Loading pattern creation device and method
The described technology addresses the challenge of optimizing cargo loading with unknown sizes by estimating and integrating these sizes into the loading pattern creation process, ensuring efficient cargo placement.
Patent Information
- Application Number
- PCT/JP2025/016767
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2025-05-07
- Publication Date
- 2026-11-12
AI Technical Summary
Conventional loading pattern creation techniques fail to optimize cargo loading when cargoes of unknown sizes are included, as they rely solely on deterministic information.
A loading pattern creation device and method that acquires cargo information for both known and unknown sizes, estimates the unknown sizes, and creates optimized loading patterns using a combination of estimation and optimization algorithms like DBLF and genetic algorithms.
Enables the creation of loading patterns that optimize objectives such as loading efficiency even when cargo sizes are unknown, by accurately estimating and incorporating unknown sizes into the loading process.
Smart Images

Figure JP2025016767_12112026_PF_FP_ABST
Abstract
Description
Loading Pattern Creation Apparatus and Method
[0001] The disclosed technique relates to a loading pattern creation apparatus and a loading pattern creation method.
[0002] Conventionally, there is a technique for creating a loading pattern that indicates in what order and at which position each piece of cargo is to be loaded, when a plurality of cargoes are loaded into a bin, which is a container for collectively transporting and storing cargoes, such as a truck bed, a container, a pallet, and a trolley. For example, a technique has been proposed that calculates an optimal loading pattern for cargoes for the purpose of optimizing the loading rate based on information such as the size of the bin and the size of the cargoes (Non-Patent Document 1).
[0003] Karabulut, K., İnceoğlu, "A Hybrid Genetic Algorithm for Packing in 3D with Deepest Bottom Left with Fill Method", In: Yakhno, T. (eds) Advances in Information Systems. ADVIS 2004. Lecture Notes in Computer Science, vol 3261. Springer, Berlin, Heidelberg.
[0004] In the conventional technique, the loading pattern is optimized using only deterministic information related to the cargoes to be loaded. Therefore, when a cargo with an unknown size is included, a loading pattern cannot be created.
[0005] The disclosed technique has been made in view of the above points, and an object thereof is to create a loading pattern that optimizes an objective function such as a loading rate even when a cargo with an unknown size is included.
[0006] A first aspect of the present disclosure is a loading pattern creation device, comprising: an acquisition unit that acquires cargo information relating to each of a plurality of cargoes, which include cargoes of known size and cargoes of unknown size, and loading container information relating to a loading container for loading one or more cargoes; an estimation unit that estimates the size of the cargoes of unknown size based on the cargo information; and a creation unit that creates and outputs a pattern for loading the plurality of cargoes into the loading container, using the sizes included in the cargo information of the plurality of cargoes for the cargoes of known size, and the sizes estimated by the estimation unit for the cargoes of unknown size.
[0007] A second aspect of this disclosure is a method for creating a loading pattern, wherein a computer performs the following steps: acquires cargo information for each of a plurality of packages, which include a mixture of packages of known size and packages of unknown size, and load container information for a load container for loading one or more packages; estimates the size of the packages of unknown size based on the cargo information; and uses the sizes included in the cargo information of the plurality of packages for the packages of known size, and the estimated sizes for the packages of unknown size, to create and output a pattern for loading the plurality of packages into the load container.
[0008] According to the disclosed technology, it is possible to create loading patterns that optimize objective functions such as loading efficiency, even when the size of the cargo is unknown.
[0009] This is a block diagram showing the hardware configuration of the loading pattern creation device. This is a block diagram showing an example of the functional configuration of the loading pattern creation device. This is a diagram showing an example of a cargo information database. This is a diagram showing an example of a loading pattern database. This is a flowchart showing the flow of the loading pattern creation process. This is a flowchart showing the flow of the creation process.
[0010] An example of an embodiment of the disclosed technology will be described below with reference to the drawings. In each drawing, identical or equivalent components and parts are given the same reference numerals. Furthermore, the dimensional ratios in the drawings are exaggerated for illustrative purposes and may differ from actual ratios.
[0011] This embodiment describes an example of creating a loading pattern when multiple packages, each containing goods in cardboard boxes, are loaded onto a pallet, which is an example of a bin. However, the application of the disclosed technology is not limited to this example. For example, other packaging forms such as wooden boxes, drums, pallets, etc., may be used as packages. Also, other bins such as truck beds, containers, dollies, etc., may be used as bins.
[0012] Figure 1 is a block diagram showing the hardware configuration of the loading pattern creation device 10. As shown in Figure 1, the loading pattern creation device 10 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, storage 14, an input unit 15, a display unit 16, and a communication I / F (Interface) 17. Each component is connected to the others via a bus 19 so as to be able to communicate with each other.
[0013] The CPU 11 is a central processing unit that executes various programs and controls various parts. Specifically, the CPU 11 reads a program from the ROM 12 or storage 14 and executes the program using the RAM 13 as a working area. The CPU 11 controls each of the above components and performs various calculations according to the program stored in the ROM 12 or storage 14. In this embodiment, the ROM 12 or storage 14 stores a loading pattern creation program, which will be described later.
[0014] ROM 12 stores various programs and data. RAM 13 temporarily stores programs or data as a working area. Storage 14 consists of storage devices such as HDD (Hard Disk Drive) and SSD (Solid State Drive) and stores various programs and data, including the operating system.
[0015] The input unit 15 includes a pointing device such as a mouse and a keyboard, and is used for various types of input. The display unit 16 is, for example, a liquid crystal display and displays various types of information. The display unit 16 may also function as the input unit 15 by employing a touch panel system.
[0016] Communication I / F17 is an interface for communicating with other devices. For this communication, wired communication standards such as Ethernet® and FDDI, or wireless communication standards such as 4G, 5G, and Wi-Fi® may be used.
[0017] Next, the functional configuration of the loading pattern creation device 10 will be described. Figure 2 is a block diagram showing an example of the functional configuration of the loading pattern creation device 10. As shown in Figure 2, the loading pattern creation device 10 includes, as a functional configuration, an acquisition unit 21, an estimation unit 22, a determination unit 23, and a creation unit 24. Each functional configuration is realized by the CPU 11 reading a loading pattern creation program stored in the ROM 12 or storage 14, expanding it in the RAM 13, and executing it.
[0018] The acquisition unit 21 acquires information about each of several packages, which include packages of known size and packages of unknown size, and information about bins (in this case, pallets) for loading one or more packages. Note that a bin is an example of a "loading container" in this disclosure, and bin information is an example of "loading container information" in this disclosure.
[0019] Specifically, the acquisition unit 21 acquires package information from the package information DB (database) 31. In this embodiment, as described above, a package is a cardboard box containing one or more items. The package information DB 31 stores information about the items contained in the package, the size of the package (external dimensions), and other information for each package.
[0020] Figure 3 shows an example of the package information DB31. In the example in Figure 3, each row (each record) corresponds to package information for one package. In the example in Figure 3, the package information includes information such as "product code," "product name," "specifications," "quantity," "size (length)," "size (width)," and "size (height)." The "product code" is the identification information of the product to be stored in the package. The "product name" is the name of the product. The "specifications" are the size of the product pieces, the quantity of products to be stored in one box, etc. The "quantity" is the number of boxes for the package. "Size (length)," "size (width)," and "size (height)" are the external dimensions of the package. In Figure 3, if each size column is "-", it indicates that the size of the package is unknown.
[0021] In addition to the above, the cargo information may also include the product category, the weight of the cargo, and information regarding permission to rotate the cargo. Information regarding permission to rotate the cargo indicates whether it is permitted to rotate the cargo along a horizontal or vertical axis when loading it onto a pallet. For example, cargo marked "This Side Up" will have permission information indicating that rotation along the horizontal axis, i.e., turning it upside down, is prohibited.
[0022] Furthermore, the acquisition unit 21 acquires bin information specified by the user. The bin information includes the type of bin, information for identifying the bin's load capacity, etc. The type of bin can be a truck bed, container, pallet, dolly, etc. If the bin is a container, the information for identifying the bin's load capacity is the internal dimensions of the bin (length, width, height). If the bin is a container without side walls, such as a pallet, the length and width of the bin, and a predetermined height that allows for loading cargo, may be used as information for identifying the bin's load capacity.
[0023] The estimation unit 22 estimates the size of packages whose size is unknown based on the package information. Specifically, the estimation unit 22 estimates the size of packages whose size is unknown based on the similarity between the package information of packages whose size is unknown and the package information of packages whose size is unknown.
[0024] For example, the estimation unit 22 vectorizes the package information other than size, such as the product name and specifications, for each package included in the package information DB 31 using a Bag of Words or the like. The estimation unit 22 calculates, for example, a cosine similarity as the similarity between each of the first vectors for packages with known sizes and each of the second vectors for packages with unknown sizes. The estimation unit 22 estimates the size of the package to be estimated as the size of the package corresponding to the first vector that has the highest similarity with the second vector for the package to be estimated.
[0025] Alternatively, for example, the estimation unit 22 may input the package information of a package with an unknown size into a machine learning model that has been trained to output the package size when package information other than size is input, and estimate the size of the package with an unknown size.
[0026] Furthermore, for example, the estimation unit 22 may estimate the size of the package from part of the package information. For example, suppose the package information specifications specify that the piece size is 300 mm x 100 mm x 100 mm and the number of pieces is 6 (stored in 2 x 3). In this case, the estimation unit 22 estimates the size of the package to be 300 mm x 200 mm x 300 mm.
[0027] The estimation unit 22 may estimate multiple size candidates for a single package whose size is unknown. For example, in the case of estimation using the similarity described above, the estimation unit 22 may estimate the sizes corresponding to each of the first vectors whose similarity to the second vector is greater than or equal to a predetermined value, or the top predetermined number of first vectors whose similarity is greater than or equal to a predetermined value, as size candidates for the package whose size is unknown.
[0028] The decision unit 23 determines whether to create a loading pattern using a size that encompasses multiple size candidates (hereinafter referred to as the "inclusive size") or to create a loading pattern for each size candidate when there are multiple size candidates for a package whose size is unknown, as estimated by the estimation unit 22. The decision unit 23 may decide on the above policy for each package with multiple size candidates, or it may decide on a single policy for the entire process.
[0029] For example, if there are two size candidates as shown below, the determination unit 23 will calculate 200 mm x 160 mm x 180 mm as the inclusive size. Size candidate 1: 200 mm x 150 mm x 180 mm Size candidate 2: 180 mm x 160 mm x 180 mm When calculating the inclusive size, the package may be rotated based on the rotation permission information to find the smallest possible inclusive size.
[0030] The decision unit 23 may determine the policy for creating the loading pattern based on user specifications, or based on the calculated inclusion size. In this case, for example, if there is a large discrepancy between the inclusion size and the candidate size, the decision unit 23 may decide to create a loading pattern for each size candidate. The decision unit 23 may determine the magnitude of the discrepancy between the inclusion size and the candidate size based, for example, whether the sum of the differences between the inclusion size and each of the size candidates is greater than or equal to a predetermined value.
[0031] The creation unit 24 creates a loading pattern using the sizes included in the cargo information of multiple cargo stored in the cargo information DB 31 for cargo with known sizes, and the sizes estimated by the estimation unit 22 for cargo with unknown sizes. Specifically, if there are no cargo with unknown sizes included in the cargo information DB 31, the creation unit 24 creates a loading pattern using the sizes included in the cargo information. Furthermore, if there are cargo with unknown sizes included in the cargo information DB 31, and there is only one size candidate for all of them, or if the decision unit 23 has decided to use the included size, the creation unit 24 creates a loading pattern for cargo with unknown sizes using the sizes estimated by the estimation unit 22 or the included size calculated by the decision unit 23.
[0032] Furthermore, if the decision unit 23 decides to create a loading pattern for each size candidate, the creation unit 24 will create loading patterns for all combinations of size candidates. For example, suppose the sizes of packages with product codes 20 and 30 are unknown, and there are three estimated size candidates for package 20 and two estimated size candidates for package 30. In this case, the creation unit 24 will create 3 x 2 = 6 loading patterns.
[0033] The creation unit 24 calculates the loading rate for each created loading pattern, which is the ratio of the volume of the goods to be loaded to the loading capacity of the pallet. The creation unit 24 stores the created loading patterns in the loading pattern DB 32 along with the calculated loading rate. Figure 4 shows an example of the loading pattern DB 32. In the example in Figure 4, each row (each record) represents one loading pattern. The loading pattern DB 32 shown in Figure 4 stores information such as "ID", "Size", "Bin number", "Loading position", "Loading order", "Product code", and "Loading rate".
[0034] "ID" is the identification number of the loading pattern. "Size" is a list of the sizes of each item to be loaded. For items with unknown sizes, the size is an estimated size candidate or an inclusive size. "Bin number" is the identification number of the bin (in this case, pallet) on which each item is loaded. "Loading position" is a list of information that identifies the position of each item on the pallet. For example, the loading position may be the coordinate value of the item's predetermined position (e.g., the center of the bottom) within the loading area of the pallet. Alternatively, the loading position may be the coordinate value of the smallest coordinate point of the item in three-dimensional coordinate space (the point where the x, y, and z coordinates are all smallest). The loading position may also be specified by which layer the item is on, which position it is in the length direction of the pallet, and which position it is in the width direction of the pallet. "Loading order" is a list of the loading order of each item. "Product code" is a list of the product code of each item.
[0035] The method for creating a loading pattern by the creation unit 24 will be explained in detail. The creation unit 24 may create a loading pattern using, for example, DBLF (Deepest Bottom Left with Fill) (see Non-Patent Literature 1). DBLF is a method that sets an order for the cargo to be loaded and loads the cargo in that order, starting as deep as possible, then as low as possible (Bottom), and then as far to the left as possible (Left). If the cargo can no longer be loaded onto one pallet, the creation unit 24 creates a loading pattern that loads the cargo onto the next pallet. The creation unit 24 then creates an optimal loading pattern by optimizing the loading order of the cargo, for example, to optimize the loading rate.
[0036] The creation unit 24 may use, for example, a genetic algorithm as an optimization technique for the loading order. In this case, the creation unit 24 may treat the candidate loading order solutions as individuals, generate a loading order for a portion of the initial population of individuals using a predetermined method, and randomly generate the loading orders for other individuals. The predetermined method may be, for example, using the loading order of a loading pattern stored in the loading pattern DB 32 whose package size is similar to the package for which the loading pattern is to be created. The creation unit 24 may determine whether the package sizes are similar using the ratio of the sizes of each side of the package, the ratio of the volumes of the packages, etc. The creation unit 24 may also use a method of sorting the loading order in descending order by package height, package width, package volume, or package base area as a predetermined method.
[0037] The creation unit 24 identifies the optimal loading order by minimizing the fitness, which is expressed by the following equation, as the objective function: fitness = number of bins used + loading rate of the bin with the lowest loading rate.
[0038] Furthermore, the creation unit 24 may rearrange the priority of DBLF in creating the loading pattern, for example, Bottom → Deep → Left. The creation unit 24 may also use other methods as loading pattern creation techniques, such as DFTRC (Distance to the Front-Top-Right Corner) (see Reference 1). (Reference 1) José Fernando Gonçalves, Mauricio GC Resende, "A biased random key genetic algorithm for 2D and 3D bin packing problems", International Journal of Production Economics, October 2013.
[0039] The creation unit 24 may also use other optimization techniques, such as a greedy algorithm, to optimize the loading order. Furthermore, the creation unit 24 may optimize both the loading order and orientation of the cargo, taking into account the permission information for cargo rotation.
[0040] Furthermore, the creation unit 24 may create a loading pattern using a method that does not involve optimizing the loading order. For example, the creation unit 24 may identify the loading order in descending order of weight, volume, or base area of the cargo, and then apply DBLF to this loading order to create a loading pattern. In this case as well, the creation unit 24 creates a loading pattern such that if it is no longer possible to load all the cargo onto one pallet, it loads the cargo onto the next pallet, similar to when using the optimization technique described above.
[0041] Furthermore, the creation unit 24 may speed up the processing by using a part of the loading patterns with similar package sizes among the loading patterns stored in the loading pattern DB 32, in a method that does not use loading sequence optimization. For example, the creation unit 24 rearranges each data of the size, bin number, loading position, loading sequence, and product code of the loading pattern acquired from the loading pattern DB 32 in the order of loading sequence. This rearranged information is hereinafter referred to as the "first array". Similarly, the creation unit 24 rearranges the sizes and product codes of the packages that are the targets of loading pattern creation in the order of loading sequence. This rearranged information is hereinafter referred to as the "second array". When the first array and the second array match from the start up to the x-th element, the creation unit 24 directly uses the data of the size, bin number, loading position, loading sequence, and product code from the start up to the x-th element of the first array, and creates a loading pattern starting from the (x+1)-th package.
[0042] However, when using a loading pattern in which the loading position is determined depending only on the package size, the loading pattern does not change as long as the sizes are the same even if the product codes are different. Therefore, the creation unit 24 may determine the match between the first array and the second array based only on the size. In that case, the creation unit 24 creates the loading pattern using the product code of the second array.
[0043] Furthermore, even in a method that does not use loading sequence optimization, the creation unit 24 may change the priority of DBLF to, for example, Bottom→Deep→Left, or may use other methods such as DFTRC.
[0044] The creation unit 24 outputs the loading patterns stored in the loading pattern DB 32. When a plurality of loading patterns are created, the creation unit 24 may output all of them, or may output one loading pattern with the maximum loading rate. In addition, the creation unit 24 may output loading patterns whose loading rate is equal to or higher than a predetermined value, or a predetermined number of loading patterns with the top-ranked loading rates.
[0045] Next, the operation of the loading pattern creation device 10 will be described. FIG. 5 is a flowchart showing the flow of loading pattern creation processing by the loading pattern creation device 10. The CPU 11 reads out a loading pattern creation program from the ROM 12 or the storage 14, develops it in the RAM 13, and executes it, whereby the loading pattern creation processing is performed. Note that the loading pattern creation processing is an example of the "loading pattern creation method" of the present disclosure.
[0046] In step S11, the CPU 11, acting as the acquisition unit 21, acquires package information from the package information DB 31. Further, the CPU 11, acting as the acquisition unit 21, acquires bin information specified by the user.
[0047] Next, in step S12, the CPU 11, acting as the estimation unit 22, estimates the size of a package with an unknown size based on the similarity between package information other than the size of a package with a known size and the package information of the package with an unknown size.
[0048] Next, in step S13, the CPU 11, acting as the determination unit 23, determines whether there exists a package for which there are a plurality of size candidates estimated in step S12 described above. If no such package exists, the process proceeds to step S15, and if such a package exists, the process proceeds to step S16.
[0049] In step S15, the CPU 11, acting as the determination unit 23, determines to create a loading pattern using the size estimated in step S12 described above as the size of the package with an unknown size, and the process proceeds to step S20.
[0050] In step S16, the CPU 11, acting as the determination unit 23, calculates an inclusive size that includes the plurality of size candidates for each package that has a plurality of size candidates. Next, in step S17, the CPU 11, acting as the determination unit 23, determines a policy of creating a loading pattern with the inclusive size or creating a loading pattern for each size candidate, for example, based on the magnitude of the deviation between the inclusive size and the candidate sizes.
[0051] Next, in step S20, creation processing is executed. Here, the creation processing will be described with reference to FIG. 6.
[0052] In step S21, the CPU 11, as the creation unit 24, calculates the number of loading patterns to be created N. If the process moves from step S15 to this step, or if in step S17 it is decided to create loading patterns for all packages with multiple size candidates using the included size, the CPU 11, as the creation unit 24, calculates N = 1. If in step S17 it is decided to create loading patterns for each size candidate for some packages, the CPU 11, as the creation unit 24, calculates the total number of size candidate combinations for each package as N.
[0053] Next, in step S22, the CPU 11, acting as the creation unit 24, sets the variable i, which will be the identification number (ID) of the loading pattern, to 0. Then, in step S23, the CPU 11, acting as the creation unit 24, determines whether i is 0 or not. If i = 0, the process proceeds to step S25; otherwise, the process proceeds to step S24.
[0054] In step S24, the CPU 11, acting as a creation unit 24, retrieves a loading pattern from the loading pattern DB 32 whose size is similar to the size of the cargo for which the loading pattern is to be created.
[0055] Next, in step S25, the CPU 11, as the creation unit 24, creates a loading pattern, for example, using DBLF, and uses a genetic algorithm to optimize the loading order of the cargo, with the candidate solutions for the loading order as individuals, thereby creating the optimal loading pattern. The CPU 11, as the creation unit 24, may apply the loading order of the loading pattern obtained in step S24 as part of the initial population of individuals in the genetic algorithm.
[0056] Next, in step S26, the CPU 11, acting as the creation unit 24, calculates the loading rate for each of the created loading patterns. The CPU 11 then assigns the variable i as the ID of the created loading pattern and stores the loading pattern along with the loading rate in the loading pattern DB 32.
[0057] Next, in step S27, the CPU 11, acting as the creation unit 24, increments the variable i by 1. Then, in step S28, the CPU 11, acting as the creation unit 24, determines whether the variable i is less than the number of stacking patterns N calculated in step S21. If i < N, the process returns to step S23; if i ≥ N, the creation process ends and the process returns to the stacking pattern creation process (Figure 5).
[0058] Next, in step S30, the CPU 11, acting as the creation unit 24, outputs the loading pattern stored in the loading pattern DB 32, and the loading pattern creation process ends.
[0059] As described above, the loading pattern creation device according to this embodiment acquires cargo information for each of a plurality of packages, which include packages of known size and packages of unknown size, and loading container information for a loading container for loading one or more packages. The loading pattern creation device also estimates the size of packages of unknown size based on the acquired cargo information. The loading pattern creation device then uses the cargo information of the plurality of packages, and the estimated size for packages of unknown size, to create and output a loading pattern for loading multiple packages into a loading container. This makes it possible to create a loading pattern that optimizes objective functions such as loading rate, even when packages of unknown size are included.
[0060] In addition, the loading pattern creation process, which is executed by the CPU after reading the software (program) in the above embodiment, may be executed by various processors other than the CPU. Examples of such processors include dedicated electrical circuits, which are processors with circuit configurations specifically designed to execute particular processes, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices) whose circuit configuration can be changed after manufacturing, and ASICs (Application Specific Integrated Circuits). Furthermore, the loading pattern creation process may be executed by one of these various processors, or by a combination of two or more processors of the same or different types (for example, multiple FPGAs, and a combination of a CPU and an FPGA). More specifically, the hardware structure of these various processors is an electrical circuit that combines circuit elements such as semiconductor elements.
[0061] Furthermore, although the above embodiment describes a configuration in which the loading pattern creation program is pre-stored (installed) in the storage 14, the invention is not limited to this configuration. The program may be provided in a form stored on a non-transitor storage medium such as a CD-ROM (Compact Disk Read Only Memory), DVD-ROM (Digital Versatile Disk Read Only Memory), or USB (Universal Serial Bus) memory. Alternatively, the program may be provided in a form that can be downloaded from an external device via a network.
[0062] The following additional information is disclosed regarding the embodiments described above.
[0063] (Note 1) A loading pattern creation device comprising: a memory; and at least one processor connected to the memory, wherein the processor acquires: luggage information for each of a plurality of packages, which include luggage of known size and luggage of unknown size; and loading container information for a loading container for loading one or more luggage; estimates the size of the luggage of unknown size based on the luggage information; and creates and outputs a pattern for loading the plurality of luggage into the loading container, using the sizes included in the luggage information of the plurality of luggage for the luggage of known size, and the estimated size for the luggage of unknown size.
[0064] (Appendix 2) A non-temporary storage medium storing a program executable by a computer to perform a loading pattern creation process, wherein the loading pattern creation process acquires: cargo information for each of a plurality of packages, which include a mixture of packages of known size and packages of unknown size; and loading container information for a loading container for loading one or more packages; estimates the size of the packages of unknown size based on the cargo information; and creates and outputs a pattern for loading the plurality of packages into the loading container, using the sizes included in the cargo information of the plurality of packages for the packages of known size, and the estimated size for the packages of unknown size.
[0065] 10 Loading pattern creation device 11 CPU 12 ROM 13 RAM 14 Storage 15 Input unit 16 Display unit 17 Communication I / F 19 Bus 21 Acquisition unit 22 Estimation unit 23 Determination unit 24 Creation unit 31 Cargo information DB 32 Loading pattern DB
Claims
1. A loading pattern creation device comprising: an acquisition unit that acquires loading information for each of a plurality of packages, which include packages of known size and packages of unknown size, and loading container information for a loading container for loading one or more packages; an estimation unit that estimates the size of the packages of unknown size based on the loading information; and a creation unit that creates and outputs a pattern for loading the plurality of packages into the loading container, using the sizes included in the loading information of the plurality of packages for the packages of known size, and the sizes estimated by the estimation unit for the packages of unknown size.
2. The loading pattern creation device according to claim 1, wherein the estimation unit estimates the size of a package whose size is unknown based on the similarity between the package information other than the size of a package whose size is known and the package information of a package whose size is unknown.
3. When there are multiple candidate sizes estimated by the estimation unit, the loading pattern creation device according to claim 1 or 2 further includes a determination unit that determines whether to create a pattern with a size encompassing the multiple candidate sizes as the size of the package whose size is unknown, or to create a pattern for each of the multiple candidates, wherein the creation unit creates the pattern according to the determination by the determination unit.
4. A method for creating a loading pattern, in which a computer performs the following processes: acquiring cargo information for each of a group of packages, which include packages of known size and packages of unknown size, and loading container information for a loading container for loading one or more packages; estimating the size of the packages of unknown size based on the cargo information; and creating and outputting a loading pattern for the group of packages in the loading container using the cargo information for the group of packages and, in the case of the packages of unknown size, the estimated size.
