearest neighbor solution The system includes a neighbor evaluation value calculation unit that calculates a neighbor evaluation value indicating the degree of workload of the shipping operation in the solution, and a setting unit that sets the loading information and loading order used in the loading plan based on the initial evaluation value and the neighbor evaluation value, wherein the initial evaluation value calculation unit and the neighbor evaluation value calculation unit calculate the initial evaluation value and the neighbor evaluation value by considering at least one of the following: multiple gripping, in which multiple items are gripped and loaded in one loading operation; repurposing, in which a supply pallet loaded with inventory items is repurposed as a shipping pallet for shipment; and reverse loading, in which an unnecessary number of items are removed from the supply pallet and the supply pallet is repurposed as the shipping pallet.

[0010] According to this disclosure, an appropriate savings plan can be created according to the nature of the work.

[0011] Figure 1 is a schematic diagram of the stacking system according to the first embodiment. Figure 2 is a schematic block diagram of the information processing device according to the first embodiment. Figure 3 is a flowchart illustrating the setting flow of stacking information and stacking order. Figure 4 is a schematic diagram showing an example of an initial solution in the second embodiment. Figure 5 is a schematic diagram showing an example of a first neighboring solution in the second embodiment. Figure 6 is a schematic diagram showing an example of a second neighboring solution in the second embodiment. Figure 7 is a schematic diagram showing an example of a third neighboring solution in the second embodiment. Figure 8 is a schematic diagram illustrating the setting of item information in the third embodiment. Figure 9 is a schematic diagram illustrating the setting of item information in the third embodiment. Figure 10 is a schematic diagram illustrating the setting of item information in the third embodiment. Figure 11 is a schematic diagram illustrating the setting of item information in the third embodiment. Figure 12 is a processing flow illustrating the method of assigning individual items in the third embodiment. Figure 13 is a processing flow illustrating the method of assigning all items in the third embodiment. Figure 14 is a schematic diagram for illustrating the setting of stacking information in the third embodiment. Figure 15 is a schematic diagram illustrating the setting of stacking information in the third embodiment. Figure 16 is a schematic diagram illustrating the setting of stacking information in the third embodiment. Figure 17 is a schematic diagram illustrating the setting of stacking information in the third embodiment. Figure 18 is a flowchart illustrating the processing flow for setting stacking information in the third embodiment. Figure 19 is a schematic diagram illustrating the setting of stacking order in the third embodiment. Figure 20 is a flowchart illustrating the processing flow for setting stacking order and pallet to be used in the third embodiment. Figure 21 is a schematic diagram illustrating the setting of the initial solution for stacking order in the fourth embodiment. Figure 22 is a flowchart illustrating the processing flow for setting stacking order in the fourth embodiment. Figure 23 is a flowchart illustrating the processing flow for setting pallet to be used in the fourth embodiment.

[0012] Preferred embodiments of the present disclosure will be described in detail below with reference to the attached drawings. However, this disclosure is not limited to these embodiments, and where there are multiple embodiments, they may also be combinations of these embodiments.

[0013] (First Embodiment) (Stacking System) Figure 1 is a schematic diagram of the stacking system according to the first embodiment. The stacking system 100 according to this embodiment is a system that stacks goods Q onto pallets P in equipment W based on a stacking plan. More specifically, the stacking system 100 of this embodiment transfers goods Q loaded on a supply pallet P0 onto pallet P. Equipment W is, for example, a warehouse or other equipment that is managed for logistics, but it may be any equipment on which the work of stacking goods Q onto pallets P is performed. Goods Q are items that are stacked on pallets P, for example, goods to be shipped. Here, according to the definition of the Japan Pallet Association, pallet P is "a structure that has a surface for placing cargo grouped into a single unit, and that allows for handling, transportation, and storage by manpower or a dedicated vehicle such as a forklift. This includes those with a superstructure." In this embodiment, pallet P refers to a container or stand on which goods Q are stacked. In other words, the term "pallet" here is not limited to a flat platform, but also includes carts and other similar objects, and can be of any shape. The supply pallet P0, like the pallet P, may be a container or platform (a so-called pallet) on which the goods Q are stacked. In this embodiment, only one type of goods Q is loaded onto a single supply pallet P0, whereas a pallet P may be loaded with multiple types of goods Q depending on the work content (shipping order). In this embodiment, multiple goods Q are stacked on the pallet P in layers. For example, layers of multiple goods Q are stacked on the pallet P. The types and arrangement order of goods Q in one layer of the pallet P, and the stacking order of the pallets P, are determined by the stacking plan.

[0014] As shown in Figure 1, the stacking system 100 includes an information processing device 10, a mobile unit 12, and a stacking device 14. The information processing device 10 is a device that creates a stacking plan for stacking goods Q on pallets P, and will be described in more detail later. The mobile unit 12 is a mobile unit that transports supply pallets P0 and pallets P. For example, the mobile unit 12 may be a forklift, and more specifically, it may be a so-called AGV (Automated Guided Vehicle) or AGF (Automated Guided Forklift). The stacking device 14 is a device (robot) that stacks goods Q on pallets P, and is a so-called palletizer. In this embodiment, the stacking device 14 can simultaneously grasp and stack multiple goods Q. Note that multiple mobile units 12 and stacking devices 14 may be deployed in the facility W.

[0015] (Equipment) Equipment W includes an item storage area AR1 where item Q is placed, a supply pallet placement area AR2 where supply pallets P0 are placed, a pallet placement area AR3 where pallets P are placed, a shipment placement area AR4 where stacked pallets P are placed, and an empty pallet placement area AR5 where empty supply pallets P0 are placed. Item Q that is to be stacked on pallets P is stored in item storage area AR1. In the example in Figure 1, item Q is stored in item storage area AR1 in a state where it is loaded on supply pallets P0, but it is not limited to this, and item Q that is not loaded on supply pallets P0 may also be stored. In this case, for example, a stacking device 14 may also be provided in item storage area AR1 to perform the operation of loading item Q onto supply pallets P0. Supply pallet placement area AR2 is the area where supply pallets P0 loaded with item Q are placed, and pallet placement area AR3 is the area where pallets P on which item Q is stacked are placed. The stacking device 14 picks up items Q from supply pallets P0 located in the supply pallet placement area AR2 and stacks the picked-up items Q on pallets P located in the pallet placement area AR3. The shipping item placement area AR4 is where pallets P that have completed the stacking work in the pallet placement area AR3 are placed, and pallets P may be shipped from here. The empty pallet placement area AR5 is where pallets (supply pallets P0) that are empty and do not have items Q stacked on them are placed. That is, for example, a supply pallet P0 that has become empty after the transshipment of items Q is completed is placed in the empty pallet placement area AR5. The transport of supply pallets P0 and pallets P between each area may be carried out by the mobile body 12. Note that the above layout of the equipment W is just an example and is not limited to the above description.

[0016] (Information Processing Device) Figure 2 is a schematic block diagram of an information processing device according to the first embodiment. The information processing device 10 is a device that generates a stacking plan for stacking goods Q on a pallet P. In the example of Figure 1, the information processing device 10 is installed in the facility W, but is not limited to that and may be installed in any location. The information processing device 10 is a computer and, as shown in Figure 2, has a communication unit 20, a storage unit 22, and a control unit 24. The information processing device 10 may also have an input unit that receives input from a user and an output unit that outputs information (for example, a display unit that displays an image). The information processing device 10 may be configured as a standalone device, may be configured as an integral part of other devices, or may be configured as a system combining various devices such as arithmetic circuits and data servers, and is not particularly limited.

[0017] The communication unit 20 is a communication module that communicates with external devices, such as an antenna. The information processing device 10 communicates with external devices via wireless communication, but it may also use wired communication, and the communication method is arbitrary. The storage unit 22 is a memory that stores various information such as the calculation contents and programs of the control unit 24, and includes at least one of the following: a main memory device such as RAM (Random Access Memory) and ROM (Read Only Memory), and an external memory device such as an HDD (Hard Disk Drive). The program for the control unit 24 stored in the storage unit 22 may be stored on a recording medium that the information processing device 10 can read.

[0018] The control unit 24 is a processing unit that performs calculations and includes calculation circuits such as a CPU (Central Processing Unit). The control unit 24 includes a work information acquisition unit 30, an initial solution calculation unit 32, a neighboring solution calculation unit 34, an initial evaluation value calculation unit 36, a neighboring evaluation value calculation unit 38, a setting unit 40, and an output control unit 42. The control unit 24 reads a program (software) from the storage unit 22 and executes it to realize the work information acquisition unit 30, the initial solution calculation unit 32, the neighboring solution calculation unit 34, the initial evaluation value calculation unit 36, the neighboring evaluation value calculation unit 38, the setting unit 40, and the output control unit 42, and executes their respective processes. The control unit 24 may execute these processes using one CPU, or it may have multiple CPUs and execute the processes using those multiple CPUs. Furthermore, at least a portion of the processing performed by the work information acquisition unit 30, the initial solution calculation unit 32, the neighboring solution calculation unit 34, the initial evaluation value calculation unit 36, the neighboring evaluation value calculation unit 38, the setting unit 40, and the output control unit 42 may be implemented by hardware circuits.

[0019] The work information acquisition unit 30 acquires work information indicating the work content for generating a stacking plan. The work information includes information indicating the item Q to be shipped to the shipping destination. The initial solution calculation unit 32 calculates an initial solution indicating stacking information and stacking order based on the work information. The neighboring solution calculation unit 34 calculates a neighboring solution based on the initial solution. A neighboring solution refers to a solution in which some of the stacking information and stacking order have been changed from the initial solution. The initial evaluation value calculation unit 36 ​​calculates an initial evaluation value indicating the load of the shipping work in the initial solution based on the initial solution. The neighboring evaluation value calculation unit 38 calculates a neighboring evaluation value indicating the load of the shipping work in the neighboring solution based on the neighboring solution. The setting unit 40 sets the stacking information and stacking order to be used in the stacking plan based on the initial evaluation value and neighboring evaluation value. The output control unit 42 outputs the stacking plan information, including the stacking information and stacking order, to an external device via the communication unit 20. For example, the output control unit 42 transmits information about the stacking plan to the mobile unit 12 and the stacking device 14, causing the mobile unit 12 and the stacking device 14 to perform the work based on the stacking plan.

[0020] (Processing by the Information Processing Device) The specific processing performed by the information processing device 10 will be explained below.

[0021] (Acquisition of work information) The work information acquisition unit 30 acquires work information indicating the work content for generating a stacking plan. The work information acquisition unit 30 may acquire work information by any method, but for example, it may acquire work information from a device (higher-level system) that manages the equipment W via the communication unit 20. The work information is information used to set the stacking information. The work information may be any information that can be used to set the stacking information, but it is preferable that it includes at least information indicating the goods Q to be shipped (goods information). More specifically, in this embodiment, the work information includes destination information, goods information indicating the goods Q to be shipped to that destination, supply pallet information indicating the supply pallet P0 on which the goods Q to be shipped are loaded, pallet information indicating the type of pallet P, and calculation conditions for calculating the initial evaluation value and neighboring evaluation value described later. Destination information is information indicating the destination for the current work. Goods information is information indicating the type and number of goods Q scheduled to be shipped for each destination. The item information may include, in addition to the type and number of items Q, at least one (preferably all) of the following: information indicating the size of the items Q to be shipped (e.g., length in the vertical, horizontal, and height directions); information indicating the weight of the items Q; information indicating the maximum number of items Q that can be placed in one layer on the pallet P; information indicating the maximum number of items Q that can be loaded onto the pallet P; and information indicating whether or not n items (where n is any integer) of items Q can be gripped simultaneously by the loading device 14. The supply pallet information includes information on the type and number of items Q to be loaded onto the supply pallet P0. The pallet information includes information indicating the width of one layer of the pallet P (e.g., length in the vertical and horizontal directions).

[0022] (Calculation of initial solution) The initial solution calculation unit 32 calculates an initial solution that shows the stacking information and stacking order based on the work information. The stacking information is information that shows the items Q (types of items Q) to be assigned to pallet P and the stacking order of items Q on that pallet P. In other words, the stacking information includes item information that shows the types and number of items Q to be assigned to pallet P, and stacking information that shows how to stack the items Q indicated by the pallet information on pallet P. The stacking order is information that shows the order in which stacking is performed on pallet P, or in other words, the order in which the pallets P for shipment are completed.

[0023] The initial solution calculation unit 32 sets the stacking information (item information and stacking information) for each of the multiple pallets P used for shipping, based on the work information. The initial solution calculation unit 32 may set the stacking information for the initial solution in any way based on the work information. For example, the initial solution calculation unit 32 may calculate the number of pallets P used for each shipping destination and the types and numbers of items Q to be assigned to each pallet P used for the shipping destination, based on the types and numbers of items Q for each shipping destination indicated by the work information, and use this as the item information for the initial solution. Then, the initial solution calculation unit 32 may calculate the stacking order (stacking information) of the assigned items Q for each pallet P used for the shipping destination and use this as the stacking information for the initial solution.

[0024] Furthermore, it is preferable that the initial solution calculation unit 32 also includes pallet information indicating the supply pallet P0 that will be the source of the item Q to be assigned to pallet P in the initial solution. Pallet information is set for each item Q assigned to pallet P in the initial solution. The initial solution calculation unit 32 may set the pallet information in any way based on the work information. For example, the initial solution calculation unit 32 may set the pallet information by identifying the supply pallet P0 on which the item Q assigned to pallet P in the initial solution is loaded, based on the work information.

[0025] (Calculation of Neighboring Solutions) The neighboring solution calculation unit 34 calculates neighboring solutions that show stacking information and stacking order based on the initial solution. The neighboring solutions are the same as the initial solution in that they include stacking information and stacking order, but some of the stacking information and stacking order differ from the initial solution. The neighboring solution calculation unit 34 may calculate neighboring solutions using any method based on the initial solution. For example, the neighboring solution calculation unit 34 may calculate neighboring solutions in which some of the types and number of items Q loaded on the pallet P, the stacking order of items Q on the pallet P, and the order in which the pallet P is finished differ from the initial solution. The neighboring solution calculation unit 34 may calculate multiple types of neighboring solutions. That is, the neighboring solution calculation unit 34 may calculate multiple types of neighboring solutions such that some of the stacking information and stacking order differ from the initial solution, and some of the stacking information and stacking order also differ from one neighboring solution to another.

[0026] (Calculation of initial evaluation value) The initial evaluation value calculation unit 36 ​​calculates an initial evaluation value based on the initial solution, indicating the degree of workload for shipping operations in the initial solution. The initial evaluation value is an index for evaluating the initial solution. As will be described in more detail later, the information processing device 10 also calculates a neighboring evaluation value for evaluating neighboring solutions, and uses the solution with the higher evaluation between the initial evaluation value and the neighboring evaluation value in the stacking plan.

[0027] The initial evaluation value is an index indicating the degree of workload when it is assumed that the work of stacking goods Q on pallet P according to the stacking information and stacking order shown in the initial solution and then shipping pallet P is performed. The initial evaluation value calculation unit 36 ​​may calculate the initial evaluation value using any method based on the initial solution, but in this embodiment, the initial evaluation value is calculated based on at least one of the following: the workload of the stacking device 14 that stacks goods Q on pallet P, the workload of the mobile body 12 that transports pallet P, and the number of small batches.

[0028] (Workload of the stacking device) When calculating the initial evaluation value based on the workload of the stacking device 14, the initial evaluation value calculation unit 36 ​​calculates the workload of the stacking device 14 based on the initial solution. The workload of the stacking device 14 here refers to the degree of workload of the work performed by the stacking device 14, assuming that the goods Q are stacked on the pallet P according to the stacking information and stacking order shown in the initial solution. For example, the initial evaluation value calculation unit 36 ​​may calculate the total number of times goods Q are stacked from the supply pallet P0 to the pallet P (the pallet for shipping), assuming that the goods Q are stacked on the pallet P according to the stacking information and stacking order shown in the initial solution, and use that total number as the initial evaluation value.

[0029] In this embodiment, the initial evaluation value calculation unit 36 ​​calculates an initial evaluation value based on the item information, supply pallet information, shipping destination information, and calculation conditions acquired by the work information acquisition unit 30. The total number of times the stacking device 14 stacks item Q from the supply pallet P0 onto the shipping pallet P is calculated using the initial evaluation value, which takes into account at least one of multiple gripping, repositioning, and reverse stacking.

[0030] Multiple gripping refers to the stacking device 14 picking up two or more items Q simultaneously. The initial evaluation value calculation unit 36 ​​determines whether n items of item Q can be gripped based on the item information's gripping capability information. The initial evaluation value calculation unit 36 ​​also determines that n items of item Q can be gripped if n or more items Q are stacked on the same layer of the supply pallet P0.

[0031] Repurposing refers to the reuse of a supply pallet P0 loaded with goods Q as a shipping pallet P. The initial evaluation value calculation unit 36 ​​determines whether a supply pallet P0 stored in the goods storage area AR1, or a supply pallet P0 in which goods Q remain after being picked up in the supply pallet placement area AR2, can be reused as a shipping pallet P. The initial evaluation value calculation unit 36 ​​determines that a supply pallet P0 can be reused if it contains all the goods Q required for the shipping pallet P. A supply pallet P0 determined to be reusable is transported as a shipping pallet P to the shipping item placement area AR4. The initial evaluation value calculation unit 36 ​​also determines that a supply pallet P0 can be reused if it contains the required number of goods Q to be stacked on the bottom layer of the shipping pallet P. A supply pallet P0 determined to be reusable is placed in the pallet placement area AR3 as a shipping pallet P. When a supply pallet P0 is repurposed as a shipping pallet P, the number of times the goods Q that were loaded onto this supply pallet P0 are loaded is counted as 0.

[0032] Reverse stacking refers to the process of removing an unnecessary number of items Q from a supply pallet P0 loaded with items Q and transferring them to another pallet P, thereby repurposing the original supply pallet P0 as a shipping pallet P. In this case, the number of items Q removed to the other pallet P (reverse stacking number) is the number of items Q originally loaded on the supply pallet P0 (inventory number) minus the number of items Q required for the shipping pallet P (shipping number). In other words, if the number of items Q removed to the other pallet P is less than the number of items Q required for the shipping pallet P, the number of stacking operations can be reduced. However, if the other pallet P from which the unnecessary items Q have been removed needs to be removed from the pallet placement area AR3, the number of transport operations and transport time for reverse stacking will be added.

[0033] The initial evaluation value calculation unit 36 ​​determines, for example, whether to reverse stack the supply pallet P0 stored in the item storage area AR1, or the supply pallet P0 on which item Q remains after item Q has been picked up in the supply pallet placement area AR2, if it determines that it cannot be repurposed as a pallet P for shipping. The initial evaluation value calculation unit 36 ​​determines whether the supply pallet P0 after reverse stacking can be repurposed as a pallet P for shipping. If the initial evaluation value calculation unit 36 ​​determines that the supply pallet P0 after reverse stacking cannot be repurposed as a pallet P for shipping, it then calculates the stacking time required for normal stacking and the stacking time required for reverse stacking. Normal stacking refers to stacking the required number of items Q from the supply pallet P0 on which the items Q are stacked onto the pallet P for shipping. The time required for normal stacking is calculated as (number of items to be shipped) × (time required to stack one item Q). If item Q has n items that can be gripped, the time required for reverse stacking is calculated as (number of items shipped - (number of items Q that can be gripped n times) + (number of items Q that can be gripped n times) / n) × (time required to stack one item Q). The time required for reverse stacking is calculated as (number of items to be reverse stacked) × (time required to stack one item Q) + (transportation time for reverse stacking). If item Q has n items that can be gripped, the time required for reverse stacking is calculated as (number of items to be reverse stacked - (number of items Q that can be gripped n times) + (number of items Q that can be gripped n times) / n) × (time required to stack one item Q) + (transportation time for reverse stacking). The initial evaluation value calculation unit 36 ​​compares the time required for forward stacking with the time required for reverse stacking, and if the time required for reverse stacking is shorter, it determines to perform reverse stacking.

[0034] In other words, the initial evaluation value calculation unit 36 ​​calculates the initial evaluation value assuming reverse stacking is performed if s × Tp > (a - s) × Tp + Te, where a is the inventory quantity, s is the number of items to be shipped, Tp is the time required for one stacking operation, and Te is the transport time for reverse stacking. Furthermore, the initial evaluation value calculation unit 36 ​​calculates the initial evaluation value assuming multiple gripping and reverse stacking are performed if m1 is the number of items Q that can be gripped simultaneously in n units when performing forward stacking, and m2 is the number of items Q that can be gripped simultaneously in n units when performing reverse stacking, where (s - m1 + m1 / n) × Tp > ((a - s) - m2 + m2 / n) × Tp + Te. It is assumed that the inventory quantity is greater than or equal to the number of items to be shipped (a ≥ s).

[0035] (Workload of the mobile body) When calculating the initial evaluation value based on the workload of the mobile body 12, the initial evaluation value calculation unit 36 ​​calculates the workload of the mobile body 12 based on the initial solution. The workload of the mobile body 12 here refers to the degree of workload of the work performed by the mobile body 12, assuming that the goods Q are stacked on the pallet P according to the stacking information and stacking order shown in the initial solution. For example, the initial evaluation value calculation unit 36 ​​may calculate the total number of times the mobile body 12 transports the pallet P, assuming that the goods Q are stacked on the pallet P according to the stacking information and stacking order shown in the initial solution, and use that total number as the initial evaluation value.

[0036] In addition, the facility W may be equipped with both AGVs and AGFs as mobile units 12. An AGV is a mobile unit that does not have the function of a forklift and does not have forks capable of lifting pallets P. An AGF is a mobile unit that has the function of a forklift. The AGV performs the task of transporting pallets P between, for example, the goods storage area AR1, the supply pallet placement area AR2, the pallet placement area AR3, the shipment item placement area AR4, and the empty pallet placement area AR5, while the AGF performs the task of transporting pallets P (pallets for shipment) placed in area AR4 to shelves. However, the tasks performed by the AGV and AGF are not limited to these and may be set as appropriate. When both AGVs and AGFs are equipped in this manner, the initial evaluation value calculation unit 36 ​​may calculate at least one of the workload of the AGV and the workload of the AGF as the workload of the mobile unit 12. The workload for the AGV could be, for example, the total number of times the AGV transports pallet P, assuming that goods Q are loaded onto pallet P according to the loading information and loading order shown in the initial solution. The workload for the AGF could be, for example, the total number of times the AGF transports pallet P, assuming that goods Q are loaded onto pallet P according to the loading information and loading order shown in the initial solution.

[0037] (Small Lot Consecutive Count) The small lot consecutive count refers to the number of times that items Q (small lot items Q) with a predetermined number or less allocated to pallet P are stacked consecutively. In other words, if the predetermined number is 2, it refers to the number of times that items Q with a allocation of 2 or less allocated to pallet P are stacked consecutively. If the number of items Q allocated to pallet P is small, the time it takes for the stacking device 14 to stack those items Q will be shorter. On the other hand, if there are many types of items Q allocated to pallet P, the number of times the supply pallet P0 is transported to the supply pallet placement area AR2 will increase. Therefore, if the small lot consecutive count increases, the work time for each item Q of the stacking device 14 will decrease, but the number of times the supply pallet P0 is transported will increase, which increases the risk that the supply of supply pallets P0 will not keep up when the stacking device 14 is ready to start the next operation. Thus, as the number of consecutive small lots increases, the stacking device 14 has to wait for the arrival of the supply pallet P0, and the number of transports by the mobile body 12 increases. Therefore, the number of consecutive small lots can be said to be an indicator that corresponds to the degree of workload of the shipping operation.

[0038] When calculating the initial evaluation value based on the number of consecutive small lots, the initial evaluation value calculation unit 36 ​​calculates the number of consecutive small lots based on the initial solution. For example, the initial evaluation value calculation unit 36 ​​may calculate the number of consecutive small lots assuming that goods Q are stacked on pallet P according to the stacking information and stacking order shown in the initial solution, and use that number of consecutive small lots as the initial evaluation value.

[0039] (Combinations of initial evaluation values) The initial evaluation value calculation unit 36 ​​may calculate the initial evaluation value based on at least one of the following: the workload of the stacking device 14, the workload of the mobile body 12, and the number of consecutive small lots. That is, the initial evaluation value calculation unit 36 ​​may calculate the initial evaluation value using only one of the workload of the stacking device 14, the workload of the mobile body 12, and the number of consecutive small lots, or it may calculate the initial evaluation value using a combination of these, or it may calculate the initial evaluation value using all of these.

[0040] More specifically, priorities may be set for the workload of the stacking device 14, the workload of the mobile body 12, and the number of consecutive small lots. In this case, the initial evaluation value calculation unit 36 ​​may use the one with the highest priority among these to calculate the initial evaluation value. In this case, as will be described in more detail later, the information processing device 10 may compare the calculated initial evaluation value with the neighboring evaluation values ​​described later, and if the evaluations are the same, it may use the one with the next highest priority to calculate the initial evaluation value and repeat the same process.

[0041] (Calculation of Neighborhood Evaluation Value) The Neighborhood Evaluation Value Calculation Unit 38 calculates a neighborhood evaluation value, which indicates the degree of workload for shipping operations in the neighboring solutions, based on the neighboring solutions. If multiple neighboring solutions are calculated, the Neighborhood Evaluation Value Calculation Unit 38 calculates a neighborhood evaluation value for each neighboring solution.

[0042] The neighboring evaluation value is an index that indicates the degree of workload when it is assumed that the work of loading goods Q onto pallet P according to the loading information and loading order shown in the neighboring solution and then shipping pallet P is performed. The neighboring evaluation value calculation unit 38 calculates the neighboring solution in the same way as the initial evaluation value calculation method, except that the initial solution is replaced with the neighboring solution. In other words, in this embodiment, the neighboring evaluation value calculation unit 38 calculates the neighboring evaluation value based on at least one of the following: the workload of the loading device 14 that loads goods Q onto pallet P, the workload of the mobile body 12 that transports pallet P, and the number of small lot consecutive orders.

[0043] (Workload of the stacking device) When calculating the neighbor evaluation value based on the workload of the stacking device 14, the neighbor evaluation value calculation unit 38 calculates the workload of the stacking device 14 based on the neighbor solution. For example, the neighbor evaluation value calculation unit 38 may calculate the total number of times the item Q is stacked from the supply pallet P0 to pallet P (the pallet for shipping), assuming that the item Q is stacked on pallet P according to the stacking information and stacking order shown in the neighbor solution, and use that total number as the neighbor evaluation value.

[0044] In the present embodiment, the neighborhood evaluation value calculation unit 38 calculates a neighborhood evaluation value based on the article information, supply pallet information, shipping destination information, and calculation conditions acquired by the work information acquisition unit 30. The total number of times the stowage device 14 stows articles Q from the supply pallet P0 onto the shipping pallet P uses a neighborhood evaluation value calculated in consideration of at least one of multiple gripping, diversion, and reverse stowage. The neighborhood evaluation value calculation unit 38 performs determination and calculation regarding multiple gripping, diversion, and reverse stowage by the same method as the initial evaluation value calculation unit 36.

[0045] (Workload of moving body) When calculating a neighborhood evaluation value based on the workload of the moving body 12, the neighborhood evaluation value calculation unit 38 calculates the workload of the moving body 12 based on the neighborhood solution. For example, the neighborhood evaluation value calculation unit 38 may calculate the total number of times the moving body 12 conveys the pallet P when it is assumed that articles Q are stowed onto the pallet P according to the stowage information and stowage order indicated by the neighborhood solution, and use this total number of times as the neighborhood evaluation value.

[0046] Note that when both an AGV and an AGF are provided, the neighborhood evaluation value calculation unit 38 may calculate at least one of the workload of the AGV and the workload of the AGF as the workload of the moving body 12. Examples of the AGV's workload include the total number of times the AGV conveys the pallet P when it is assumed that articles Q are stowed onto the pallet P according to the stowage information and stowage order indicated by the neighborhood solution. Examples of the AGF's workload include the total number of times the AGF conveys the pallet P when it is assumed that articles Q are stowed onto the pallet P according to the stowage information and stowage order indicated by the neighborhood solution.

[0047] (Number of consecutive small lots) When calculating a neighborhood evaluation value based on the number of consecutive small lots, the neighborhood evaluation value calculation unit 38 calculates the number of consecutive small lots based on the neighborhood solution. For example, the neighborhood evaluation value calculation unit 38 may calculate the number of consecutive small lots when it is assumed that articles Q are stowed onto the pallet P according to the stowage information and stowage order indicated by the neighborhood solution, and use this number of consecutive small lots as the neighborhood evaluation value.

[0048] (Combination of Neighbor Evaluation Values) Note that the neighbor evaluation value calculation unit 38 may calculate the neighbor evaluation value based on at least any one of the workload of the stowage device 14, the workload of the moving body 12, and the continuous number of small lots.

[0049] More specifically, a priority order may be set for each of the workload of the stowage device 14, the workload of the moving body 12, and the continuous number of small lots. In this case, the neighbor evaluation value calculation unit 38 may calculate the neighbor evaluation value using the one with the highest priority among these. In this case, as will be described in detail later, the information processing apparatus 10 compares the initial evaluation value having the highest priority with the neighbor evaluation value, and if the evaluations are the same, the neighbor evaluation value may be calculated using the one having the next highest priority, and the same process may be repeated.

[0050] (Setting of Stowage Information and Stowage Order) The setting unit 40 sets stowage information and stowage order to be used in a stowage plan based on the initial evaluation value and the neighbor evaluation value. Specifically, the setting unit 40 sets the stowage information and stowage order included in the solution having a higher evaluation value among the initial solution and the neighbor solution as the stowage information and stowage order to be used in the stowage plan. That is, for example, when the initial evaluation value has a higher evaluation than the neighbor evaluation value, the setting unit 40 sets the stowage information and stowage order included in the initial solution as the stowage information and stowage order to be used in the stowage plan. On the other hand, when the neighbor evaluation value has a higher evaluation than the initial evaluation value, the setting unit 40 sets the stowage information and stowage order included in the neighbor solution as the stowage information and stowage order to be used in the stowage plan. In the example of the present embodiment, the initial evaluation value and the neighbor evaluation value are set such that the lower the evaluation value is, the higher the evaluation becomes. However, the present invention is not limited thereto, and the setting may be made such that the lower the evaluation value is, the higher the evaluation becomes.

[0051] Furthermore, if multiple neighboring solutions are calculated, the setting unit 40 performs the same processing for each neighboring solution. That is, for example, the setting unit 40 selects the solution with the higher evaluation among the initial solution and the neighboring solutions as the optimal solution. Then, the setting unit 40 compares the evaluation value of that optimal solution with the evaluation value of another neighboring solution and selects the solution with the higher evaluation as the optimal solution. That is, the setting unit 40 compares the evaluation value of the current optimal solution with the evaluation value of an unevaluated neighboring solution and selects the solution with the higher evaluation as the optimal solution. Once the evaluation comparison for all neighboring solutions is complete, the stacking information and stacking order included in the optimal solution at that point are set as the stacking information and stacking order to be used in the stacking plan.

[0052] (Processing Flow) The setting flow for the stacking information and stacking order used in the stacking plan described above will now be explained. Figure 3 is a flowchart illustrating the setting flow for stacking information and stacking order. As shown in Figure 3, the information processing device 10 acquires work information using the work information acquisition unit 30 (step S2). Then, the information processing device 10 calculates an initial solution for stacking information and stacking order based on the work information using the initial solution calculation unit 32 (step S4), and calculates an initial evaluation value corresponding to the initial solution using the initial evaluation value calculation unit 36 ​​(step S6). Furthermore, the information processing device 10 calculates a neighboring solution for stacking information and stacking order based on the initial solution using the neighboring solution calculation unit 34 (step S8), and calculates a neighboring evaluation value corresponding to the neighboring solution using the neighboring evaluation value calculation unit 38 (step S10).

[0053] The information processing device 10, using the setting unit 40, compares evaluation values ​​(in this case, the initial evaluation value and the neighboring evaluation value). If the evaluation value corresponding to the new solution is higher (step S12; Yes), the new solution is set as the optimal solution (step S14). For example, if the neighboring evaluation value is higher than the initial evaluation value, the neighboring solution corresponding to that neighboring evaluation value is set as the optimal solution. If the computation time so far is below a threshold (step S16; Yes), and it is possible to create other neighboring solutions that have not been calculated so far (step S18; Yes), the process returns to step S8, calculates other neighboring solutions that have not been calculated so far, and repeats the subsequent processing. In this case, the neighboring evaluation value corresponding to the newly calculated neighboring solution is compared with the evaluation value corresponding to the optimal solution. If the neighboring evaluation value corresponding to the newly calculated neighboring solution is higher, the newly calculated neighboring solution is updated as the optimal solution. On the other hand, if the neighboring evaluation value corresponding to the newly calculated neighboring solution is lower, the optimal solution is not updated.

[0054] Furthermore, if the evaluation value corresponding to the original solution is higher than the evaluation value of the new solution (Step S12; No), the original solution is set as the optimal solution (Step S15). That is, for example, in the first calculation, if the initial evaluation value is higher than the neighboring evaluation value, the initial evaluation value is set as the optimal solution. Also, for example, when the calculation is repeated, if the evaluation value corresponding to the set optimal solution is higher than the newly calculated neighboring evaluation value, that optimal solution is used as the optimal solution.

[0055] If the calculation time so far exceeds the threshold (Step S16; No.), this process will be terminated. In this case, the stacking information and stacking order of the optimal solution at that point will be set as the stacking information and stacking order used in the stacking plan.

[0056] Similarly, if it is impossible to create other nearby solutions (Step S18; No), this process is terminated. In this case, the stacking information and stacking order of the optimal solution at that point are set as the stacking information and stacking order used in the stacking plan.

[0057] The determination of whether or not it is possible to create other neighboring solutions, as shown in step S18, may be made using any method. For example, using the neighboring solution calculation method currently in use, it is determined whether it is possible to calculate a neighboring solution that differs in some aspects (stack information or stacking order) from the neighboring solutions already calculated. If such a neighboring solution is possible to calculate, it is determined that it is possible to create other neighboring solutions, and the other neighboring solutions are calculated using the neighboring solution calculation method currently in use.

[0058] On the other hand, if it is impossible to calculate a nearby solution with some differences (stack information or stacking order) using the currently used method for calculating nearby solutions, the system will determine if there is a calculation method that is different from the currently used method and has not been used before. If such a calculation method exists, the system will determine that it is possible to create other nearby solutions, switch to that calculation method, and continue calculating nearby solutions using that method. On the other hand, if there is no calculation method that has not been used before (i.e., all calculation methods have been used), the system will determine that it is impossible to create other nearby solutions. When using multiple calculation methods in this way, the calculation methods may be given priority and applied in order of highest priority.

[0059] Furthermore, when calculating the evaluation value, a priority order may be set for which to use: the workload of the stacking device 14, the workload of the moving body 12, or the number of consecutive small lots. In this case, the information processing device 10 uses the one with the highest priority among these to calculate the evaluation value (initial evaluation value and neighboring evaluation value). Then, as shown in steps S12 to S16, if the evaluations of the calculated evaluation values ​​differ, the solution with the higher evaluation value is set as the optimal solution. On the other hand, if the evaluations of the calculated evaluation values ​​are the same, the information processing device 10 uses the one with the next highest priority to recalculate the evaluation value (initial evaluation value and neighboring evaluation value) and repeats the same process.

[0060] The priority order in calculating the evaluation value may be set as appropriate. For example, the priority may be set in the order of workload of the stacking device 14, workload of the mobile body 12, and number of consecutive small lots, or the priority may be set in the order of workload of the stacking device 14, number of consecutive small lots, and workload of the mobile body 12. Alternatively, the priority may be set in the order of workload of the mobile body 12, workload of the stacking device 14, and number of consecutive small lots, or the priority may be set in the order of workload of the mobile body 12, number of consecutive small lots, and workload of the stacking device 14. Alternatively, the priority may be set in the order of number of consecutive small lots, workload of the stacking device 14, and workload of the mobile body 12, or the priority may be set in the order of number of consecutive small lots, workload of the mobile body 12, and workload of the stacking device 14. Furthermore, as described above, the workload of the mobile unit 12 may be divided into the workload of the AGV and the workload of the AGF, and priorities may be set for the workload of the AGV and the workload of the AGF.

[0061] (Execution of stacking work) The information processing device 10 calculates the initial solution and the nearest solution as described above, and sets the stacking information and stacking order to be used in the stacking plan based on their evaluation values ​​(initial evaluation value and nearest evaluation value). The stacking plan may also include information on the supply pallet P0 that supplies the goods Q to the pallet P. The output control unit 42 of the information processing device 10 outputs the generated stacking plan (information including stacking information and stacking order) to an external device. For example, in this embodiment, the output control unit 42 outputs the stacking plan to another device of the stacking system 100, and the other device executes the stacking work (stacking goods Q onto pallet P) according to the stacking plan. That is, for example, the output control unit 42 transmits the stacking plan to a control unit (not shown) that controls the operation of the stacking device 14, and the control unit controls the operation of the stacking device 14 to execute the stacking work according to the stacking plan. For example, the output control unit 42 transmits a stacking plan to a control unit (not shown) that controls the operation of the mobile body 12, and that control unit controls the operation of the mobile body 12 to perform stacking work according to the stacking plan.

[0062] (Effects) As described above, the information processing device 10 according to this embodiment calculates an initial solution and a neighboring solution, and sets the stacking information and stacking order to be used in the stacking plan based on their evaluation values ​​(initial evaluation value and neighboring evaluation value). Therefore, according to this embodiment, an appropriate stacking plan can be created according to the work content. That is, by creating a neighboring solution that is partially modified from the initial solution and evaluating it, an optimized stacking plan can be set according to the work content. Furthermore, by using a neighboring solution that is partially modified from the initial solution without changing the entire initial solution, an optimized stacking plan can be set according to the work content while suppressing excessive computational load.

[0063] (Other) In this embodiment, the nearest neighbor solution was always calculated and the stacking plan was created, but this is not limited to this, and it is possible to decide whether or not to use the nearest neighbor solution. That is, for example, the information processing device 10 decides whether or not to calculate the nearest neighbor solution, and if it decides to calculate the nearest neighbor solution, it creates the stacking plan in the manner described in the above embodiment. On the other hand, if it decides not to calculate the nearest neighbor solution, the information processing device 10 does not calculate the nearest neighbor solution and the nearest neighbor evaluation value, and sets the stacking information and stacking order in the initial solution as the solution to be used in the stacking plan. The method for deciding whether or not to calculate the nearest neighbor solution is arbitrary. For example, information on whether or not to calculate the nearest neighbor solution may be input to the information processing device 10 by the user, and the information processing device 10 may decide whether or not to calculate the nearest neighbor solution from the input result. Alternatively, for example, the information processing device 10 may automatically decide whether or not to calculate the nearest neighbor solution. In this case, for example, the information processing device 10 may decide whether or not to use the nearest neighbor solution based on the work information.

[0064] (Second Embodiment) Next, a second embodiment will be described. The second embodiment differs from the first embodiment in that it specifies a concrete method for calculating the nearest solution. In the second embodiment, the parts that are common with the first embodiment will not be explained.

[0065] (Calculation of Initial Solution) Figure 4 is a schematic diagram showing an example of an initial solution in the second embodiment. In the example in Figure 4, the stacking information in the initial solution is set such that items Qa, Qb, and Qc are stacked on pallet Pa in this order, items Qd, Qe, and Qf are stacked on pallet Pb in this order, and items Qg, Qh, and Qi are stacked on pallet Pc in this order.

[0066] The initial solution calculation unit 32 sets the stacking order (the order in which the pallets P are finished) in the initial solution based on the work information. The initial solution calculation unit 32 may set the stacking order in the initial solution in any way based on the work information. For example, the initial solution calculation unit 32 may set the stacking order in the initial solution based on the stacking information in the initial solution calculated based on the work information. That is, the initial solution calculation unit 32 may calculate the stacking order in the initial solution based on the type and number of items Q assigned to each pallet P (item information in the initial solution) and the stacking order of items Q on each pallet P (stacking information in the initial solution).

[0067] In the example in Figure 4, the stacking order in the initial solution is set so that the stacking work is performed in the order of pallet Pa, pallet Pb, and pallet Pc. In the example in Figure 4, it is preferable to also include pallet information indicating the supply pallet P0 that will supply the items Qa to Qi assigned to pallet P in the initial solution. Pallet information is set for each item Q assigned to pallet P in the initial solution. The initial solution calculation unit 32 may set the pallet information in any way based on the work information. For example, the initial solution calculation unit 32 may set the pallet information by identifying the supply pallet P0 on which the items Q assigned to pallet P in the initial solution are loaded, based on the work information.

[0068] (Calculation of Neighboring Solutions) The neighboring solution calculation unit 34 may calculate neighboring solutions using any method based on the initial solution, but in this embodiment, the neighboring solution calculation unit 34 calculates one of the first to fourth neighboring solutions, which will be described below, as the initial solution.

[0069] (First Neighborhood Solution) Figure 5 is a schematic diagram showing an example of a first neighborhood solution in the second embodiment. The neighborhood solution calculation unit 34 may calculate a first neighborhood solution as a neighborhood solution, which is a solution obtained by swapping the items Q assigned to multiple pallets P that have the same destination as the initial solution. That is, the neighborhood solution calculation unit 34 swaps the items Q assigned to a pallet P with the items Q assigned to other pallets P that have the same destination as that pallet P, with respect to the stacking information in the initial solution, and sets new stacking information (first stacking information). The neighborhood solution calculation unit 34 calculates a solution obtained by changing the stacking information in the initial solution to the first stacking information as the first neighborhood solution. The first neighborhood solution is the same as the initial solution except that the stacking information has been changed to the first stacking information. That is, the neighborhood solution calculation unit 34 calculates the first neighborhood solution in such a way that it does not change from the initial solution except that the items Q on multiple pallets P that have the same destination as the initial solution are swapped. Note that there are cases where the neighborhood solution is calculated multiple times. In this case, as will be explained in more detail later, the solution with the higher evaluation value between the initial solution and the neighboring solutions is set as the optimal solution, and a new neighboring solution is calculated based on the optimal solution. Therefore, it can be said that the neighboring solution calculation unit 34 calculates a solution as the first neighboring solution by swapping the items Q assigned to multiple pallets P that have the same destination, for the optimal solution (the initial solution with a higher evaluation value than the already calculated neighboring solution, or the already calculated neighboring solution with a higher evaluation value than the initial solution).

[0070] The neighboring solution calculation unit 34 may swap items Q between three or more pallets P that have the same destination, relative to the initial solution (optimal solution). However, it is preferable to calculate the first neighboring solution as the solution obtained by swapping items Q assigned to two pallets P that have the same destination, relative to the initial solution. Furthermore, the neighboring solution calculation unit 34 may swap multiple types of items Q relative to the initial solution (optimal solution). However, it is preferable to calculate the first neighboring solution as the solution obtained by swapping one type of item Q. That is, it is preferable for the neighboring solution calculation unit 34 to calculate the first neighboring solution by swapping one type of item Q assigned to two pallets P that have the same destination using a so-called swap neighborhood. Furthermore, it is preferable that the items Q to be swapped are of different types, but items Q of the same type but different numbers may also be swapped.

[0071] Figure 5 shows an example of a first-neighborhood solution in which items Q assigned to pallets Pa and Pb, which are destined for the same destination, are swapped. In other words, in the example in Figure 4, in the initial solution, items Qa, Qb, and Qc are assigned to pallet Pa in that order, and items Qd, Qe, and Qf are assigned to pallet Pb in that order. In contrast, in the first-neighborhood solution shown in Figure 5, item Qb assigned to pallet Pa and item Qe assigned to pallet Pb are swapped. In other words, in the first-neighborhood solution, items Qa, Qe, and Qc are assigned to pallet Pa in that order, and items Qd, Qb, and Qf are assigned to pallet Pb in that order. That is, in the first-neighborhood solution, the position of the swapped item Q on pallet P (the stacking position of item Qe in Figure 5) is the same as the position of the item Q on pallet P before the swap in the initial solution (the stacking position of item Qb in Figure 4).

[0072] (Second Neighborhood Solution) Figure 6 is a schematic diagram showing an example of a second neighborhood solution in the second embodiment. The neighborhood solution calculation unit 34 may calculate a second neighborhood solution as a neighborhood solution by changing the stacking order of items Q on the pallet P compared to the initial solution. That is, the neighborhood solution calculation unit 34 changes the stacking order of items Q assigned to the pallet P compared to the stacking information in the initial solution and sets new stacking information (second stacking information). The neighborhood solution calculation unit 34 calculates a solution in which the stacking information in the initial solution has been changed to the second stacking information as the second neighborhood solution. The second neighborhood solution is the same as the initial solution except that the stacking information has been changed to the second stacking information. That is, the neighborhood solution calculation unit 34 calculates the second neighborhood solution in such a way that it does not change from the initial solution except that the stacking order of items Q on the pallet P is rearranged. Furthermore, as mentioned above, since there are cases where multiple neighboring solutions are calculated, the neighboring solution calculation unit 34 calculates a solution as a second neighboring solution by changing the stacking order of items Q on the pallet P with respect to the optimal solution (the initial solution with a higher evaluation value than the already calculated neighboring solution, or the already calculated neighboring solution with a higher evaluation value than the initial solution).

[0073] The neighboring solution calculation unit 34 may swap the stacking order of three or more items Q assigned to the same pallet P relative to the initial solution (optimal solution), but it is preferable to calculate a solution in which the stacking order of two items Q assigned to the same pallet P is swapped relative to the initial solution as the second neighboring solution. Furthermore, the neighboring solution calculation unit 34 may swap the stacking order of items Q on multiple pallets P relative to the initial solution (optimal solution), but it is preferable to calculate a solution in which the stacking order of items Q assigned to one pallet P is swapped as the second neighboring solution. In other words, it is preferable for the neighboring solution calculation unit 34 to calculate the second neighboring solution by swapping the stacking order of two items Q on one pallet P using a so-called swap neighborhood.

[0074] Figure 6 shows an example of a second-neighborhood solution in which the stacking order of items Q assigned to pallet Pa is reversed. That is, in the example in Figure 4, in the initial solution, items Qa, Qb, and Qc are assigned to pallet Pa in this order. In contrast, in the second-neighborhood solution shown in Figure 6, the stacking order of items Qb and Qc is reversed, and items Qa, Qc, and Qb are assigned to pallet Pa in this order.

[0075] (Third Neighborhood Solution) Figure 7 is a schematic diagram showing an example of a third neighborhood solution in the second embodiment. The neighborhood solution calculation unit 34 may calculate a third neighborhood solution as a neighborhood solution by changing the order of stacking pallets P with respect to the initial solution. That is, the neighborhood solution calculation unit 34 calculates a solution with a changed stacking order in the initial solution as the third neighborhood solution. The third neighborhood solution is the same as the initial solution except that the stacking order has been changed. That is, the neighborhood solution calculation unit 34 calculates the third neighborhood solution in such a way that it does not change from the initial solution except that the stacking order is changed. As mentioned above, there are cases where neighborhood solutions are calculated multiple times, so it can be said that the neighborhood solution calculation unit 34 calculates a solution as the third neighborhood solution by changing the order of stacking pallets P with respect to the optimal solution (the initial solution with a higher evaluation value than the calculated neighborhood solution, or the calculated neighborhood solution with a higher evaluation value than the initial solution).

[0076] The neighborhood solution calculation unit 34 may swap the stacking order (finishing order) of three or more pallets P with respect to the initial solution (optimal solution), but it is preferable to calculate a solution in which the stacking order of two pallets P is swapped with respect to the initial solution as the third neighborhood solution. That is, it is preferable for the neighborhood solution calculation unit 34 to calculate the third neighborhood solution by swapping the stacking order of two pallets P using a so-called swap neighborhood.

[0077] Figure 7 shows an example of a third-neighborhood solution where the stacking order of pallets Pa and Pb is reversed. That is, in the example in Figure 4, the initial solution is stacked in the order of pallets Pa, Pb, and Pc, but in the third-neighborhood solution shown in Figure 7, the stacking order is pallets Pb, Pa, and Pc.

[0078] (Fourth Neighborhood Solution) The neighborhood solution calculation unit 34 may calculate a solution as a neighborhood solution in which the supply pallet P0 that supplies the item Q assigned to pallet P has been changed from the initial solution. That is, the neighborhood solution calculation unit 34 calculates a solution as the fourth neighborhood solution in which the pallet information (information indicating the supply pallet P0 that will be the source of the item Q) in the initial solution has been changed. The third neighborhood solution is the same as the initial solution except that the pallet information has been changed. That is, the neighborhood solution calculation unit 34 calculates the fourth neighborhood solution in which the supply pallet P0 has been changed from the initial solution except that the supply pallet P0 has been changed. As mentioned above, there are cases in which neighborhood solutions are calculated multiple times, so it can be said that the neighborhood solution calculation unit 34 calculates a solution as the fourth neighborhood solution in which the supply pallet P0 has been changed from the optimal solution (the initial solution with a higher evaluation value than the calculated neighborhood solution, or the calculated neighborhood solution with a higher evaluation value than the initial solution).

[0079] The nearest neighbor solution calculation unit 34 may change the supply pallets P0 for multiple types of items Q in the initial solution (optimal solution), but it is preferable to calculate the fourth nearest neighbor solution by changing only the supply pallet P0 for one type of item Q in the initial solution. Alternatively, it is preferable for the nearest neighbor solution 34 to calculate the fourth nearest neighbor solution by replacing the supply pallet P0 set in the initial solution (optimal solution) with another supply pallet P0 that has the same number of the same items Q. In other words, it is preferable for the nearest neighbor solution 34 to calculate the fourth nearest neighbor solution by swapping two supply pallets P0 that have the same number of the same items Q loaded on them using a so-called swap neighborhood.

[0080] In the example shown in Figure 4, in the initial solution, item Qa is assigned to pallet Pa. In this case, for example, a solution obtained by changing the supply pallet P0 of item Qa assigned to pallet Pa from the initial solution may be calculated as the fourth-neighbor solution.

[0081] (Combinations of each neighborhood solution) When calculating multiple types of neighborhood solutions, the neighborhood solution calculation unit 34 may use at least one of the first to fourth neighborhood solutions to calculate multiple types of neighborhood solutions. That is, the neighborhood solution calculation unit 34 may use only one of the first to fourth neighborhood solutions to calculate multiple types of neighborhood solutions, or it may use a combination of the first to fourth neighborhood solutions to calculate multiple types of neighborhood solutions, or it may use all of the first to fourth neighborhood solutions to calculate multiple types of neighborhood solutions.

[0082] More specifically, a priority order may be set for each of the first to fourth nearest neighbor solutions. In this case, the nearest neighbor solution calculation unit 34 calculates multiple nearest neighbor solutions using the nearest neighbor solution calculation method with the highest priority among the first to fourth nearest neighbor solutions. If, before a predetermined time has elapsed (i.e., while the total calculation time so far is below a threshold), the nearest neighbor solution calculation unit 34 has calculated the nearest neighbor solutions for all combinations using the highest priority calculation method, it may continue calculating the nearest neighbor solutions using the nearest neighbor solution calculation method with the next highest priority. For example, if the first nearest neighbor solution has the highest priority, the nearest neighbor solution calculation unit 34 calculates multiple first nearest neighbor solutions by changing the pallets P and items Q to be swapped. If, before a predetermined time has elapsed (i.e., while the total calculation time so far is below a threshold), the calculation of the first nearest neighbor solutions for all combinations has been completed, it continues the same calculation using the nearest neighbor solution calculation method with the next highest priority.

[0083] (Effects) As described above, in the second embodiment, at least one of the first, second, third, and fourth nearest neighbor solutions is calculated as a nearest neighbor solution to the initial solution or the optimal solution, which is a calculated nearest neighbor solution with a higher evaluation value than the initial solution. The first nearest neighbor solution is a solution obtained by swapping the items Q assigned to multiple pallets P that have the same destination, with respect to the stacking information of the initial solution or the optimal solution. The second nearest neighbor solution is a solution obtained by changing the stacking order of items Q on the pallets P with respect to the stacking information of the initial solution or the optimal solution. The third nearest neighbor solution is a solution obtained by changing the order of the pallets P on which stacking is performed with respect to the stacking order of the initial solution or the optimal solution. The fourth nearest neighbor solution is a solution obtained by changing the supply pallet P0 that supplies the items Q assigned to the pallets P with respect to the initial solution or the optimal solution. By using such nearest neighbor solutions, a more optimized stacking plan can be set according to the work content.

[0084] (Third Embodiment) Next, the third embodiment will be described. The third embodiment differs from the second embodiment in that it specifies a concrete method for calculating the initial solution. In the third embodiment, the parts that are common with the second embodiment will not be explained.

[0085] In the third embodiment, the initial solution calculation unit 32 calculates, as stacking information in the initial solution, item information indicating the type and number of items Q to be assigned to the pallet P, and stacking information indicating how the items Q indicated by the pallet information will be stacked on the pallet P.

[0086] (Setting of item information) (Calculation of the number of individual items) Figures 8 to 11 are schematic diagrams illustrating the setting of item information in the third embodiment. When setting item information in the initial solution, the initial solution calculation unit 32 classifies the items Q to be shipped to the target destination into overall items and individual items, and calculates the number of items Q classified as individual items (number of individual items). Overall items refer to items Q that are stacked to occupy the entirety of each layer, assuming that items Q are stacked on the pallet P so that they occupy the entirety of each layer. Individual items refer to items Q that are left over because they do not occupy the entirety of one layer in that case. In other words, overall items are items Q that are stacked over the entire area where stacking is possible in one layer of the pallet P, and individual items are items Q that are stacked only in a part of the area where stacking is possible in one layer of the pallet P. Based on the information of the number of items Q to be shipped to the target destination and the information of the maximum number of items Q that can be placed, as indicated in the work information, the initial solution calculation unit 32 classifies the items Q into overall items and individual items, and calculates the number of individual items for that item Q. In this embodiment, the initial calculation unit 32 calculates the number of individual items as the remainder when the number of items Q to be shipped to the destination is divided by the maximum number of items that can be placed. That is, for example, if 18 items Q are shipped and the maximum number of items that can be placed is 4, and the items Q are stacked on the pallet P so that each layer is entirely occupied by items Q, then 4 layers will be stacked and 2 items Q will remain. Therefore, in this example, the total number of items will be 16 (4 layers), and the number of individual items will be 2. Note that if there is no remainder when the number of items Q is divided by the maximum number of items that can be placed (i.e., it is divisible), then all items are classified as total items, and the number of individual items will be zero.

[0087] The initial calculation unit 32 classifies the items Q to be shipped to the destination into general items and individual items, and calculates the number of individual items for each item Q. Figure 8 shows an example where 18 items QA, 2 items QB, 12 items QC, and 6 items QD are scheduled to be shipped to a certain destination, illustrating the case where the maximum number of items QA to be placed is 4, the maximum number of items QB to be placed is 2, and the maximum number of items QC and QD to be placed is 3. In this case, in the example of Figure 8, item QA is classified into 16 general items QA1 and 2 individual items QA2, and items QB, QC, and QD, excluding individual items, are all classified into general items QB1, QC1, and QD1.

[0088] (Assignment of individual items) Next, the initial solution calculation unit 32 assigns individual items (items Q in the number of items classified as individual items) to pallet P. Assigning items Q to pallet P means selecting the items Q to be stacked on that pallet P (selecting the pallet P to be stacked with those items Q). The initial solution calculation unit 32 searches whether each individual item of one type of item Q can be placed in one layer of pallet P. Based on the size of the items Q shown in the work information and the width of one layer of pallet P, the initial solution calculation unit 32 searches whether there is an arrangement in one layer where the individual items do not overlap with each other. The search method here can be arbitrary, but for example, the Bottom-Left method may be used. If there is an arrangement in one layer where the individual items do not overlap with each other, the initial solution calculation unit 32 assigns the individual items of those items Q to that pallet P. On the other hand, if the initial solution calculation unit 32 determines that there is no arrangement in which individual items Q do not overlap in a single layer, it will not assign those individual items Q to the pallet P. For example, Figure 9 shows an example in which it was determined that two individual items QA2 could be placed in a single layer on pallet PA, and therefore two individual items QA2 (all individual items QA2) were assigned to pallet PA. If it is determined that an individual item should not be assigned to pallet P, the unit may search to see if it can be assigned to another pallet P, or it may divide the total number of individual items into multiple groups and determine whether it is possible to assign each group to a pallet P.

[0089] Thus, in this embodiment, an individual item Q of a single type is assigned to a pallet PA when it can be arranged in one layer; however, the criteria for assignment are not limited to this. For example, if there is an arrangement in one layer where individual items do not overlap, and the sum of the height of that individual item and the height of all stacked items of the same type (total item height) is less than or equal to a predetermined threshold height, then that individual item may be assigned to a pallet PA. In other words, even if there is an arrangement in one layer where individual items do not overlap, if the total height when stacked with all stacked items of the same type exceeds the threshold height, then that individual item does not need to be assigned to a pallet PA. The total item height can be calculated arbitrarily, but for example, it may be the value obtained by multiplying the total number of stacked items (calculated by dividing the total number of items by the maximum number of items that can be arranged) by the height of the total items, and the same applies hereafter. In other words, for example, in the example in Figure 8, stacking all of the items QA1 results in four layers. Therefore, the total height is the sum of the total height of all four layers of items QA1 and the height of the individual item QA2. If this total height exceeds the threshold height, the individual item QA2 does not need to be assigned to that pallet PA.

[0090] If there are individual items for multiple types of goods Q for a single shipping destination, the initial calculation unit 32 allocates the individual items to the pallet P for each type of goods Q in the same manner. The order in which the individual items are allocated for each type of goods Q can be arbitrary, but it is preferable to start the allocation with the individual items of the goods Q with the lowest height among the multiple types of goods Q.

[0091] Furthermore, if there are individual items for multiple types of items Q, the initial solution calculation unit 32 may assign multiple types of individual items to one layer of a single pallet P. Specifically, the initial solution calculation unit 32 searches for an arrangement of individual items in one layer that does not overlap with already assigned individual items and does not overlap with the individual items to be assigned this time. If the initial solution calculation unit 32 finds an arrangement that does not overlap with already assigned individual items and does not overlap with the individual items to be assigned this time, it assigns those individual items to that pallet P. On the other hand, if there is no such arrangement, the initial solution calculation unit 32 does not assign those individual items to that pallet P, but assigns them to, for example, another pallet P. That is, for example, if two individual items of a different type can be placed on the same layer as the two individual items QA2 on a pallet PA that already has two individual items QA2 assigned, then those two individual items will also be assigned to the same layer as the individual items QA2 on pallet PA. On the other hand, if those individual items cannot be placed on the same layer as individual item QA2 without overlapping, those individual items are assigned to another pallet P.

[0092] The initial solution calculation unit 32 performs the process of assigning individual items to pallet P for each individual item.

[0093] (Assignment of total items to pallets to which individual items have already been assigned) The initial calculation unit 32 assigns total items (the number of items Q classified as total items) to pallets P to which individual items have already been assigned. The initial calculation unit 32 prioritizes assigning total items of the same type as the assigned individual items to that pallet P. Specifically, the initial calculation unit 32 selects total items of the same type as the individual items from among the total items for the target shipping destination and determines whether the selected total items can be assigned to the pallet P to which the individual items have been assigned. The initial calculation unit 32 calculates the total item height of the selected total items (the height when all total items are stacked) and the total height of the assigned individual items, and if the total height is less than or equal to the threshold height, it determines that all of the selected total items can be assigned to that pallet PA and performs the assignment. On the other hand, if the total height is higher than the threshold height, the initial calculation unit 32 does not assign the selected total items to that pallet PA and assigns them to other pallets P. In the example shown in Figure 10, the total height of the four layers containing all of the total items QA1, plus the height of the individual items QA2, is less than or equal to the threshold height. Therefore, all of the total items QA1 are assigned to pallet PA. By prioritizing the assignment of total items of the same type as individual items in this way, it is possible to suppress the increase in the variety of items Q stacked on pallet PA, thereby reducing the number of times the supply pallet P0 for stacking needs to be replaced and reducing working time.

[0094] The initial calculation unit 32 may assign other types of whole items to a pallet P that has already been assigned individual items and whole items. In this case, the initial calculation unit 32 selects a type of whole item that has not yet been assigned from among the whole items for the target shipping destination and determines whether the selected whole item can be assigned to the pallet P that has already been assigned individual items and whole items. The initial calculation unit 32 calculates the total height of the selected whole item (the height when all whole items are stacked), the total height of the assigned whole items, and the height of the assigned individual items. If the total height is less than or equal to the threshold height, the initial calculation unit 32 determines that all of the selected whole items can be assigned to that pallet P and performs the assignment. On the other hand, if the total height is higher than the threshold height, the initial calculation unit 32 does not assign the selected whole items to that pallet P and assigns them to another pallet P. In the example shown in Figure 10, the total height of one layer containing all of the total items QB1, the height of four layers of the total items QA1, and the height of individual items QA2 is less than or equal to the threshold height. Therefore, all of the total items QB1 are allocated to pallet PA.

[0095] Furthermore, if a pallet P to which an individual item has been assigned cannot be assigned a whole item of the same type as that individual item, a whole item of a different type may be assigned to that pallet P. In this case, the initial calculation unit 32 selects a whole item of a type whose assignment has not been determined from among the whole items for the target shipping destination, calculates the sum of the height of the selected whole item (the height when all whole items are stacked) and the height of the assigned individual item, and assigns all of the selected whole item to that pallet PA if the sum is less than or equal to the threshold height. On the other hand, if the sum is higher than the threshold height, the initial calculation unit 32 does not assign the selected whole item to that pallet P, but assigns it to another pallet P.

[0096] The initial calculation unit 32 sequentially selects the types of items for which the assignment has not yet been determined, and repeats the above process for the pallets P to which individual items have been assigned, thereby assigning the total items to the pallets P to which individual items have been assigned.

[0097] The initial calculation unit 32 performs the overall item allocation process described above for each pallet P to which individual items are assigned.

[0098] (Assignment of total items to pallets to which individual items have not been assigned) Once the process of assigning total items to pallets P to which individual items have been assigned has been completed, the initial calculation unit 32 assigns total items (the number of items Q classified as total items) to pallets P to which individual items have not been assigned. The initial calculation unit 32 selects total items of a type to which an assignment has not been decided from among the total items for the target shipping destination and determines whether the selected total items can be assigned to pallets P to which individual items have not been assigned. The initial calculation unit 32 calculates the total item height of the selected total items (the height when all total items are stacked) and determines that all of the selected total items can be assigned to that pallet P if the total item height is less than or equal to the threshold height, and performs the assignment. On the other hand, if the total item height is higher than the threshold height, the initial calculation unit 32 does not assign the selected total items to that pallet P. In this way, by prioritizing the assignment of total items of the same type, it is possible to suppress the number of types of items Q stacked on pallets PA, thereby reducing the number of times the supply pallet P0 for stacking is replaced and reducing working time.

[0099] The initial calculation unit 32 may assign other types of overall items to a pallet P that has already been assigned overall items but no individual items have been assigned to it. In this case, the initial calculation unit 32 selects an overall item of a type whose assignment has not been decided from among the overall items for the target shipping destination and determines whether the selected overall item can be assigned to the pallet P that has already been assigned overall items. The initial calculation unit 32 calculates the total height of the selected overall item and the total height of the already assigned overall item, and if the total height is less than or equal to the threshold height, it determines that all of the selected overall items can be assigned to that pallet P and performs the assignment. On the other hand, if the total height is higher than the threshold height, the initial calculation unit 32 does not assign the selected overall items to that pallet P and assigns them to another pallet P. In the example in Figure 11, the total height of the four layers containing all of the overall items QC1 and the total height of the two layers containing all of the overall items QC1 is less than or equal to the threshold height, so all of the overall items QC1 and QD1 are assigned to pallet PB.

[0100] The initial solution calculation unit 32 sequentially selects all items of a type to which an assignment has not yet been determined, and repeats the above process for each pallet P to which individual items have been assigned, thereby assigning all items to pallets P to which no individual items have been assigned. In other words, if a new all item cannot be assigned to one pallet P, a new pallet P is selected, and the all items are assigned to the newly selected pallet P in the same manner. In the above explanation, the entire number of all items of one type was assigned to one pallet P, but if the entire number of all items cannot be assigned to any pallet P, the entire number of all items may be divided into layers and assigned to each layer on a pallet P.

[0101] The initial calculation unit 32, using the method described above, assigns individual items and the total number of items to pallets P for each shipping destination so that all of the items Q for that destination are assigned to one of the pallets P. As a result, item information indicating the type and number of items Q to be assigned to each pallet P is set for each shipping destination.

[0102] Furthermore, in the above explanation, the overall items were selected and assigned sequentially to pallets P that already had individual items assigned and pallets P that did not have individual items assigned. The order in which the overall items were selected was arbitrary, but for example, the initial calculation unit 32 may alternately select the type of overall item that has the maximum overall item height (the height when all overall items are stacked) and the type of overall item that has the minimum overall item height from among the types of overall items that have not yet been selected. That is, for example, the overall item with the maximum overall item height is selected from among the unselected overall items, then the overall item with the minimum overall item height is selected from among the remaining (unselected) overall items, and then the overall item with the maximum overall item height is selected from among the remaining overall items. By alternately selecting the overall items with the maximum and minimum overall item height in this way, it becomes possible to stack the overall items as high as possible on the pallets P, and the number of pallets P used can be reduced. Alternatively, for example, the initial calculation unit 32 may alternately select the type of overall item with the maximum number of items and the type of overall item with the minimum number of items from among the types of overall items that have not yet been selected. In other words, for example, the system selects the item with the largest quantity from the unselected items, then selects the item with the smallest quantity from the remaining (unselected) items, and then selects the item with the largest quantity from the remaining items. By alternately selecting the items with the largest and smallest quantities in this way, it becomes possible to stack as many items as possible on pallet P, thereby reducing the number of pallets P used.

[0103] Furthermore, if we define a first selection method as alternately selecting the overall item with the maximum overall height and the overall item with the minimum overall height, and a second selection method as alternately selecting the overall item with the maximum number and the overall item with the minimum number, the initial calculation unit 32 may determine which of the first and second selection methods to use based on the number of pallets P required when all items Q for one shipping destination are allocated. In this case, for example, the initial calculation unit 32 may select the unselected overall items using the first selection method and set the item information so that all items Q for the shipping destination are allocated to one of the pallets P using the method described above (performing the allocation between individual items and overall items). Alternatively, the initial calculation unit 32 may select the unselected overall items using the second selection method and set the item information so that all items Q for the shipping destination are allocated to one of the pallets P using the method described above (performing the allocation between individual items and overall items). The initial solution calculation unit 32 then compares the total number of pallets P required when all items Q are allocated using the first selection method with the total number of pallets P required when all items Q are allocated using the second selection method, and may treat the item information set by the selection method with the smaller total number of pallets P as the item information to be used in the stacking plan.

[0104] (Item Information Setting Flow) The processing flow for setting item information in the initial solution, as described above, is now explained. Figure 12 is a processing flow illustrating the method for assigning individual items in the third embodiment, and Figure 13 is a processing flow illustrating the method for assigning all items in the third embodiment. As shown in Figure 12, when assigning individual items to pallet P, the initial solution calculation unit 32 selects individual items (step S22), determines whether the total number of selected individual items can be placed in one layer of the selected pallet P (step S24), and if they can be placed in one layer (step S24; Yes), assigns the total number of those individual items to the pallet P (step S26). If there are no other individual items (step S28; No), this process ends. On the other hand, if there are other individual items (step S28; Yes), the process returns to step S22 and continues the process of selecting other individual items and assigning them to pallet P. If the selected individual items cannot be placed in one layer of pallet P (step S24; No), the process proceeds to step S28. If the selected individual item cannot be placed in one layer of the selected pallet P, another pallet P may be selected, and it may be determined whether the individual item can be placed in one layer on the other pallet P.

[0105] As shown in Figure 13, when all items are to be allocated to pallet P, the initial calculation unit 32 selects pallet P and all items (step S30), determines whether the entire number of selected items can be allocated to the selected pallet P (step S32), and if they can be allocated (step S32; Yes), allocates the entire number of those items to that pallet P (step S34). If there are no other items to allocate (step S36; No), the process ends. On the other hand, if there are other items to allocate (step S36; Yes), the process returns to step S30 and continues the process of selecting other individual items and allocating them to pallet P. Also, if the selected items cannot be allocated to the selected pallet P (step S32; No), the process proceeds to step S36. After determining whether all items can be allocated to the selected pallet P, if there are still items whose allocation has not been decided, the process returns to step S30, selects another pallet P, and determines whether the unallocated items can be allocated to the other pallet P.

[0106] (Setting of stacking information) The item information set in the above description indicates the type and number of items Q (individual items and overall items) to be assigned to pallet P. Therefore, it is not yet determined how the assigned items Q will be arranged on pallet P. For this reason, in this embodiment, after setting the item information in the initial solution, the initial solution calculation unit 32 sets stacking information in the initial solution, which indicates how the items Q will be stacked on pallet P. This will be explained in detail below.

[0107] Figures 14 to 17 are schematic diagrams illustrating the setting of stacking information in the third embodiment. The initial calculation unit 32 sets the stacking order of the same type of item Q assigned to one pallet P as stacking information, treating the same type of item Q as one layer group L. Here, a layer group L can be said to be a collection of layers containing the same type of item Q, and one layer group L may contain one layer of the same type of item Q or multiple layers. The number of layers included in a layer group L is, for example, the value obtained by dividing the total number of items Q by the maximum number of items that can be placed, and if individual items are included, it is the sum of the value obtained by dividing the total number of items Q by the maximum number of items that can be placed and the number of layers of the individual items. That is, for example, the layer group L of item QA shown in Figure 10 contains a total of 5 layers, consisting of 4 layers of total items and 1 layer of individual items.

[0108] The initial solution calculation unit 32 determines whether individual items are assigned to pallet P based on the item information of pallet P. If individual items are assigned, the initial solution calculation unit 32 sets the layer group L of items Q, which includes the individual items, as the top layer group, which is the layer group to be placed on top of pallet P. On the other hand, for pallet P that does not have individual items assigned, the top layer group is not set at this stage. In this way, by placing the layer group L containing individual items as the top layer, items Q can be stacked appropriately. In the layer group L of items Q that includes individual items, the layer of the individual items is set as the top layer among the layers of item Q.

[0109] Furthermore, the initial solution calculation unit 32 determines, based on the item information of the pallet P, whether there is a maximum weight layer group among the layer groups L contained in the pallet P where the weight of one layer of item Q is maximized. If there is a maximum weight layer group, the initial solution calculation unit 32 sets the maximum weight layer group as the bottom layer group, which is the layer group to be placed at the very bottom of the pallet P. In this case, for example, the initial solution calculation unit 32 can determine whether there is a maximum weight layer group by calculating the weight of one layer of item Q for each layer group L from the weight of item Q and the maximum number of items that can be placed, and comparing them. Note that if there is no maximum weight layer group, that is, if there are multiple layer groups L where the weight of one layer of item Q is maximized, it is not necessary to set a bottom layer group. In this way, by placing the heaviest layer at the bottom, it becomes possible to stack items while maintaining weight balance.

[0110] In the example in Figure 14, pallet PA is assigned item QA, which includes individual item QA2, and item QB, which does not include individual item QA. Therefore, in the example in Figure 14, layer group L2 of item QA is set as the uppermost layer group of pallet PA, and layer group L1 of the remaining item QB is set as the layer group below layer group L2 (the lowest layer group in this example). In the example in Figure 15, pallet PB is assigned items QC and QD, which do not include individual item QA, and layer group QC is the layer group with the maximum weight. Therefore, in the example in Figure 15, layer group L1 of item QC is set as the lowest layer group of pallet PB, and layer group L2 of the remaining item QD is set as the layer group above layer group L1 (the uppermost layer group in this example).

[0111] Next, the initial calculation unit 32 extracts popular items (target items) from among the items Q assigned to each destination pallet P. Here, popular items refer to the item Q that has the largest number of assigned destinations. That is, for example, if item QD is shipped to three destinations and items QA, QB, and QD are shipped to two destinations, then item QD will be the popular item. However, popular items are not limited to the item Q that has the largest number of assigned destinations, but may refer to any type of item Q selected by any method from the items Q assigned to each destination pallet P, for example, the item Q with the largest total quantity per destination.

[0112] Once popular items have been extracted, the initial calculation unit 32 selects the first pallet from among the pallets P for each shipping destination that has the largest number of assigned popular items. The initial calculation unit 32 sets the layer group L containing the popular items assigned to the selected first pallet as the lowest layer group of the selected first pallet. If there are multiple pallets P that have the largest number of assigned popular items, any one of them may be selected as the first pallet. If the lowest layer group of the selected first pallet has already been set, the unit selects a new first pallet from among the unselected pallets P excluding that first pallet that has the largest number of popular items, and performs the same process. For example, in the example in Figure 16, pallet PC has the largest number of popular items QD among the pallets P, so the layer group L1 of item QD is set as the lowest layer group.

[0113] Furthermore, the initial calculation unit 32 selects a second pallet from among the pallets P for each shipping destination that has the smallest number of popular items assigned to it. The initial calculation unit 32 sets the layer group L containing the popular items assigned to the selected second pallet as the second-to-last second layer group of the selected second pallet. The second layer group can also be said to be the layer group directly above the lowest layer group. If there are multiple pallets P that have the smallest number of popular items assigned to them, any one of them may be selected as the second pallet. Also, if the second layer group of the selected second pallet has already been set, the unit selects a new second pallet from among the unselected pallets P excluding that second pallet that has the smallest number of popular items and performs the same process. For example, in the example in Figure 17, pallet PD has the smallest number of popular items QD among the pallets P, so the layer group L2 of item QD is set as the second layer group. In the example in Figure 17, the layer group L1 of item QA is the heaviest layer group, so it is set as the lowest layer group.

[0114] Once the bottom layer of the first pallet and the second layer of the second pallet have been set, the initial calculation unit 32 selects the first and second pallets from among the unselected pallets P that have not been selected as the first and second pallets so far, and repeats the process of setting the bottom layer of the first pallet and the process of setting the second layer of the second pallet. Furthermore, once all pallets P have been selected for the extracted popular items, the initial calculation unit 32 extracts the item Q with the largest number of assigned shipping destinations from among the items Q excluding the popular items extracted so far, and repeats the process of selecting the first pallet and setting its bottom layer, and the process of selecting the second pallet and setting its second layer. In this way, by making the bottom layer of the first pallet and the second layer of the second pallet the same popular items, it becomes possible to stack those popular items consecutively on the second pallet and the first pallet, thereby reducing the time required for stacking and lowering the workload of the stacking work.

[0115] Next, the initial solution calculation unit 32 extracts pallets P whose stacking order for all layer groups L has not been determined (pallets P whose stacking order for at least some of the layer groups L has not been determined). For the extracted pallets P, the initial solution calculation unit 32 selects the largest layer group from among the layer groups L whose stacking order has not been determined that has the maximum number of items Q, and sets the selected largest layer group as the lowest layer group among the layer groups L whose stacking order has not been determined. That is, for example, if the lowest layer group has been determined and the second layer group (the layer group directly above the lowest) has not been determined, the largest layer group will be set as the second layer group, and if the lowest layer group and the second layer group have been determined, the largest layer group will be set as the layer group directly above the second layer group. If there are multiple layer groups L that have the maximum number of items Q, any one of them may be selected as the largest layer group.

[0116] Next, the initial solution calculation unit 32 selects the smallest layer group from the undecided layer group L for the extracted pallet P, which has the minimum number of items Q, and sets the selected smallest layer group as the bottom layer group among the undecided layer group L. Next, the initial solution calculation unit 32 newly selects the largest layer group from the undecided layer group L for the extracted pallet P, which has the maximum number of items Q, and sets the selected largest layer group as the bottom layer group among the undecided layer group L. In other words, the initial solution calculation unit 32 sets the stacking order for the undecided layer group L such that the largest layer group and the smallest layer group are stacked alternately from the bottom. For example, in the example in Figure 16, in pallet PC, the largest of the undetermined layer groups L is layer group L2 containing item QB, and the smallest layer group is layer group L3 containing item QC, so the stacking order is set in the order of layer group L1, layer group L2, and layer group L3. Also, for example, in the example in Figure 17, in pallet PD, the largest of the undetermined layer groups L is layer group L3 containing item QC, and the smallest layer group is layer group L4 containing item QB, so the stacking order is set in the order of layer group L1, layer group L2, layer group L3, and layer group L4. In this way, by stacking the largest and smallest layer groups alternately on a single pallet P, it becomes possible to average the replacement time of the supply pallets P0 stacked on a single pallet P, reducing the time required for stacking and lowering the workload of the stacking work. In other words, for example, when stacking layers L with a small number of items Q in succession, the stacking of layer L is completed early in succession, which may result in not being able to replace the supply pallet P0 in time, and stacking may have to wait until the replacement of the supply pallet P0 is completed. In contrast, by stacking the largest layer group and the smallest layer group alternately, the opportunities to wait until the replacement of the supply pallet P0 is completed can be reduced.

[0117] The initial solution calculation unit 32 performs the above processing for all pallets P, thereby setting stacking information for all pallets P.

[0118] (Setting Flow of Stacking Information) The processing flow for setting stacking information in the initial solution described above will now be explained. Figure 18 is a flowchart illustrating the processing flow for setting stacking information in the third embodiment. As shown in Figure 18, the initial solution calculation unit 32 sets the layer group L containing individual items as the uppermost layer group and the heaviest layer group as the lowest layer group for each pallet P in which item information has been set (step S40). Then, the initial solution calculation unit 32 extracts popular items (step S41), selects a first pallet containing the maximum number of popular items and a second pallet containing the minimum number of popular items from each pallet P (step S42), sets the lowest layer group of the first pallet as the layer group of popular items, and sets the second layer group of the second pallet as the layer group of popular items (step S43). Subsequently, it is determined whether all pallets have been selected as either the first or second pallet for the extracted popular items (step S44). If the selection is not complete (step S44; No), the process returns to step S42 and continues selecting the first and second pallets from the unselected pallets P. If all pallets have been selected (step S44; Yes), it is determined whether all items Q have been extracted as popular items from each item Q (step S45). If the extraction is not complete (step S45; No), the process returns to step S41 and continues extracting popular items from the remaining items Q that have not been extracted so far. If the extraction is complete (step S45; Yes), the initial solution calculation unit 32 sets the stacking order for each pallet P so that the largest and smallest layer groups are arranged alternately (step S46).

[0119] (Setting the stacking order) As explained above, the initial solution calculation unit 32 sets, for each pallet P, item information indicating the items Q to be stacked on that pallet P, and stacking information indicating the stacking order of the items Q, as the initial solution. Once the stacking information in the initial solution is set, the initial solution calculation unit 32 sets the stacking order in the initial solution, which indicates the order in which the stacking work for each pallet P is performed, based on the stacking information for each pallet P. The initial solution calculation unit 32 also sets the supply pallet P0 to be used to supply items Q to the pallet P, based on the work information (supply pallet information). The following will explain this in detail.

[0120] The initial solution calculation unit 32 determines, based on the stacking information, whether there are two pallets P to which layers of the same type of item Q are assigned. If there are two pallets P to which layers of the same type of item Q are assigned, the stacking order of those pallets P is set to be consecutive. In other words, in this case, the stacking order is set so that as soon as the stacking of the item Q on one pallet P to which layers of the same type of item Q are completed, the item Q on the other pallet P is stacked. This makes it possible to supply the same type of item Q to two pallets P with consecutive stacking orders using one supply pallet P0, thereby reducing the number of times the supply pallet P0 placed in the supply pallet placement area AR2 is changed, for example, and thus reducing the workload and working time.

[0121] (Selection of the First Pallet) Figure 19 is a schematic diagram illustrating the setting of the stacking order in the third embodiment. More preferably, in the third embodiment, the initial calculation unit 32 extracts popular items (target items) from among the items Q assigned to each destination's pallet P. The initial calculation unit 32 extracts as the first pallet the pallet P in which popular items are assigned to a layer above the lowest layer group (a layer group of the second layer or higher) among the pallets P. If there are multiple pallets P in which popular items are assigned to a layer above the lowest layer group, any one of them may be selected as the first pallet. The popular items here refer to the item Q with the largest number of assigned destinations, as described above, but are not limited to that. They may refer to any type of item Q selected by any method from the items Q assigned to each destination's pallet P, for example, the item Q with the largest total number per destination.

[0122] (Selection of Pallet to Use for First Pallet) The initial calculation unit 32 selects a supply pallet P0 (pallet to use) to supply items Q other than popular items assigned to the first pallet to the first pallet, based on the stacking information for the first pallet and the supply pallet information included in the work information. The initial calculation unit 32 selects a pallet to use from among the respective supply pallets P0 to supply items Q other than popular items, based on the types and number of items Q other than popular items assigned to the first pallet as shown in the stacking information and the types and number of items Q to be loaded on the supply pallet P0 as shown in the supply pallet information. More specifically, in this embodiment, the initial calculation unit 32 determines whether there is a supply pallet P0 that has the same number of items Q (items of the same type) as the number of items Q other than popular items assigned to the first pallet. If there is a supply pallet P0 that has the same number of items Q of the same type, the initial calculation unit 32 selects that supply pallet P0 as the pallet to use to supply items Q to the first pallet. If there are multiple supply pallets P0 that contain the same number of the same type of item Q, any one of them may be selected as the pallet to use. On the other hand, if there are no supply pallets P0 that contain the same number of the same type of item Q, the initial calculation unit 32 selects the supply pallet P0 that contains the smallest number of the item Q from among the supply pallets P0 as the pallet to use.

[0123] Once the initial calculation unit 32 has selected a pallet to be used, it sets the same number of items Q that will be loaded onto that pallet as the number of items Q that will be allocated to the first pallet, as items Q to be supplied to that first pallet.

[0124] The initial calculation unit 32 selects a pallet to use for each of the items Q other than popular items that are assigned to the first pallet, using the method described above. For the supply pallet P0 selected as the pallet to use, the number obtained by subtracting the number of items Q to be supplied to pallet P (in this case, the first pallet) from the original number of items Q to be loaded onto the supply pallet P0 is used as the number of items Q to be loaded onto the supply pallet P0, and the subsequent pallet selection process is carried out in the same manner thereafter.

[0125] (Selection of the second pallet and setting of the stacking order) The initial calculation unit 32 extracts from each pallet P the pallet P to which popular items are assigned to the bottom layer group as the second pallet. If there are multiple pallets P to which popular items are assigned to the bottom layer group, any one of them may be selected as the second pallet. More specifically, in this embodiment, if the sum of the number of popular items in the bottom layer group of pallet P and the number of popular items in the first pallet is the same as the number of popular items to be loaded onto the supply pallet P0 indicated in the work information (supply pallet information), then that pallet P (pallet P to which popular items are assigned to the bottom layer group) is selected as the second pallet. In other words, if there is a supply pallet P0 to which the same number of popular items as the above-mentioned sum is loaded, the initial calculation unit 32 selects the pallet P used to calculate that sum as the second pallet.

[0126] The initial solution calculation unit 32, once it has selected the second pallet, determines the stacking order so that the stacking of the second pallet is performed immediately after the stacking of the first pallet. The initial solution calculation unit 32 also selects a supply pallet P0 as the pallet to supply popular items to the first and second pallets, which will have the same number of popular items as the sum of the number of popular items assigned to the first and second pallets. For example, in the example shown in Figure 19, popular items QD are assigned to the second layer group L2 (the layer above the bottom layer group) of pallet PA, and popular items QD are assigned to the bottom layer group L1 of pallet PB. The supply pallet P0A will then have the same number of items QD (3 layers) as the sum of the number of items QD in the second layer group L2 of pallet PA (1 layer) and the number of items QD in the bottom layer group L1 of pallet PB (2 layers). Therefore, in the example shown in Figure 19, the stacking order is set so that stacking is performed on pallet PB (second pallet) immediately after pallet PA (first pallet), and supply pallet P0A is selected as the pallet used to supply goods QD to pallets PA and PB. In this way, by setting supply pallet P0, which can use up all of the popular goods on the first and second pallets, as the pallet used, and by stacking the first and second pallets consecutively, it becomes possible to repurpose supply pallet P0, which has used up all of the popular goods, as pallet P (in this case, the second pallet). This reduces the transport load on supply pallet P0, and thus can suitably reduce working time and workload.

[0127] On the other hand, in this embodiment, if there is no supply pallet P0 that has the same number of popular items as the total value mentioned above, the pallet P with the largest number of popular items assigned to the bottom layer is selected as the second pallet. The initial calculation unit 32 determines the stacking order so that the stacking of the second pallet is performed immediately after the stacking of the first pallet. Then, if the initial calculation unit 32 has selected the pallet P with the largest number of popular items as the second pallet, it selects the supply pallet P0 with the largest number of popular items to be loaded from among the respective supply pallets P0 as the pallet used to supply popular items to the first and second pallets. That is, for example, if there is no supply pallet P0A that has three layers' worth of items QD loaded, as shown in Figure 19, the supply pallet P0 with the largest number of items QD loaded from among the respective supply pallets P0 is selected as the pallet used to supply items QD to pallets PA and PB. Alternatively, one supply pallet P0 may be used to supply popular items to both the first and second pallets, or a separate pallet may be selected to supply popular items to the first pallet and another to supply popular items to the second pallet.

[0128] Furthermore, the initial calculation unit 32 also selects a pallet to use for supplying items Q other than the popular items assigned to the second pallet to the second pallet. The method for selecting the pallet to use here is the same as the method for selecting the pallet to use for supplying items Q other than the popular items to the first pallet, so an explanation is omitted.

[0129] As described above, in this embodiment, a first pallet and a second pallet are selected, the stacking order is set so that the stacking of the second pallet is performed immediately after the stacking of the first pallet, and the pallets to be used for the first and second pallets are selected. Subsequently, the initial calculation unit 32 selects the second pallet as the first pallet, sets the items Q included in the layers above the lowest layer group of the newly selected first pallet (the original second pallet) as popular items, and performs the same processing as described above. The initial calculation unit 32 continues this process until all pallets are selected as either the first or second pallet, thereby setting the stacking order and selecting the pallets to be used for all pallets.

[0130] (Processing Flow) Next, the processing flow for setting the stacking order and pallet to be used, as described above, will be explained. Figure 20 is a flowchart illustrating the processing flow for setting the stacking order and pallet to be used in the third embodiment. As shown in Figure 20, the initial solution calculation unit 32 selects a pallet P to which a group of layers L containing popular items is assigned in layers above the bottom layer group as the first pallet (step S50), and selects a pallet to be used to supply items Q other than popular items to the first pallet (step S52). Then, the initial solution calculation unit 32 extracts a pallet P containing popular items in the bottom layer group (step S54), and determines whether there is a supply pallet P0 on which the same number of popular items as the sum of the number of popular items on the extracted pallet P and the first pallet exists (step S56). If there is a supply pallet P0 that contains the same number of popular items as the total value (step S56; Yes), the initial calculation unit 32 selects the extracted pallet P as the second pallet (step S58), sets the stacking order in the order of the first pallet and then the second pallet, and selects the supply pallet P0 that contains the same number of popular items as the number of popular items as the pallet to supply popular items to the first and second pallets (step S60). After that, the initial calculation unit 32 selects a pallet to supply items Q other than popular items to the second pallet (step S62), and if the stacking order has been set for all pallets P (step S64; Yes), this process ends, and if the stacking order has not been set for all pallets P (step S64; No), it returns to step S52, sets the second pallet as the new first pallet, and continues the process.

[0131] Furthermore, if there is no supply pallet P0 that has the same number of popular items as the total value (step S56; No), the initial calculation unit 32 selects the pallet P with the maximum number of popular items from among the extracted pallets P as the second pallet (step S66), sets the stacking order in the order of the first pallet and then the second pallet, and selects the supply pallet P0 with the maximum number of popular items to be loaded as the pallet to be used to supply popular items to the first and second pallets (step S68), and proceeds to step S62.

[0132] (Effects) As described above, in the third embodiment, an initial solution for the stacking order of each pallet P is set based on the stacking information of a plurality of pallets P. According to this embodiment, since the initial solution for the stacking order of pallets P is set according to the type, number, and stacking order of the items Q to be stacked on each pallet P, an appropriate stacking plan can be created according to the work content. Furthermore, in this embodiment, the stacking order of the first pallet, to which popular items are assigned to the layers above the bottom layer, and the second pallet, to which popular items are assigned to the bottom layer, is made continuous. This further reduces the number of times the supply pallets P0 placed in the supply pallet placement area AR2 are replaced, thereby reducing the workload and working time. In addition, in this embodiment, a supply pallet P0 that can just use up all the popular items in the first and second pallets is set as the pallet to be used, and the first and second pallets are stacked continuously. This makes it possible to repurpose the supply pallet P0 that has used up all the popular items as the second pallet, for example, reducing the transport load of the supply pallet P0 and suitably reducing working time and workload.

[0133] (Fourth Embodiment) Next, the fourth embodiment will be described. In the fourth embodiment, the method for setting the stacking order in the initial solution differs from that of the third embodiment. In the fourth embodiment, parts that have the same configuration as the third embodiment will not be explained.

[0134] (Setting of stacking information) In the third embodiment, when setting the stacking information in the initial solution, the initial solution calculation unit 32 sets the layer group L containing popular items as the lowest layer group for the first pallet with the maximum number of popular items, and sets the layer group L containing popular items as the second layer group (second to last layer group) for the second pallet with the minimum number of popular items. In contrast, in the fourth embodiment, the initial solution calculation unit 32 sets the layer group L containing popular items as the lowest layer group for the first pallet, and sets the layer group L containing popular items as the uppermost layer group for the second pallet. Then, the second pallet is selected as a new first pallet, the items Q in the uppermost layer group of the new first pallet are set as popular items, and the pallet P containing popular items is selected as a new second pallet, and the process of setting the popular items of that second pallet as the lowest layer group is repeated. The method of setting stacking information other than this point is the same as in the third embodiment, so the explanation is omitted.

[0135] (Setting the stacking order) Figure 21 is a schematic diagram illustrating the setting of the initial solution for the stacking order in the fourth embodiment. In the fourth embodiment as well, the initial solution calculation unit 32 sets the stacking order based on the stacking information of each pallet P, similar to the second embodiment. Preferably, if there are two pallets P to which layers of the same type of article Q are assigned, the stacking order of those pallets P is set to be consecutive. However, in the fourth embodiment, the more preferred method for setting the stacking order and the pallets to be used differs from that of the third embodiment. This will be explained in detail below.

[0136] In the fourth embodiment, the initial calculation unit 32 selects the pallet P from among the various pallets P that contains the largest number of items Q in the uppermost layer as the first pallet. However, the method of selecting the first pallet is not limited to this, and the first pallet may be selected from each of the pallets P in any way.

[0137] The initial solution calculation unit 32 selects as the second pallet a pallet P whose lowest layer contains the same type of item Q as the item Q contained in the uppermost layer of the first pallet. If there are multiple pallets P whose lowest layer contains the same type of item Q as the item Q contained in the uppermost layer of the first pallet, any one of them may be selected as the second pallet.

[0138] The initial sorting unit 32, once it has selected the second pallet, determines the stacking order so that the stacking of the second pallet is performed immediately after the stacking of the first pallet. Subsequently, the initial sorting unit 32 selects the second pallet as the new first pallet, and then selects a new second pallet that contains the items Q included in the top layer of the newly selected first pallet (the original second pallet) in its bottom layer, and performs the same process as above. The initial sorting unit 32 continues this process until all pallets are selected as either the first or second pallet, thereby setting the stacking order for all pallets. In this way, by having the first and second pallets, which contain the same items Q in their top and bottom layers, stacking can be performed consecutively, allowing the supply pallet P0 that has been transferred to the first pallet to be used as the second pallet. This reduces the number of times the supply pallet is moved from the supply pallet placement area AR2 to another location to continue stacking, thereby suitably reducing working time and workload.

[0139] (Selection of pallets to use) After setting the stacking order, the initial solution calculation unit 32 selects the pallets to use to supply the items Q to each pallet P.

[0140] Specifically, the initial solution calculation unit 32 extracts one type of item Q1 from among the items Q assigned to each destination pallet P using any method. Then, from among the pallets P for which the stacking order has been set, the initial solution calculation unit 32 extracts a pallet P that contains the extracted item Q1 in the lowest layer group, and from among the pallets P that contain item Q1 in the lowest layer group, selects the pallet P with the largest number of items Q1 as the first pallet.

[0141] The initial calculation unit 32 selects a pallet P as the second pallet that is stacked in front of the first pallet and to which items Q1 included in the bottom layer of the first pallet are assigned. The initial calculation unit 32 then determines whether there is a supply pallet P0 that has the same number of items Q1 as the sum of the number of items Q1 in the bottom layer of the first pallet and the number of items Q1 included in the second pallet. If there is a supply pallet P0 that has the same number of items Q1, the initial calculation unit 32 selects that supply pallet P0 as the pallet to be used to supply those items Q1 to the first and second pallets. If there is no supply pallet P0 that has the same number of items Q1, another pallet P that is stacked in front of the first pallet and to which items Q1 included in the bottom layer of the first pallet are assigned is newly selected as the second pallet, and the same process is performed.

[0142] For example, in the example shown in Figure 21, the top layer L1 of pallet PA is assigned to the item QD, and the bottom layer L1 of pallet PB is assigned to the item QD. The supply pallet P0A is then loaded with the same number of item QDs (four layers) as the sum of the number of item QDs in the top layer L1 of pallet PA (two layers) and the number of item QDs in the bottom layer L1 of pallet PB (two layers). Therefore, in the example shown in Figure 21, the stacking order is set so that the stacking operation is performed on pallet PB (second pallet) immediately after pallet PA (first pallet), and the supply pallet P0A is selected as the pallet used to supply the item QD to pallets PA and PB. In this way, by setting a supply pallet P0 that can use up exactly the amount of goods Q1 on both the first and second pallets as the pallet to be used, it becomes possible to repurpose the supply pallet P0, which has used up the goods Q1 that have been transferred to the first pallet as needed, as pallet P (in this case, the second pallet). This reduces the transport load on the supply pallet P0, and thus can suitably reduce working time and workload.

[0143] The initial solution calculation unit 32 determines whether all pallets P belong to the first pallet or the second pallet, and for those that belong to the first or second pallet, it performs the same processing as above to assign a pallet to be used. On the other hand, for pallets P that do not belong to the first or second pallet, a pallet to be used is assigned using the method described below. Hereinafter, pallets P that do not belong to the first or second pallet will be referred to as remaining pallets as appropriate, and the method of assigning pallets to be used will be explained.

[0144] The initial solution calculation unit 32 extracts multiple (preferably two) remaining pallets to which item Q1 will be assigned. The initial solution calculation unit 32 then determines whether there is a supply pallet P0 that contains the same number of items Q1 as the sum of the number of items Q1 assigned to the extracted multiple remaining pallets. If there is a supply pallet P0 that contains the same number of items Q1, the initial solution calculation unit 32 selects that supply pallet P0 as the pallet to be used to supply the items Q1 to the extracted multiple remaining pallets. The initial solution calculation unit 32 repeats this process until there is no longer a supply pallet P0 that contains the same number of items Q1 as the sum of the number of items Q1 for all remaining pallets.

[0145] Then, if there are any remaining pallets that have not been assigned a pallet to use, the initial calculation unit 32 selects a supply pallet P0 that has a larger number of items Q1 loaded on it than the number of items Q1 that will be assigned to the remaining pallet, as the pallet to use for supplying items Q1 to that remaining pallet.

[0146] The initial calculation unit 32 selects the pallets to supply item Q1 to each pallet P to which item Q1 is assigned, using the method described above. Subsequently, the initial calculation unit 32 extracts items Q that have not yet been extracted as item Q1 from among the items Q assigned to each destination pallet P, and performs the process of selecting the pallets to be used in the same manner as described above. The initial calculation unit 32 performs this process until all items Q have been extracted as item Q1, thereby selecting the pallets to be used for all items Q.

[0147] (Processing Flow) Next, the processing flow for setting the stacking order and the pallets to be used in the initial solution in the fourth embodiment described above will be explained. Figure 22 is a flowchart illustrating the processing flow for setting the stacking order in the fourth embodiment, and Figure 23 is a flowchart illustrating the processing flow for setting the pallets to be used in the fourth embodiment.

[0148] As shown in Figure 22, when setting the stacking order, the initial solution calculation unit 32 selects a first pallet from among the pallets P (step S70), determines if there is a pallet P in which the same item Q as the top group of items Q of the first pallet is assigned to the bottom group (step S72), and if there is (step S72; Yes), selects that pallet P as the second pallet (step S74). Then, the initial solution calculation unit 32 sets the stacking order in the order of the first pallet and then the second pallet (step S76), and if the setting of the stacking order is completed for all pallets P (step S78; Yes), this process ends, and if it is not completed (step S78; No), returns to step S70 and repeats the process. If there is no pallet P in which the same item Q as the top group of items Q of the first pallet is assigned to the bottom group (step S72; No), proceeds to step S78.

[0149] As shown in Figure 23, when setting the pallets to be used, the initial calculation unit 32 extracts the items Q1 and selects a first pallet from among the pallets P that contain items Q1 in the bottom layer, the pallet with the maximum number of items Q1 in the bottom layer (step S80), and selects a second pallet to which items Q1 will be assigned from among the pallets P that are stacked in front of the first pallet (step S82). Then, the initial calculation unit 32 determines whether there is a supply pallet P0 that has the same number of items Q1 as the sum of the number of items Q1 in the first and second pallets (step S84), and if there is (step S84; Yes), selects that supply pallet P0 as the pallet to be used to supply items Q1 to the first and second pallets (step S86). Furthermore, it is preferable to use this supply pallet P0 as the supply pallet for the first pallet, and then repurpose the supply pallet P0 after the supply of items Q1 to the first pallet is complete for the second pallet. In other words, once the supply pallet P0 has finished supplying goods Q1 to the first pallet, it becomes a source of goods Q1 for the second pallet, and can therefore be called a supply pallet for the second pallet. Since it is used as the second pallet without transshipment, it can also be said to be repurposed as the second pallet itself. The process then proceeds to step S88. If not all pallets P have been selected as either the first or second pallet (step S88; No), the process returns to step S80. If all pallets P have been selected as either the first or second pallet (step S88; Yes), the process proceeds to step S90. The process also proceeds to step S88 if there is no supply pallet P0 that can hold the same number of goods Q1 as the sum of the number of goods Q1 on the first and second pallets (step S84; No).

[0150] In step S90, the initial calculation unit 32 extracts multiple remaining pallets to which item Q1 will be assigned from the remaining pallets to which no pallet has been selected for supplying item Q1. The initial calculation unit 32 then determines (step S90) whether there is a supply pallet P0 that will have the same number of items Q1 as the sum of the number of items Q1 assigned to the multiple remaining pallets to which item Q1 will be assigned. If there is (step S90; Yes), the unit selects that supply pallet P0 as the pallet to which item Q1 will be supplied to those remaining pallets (step S92). After step S92 is executed, the unit returns to step S90 and repeats the process until there are no more supply pallets P0 that will have the same number of items Q1 as the sum of the number of items Q1 assigned to the multiple remaining pallets. On the other hand, if there is no supply pallet P0 that has the same number of items Q1 as the total number of items Q1 on multiple remaining pallets (step S90; No), the initial calculation unit 32 selects a supply pallet P0 that has a greater number of items Q1 than the number of items Q1 that would be assigned to the remaining pallets that were not assigned a pallet to be used as the pallet to supply items Q1 to those remaining pallets (step S94). Subsequently, if the selection of a pallet to supply items Q1 to all remaining pallets is not completed (step S96; No), the process returns to step S94. On the other hand, if the selection of a pallet to supply items Q1 to all remaining pallets is completed, the process proceeds to step S98. In step S98, if not all items Q have been extracted as items Q1 (step S98; No), the process returns to step S80, new items Q are extracted as items Q1, and the process is repeated. If all items Q have been extracted as items Q1 (step S98; Yes), the process ends.

[0151] Thus, in the fourth embodiment, by performing the stacking operation continuously on the first and second pallets, which contain the same items Q in the uppermost and lowermost layers, the number of times the supply pallet P0, once the transshipment is complete, is moved from the supply pallet placement area AR2 to another location to continue the stacking operation can be reduced, thereby suitably reducing the working time and workload.

[0152] (Effects) As described above, the information processing method according to the first aspect of this disclosure includes the steps of: acquiring work information indicating goods Q to be shipped to a destination; calculating an initial solution based on the work information, which indicates goods Q to be assigned to each of a plurality of pallets P, stacking information indicating the stacking order of goods Q on the pallets P, and stacking order indicating the order in which the pallets P are stacked; calculating an initial evaluation value based on the initial solution, which indicates the degree of workload of the shipping work in the initial solution; calculating a neighboring solution based on the initial solution, which is obtained by changing a part of the stacking information and stacking order from the initial solution; and calculating an information processing method based on the neighboring solution, which is obtained in the neighboring solution The method includes the steps of calculating a neighboring evaluation value indicating the workload of the shipping operation, and setting the stacking information and stacking order to be used in the stacking plan based on the initial evaluation value and the neighboring evaluation value. In the step of calculating the initial evaluation value and the neighboring evaluation value, the initial evaluation value and the neighboring evaluation value are calculated by considering at least one of the following: multiple gripping, in which multiple items Q are gripped and stacked in one stacking operation; repurposing, in which a supply pallet P0 on which inventory items Q are stacked is repurposed as a pallet P for shipping; and reverse stacking, in which an unnecessary number of items Q are removed from the supply pallet P0 and placed on another pallet P, and the supply pallet P0 is repurposed as a pallet P for shipping. According to this disclosure, by creating a neighboring solution that is partially modified from the initial solution and evaluating it, an optimized stacking plan can be set according to the work content, and an appropriate stacking plan can be created according to the work content.

[0153] The information processing method according to the second aspect of this disclosure is the information processing method according to the first aspect, wherein in the step of calculating the initial evaluation value and the neighboring evaluation value, if the number of items Q loaded on the supply pallet P0 is a, the number of items Q required for the shipping pallet P is s, the time required for one stacking is Tp, and the time required to transport another pallet P (loaded with the removed items Q) is Te, then if s × Tp > (a - s) × Tp + Te is satisfied, the initial evaluation value and the neighboring evaluation value are calculated assuming that reverse stacking is performed. According to this disclosure, by using such a calculation formula, a more optimized stacking plan can be set according to the work content.

[0154] The information processing method according to the third aspect of this disclosure is the information processing method according to the first aspect, wherein in the step of calculating the initial evaluation value and the neighboring evaluation value, if a is the number of items Q loaded on the supply pallet P0, s is the number of items Q required for the shipping pallet P, m1 is the number of items Q that can be simultaneously grasped n at a time when reverse stacking is not performed, m2 is the number of items Q that can be simultaneously grasped n at a time when reverse stacking is performed, Tp is the time required for one stacking, and Te is the time required to transport another pallet P (loaded with the removed items Q), then if (s - m1 + m / n) × Tp > ((a - s) - m2 + m2 / n) × Tp + Te is satisfied, the initial evaluation value and the neighboring evaluation value are calculated assuming that multiple grasping and reverse stacking are performed. According to this disclosure, by using such calculation formulas, a more optimized stacking plan can be set according to the work content.

[0155] The information processing method according to the fourth aspect of this disclosure is an information processing method according to any of the first to third aspects, wherein in the step of calculating the initial evaluation value and the nearest evaluation value, the initial evaluation value and the nearest evaluation value are calculated based on at least one of the following: the workload of the stacking device 14 that stacks goods Q on pallet P, the workload of the mobile body 12 that transports pallet P, and the number of times goods Q, which are assigned to pallet P in quantities less than or equal to a predetermined number, are stacked consecutively (number of consecutive small lots). According to this disclosure, by using such evaluation values, a more optimized stacking plan can be set according to the work content.

[0156] The information processing method relating to the fifth aspect of this disclosure is an information processing method relating to any of the first to fourth aspects, wherein in the step of setting the stacking information and stacking order, if the evaluation of the neighboring evaluation value is higher than the evaluation of the initial evaluation value, the stacking information and stacking order in the neighboring solution are set as the solution to be used for the stacking plan. According to this disclosure, by adopting the neighboring solution when the evaluation of the neighboring solution is high, an appropriate stacking plan can be created according to the degree of workload of the shipping work.

[0157] The information processing method according to the sixth aspect of this disclosure is an information processing method according to any of the first to fifth aspects, further including a step of deciding whether or not to calculate a neighboring solution. If it is decided not to calculate a neighboring solution, the neighboring solution and neighboring evaluation value are not calculated, and the stacking information and stacking order in the initial solution are set as the solution to be used in the stacking plan. According to this disclosure, it is possible to decide whether or not to use a neighboring solution as needed, so that the stacking plan can be set appropriately according to the work content and computation load.

[0158] The stacking method according to the seventh aspect of this disclosure stacks goods Q onto multiple pallets P based on the stacking information and stacking order set in the information processing method according to any of the first to sixth aspects. According to this disclosure, goods Q can be stacked appropriately.

[0159] The program according to the eighth aspect of this disclosure includes the steps of: acquiring work information indicating goods Q to be shipped to a destination; calculating an initial solution based on the work information, which includes goods Q to be assigned to each of a plurality of pallets P, stacking information indicating the stacking order of goods Q on the pallets P, and stacking order indicating the order in which the pallets P are stacked; calculating an initial evaluation value based on the initial solution, which indicates the degree of workload of the shipping work in the initial solution; calculating a neighboring solution based on the initial solution, which is obtained by changing a part of the stacking information and stacking order from the initial solution; and calculating the degree of workload of the shipping work in the neighboring solution based on the neighboring solution. The computer is made to perform the following steps: calculate the indicated neighboring evaluation value; set the stacking information and stacking order to be used in the stacking plan based on the initial evaluation value and the neighboring evaluation value; and in the step of calculating the initial evaluation value and the neighboring evaluation value, the computer is made to calculate the initial evaluation value and the neighboring evaluation value by considering at least one of the following: multiple gripping, in which multiple items Q are gripped and stacked in one stacking operation; repurposing, in which a supply pallet P0 on which inventory items Q are stacked is repurposed as a shipping pallet P; and reverse stacking, in which an unnecessary number of items Q are removed from the supply pallet P0 and placed on another pallet P, and the supply pallet P0 is repurposed as a shipping pallet P. According to this disclosure, an appropriate stacking plan can be created according to the work content.

[0160] The information processing apparatus according to the ninth aspect of this disclosure includes: a work information acquisition unit 30 that acquires work information indicating goods Q to be shipped to a destination; an initial solution calculation unit 32 that calculates an initial solution based on the work information, indicating goods Q to be assigned to each of a plurality of pallets P, stacking information indicating the stacking order of goods Q on the pallets P, and stacking order indicating the order in which the pallets P are stacked; an initial evaluation value calculation unit 36 ​​that calculates an initial evaluation value indicating the degree of workload of the shipping work in the initial solution based on the initial solution; a nearest neighbor solution calculation unit 34 that calculates a nearest neighbor solution based on the initial solution, with some of the stacking information and stacking order changed from the initial solution; and a nearest neighbor solution calculation unit that calculates a nearest neighbor solution based on the nearest neighbor solution. The system includes a proximity evaluation value calculation unit 38 that calculates a proximity evaluation value indicating the degree of workload for shipping operations, and a setting unit 40 that sets stacking information and stacking order to be used in the stacking plan based on the initial evaluation value and the proximity evaluation value. The initial evaluation value calculation unit 36 ​​and the proximity evaluation value calculation unit 38 calculate the initial evaluation value and the proximity evaluation value by considering at least one of the following: multiple gripping, where multiple items Q are gripped and stacked in one stacking operation; repurposing, where a supply pallet P0 loaded with inventory items Q is repurposed as a shipping pallet P; and reverse stacking, where an unnecessary number of items Q are removed from the supply pallet P0 and placed on another pallet P, and the supply pallet P0 is repurposed as a shipping pallet P. According to this disclosure, an appropriate stacking plan can be created according to the work content.

[0161] Although embodiments of the present disclosure have been described above, the embodiments are not limited to those described herein. Furthermore, the aforementioned components include those that can be easily conceived by those skilled in the art, those that are substantially the same, and those that fall within the so-called equivalent range. Moreover, the aforementioned components can be combined as appropriate. Furthermore, various omissions, substitutions, or modifications of the components can be made without departing from the gist of the embodiments described above.

[0162] 10 Information processing device 30 Work information acquisition unit 32 Initial solution calculation unit 34 Neighboring solution calculation unit 36 ​​Initial evaluation value calculation unit 38 Neighboring evaluation value calculation unit 40 Setting unit 100 Stacking system P Pallet Q Goods

Claims

1. The process includes the steps of: acquiring work information indicating the goods to be shipped to the destination; calculating an initial solution based on the work information, which includes stacking information indicating the goods to be assigned to each of a plurality of pallets, the stacking order of the goods on the pallets, and the stacking order indicating the order in which the pallets are stacked; calculating an initial evaluation value indicating the degree of workload of the shipping work in the initial solution based on the initial solution; calculating a neighboring solution based on the initial solution, which is obtained by changing a part of the stacking information and the stacking order from the initial solution; calculating a neighboring evaluation value indicating the degree of workload of the shipping work in the neighboring solution based on the neighboring solution; and setting the stacking information and the stacking order to be used in the stacking plan based on the initial evaluation value and the neighboring evaluation value. An information processing method that, in the step of calculating the initial evaluation value and the neighboring evaluation value, considers at least one of the following: multiple gripping, in which multiple items are gripped and stacked in a single stacking operation; repurposing, in which a supply pallet loaded with inventory items is repurposed as a shipping pallet for shipment; and reverse stacking, in which an unnecessary number of items are removed from the supply pallet and the supply pallet is repurposed as the shipping pallet.

2. The information processing method according to claim 1, in the step of calculating the initial evaluation value and the neighboring evaluation value, if a is the number of items loaded on the supply pallet, s is the number of items required for the shipping pallet, Tp is the time required for one loading, and Te is the time required to transport the other pallet, then if s × Tp > (a - s) × Tp + Te is satisfied, the initial evaluation value and the neighboring evaluation value are calculated assuming that the reverse loading is performed.

3. The information processing method according to claim 1, wherein in the step of calculating the initial evaluation value and the neighboring evaluation value, if a is the number of items loaded on the supply pallet, s is the number of items required for the shipping pallet, m1 is the number of items that can be simultaneously gripped n at a time when reverse stacking is not performed, m2 is the number of items that can be simultaneously gripped n at a time when reverse stacking is performed, Tp is the time required for one stacking, and Te is the time required to transport the other pallet, then if (s - m1 + m1 / n) × Tp > ((a - s) - m2 + m2 / n) × Tp + Te is satisfied, the initial evaluation value and the neighboring evaluation value are calculated assuming that multiple gripping and reverse stacking are performed.

4. The information processing method according to claim 1, wherein in the step of calculating the initial evaluation value and the neighboring evaluation value, the initial evaluation value and the neighboring evaluation value are calculated based on at least one of the following: the workload of a stacking device that stacks the articles on the pallet, the workload of a mobile body that transports the pallet, and the number of times the articles, which are assigned to the pallet in quantities less than or equal to a predetermined number, are stacked consecutively.

5. The information processing method according to claim 1, wherein in the step of setting the stacking information and the stacking order, if the evaluation of the neighboring evaluation value is higher than the evaluation of the initial evaluation value, the stacking information and the stacking order in the neighboring solution are set as the solution to be used for the stacking plan.

6. The information processing method according to claim 1, further comprising the step of determining whether or not to calculate the nearest neighbor solution, wherein if it is determined not to calculate the nearest neighbor solution, the nearest neighbor solution and the nearest neighbor evaluation value are not calculated, and the stacking information and stacking order in the initial solution are set as the solution to be used in the stacking plan.

7. A stacking method for stacking articles on a plurality of pallets based on the stacking information and stacking order set in the information processing method according to any one of claims 1 to 6.

8. The computer is made to perform the following steps: acquire work information indicating the goods to be shipped to the destination; calculate an initial solution based on the work information, which indicates the goods to be assigned to each of a plurality of pallets, stacking information indicating the stacking order of the goods on the pallets, and stacking order indicating the order in which the pallets are stacked; calculate an initial evaluation value indicating the degree of workload of the shipping work in the initial solution based on the initial solution; calculate a neighboring solution based on the initial solution, which is obtained by changing a part of the stacking information and the stacking order from the initial solution; calculate a neighboring evaluation value indicating the degree of workload of the shipping work in the neighboring solution based on the neighboring solution; and set the stacking information and the stacking order to be used in the stacking plan based on the initial evaluation value and the neighboring evaluation value. A program that, in the step of calculating the initial evaluation value and the neighboring evaluation value, considers at least one of the following: multiple gripping, in which multiple items are gripped and stacked in a single stacking operation; repurposing, in which a supply pallet loaded with inventory items is repurposed as a shipping pallet; and reverse stacking, in which an unnecessary number of items are removed from the supply pallet and the supply pallet is repurposed as the shipping pallet.

9. Includes: a work information acquisition unit that acquires work information indicating goods to be shipped to a destination; an initial solution calculation unit that calculates an initial solution based on the work information, indicating the goods to be assigned to each of a plurality of pallets, stacking information indicating the stacking order of the goods on the pallets, and stacking order indicating the order in which the pallets are stacked; an initial evaluation value calculation unit that calculates an initial evaluation value indicating the degree of workload of the shipping work in the initial solution based on the initial solution; a neighboring solution calculation unit that calculates a neighboring solution based on the initial solution, with some of the stacking information and stacking order changed from the initial solution; a neighboring evaluation value calculation unit that calculates a neighboring evaluation value indicating the degree of workload of the shipping work in the neighboring solution based on the neighboring solution; and a setting unit that sets the stacking information and stacking order to be used in the stacking plan based on the initial evaluation value and the neighboring evaluation value. An information processing device which calculates the initial evaluation value and the neighboring evaluation value by considering at least one of the following: multiple gripping, in which multiple items are gripped and stacked in a single stacking operation; repurposing, in which a supply pallet loaded with inventory items is repurposed as a shipping pallet for shipment; and reverse stacking, in which an unnecessary number of items are removed from the supply pallet and the supply pallet is repurposed as the shipping pallet.

Images