Data mapping device and data mapping method

Abstract

To efficiently perform accurate data mapping between different data systems. [Solution] The data mapping device 100 is configured to include a storage device 101 that holds information on the data architecture in each system, and a processor 105 that identifies N-to-M correspondences between columns of each table between one system 300 and another system 400 based on data architecture information 1013 related to one system 300 that is the target of data mapping and data architecture information 1014 related to the other system 400 that is the target of data mapping, and outputs N-to-M mapping information 1017 that includes the correspondences.

Description

[Technical Field]

[0001] The present invention generally relates to a data mapping device and a data mapping method, and more particularly to a technology that enables efficient execution of accurate data mapping between different data systems. [Background technology]

[0002] In the manufacturing industry, there is a demand for variable-mix, variable-volume production, which allows a wide variety of products to be produced and supplied in just the required quantities. As a result, there is growing interest in flexible production, which can flexibly respond to fluctuations in demand and production adjustments. This flexible production aims to simultaneously achieve both high-efficiency production using dedicated equipment and conveyors that limit the number of products produced in line production, and high-mix production using multi-skilled workers in cell production.

[0003] A key feature of flexible production is the realization of variable-mix, variable-volume production. As a result, both the total production volume and the number of different types of products produced change in line with daily fluctuations in supply and demand. Furthermore, production plans must be formulated that take into account the constraints of the different production conditions, such as cycle time, required resources, and whether or not changeovers are required, for each type of product. This means that the number of combinations of various conditions and information, such as which product type to produce, on which equipment, and in what work order, can easily become enormous.

[0004] Therefore, for example, when a solution for transitioning to or linking with a new production format as described above is applied to a production site primarily based on conventional line production, performing data mapping (e.g., identifying and migrating correspondences between systems for various data used in production planning, etc.) between the system after the solution is applied and the system to which the solution is applied (such as the system at the existing production site) requires a cumbersome and extensive amount of work. Patent Document 1, for example, has proposed a conventional technique for improving the efficiency of data mapping as described above. In Patent Document 1, in light of the conventional problem that only users who know the contents of the data themselves can define correspondences between data, the technique discloses a technique capable of automatically generating mapping rules, which are definitions of correspondences between data, based on the similarity between the names of source data and the names of destination data.

[0005] Furthermore, with existing methods, in order to improve accuracy, it is necessary to train a machine learning model for each task or matching target, and re-learning is required when matching other tasks or matching targets, which increases the learning cost. A technology has also been disclosed that addresses this problem and improves the judgment accuracy of matching tasks while maintaining versatility (see Patent Document 2). [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Laid-Open No. 2002-99561 [Patent Document 2] Patent No. 7454156 Summary of the Invention [Problem to be solved by the invention]

[0007] However, the data structure of the existing system and the data structure of the system after the solution is applied do not all match in terms of data items, inclusion relationships, citation relationships, etc. Therefore, when performing data mapping work, appropriate knowledge and experience regarding the data structures of each system is required. Therefore, if this work is carried out by an unskilled person without such knowledge and experience, it is expected that trial and error and rework will be repeated, resulting in an increase in labor costs.

[0008] Therefore, the present invention has been made in consideration of the above-mentioned problems, and an object of the present invention is to provide a technology that enables efficient execution of accurate data mapping between different data systems. [Means for solving the problem]

[0009] A data mapping device according to one aspect of the present invention is characterized by comprising: a storage device for storing information on data architectures in each system; and a processor for identifying N-to-M correspondences of columns of each table between one system and the other system based on the information on the data architectures of the one system to be data mapped and the information on the data architectures of the other system to be data mapped, and outputting N-to-M mapping information including the correspondences. [Effects of the Invention]

[0010] According to the present invention, accurate data mapping between different data systems can be efficiently performed. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a diagram illustrating a system configuration including a data mapping device according to an embodiment. [Figure 2] FIG. 1 is a diagram illustrating an example of the configuration of a data mapping device according to an embodiment. [Figure 3] FIG. 2 illustrates an example of the configuration of a management terminal according to an embodiment. [Figure 4]FIG. 10 is a diagram illustrating an example of a configuration of source data model information according to the embodiment. [Figure 5] FIG. 10 is a diagram illustrating an example of the configuration of converted data model information according to the embodiment. [Figure 6] FIG. 10 is a diagram illustrating an example of the configuration of presence probability information according to the embodiment. [Figure 7] FIG. 10 is a diagram illustrating an example of the configuration of one-to-one mapping information in the embodiment. [Figure 8] FIG. 10 is a diagram illustrating an example of the configuration of N-to-M mapping information in the embodiment. [Figure 9] FIG. 10 is a diagram illustrating an example of a main flow of a data mapping method according to an embodiment. [Figure 10] FIG. 10 is a diagram illustrating an example of the flow of one-to-one mapping processing in the embodiment. [Figure 11A] FIG. 10 is a diagram illustrating an example of acquiring a conversion destination column group in the embodiment. [Figure 11B] FIG. 10 is a diagram illustrating an example of a prompt in an embodiment. [Figure 11C] FIG. 10 is a diagram showing an example of an LLM answer in an embodiment. [Figure 11D] FIG. 10 is a diagram illustrating an example of one-to-one mapping information in the embodiment. [Figure 12] FIG. 10 is a diagram illustrating an example of the flow of N-to-M mapping processing in the embodiment. [Figure 13A] 10A and 10B are diagrams illustrating an example of acquiring the shortest path (transformation destination) between columns in the embodiment. [Figure 13B] 10A and 10B are diagrams illustrating an example of acquiring the shortest path (source of conversion) between columns in the embodiment. [Figure 13C] FIG. 10 is a diagram illustrating an example of N-to-M mapping information in the embodiment. [Figure 14] FIG. 10 is a diagram illustrating a conceptual example of a path search according to an embodiment. [Figure 15] FIG. 10 is a diagram illustrating an example of a display screen of a mapping log in the embodiment. [Figure 16] FIG. 10 is a diagram illustrating a system configuration including a data mapping device according to another embodiment. [Figure 17]FIG. 10 is a diagram illustrating an example of source master data according to another embodiment. [Figure 18] FIG. 10 is a diagram illustrating an example of converted master data according to another embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] In the following description, a processor is an arithmetic unit and may be one or more processor devices. The at least one processor device may typically be a microprocessor device such as a CPU (Central Processing Unit), but may also be another type of processor device such as a GPU (Graphics Processing Unit). The at least one processor device may be a single-core or multi-core. The at least one processor device may also be a processor core. The at least one processor device may also be a processor device in a broader sense, such as a hardware circuit that performs part or all of the processing (e.g., an FPGA (Field-Programmable Gate Array), a CPLD (Complex Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit)).

[0013] In the following description, a "memory" refers to one or more memory devices, which are an example of one or more storage devices. At least one of the memory devices may be a volatile memory device or a non-volatile memory device.

[0014] In the following description, a "storage device" may refer to one or more persistent storage devices, which are an example of one or more storage devices. A persistent storage device may typically be a non-volatile storage device, specifically, for example, a hard disk drive (HDD), a solid state drive (SSD), or a non-volatile memory express (NVMe) drive.

[0015] In the following description, information that provides an output in response to an input may be described using expressions such as "xxx table" or "xxx database." However, this information may be data of any structure (for example, structured data or unstructured data), or may be a learning model such as a neural network, genetic algorithm, or random forest that generates an output in response to an input. Therefore, "xxx table" or "xxx database" may be referred to as "xxx information." In the following description, the structure of each database or table is an example, and one database or table may be divided into two or more databases or tables, or all or part of two or more databases or tables may be one database or table.

[0016] In the following description, a communication device may be one or more communication interface devices, which may be one or more homogeneous communication interface devices (e.g., one or more NICs (Network Interface Cards)) or two or more heterogeneous communication interface devices (e.g., an NIC and an HBA (Host Bus Adapter)).

[0017] In the following description, processing may be described using a "program" as the subject. However, because a program is executed by a processor to perform a predetermined process using a storage device and / or an interface device, etc., as appropriate, the subject of the process may also be the processor (or a device such as a controller having the processor). A program may be installed in a device such as a computer from a program source. The program source may be, for example, a program distribution server or a computer-readable (e.g., non-transitory) recording medium. In the following description, two or more programs may be realized as one program, or one program may be realized as two or more programs.

[0018] In addition, in the following description, when describing elements of the same type without distinguishing between them, common parts of the reference symbols may be used, and when describing elements of the same type with distinction between them, reference symbols or element identifiers may be used. <System configuration including data mapping device>

[0019] FIG. 1 is a diagram showing an example of a system configuration including a data mapping device 100 according to this embodiment.

[0020] The data mapping device 100 of this embodiment is a device that can efficiently perform accurate data mapping between different data systems, and is connected to a large-scale language model 5, a management terminal 200, a source system 300, and a destination system 400 via a network N. Therefore, these may be referred to as a data mapping system 10.

[0021] The data mapping device 100 executes a process of identifying at least N-to-M correspondences between the columns of each table between the source system 300 and the destination system 400, based on information on the data structure (source data model information 1013) related to one system to be data mapped, i.e., the source system 300, and information on the data structure (destination data model information 1014) related to the other system to be data mapped, i.e., the destination system 400, and outputs N-to-M mapping information including the correspondences.

[0022] The source data model information 1013 and destination data model information 1014 are acquired by the data mapping device 100 via the network N from the source master data 310 of the source system 300 and the destination master data 410 of the destination system 400, or from the management terminal 200.

[0023] Furthermore, the data mapping device 100 is communicably connected to a large-scale language model 5 and a management terminal 200 via an appropriate network N. Of these, the large-scale language model 5 becomes a large-scale language model for identifying one-to-one correspondence between columns of each table between the source system 300 and the destination system 400 by adding the source data model information 1013 and destination data model information 1014.

[0024] Therefore, the large-scale language model 5 of this embodiment is a model that has been trained using training data related to the correspondence. Such a large-scale language model 5 may be implemented as a service that is made available via an API (Application Programming Interface) server on the network N, or may be held by the data mapping device 100 itself and available for use as needed.

[0025] The management terminal 200 is a terminal used by, for example, an administrator in charge of production planning. The administrator, who is the user of this management terminal 200, connects the management terminal 200 to the data mapping device 100 via the network N and transmits, for example, instructions and data required for flexible production planning to the data mapping device 100.

[0026] The data transmitted by this management terminal 200 may include source master data 310 of source system 300 and destination master data 410 of destination system 400. In the above example, source system 300 is a system that manages product manufacturing, etc., primarily based on line production before the transition to flexible manufacturing. In the same example, destination system 400 is a system after the transition to a flexible manufacturing system. Of course, these assumptions are merely examples, and the present invention can be applied to various systems and situations requiring data mapping.

[0027] The data mapping device 100 shown in Figures 1 and 2 includes a storage device 101 consisting of a non-volatile storage device such as a hard disk drive or an embedded multimedia card, an I / O 102 for connecting to an input device such as a keyboard and an output device such as a display, a memory 103 consisting of a volatile storage device such as a RAM (Random Access Memory), a communication device 104 connected to a network N for communicating with various devices, and a processor 105 that calls a program 1010 stored in the storage device 101 into the memory 103 to perform overall control of the device itself and perform various judgments, calculations, and control processing.

[0028] The functions implemented in the processor 105 of the data mapping device 100 include a one-to-one mapping processing unit 110, an N-to-M mapping processing unit 111, and a production planning unit 112 (see FIG. 1). Details of each of these functions will be described later.

[0029] The programs 1010 stored in the storage device 101 may include a general operating system (OS) and applications that implement the above functions, as well as a large-scale language model 1011 and a solver 1012. As already mentioned, the large-scale language model 1011 specifies a one-to-one correspondence between columns in each table between the source system 300 and the destination system 400 by providing source data model information 1013 and destination data model information 1014. The solver 1012 is a tool for solving optimization problems such as path search. The solver 1012 processes the source data model information 1013 and the destination data model information 1014, and obtains a solution based on the condition of the shortest path that directly or indirectly (meaning by connecting via another table) connects specific columns in each table between the source system 300 and the destination system 400.

[0030] Of course, a configuration may be adopted in which only one of the large-scale language model 1011 and the solver 1012 is provided to implement various functions such as one-to-one mapping processing in the data mapping function and path search.

[0031] Note that part of the processing performed by the processor 105 when it executes the program 1010 may be executed by another arithmetic device (for example, hardware such as an ASIC or FPGA). The memory 103 may be a ROM (Read Only Memory), which is a non-volatile storage element. The ROM stores unchanging programs (for example, BIOS). The RAM is a high-speed, volatile storage element such as a DRAM (Dynamic Random Access Memory), and temporarily stores the program 1010 executed by the processor 105 and data used when the program 1010 is executed.

[0032] The input and output devices may also be provided by the large-scale language model 5 (providing server), the management terminal 200, the source system 300, or the destination system 400, all of which are connected to the data mapping device 100 via a network N. The program 1010 executed by the processor 105 is provided to the data mapping device 100 from removable media (CD-ROM, flash memory, etc.) or via the network N, and is stored in a non-volatile storage device 101, which is a non-transitory storage medium.

[0033] Therefore, the data mapping device 100 reads data from the removable media via the I / O 102. The data mapping device 100 is a computer system configured on one physical computer or on multiple logically or physically configured computers, and may operate on a virtual computer built on multiple physical computer resources. The data mapping device 100 is preferably configured on a cloud, but may also be on-premise configured on a specific computer (hardware).

[0034] The network N connecting the data mapping device 100 with the large-scale language model 5, the management terminal 200, the source system 300, and the destination system 400 may be the Internet, a LAN (Local Area Network), a WAN (Wide Area Network), or a mobile phone network, but is not limited to these.

[0035] The management terminal 200 connected to the network N is specifically implemented in the form of a PC (Personal Computer), a tablet terminal, a smartphone, etc. Similar to the data mapping device 100, such a management terminal 200 includes a storage device 201, an I / O 202, a memory 203, a communication device 204, and a processor 205, as illustrated in FIG.

[0036] Data exchange between the data mapping device 100 and the large-scale language model 5, management terminal 200, source system 300, and destination system 400 may be performed according to, for example, an API (Application Programming Interface) protocol. In this case, it is assumed that each device is pre-implemented with the functions and configurations for executing each process of request and response by the API. <Example of data configuration>

[0037] Next, specific examples of data held and managed by the data mapping device 100 will be described with reference to FIGS.

[0038] FIG. 4 is a diagram showing an example of the configuration of the source data model information 1013 in this embodiment.

[0039] This source data model information 1013 is a table that stores the names of the tables, column names, primary keys, and foreign key values ​​managed in the source system 300 .

[0040] Of the above values, the table name is the name of the table, and is information that defines information concepts such as "work" and "equipment" that are primarily required for production planning. The column name is information that indicates the names of the columns that make up the table. The primary key is identification information that uniquely identifies the table. The foreign key is identification information that associates the column with a column (primary key) in another table.

[0041] Next, the converted data model information 1014 of this embodiment will be described.

[0042] FIG. 5 is a diagram showing an example of the configuration of the converted data model information 1014 in this embodiment.

[0043] The conversion destination data model information 1014 is a table that stores the names of the tables, column names, primary keys, and foreign key values ​​that are managed in the conversion destination system 400 .

[0044] Of the above values, the table name is the name of the table, and is information that defines information concepts such as "Task," "Machines," and "Machine," which are mainly required for production planning in flexible manufacturing. The column name is information that indicates the names of the columns that make up the table. The primary key is identification information that uniquely identifies the table. The foreign key is identification information that associates the column with a column (primary key) in another table.

[0045] Next, the existence probability information 1015 of this embodiment will be described.

[0046] FIG. 6 is a diagram showing an example of the configuration of the presence probability information 1015 in this embodiment.

[0047] The existence probability information 1015 is a table that stores information about the probability that a column in a table of the conversion destination system 400 also exists in a table of the conversion source system 300. As shown in Fig. 6, the existence probability information 1015 is composed of a table name, a column name, and a probability value.

[0048] Among these, the table name is the name of a table in at least the flexible manufacturing management system of the destination system 400. The column name is the name of a column in the table. The accuracy is information indicating the probability that the column name also exists in the source system 300. The existence accuracy information 1015 is, for example, defined by a person with appropriate knowledge or defined and generated by the large-scale language model 5. In this case, the large-scale language model 5 includes a model that has learned the appearance tendency of common column names for each column name and its similarity range in combinations of the source system 300 and the destination system 400 that have been previously subjected to data mapping. The large-scale language model 5 outputs the accuracy of each column name by adding information about the destination system 400 that is the current processing target, i.e., the destination data model information 1014, and generates the existence accuracy information 1015.

[0049] Next, the one-to-one mapping information 1016 of this embodiment will be described.

[0050] FIG. 7 is a diagram showing an example of the configuration of the one-to-one mapping information 1016 in this embodiment.

[0051] The one-to-one mapping information 1016 is information indicating the processing result by the one-to-one mapping processing unit 110 of the data mapping device 100. As shown in Fig. 7, the one-to-one mapping information 1016 is composed of values ​​of a source table name, a source column name, a destination table name, and a destination column name. In other words, it is a collection of records in which the table names and column names in the source system 300 are linked to the corresponding table names and column names in the destination system 400.

[0052] The one-to-one mapping processing unit 110 outputs one-to-one mapping information 1016 including one-to-one correspondences between columns of each table between the source system 300 and the destination system 400 by assigning the source data model information 1013 and the destination data model information 1014 to, for example, the large-scale language model 5. At this time, the one-to-one mapping processing unit 110 refers to the existence probability information 1015 shown in Fig. 6 and identifies one-to-one correspondences between columns in the source system 300 and columns in the destination system 400 whose "probability" is higher than a reference value.

[0053] Next, the N-to-M mapping information 1017 of this embodiment will be described.

[0054] FIG. 8 is a diagram showing an example of the configuration of the N-to-M mapping information 1017 in this embodiment.

[0055] The N-to-M mapping information 1017 is information indicating the processing results by the N-to-M mapping processing unit 111 of the data mapping device 100. Such N-to-M mapping information 1017 is composed of values ​​of the source table name, source column name, destination table name, and destination column name, as shown in Fig. 8. However, as shown in the figure, the source column name "Feasible Work ID" of the source table name "Equipment" is linked to the column names "Task," "Machines," and "Machine" ("machines_id," "machines_id," "machine_id") as the destination table names, i.e., an N-to-M correspondence is defined.

[0056] The N-to-M mapping processing unit 111 identifies the N-to-M correspondence relationship by using the search results for paths connecting columns in the source system 300, which are included in the one-to-one mapping information 1016, and the search results for paths connecting columns in the destination system 400, which are included in the one-to-one mapping information 1016. The processing contents of the N-to-M mapping processing unit 111 will be described later. <Data mapping method: Main flow>

[0057] Next, the processing flow of the data mapping method of this embodiment will be described together with each functional unit in the data mapping device 100.

[0058] 9 is a diagram showing an example of the flow of the data mapping method in this embodiment. In explaining this example of the flow, each functional unit of the data mapping device 100 shown in FIG. 1 will also be explained as appropriate.

[0059] First, the one-to-one mapping processing unit 110 of the data mapping device 100 acquires the source master data 310 and the destination master data 410 and the existence probability information 1015 from the source system 300 and the destination system 400 or the management terminal 200 (S1). At this time, the one-to-one mapping processing unit 110 stores the acquired source master data 310 in the source data model information 1013, and stores the acquired destination master data 410 in the destination data model information 1014. The one-to-one mapping processing unit 110 also stores the existence probability information 1015 in the storage device 101.

[0060] Next, the one-to-one mapping processing unit 110 of the data mapping device 100 generates one-to-one mapping information 1016 by assigning the source data model information 1013 and destination data model information 1014 obtained in S1 above, and the existence probability information 1015, to the large-scale language model 5 (S2). Details of this one-to-one mapping process (S2) will be described later with reference to Fig. 10 and Figs. 11A to 11D.

[0061] Next, the N-to-M mapping processing unit 111 of the data mapping device 100 identifies the N-to-M correspondence by using the search results for paths connecting columns in the source data model information 1013 and the destination data model information 1014, which are included in the one-to-one mapping information 1016 obtained in S2 above, and the search results for paths connecting columns in the destination data model information 1014, which are included in the one-to-one mapping information 1016, and generates N-to-M mapping information 1017 (S3). Details of this N-to-M mapping process (S3) will be described later with reference to FIGS. 12 to 15. <One-to-one mapping process>

[0062] Next, the one-to-one mapping process (S2) in the above flow will be described in detail with reference to FIGS. 10 to 11D.

[0063] FIG. 10 is a diagram showing an example of the flow of one-to-one mapping processing in this embodiment.

[0064] Here, the one-to-one mapping processing unit 110 of the data mapping device 100 identifies a group of converted columns having an existence probability higher than a reference value based on the existence probability information 1015 (S21). When referring to the existence probability information 1015 illustrated in Fig. 6, the one-to-one mapping processing unit 110 searches for records in which the value of the "Certainty" field in the existence probability information 1015 is "High," and identifies a group of converted columns including, for example, the column "task_id" of the table "Task" and the column "machine_id" of the table "Machine" (see Fig. 11A).

[0065] Also, the one-to-one mapping processing unit 110 of the data mapping device 100 generates a prompt (see FIG. 11B) including the destination column group specified in S21 (S22). This prompt is given to the large language model 5 to make the large language model 5 answer with the column names in the source system 300 that correspond to the destination column group in the destination system 400, that is, the column names that are used with the same meaning as data. Such a prompt is generated by instructing the large language model for prompt generation to generate it, or by setting each value of the destination column group in a prompt template that the one-to-one mapping processing unit 110 holds in advance.

[0066] Also, the one-to-one mapping processing unit 110 of the data mapping device 100 obtains an answer corresponding to the above prompt (see FIG. 11C) from the large language model 5 (S23). As shown in FIG. 11C, this answer lists the names of the corresponding columns in the source system 300 for each column name included in the destination column group.

[0067] Also, the one-to-one mapping processing unit 110 of the data mapping device 100 generates one-to-one mapping information 1016 based on the answer obtained in S23 (S24), stores this in the storage device 101, and ends this flow. In the case of the above example, an example of the one-to-one mapping information 1016 is shown that defines that the column "task_id" of the table "m_task" in the destination system 400 corresponds one-to-one with the column "work ID" of the table "work" in the source system 300, and the column "machine_id" of the table "m_machine" in the destination system 400 corresponds one-to-one with the column "equipment ID" of the table "equipment" in the source system 300. <N-to-M Mapping Processing>

[0068] Subsequently, the details of the N-to-M mapping processing (S3) in the above flow will be described based on FIGS. 12 to 15.

[0069] FIG. 12 is a diagram showing an example of the flow of N-to-M mapping processing in this embodiment.

[0070] Here, the N-to-M mapping processing unit 111 of the data mapping device 100 uses the foreign keys of the converted data model information 1014 to identify the shortest path connecting columns whose "probability" is "high" in the existence probability information 1015, i.e., a column group (S31). In this case, the N-to-M mapping processing unit 111 identifies, for example, the column "task_id" of the table "Task" and the column "machine_id" of the table "Machine" as records in which the foreign key value is set in the converted data model information 1014 illustrated in FIG. 5. This identification concept is shown as (1) in the ER diagram 1401 and its visualization graph 1402 in FIG. 14.

[0071] Furthermore, the N-to-M mapping processing unit 111 references the foreign keys of each column in the converted data model information 1014, and searches for the shortest path connecting the columns with the highest "accuracy" through a series of paths (see FIG. 13A) from the column "task_id" of the table "Task" to the column "machines_id" of the table "Task", the column "machines_id" of the table "Machines", the column "machine_id" of the table "Machines", and the column "machine_id" of the table "Machine". It is preferable to perform such a search using, for example, the solver 1012 (the same applies below).

[0072] Furthermore, the N-to-M mapping processing unit 111 of the data mapping device 100 uses the foreign keys of the source data model information 1013 to identify the shortest path connecting the columns included in the one-to-one mapping information 1016, i.e., a group of columns (S32). In this case, the N-to-M mapping processing unit 111 searches for a series of paths (FIG. 13B) from the column "Task ID" of the table "Task" to the column "Available Task ID" of the table "Equipment" and to the column "Equipment ID" of the table "Equipment" based on the linkage between the columns indicated by the foreign key values ​​of each record in the source data model information 1013 illustrated in FIG. 4, as the shortest path connecting the source column "Task ID" of the table "Task" and the source column "Equipment ID" of the table "Equipment" shown in FIG. 7. This identification concept is shown as (2) in the ER diagram 1401 and its visualization graph 1402 in FIG. 14.

[0073] Furthermore, the N-to-M mapping processing unit 111 of the data mapping device 100 saves all edges connecting each column group of the conversion destination system 400 and the conversion source system 300 in the N-to-M mapping information 1017 (S33), and ends this flow. In this case, the N-to-M mapping processing unit 111 determines that the columns of the conversion destination system 400 for which the "confidence" of the existence confidence information 1015 is "low", for example, the column "machines_id" of the table "Task" and the columns "machines_id" and "machine_id" of the table "Machines", are tables and columns that can exist in the path between the column "task_id" of the table "Task" of the conversion destination system 400 and the column "Available Work ID" of the table "Equipment" of the conversion source system 300, as identified in S32 above.

[0074] Therefore, it is presumed that the column "Available Work ID" in the table "Equipment" of the source system 300 corresponds to the column "machines_id" in the table "Task" and the columns "machines_id" and "machine_id" in the table "Machines" and should be connected by an edge.

[0075] Based on the information obtained in the flow up to this point, the data mapping device 100 generates the mapping log display screen G10 of Fig. 15, and distributes this to, for example, the management terminal 200 for display. The mapping log display screen G10 of Fig. 15 is composed of a one-to-one mapping processing result G110 and an N-to-M mapping processing result G120.

[0076] Among these, the one-to-one mapping processing result G110 shows that the column "Task ID" of the table "Task" in the source system 300 has a one-to-one correspondence with the column "task_id" with high "confidence" in the table "Task" in the destination system 400, and that the column "Equipment ID" with high "confidence" in the table "Equipment" in the source system 300 has a one-to-one correspondence with the column "machine_id" in the table "Machine" in the destination system 400.

[0077] Furthermore, the N-to-M mapping processing result G120 indicates that in addition to the one-to-one correspondence shown in the one-to-one mapping processing result G110, the column "Performable Work ID" with low "Probability" in the table "Equipment" in the source system 300 corresponds to the column "machines_id" with low "Probability" in the table "Task" in the destination system 400, the column "machines_id" in the table "Machine", and the column "machines_id". <Production planning>

[0078] Next, a mode in which the data mapping device 100 creates a production plan using the production plan creation unit 112 will be described with reference to FIGS.

[0079] FIG. 16 is a diagram showing the configuration of a system including a data mapping device 100 according to another embodiment.

[0080] In this case, as shown in FIG. 16, the data mapping device 100 includes an extracting unit 113 and a data converting unit 114 in addition to the components shown in FIG.

[0081] The one-to-one mapping information 1016 and the N-to-M mapping information 1017 correspond to data conversion specifications, and the existence of such mapping information makes it possible to convert master data into a desired destination format. In other words, destination master data can be created. Creating destination master data allows the production planning unit 112 to output production plan information using the destination master data as input.

[0082] Therefore, the extraction unit 113 extracts table names, column names, primary keys, and foreign keys from the source master data 310A (see FIG. 17) and generates source data model information 1013. The source master data 310A shown in FIG. 17 is configured to include an operation table 311 and an equipment table 312. Of these, operation table 311 can be said to be a table that specifies that process A01 and process A02 must be performed to produce product A, and that process A01 includes two operations, operation A0101 and operation A0102. Furthermore, equipment table 312 can be said to be a table that specifies that equipment A can perform both operation A0101 and operation A0102, but equipment B can only perform operation A0101.

[0083] Furthermore, the data conversion unit 114 converts the source master data 310A into data converted into the format of the converted master data 410, i.e., converted master data 410A (see FIG. 18), based on the source master data 310A and the mapping results (one-to-one mapping information 1016 and N-to-M mapping information 1017) and the conversion specifications (corresponding relationships between tables and columns) specified in the mapping results. The production planning unit 112 inputs the converted master data 410A and applies it to a predetermined production planning engine (e.g., a known one), thereby formulating a production plan in line with, for example, a flexible production system.

[0084] As described above, the data mapping device of this embodiment makes it possible to efficiently perform accurate data mapping between different data systems.

[0085] The present invention is not limited to the above-described embodiments, and includes various modifications and equivalent configurations within the spirit of the appended claims. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to configurations including all of the described configurations. Furthermore, part of the configuration of one embodiment may be replaced with the configuration of another embodiment. Furthermore, the configuration of another embodiment may be added to the configuration of one embodiment. Furthermore, part of the configuration of each embodiment may be added, deleted, or replaced with other configurations.

[0086] Furthermore, the aforementioned configurations, functions, processing units, processing means, etc. may be realized in part or in whole in hardware, for example by designing them as integrated circuits, or may be realized in software by a processor interpreting and executing a program that realizes each function.

[0087] Information such as programs, tables, and files that realize each function can be stored in a storage device such as a memory, a hard disk, or an SSD (Solid State Drive), or in a recording medium such as an IC card, an SD card, or a DVD.

[0088] In addition, the control lines and information lines shown are those that are considered necessary for the explanation, and do not necessarily show all the control lines and information lines that are necessary for implementation. In reality, it can be considered that almost all components are interconnected.

[0089] The various explanations above can be summarized as follows. The following summary may include supplementary explanations and explanations of variations of the above explanations. In the data mapping device of this embodiment, the storage device may hold source data model information including table names, column names, and relationships between columns as the information of the data structure related to at least the one system, and may hold destination data model information including table names, column names, column descriptions, and relationships between columns as the information of the data structure related to at least the other system.

[0090] This allows for accurate identification of corresponding columns between systems by considering the relationship between columns in a table and their descriptions, which in turn enables more efficient and accurate data mapping between different data systems.

[0091] In addition, in the data mapping device of this embodiment, the processor may output one-to-one mapping information including one-to-one correspondence between columns of each table between the one system and the other system based on the source data model information and the destination data model information.

[0092] This makes it possible to efficiently identify correspondences between columns based on, for example, the tendency for column names to match / similar, etc. This in turn makes it possible to more efficiently perform accurate data mapping between different data systems.

[0093] In addition, in the data mapping device of this embodiment, when outputting the one-to-one mapping information, the processor may specify the one-to-one correspondence by adding the source data model information and the destination data model information to a large-scale language model.

[0094] This allows accurate and efficient identification of correspondences between columns using a large-scale language model that has already learned synonym relationships, without the need to prepare an exhaustive thesaurus for column names, thereby enabling more efficient and accurate data mapping between different data systems.

[0095] Furthermore, in the data mapping device of this embodiment, the storage device may hold existence probability information indicating the probability that a column of a data system in the one system exists in the data system in the other system, and when outputting the one-to-one mapping information, the processor may identify the one-to-one correspondence with a column in the one system, targeting a column in the other system whose existence probability indicated by the existence probability information is higher than a standard.

[0096] This makes it possible to identify correspondences between columns based on keywords that are unlikely to change regardless of the system or its operator, such as keywords that indicate essential items for production planning and management. This in turn makes it possible to more efficiently perform accurate data mapping between different data systems.

[0097] Furthermore, in the data mapping device of this embodiment, when outputting the N-to-M mapping information, the processor may identify the N-to-M correspondence by using a search result for paths connecting columns in one system, which is included in the one-to-one mapping information, and a search result for paths connecting columns in the other system, which is included in the one-to-one mapping information.

[0098] This makes it possible to efficiently identify which columns in the destination system can constitute a column in one system (i.e., the source system) that does not exist in the other system (i.e., the destination system) based on the correspondence between columns in one system and the correspondence between columns in the other system, i.e., the N-to-M correspondence, which ultimately makes it possible to more efficiently perform accurate data mapping between different data systems.

[0099] Furthermore, in the data mapping device of this embodiment, the storage device may store, in the existence probability information, classification information indicating at least high or low probability of the existence of a column in the data system of one of the systems in the other system.

[0100] This allows for efficient identification of inter-column correspondences by focusing only on columns with relatively high accuracy, which in turn enables more efficient and accurate data mapping between different data systems.

[0101] In addition, in the data mapping device of this embodiment, the storage device may store, in the existence probability information, information on at least two columns in the other system that have a higher probability of being present in the one system than a standard.

[0102] This makes it possible to more efficiently identify at least one-to-multiple column correspondences between systems, which in turn makes it possible to more efficiently perform accurate data mapping between different data systems.

[0103] Furthermore, in the data mapping device of this embodiment, the storage device may include, in the existence probability information, information on production equipment, work, and the required time for the work, which is used in production planning, for columns in the other system whose probability is higher than a standard.

[0104] This allows you to identify columns related to equipment, operations, and required times that are essential for production planning in advance, and makes it possible to utilize the correspondence between columns in particular, thereby enabling more efficient and accurate data mapping between different data systems.

[0105] In the data mapping device of this embodiment, the processor may execute a production planning process based on at least the N-to-M mapping information.

[0106] This enables accurate and efficient production planning based on the correspondence between the data systems of the source and destination systems, i.e., the data mapping results, and ultimately enables more efficient and accurate data mapping between different data systems. [Explanation of symbols]

[0107] 5: Large-scale language model, 10: Data mapping system, 100: Data mapping device, 110: 1-to-1 mapping processing unit, 111: N-to-M mapping processing unit, 112: Production planning unit, 200: Management terminal, 300: Source system, 310: Source master, 400: Destination system, 410: Destination master

Claims

1. a storage device for storing information on the data structure of each system; a processor that specifies an N-to-M correspondence relationship between columns of each table between one system and the other system based on information on the data architecture of one system that is a target of data mapping and information on the data architecture of the other system that is a target of data mapping, and outputs N-to-M mapping information including the correspondence relationship; A data mapping device comprising:

2. The storage device includes: Maintain source data model information including table names, column names, and relationships between columns as information on the data architecture related to at least one of the systems, and maintain destination data model information including table names, column names, column descriptions, and relationships between columns as information on the data architecture related to at least the other system; 2. The data mapping device according to claim 1.

3. The processor: outputting one-to-one mapping information including one-to-one correspondences between columns of each table between the one system and the other system based on the source data model information and the destination data model information; 3. The data mapping device according to claim 2.

4. The processor: When outputting the one-to-one mapping information, the one-to-one correspondence is specified by adding the source data model information and the destination data model information to a large-scale language model.

4. The data mapping device according to claim 3.

5. The storage device includes: maintaining existence probability information indicating the probability that a column of a data system in the other system exists in a data system in the one system; The processor: When outputting the one-to-one mapping information, the one-to-one correspondence with the columns in the one system is specified for columns in the other system whose existence probability indicated by the existence probability information is higher than a reference value.

4. The data mapping device according to claim 3.

6. The processor: When outputting the N-to-M mapping information, the N-to-M correspondence is identified by using a search result for paths connecting columns in the one system, which is included in the one-to-one mapping information, and a search result for paths connecting columns in the other system, which is included in the one-to-one mapping information.

6. The data mapping device according to claim 5.

7. The storage device includes: the existence probability information holds at least high or low classification information of the probability that a column exists in the data system of the one system in the other system; 6. The data mapping device according to claim 5.

8. The storage device includes: The presence probability information holds information on at least two columns in the other system that have a higher probability of being present in the one system than a reference value.

7. The data mapping device according to claim 6.

9. The storage device includes: The existence probability information includes, for a column in the other system whose probability is higher than a reference value, information on production equipment, work, and the required time for the work, which is used in production planning; 7. The data mapping device according to claim 6.

10. The processor: Executing a production planning process based on at least the N-to-M mapping information; 2. The data mapping device according to claim 1.

11. Stores information on the data structure of each system, Identifying N-to-M correspondences of columns of each table between the one system and the other system based on the data architecture information of one system that is a target of data mapping and the data architecture information of the other system that is a target of data mapping, and outputting N-to-M mapping information including the correspondences. A data mapping method implemented by a computer.

Images