Traceability information creation support system and traceability information creation support method

Abstract

To support creation of traceability information containing supplier information that cannot be directly grasped by the company itself.SOLUTION: A traceability information creation support system includes a processor and a storage device. The storage device retains a model which has learned a relation between an article, an article composing the article, and a supplier thereof. The processor accepts input of traceability information indicating a relation between the article, at least one article composing the article, and a supplier thereof, estimates one or more articles and suppliers thereof lacking in the input traceability information by using the model, outputs the estimated one or more articles and the suppliers thereof for confirmation by a user, stores the one or more articles and the suppliers thereof confirmed by the user in the storage device as determined traceability information, and updates the model by learning the determined traceability information.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present invention relates to a technique for supporting the creation of traceability information. [Background technology]

[0002] In order to strengthen the supply chain, there is a need to know the suppliers involved in the production of a company's products in the event of a disaster or malfunction.

[0003] As a technology for acquiring information about parts incorporated into a product, for example, there is a technology described in Japanese Patent Laid-Open Publication No. 2023-35112 (Patent Document 1).

[0004] Patent Document 1 states, "A manufacturing company that manufactures finished products by incorporating components that operate with software has a product lifecycle management system that manages information about the finished products and the components, and the scope of disclosure of data related to the software contracted with a provider company that provides the components to the manufacturing company, and the provider company has a product lifecycle management system that manages the scope of disclosure and software-related data, which is data related to the software, and the product lifecycle management system requests the product lifecycle management system to disclose software-related data within the scope of disclosure, and if the requested software-related data is within the scope of disclosure, the product lifecycle management system transmits the software-related data to the product lifecycle management system." [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2023-35112 Summary of the Invention [Problem to be solved by the invention]

[0006] Companies that manufacture products can obtain information about the suppliers (so-called Tier 1 suppliers) of the parts they directly purchase from among the parts they incorporate into their products. However, they usually cannot obtain information about the suppliers further upstream (so-called Tier 2 and above suppliers) that supply the parts or materials that are incorporated into the parts of those Tier 1 suppliers. As a result, it is not possible to identify suppliers upstream in the supply chain that are most susceptible to risks such as disasters. [Means for solving the problem]

[0007] In order to solve at least one of the above problems, the present invention provides a traceability information creation support system comprising a processor and a storage device, wherein the storage device holds a model that has learned the relationship between an item and the items that make up the item and their suppliers, and the processor accepts input of traceability information that indicates the relationship between an item and at least one item that makes up the item and its supplier, uses the model to estimate one or more items and their suppliers that are missing from the input traceability information, outputs the estimated one or more items and their suppliers for confirmation by a user, stores the one or more items and their suppliers confirmed by the user in the storage device as confirmed traceability information, and updates the model by learning the confirmed traceability information. [Effects of the Invention]

[0008] According to one aspect of the present invention, the creation of traceability information that includes information on suppliers that a company cannot directly grasp is supported.

[0009] Problems, configurations, and effects other than those described above will become apparent from the following description of the embodiments. [Brief explanation of the drawings]

[0010] [Figure 1]1 is a block diagram showing an example of the overall configuration of a system according to a first embodiment of the present invention. [Figure 2] 1 is a block diagram showing an example of the internal configuration of a system according to a first embodiment of the present invention. [Figure 3] 1 is a block diagram showing an example of a hardware configuration of a system according to a first embodiment of the present invention. [Figure 4] FIG. 1 is an explanatory diagram illustrating an example of an outline of processing performed by a system according to a first embodiment of the present invention. [Figure 5] FIG. 2 is an explanatory diagram showing an example of a format of traceability information held by the system according to the first embodiment of the present invention. [Figure 6] FIG. 10 is an explanatory diagram showing another example of the format of traceability information held by the system according to the first embodiment of the present invention. [Figure 7] 3 is a flowchart showing an example of processing of the traceability information creation support device according to the first embodiment of the present invention. [Figure 8] 4 is a flowchart showing a first specific example of processing of the traceability information creation support device according to the first embodiment of the present invention. [Figure 9] 10 is a flowchart showing a second specific example of the process of the traceability information creation support device according to the first embodiment of the present invention. [Figure 10] 3 is a flowchart showing an example of processing performed by the supplier risk assessment device according to the first embodiment of the present invention. [Figure 11] FIG. 2 is an explanatory diagram illustrating an example of processing performed by the supplier risk assessment device according to the first embodiment of the present invention. [Figure 12] FIG. 2 is an explanatory diagram illustrating an example of a screen displayed by the information terminal according to the first embodiment of the present invention. [Figure 13] FIG. 10 is a block diagram showing an example of the internal configuration of a system according to a second embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. [Example]

[0012] FIG. 1 is a block diagram showing an example of the overall configuration of a system according to a first embodiment of the present invention.

[0013] The system of the first embodiment of the present invention is configured, for example, by a service provider server 101, a service user terminal 104, and a wireless or wired network 106 connecting them. Here, the service user is, for example, a company that manufactures products, and requests the service provider to acquire traceability information that traces the parts incorporated into the company's products and their suppliers up the supply chain. In response to the request from the service user, the service provider estimates the traceability information and outputs the results, and further evaluates the risks of the suppliers and outputs the results.

[0014] In this example, the service provider server 101 is made up of a traceability information creation support device 102 and a supplier risk assessment device 103. On the other hand, the service user terminal 104 is made up of an information terminal 105. Each of these is realized by a computer system.

[0015] For example, the service provider server 101 may be realized by a single computer, and the traceability information creation support device 102 and the supplier risk assessment device 103 may be realized as functions of that computer. Alternatively, the traceability information creation support device 102 and the supplier risk assessment device 103 may each be realized as functions of independent computers. Alternatively, the functions of the traceability information creation support device 102 and the supplier risk assessment device 103 may be realized by distributed processing on multiple computers. Furthermore, the traceability information creation support device 102 may include some or all of the functions of the supplier risk assessment device 103, and the supplier risk assessment device 103 may include some or all of the functions of the traceability information creation support device 102.

[0016] Similarly, the information terminal 105 of the service user terminal 104 is also realized by a computer system. For example, the information terminal 105 may be a personal computer or server device used by the service user, or may be a portable information terminal such as a smartphone or tablet terminal.

[0017] 1 shows an example in which the service provider server 101 and the service user terminal 104 are separate computer systems connected by a network 106, but in reality, the functions of the traceability information creation support device 102 and the supplier risk assessment device 103 may be realized by a computer system on the service user side. In that case, the traceability information creation support device 102 and the supplier risk assessment device 103 are included in the service user terminal 104.

[0018] The entire service provider server 101, or the entire system including the service user terminal 104, may be referred to as, for example, a traceability information creation support system or a supplier risk assessment system, and the traceability information creation support device 102 and the supplier risk assessment device 103 may be provided as functions of the traceability information creation support system.

[0019] FIG. 2 is a block diagram showing an example of the internal configuration of the system according to the first embodiment of the present invention.

[0020] The traceability information creation support device 102 includes a traceability information acquisition unit 201, a traceability information estimation unit 202, a traceability information determination unit 203, a traceability information learning unit 204, and an output unit 205. The supplier risk assessment device 103 also includes a traceability information acquisition unit 207, a risk assessment unit 208, and a risk notification unit 209. These are provided as functions of a computer system that constitutes the service provider server 101, for example.

[0021] Furthermore, the traceability information creation support device 102 holds learning traceability information 206. Furthermore, the supplier risk assessment device 103 holds confirmed traceability information 210 and risk calculation information 211. These are stored in a storage device of the computer system that constitutes the service provider server 101.

[0022] The information terminal 105 comprises an input unit 212 and a display unit 213. These are provided as functions of the computer system that constitutes the service user terminal 104.

[0023] An overview of the processing of the system of this embodiment will now be described with reference to Fig. 2. The traceability information learning unit 204 creates and holds a learning model that estimates complete traceability information from incomplete traceability information by referring to the learning traceability information 206 as learning data.

[0024] The service user operates the input unit 212 of the information terminal 105 to input incomplete traceability information into the traceability information creation support device 102. Here, the service user is, for example, a company that manufactures products, and the incomplete traceability information is the traceability information related to the parts that make up the company's products that the company has at that time.

[0025] When the traceability information acquisition unit 201 acquires incomplete traceability information, the traceability information estimation unit 202 then inputs the incomplete traceability information into a learning model to estimate complete traceability information. The estimation result is output to the information terminal 105 via the output unit 205. The service user refers to the estimation result displayed by the display unit 213 and corrects the result if necessary. The correction of the estimation result is input to the traceability information creation support device 102 via the input unit 212.

[0026] The traceability information determination unit 203 determines the traceability information based on the estimation result by the traceability information estimation unit 202 and input from the service user. The determined traceability information is transmitted to the information terminal 105 via the output unit 205, and is also input to the traceability information learning unit 204 and the supplier risk assessment device 103. The display unit 213 of the information terminal 105 displays the determined traceability information. The traceability information learning unit 204 adds the determined traceability information to the learning data and updates the learning model.

[0027] The traceability information acquisition unit 207 of the supplier risk assessment device 103 stores the traceability information acquired from the traceability information determination unit 203 as determined traceability information 210. The risk assessment unit 208 evaluates the risk of each supplier included in the traceability information based on the traceability information acquired by the traceability information acquisition unit 207 and risk calculation information 211. The risk notification unit 209 transmits information on the evaluated risk to the information terminal 105. The display unit 213 of the information terminal 105 displays the information on the evaluated risk.

[0028] The above processing and the information to be referenced will be described in detail later.

[0029] FIG. 3 is a block diagram illustrating an example of a hardware configuration of a system according to the first embodiment of the present invention.

[0030] Specifically, Fig. 3 shows an example of the hardware configuration of a computer system that realizes the service provider server 101 and the service user terminal 104. The computer system 300 shown in Fig. 3 includes a processor (arithmetic unit) 301, a memory 302, a storage device 303, an input device 304, an output device 305, and a communication device 306. The above components are connected to each other by a bus (not shown).

[0031] The memory 302 and the storage device 303 are storage devices that store programs and data used by the processor 301. For example, the learning traceability information 206, the confirmed traceability information 210, and the risk calculation information 211 shown in FIG. 2 may be stored in the storage device 303, and at least some of them may be stored in the memory 302 as needed for processing.

[0032] The memory 302 is configured, for example, by a semiconductor memory, and is mainly used to hold programs and data that are currently being executed. For example, programs and data stored in the storage device 303 are loaded into the memory 302 at startup or when needed. The processor 301 executes various processes in accordance with the programs stored in the memory 302. The processor 301 operates in accordance with the programs and controls each unit in the computer system 300 as needed, thereby realizing various functional units, such as the traceability information acquisition unit 201, the traceability information estimation unit 202, the traceability information determination unit 203, the traceability information learning unit 204, the traceability information acquisition unit 207, the risk assessment unit 208, and the risk notification unit 209 shown in FIG. 2 .

[0033] The storage device 303 is configured by a large-capacity storage device such as a hard disk drive (HDD) or a solid-state drive (SSD), and is used to store programs and data for a long period of time.

[0034] Processor 301 may be comprised of a single processing unit or multiple processing units and may include single or multiple arithmetic units or multiple processing cores. Processor 301 may be implemented as one or more central processing units, microprocessors, microcomputers, microcontrollers, digital signal processors, state machines, logic circuits, graphics processing units, systems on a chip, and / or any device that manipulates signals based on control instructions.

[0035] The input device 304 is a hardware device through which a user inputs instructions and information. The output device 305 is a hardware device that presents various images for input and output, such as a display device or a printing device. The communication device 306 is an interface for connection to a network (for example, the wireless or wired network 106 shown in FIG. 1).

[0036] The computer system 300 may include two or more processors 301. Furthermore, the functions of the traceability information creation support device 102 and the supplier risk assessment device 103 can be implemented in multiple computer systems 300. In this case, the multiple computer systems 300 communicate with each other via a network. For example, some of the multiple functions of the system of this embodiment may be implemented in one computer system 300, and the other parts may be implemented in other computer systems 300.

[0037] The computer system 300 that realizes the traceability information creation support device 102 and the supplier risk assessment device 103 may be, for example, a server device or a personal computer owned by a service provider, or may be a so-called virtual server on the cloud. In the latter case, the computer system 300 shown in Fig. 3 is realized by computer resources on the cloud.

[0038] FIG. 4 is an explanatory diagram illustrating an example of an outline of processing of the system according to the first embodiment of the present invention.

[0039] First, the traceability information learning unit 204 performs learning using the traceability history 401 up to now as learning data, and creates a learning model 402 (step (0)).

[0040] 4 is an example of information included in the learning traceability information 206. The learning traceability information 206 includes traceability histories collected from, for example, companies that manufacture various final products and companies that manufacture parts. Here, each traceability history includes, for example, identification information of the final products or parts, etc. manufactured by each company, identification information of parts or materials, etc. incorporated into the final products or parts, etc., identification information of suppliers of the parts, materials, etc.

[0041] In the traceability history 401 shown in FIG. 4, for example, "Product A: OEM1" indicates "Product A" manufactured by manufacturer "OEM1," "Part 1: S1" indicates "Part 1" supplied by supplier "S1," and the lines connecting them indicate the parent-child relationship between them.

[0042] Specifically, one example of a history included in the traceability history 401 indicates that "Product A" manufactured by manufacturer "OEM1" incorporates "Part 1" procured from supplier "S1," "Part 2" procured from supplier "S2," and "Part 3" procured from supplier "S3," and that "Part 2" procured from supplier "S2" incorporates "Part 4" procured by supplier "S2" from supplier "S4." The above relationship may be described as follows: "Part 1:S1," "Part 2:S2," and "Part 3:S3" are included as child parts of "Product A:OEM1," and "Part 4:S1" is included as a child part of "Part 2:S2." The same applies to other products and parts.

[0043] This is an example of information obtained from, for example, manufacturer "OEM1" and stored in learning traceability information 206, and shows the history of how manufacturer "OEM1" actually procured these parts from their respective suppliers to manufacture "Product A."

[0044] Another example of history included in the traceability history 401 indicates that "Part 6: S4" is included as a child part of "Part 9: S1," that "Part 4: S4" and "Part 7: S7" are included as child parts of "Part 2: S2," and that "Part 6: S6" is included as a child part of "Part 5: S5." These histories may be obtained from suppliers "S1," "S2," and "S5," respectively, for example.

[0045] In reality, much more traceability history is accumulated in the learning traceability information 206. This information can be used for learning to create the learning model 402, but it is desirable to manage it so that it cannot be directly referenced by a third party for the purpose of, for example, investigating the source of the parts that make up a particular product.

[0046] Among the examples of history above, if we focus on the history related to "Product A: OEM1," from the perspective of the manufacturer "OEM1," "Part 1: S1," "Part 2: S2," and "Part 3: S3" correspond to so-called Tier 1, and "Part 4: S4" corresponds to so-called Tier 2. Each company has all the information that corresponds to Tier 1 from its own perspective, but information on Tier 2 and deeper levels is generally not known except in exceptional cases where the supplier discloses its sourcing or the information can be clearly identified by observing the product. In other words, this history can be said to be incomplete (or partial) traceability information related to "Product A: OEM1."

[0047] For the sake of convenience, in the case of a product and its constituent parts, the parts are described as belonging to a "lower level" and the product as belonging to a "higher level." Parts that constitute parts at a lower level belong to an even lower level. Suppliers of parts at a lower level are considered upstream suppliers in the supply chain.

[0048] In the above example, information on "Part 4: S4" among the child parts of "Part 2: S2" is exceptionally obtained. However, information on the child parts of "Part 1: S1," "Part 3: S3," and "Part 4: S4" and on child parts at even lower levels (information indicating whether or not there are child parts, and if so, what kind of parts they are and which supplier they are supplied from, etc.) is not included. The same is true for information on child parts of "Part 2: S2" other than "Part 4: S4."

[0049] However, in the traceability history 401 up to now, in the history other than the history related to "Product A: OEM1," information on child parts such as "Part 1: S1," "Part 2: S2," "Part 3: S3," and "Part 4: S4," as well as child parts at a level lower than those child parts, may be included. For this reason, the traceability information learning unit 204 creates a learning model that estimates complete traceability information by learning the traceability history 401 up to now as learning data. Here, complete traceability information means traceability information that can trace the child parts and child parts at a level lower than those for the product or part being estimated, down to child parts that cannot be divided any further, such as materials.

[0050] Next, when the traceability information acquisition unit 201 acquires incomplete traceability information 403 of the product to be estimated from the service user, the traceability information estimation unit 202 inputs the incomplete traceability information 403 into the learning model 402 (step (1)). Then, the traceability information estimation unit 202 receives the output of the learning model 402 as an estimation result 404 and outputs it to the service user (step (2)).

[0051] 4, the incomplete traceability information 403 includes information on the child parts "Part 1: S1," "Part 2: S2," and "Part 3: S3" of "Product B: OEM1," as well as information on the child part "Part 4: S4" of "Part 2: S2." This is an example of information that is input when a service user wants to obtain traceability information for "Product B: OEM1."

[0052] In this example, the estimation result 404 output from the learning model 402 adds information to the input incomplete traceability information 403, including "Part 8: S8" and "Part 9: S9", which are child parts of "Part 1: S1", "Part 5: S5" or "Part 7: S7", which are child parts of "Part 2: S2", "Part 10: S10", which is a child part of "Part 3: S3", and "Part 6: S6", which is a child part of "Part 9: S9" and "Part 5: S5".

[0053] As will be described later, the estimation process ends when all child parts that cannot be divided any further have been estimated. In addition, the estimation result based on the learning model 402 may include information indicating accuracy. For example, if the estimation result of a child part includes multiple candidates, the accuracy may be output together with the estimation result. In the example of Figure 4, it is estimated that either "Part 5: S5" or "Part 7: S7" is included as a child part of "Part 2: S2", and the respective accuracy rates of "40%" and "60%" are output. This accuracy can be used as a reference when the service user selects information on a child part that is more likely.

[0054] In the above example, when incomplete traceability information 403 is input, a process of inferring a child part and its supplier at a lower level is shown, but a process of inferring a higher-level part or product from a lower-level part may also be performed. For example, by generating a learning model 402 that infers a higher-level parent part or product from a lower-level child part by learning past traceability history 401 in advance, and then inputting incomplete traceability information consisting only of "Part 2: S2" and its child part "Part 4: S4" into learning model 402, it is possible to perform a process of inferring not only the child parts at the lower levels of "Part 2: S2" and "Part 4: S4," but also "Product B: OEM1," which is a higher-level product of "Part 2: S2," its child parts "Part 1: S1" and "Part 3: S3," and their further-level child parts, such as "Part 8: S8."

[0055] Next, the service user refers to the output inference result 404 and inputs information to correct it if necessary. The traceability information confirmation unit 203 corrects the inference result 404 according to the input information and stores the corrected traceability information as confirmed traceability information 210 (step (3)). For example, if the service user determines that there is an error in the inference result, the service user inputs information to correct the error. Alternatively, for example, as shown in the inference result 404, if there are multiple child component candidates and their accuracy is also displayed, the service user may input information to select one of them (or information to determine that all of them are incorrect and replace them with other information that is thought to be correct).

[0056] Next, the risk assessment unit 208 evaluates the risk of each supplier included in the determined traceability information by referring to the risk calculation information 211, and the risk notification unit 209 notifies the result (step (4)). At this time, it may be determined whether or not to notify about the risk based on the accuracy of the supplier's estimation.

[0057] 4, "Part 7: S7" has been deleted from the candidate child parts "Part 5: S5" and "Part 7: S7" included in the estimation result 404, and "Part 10: S10", a child part of "Part 3: S3", has also been deleted. Furthermore, because the accuracy of the estimation of supplier S8 and the assessed risk met predetermined standards, information indicating that there is a risk for "Part 8: S8" has been added.

[0058] In the above description, "products" and "parts" are distinguished from each other. However, this distinction is made for convenience of explanation. In reality, "products" may be treated as "parts" and "parts" may be treated as "products." For example, "Part 1" is a component of "Product B," but supplier "S1" may treat "Part 1" as a "product." Furthermore, "Product B" may be treated as a "product" by manufacturer "OEM1," but may also be treated by a higher-level manufacturer as a "part" of a product manufactured by that manufacturer. In this case, manufacturer "OEM1" is treated as a supplier by the higher-level manufacturer. In other words, in the description of the embodiments of the present invention, "parts" and "products" can be rephrased as "items."

[0059] FIG. 5 is an explanatory diagram illustrating an example of a format of traceability information held by the system according to the first embodiment of the present invention.

[0060] 5 shows an example of traceability information 500 relating to an automobile product. In this example, information identifying the model of the product as the product name, information indicating the automobile as the product category, and information indicating the product size, weight, manufacturer, the country to which the manufacturer belongs, and the production date are registered. Furthermore, similar information items are also registered for the sub-parts of the product.

[0061] The information shown in Figure 5 is an example, and the essential information is the parent-child relationship of the parts and the information that identifies the supplier. However, other information may be included to further improve the accuracy of the estimation. For example, the information may include the region to which the manufacturing plant belongs, the product manufacturing start date (when the product began to be handled), a unique ID, a company name ID, etc. Also, information used in supplier risk assessment (described later) (e.g., recycle rate, materials, manufacturing region, etc.) may be included.

[0062] FIG. 6 is an explanatory diagram illustrating another example of the format of traceability information held by the system according to the first embodiment of the present invention.

[0063] Although Fig. 5 shows an example of traceability information in a table format, in an actual computer system, it may be handled as a file in JSON format as shown in Fig. 6. Traceability information 600 shown in Fig. 6 is the same information as that shown in Fig. 5 expressed in JSON format.

[0064] The learning traceability information 206, the confirmed traceability information 210, and the incomplete traceability information input to the traceability information estimation unit 202 shown in Fig. 1 may all be written in the format shown in Fig. 5 or Fig. 6. However, Figs. 5 and 6 are only examples of the format of traceability information, and the traceability information creation support device 102 and the supplier risk assessment device 103 may use traceability information written in a format other than these.

[0065] FIG. 7 is a flowchart showing an example of processing of the traceability information creation support device 102 according to the first embodiment of the present invention.

[0066] First, the traceability information acquisition unit 201 acquires the name of a product to be identified and partial (i.e., incomplete) traceability information related to the product (step 701). For example, "Product B" shown in Fig. 4 may be acquired as the name of the product to be identified, and incomplete traceability information 403 related to the product B may be acquired as the partial traceability information.

[0067] Next, the traceability information estimation unit 202 uses the learning model 402 created by the traceability information learning unit 204 to estimate, from the partial traceability information, child parts at a level below the confirmed part and information on the suppliers that supply them (step 702). Here, confirmed parts refer to parts included in the partial traceability information acquired in step 701, and parts that have already been estimated and confirmed by the service user.

[0068] At this point, step (0) shown in Fig. 4 may have already been executed and the traceability information learning unit 204 may hold the learning model 402, or the traceability information learning unit 204 may execute step (0) in step 702 to create the learning model 402. In step 702, the traceability information estimation unit 202 may input, for example, the incomplete traceability information 403 shown in Fig. 4 into the learning model 402 and obtain an estimation result 404 of complete traceability.

[0069] Next, the traceability information determination unit 203 presents the traceability information estimated in step 702 to the service user and accepts any corrections (step 703). For example, the traceability information determination unit 203 may present the estimation result 404 to the service user and correct it according to a correction instruction from the service user.

[0070] For example, if the service user inputs information indicating that the presented traceability information does not need to be corrected, the traceability information determination unit 203 holds the presented traceability information as confirmed traceability information. Alternatively, if the service user inputs an instruction to correct the presented traceability information, the traceability information determination unit 203 corrects the presented traceability information in accordance with the instruction, and holds the corrected traceability information. Also, as shown in step (2) of FIG. 4, if multiple candidates are presented, the service user may input an instruction to select one of the candidates. In this case, the traceability information determination unit 203 holds the selected candidate as confirmed traceability information.

[0071] The traceability information determination unit 203 determines whether all of the traceability information has been determined (step 704). For example, the traceability information determination unit 203 determines whether all child parts have been estimated in step 702 and whether the service user has finished checking and making any necessary corrections for all of the estimated child parts.

[0072] As a result, if there is traceability information that has not yet been determined (step 704: No), estimation of child parts that have not yet been estimated (step 702) and confirmation of the same by the service user (step 703) are performed.

[0073] If it is determined that all of the traceability information has been confirmed (step 704: No), the traceability information learning unit 204 updates the training traceability information 206 by incorporating the confirmed traceability information, and updates the learning model 402 by re-learning the updated training traceability information 206 (step 705). At this time, the traceability information learning unit 204 may incorporate only information of the confirmed traceability information whose accuracy is higher than a predetermined standard into the training traceability information 206.

[0074] Here, the traceability information learning unit 204 may perform learning by weighting the information newly incorporated into the learning traceability information 206 more heavily than the weighting of other information. For example, the information included in the learning traceability information 206 may include information indicating when the information was added to the learning traceability information 206, and learning may be performed with weighting such that the more recently added the information is, the heavier the weight is. This allows newer information to be more strongly reflected, making it possible to create a highly accurate learning model 402.

[0075] A specific example of the processing of the traceability information creation support device 102 will be described with reference to Fig. 8 and Fig. 9. That is, the processing shown in Fig. 8 and Fig. 9 each shows an example of a specific procedure for executing the processing shown in Fig. 7.

[0076] In the process shown in Figure 8, the traceability information creation support device 102 estimates child parts at lower levels for the input partial traceability information, and if all of the parts at the lowest level become indivisible units, it presents the entire estimated traceability information to the service user and accepts corrections.

[0077] In contrast, in the process shown in Figure 9, the traceability information creation support device 102 estimates child components at lower levels for the input partial traceability information, presents the estimation results for each estimated child component (or for each level) to the service user, and accepts corrections, repeating this process until all components at the lowest level become indivisible units.

[0078] FIG. 8 is a flowchart showing a first specific example of the process of the traceability information creation support device 102 according to the first embodiment of the present invention.

[0079] Steps 801 and 802 in Fig. 8 are the same as steps 701 and 702 in Fig. 7, respectively, and therefore will not be described again. Next, the traceability information estimation unit 202 determines whether the lowest-level part in the estimated traceability information has become an indivisible unit (step 803). For example, if the lowest-level part cannot identify its child parts, such as raw materials, and there is no supplier at a lower level, it is determined that the lowest-level part has become an indivisible unit.

[0080] For example, the traceability information creation support device 102 may store in advance a database of parts that fall into indivisible units, and the traceability information estimation unit 202 may refer to the database to make the determination in step 803. Alternatively, if the learning model 402 does not output a child part with a higher accuracy than a predetermined standard, it may be determined that the lowest-level part is an indivisible unit.

[0081] In step 803, if it is determined that at least one of the lowest-level parts included in the traceability information estimated up to that point is not an indivisible unit (step 803: No), the traceability information estimation unit 202 returns to step 802 and estimates the child parts and their suppliers in the hierarchy below that part.

[0082] In step 803, if it is determined that all of the lowest-level parts included in the traceability information estimated up to that point have become indivisible units (step 803: Yes), the traceability information determination unit 203 presents the estimated traceability information, including the estimated supplier, to the service user (step 804).

[0083] Next, the traceability information determination unit 203 accepts any corrections to the presented traceability information from the service user and determines the traceability information by reflecting the corrections (step 805). The determination of Yes in step 803 and the execution of step 805 correspond to the determination of Yes in step 704 of the processing in FIG.

[0084] Step 806 to be executed next is the same as step 705 in FIG. 7, and therefore a description thereof will be omitted.

[0085] FIG. 9 is a flowchart showing a second specific example of the process of the traceability information creation support device 102 according to the first embodiment of the present invention.

[0086] Steps 901 and 902 in Fig. 9 are the same as steps 701 and 702 in Fig. 7, respectively, and therefore will not be described further. Next, the traceability information determination unit 203 presents the estimated traceability information, including the estimated supplier, to the service user (step 903).

[0087] Next, the traceability information determination unit 203 accepts any corrections to the presented traceability information from the service user, and determines the traceability information by reflecting the corrections (step 904).

[0088] Next, the traceability information estimation unit 202 determines whether the lowest-level part in the estimated traceability information is an indivisible unit (step 905). This determination is performed in the same manner as step 803 in FIG.

[0089] In step 905, if it is determined that the lowest-level part included in the estimated traceability information is not an indivisible unit (step 905: No), the traceability information estimation unit 202 returns to step 902 and estimates child parts and their suppliers in the hierarchy below that part. Thereafter, the traceability information determination unit 203 executes steps 903 and 904 on the estimation result.

[0090] In step 905, if it is determined that the lowest-level part included in the estimated traceability information has become an indivisible unit (step 905: Yes), the traceability information learning unit 204 executes step 906. A determination of Yes in step 905 corresponds to a determination of Yes in step 704 of the processing in Fig. 7. Step 906 is the same as step 705 in Fig. 7, and therefore a description thereof will be omitted.

[0091] 7 to 9, a process of inferring a lower-level part from a higher-level product or part has been described. However, as described with reference to Fig. 4, a process of inferring a higher-level part or product from a lower-level part may also be performed. In that case, for example, in steps 702, 802, and 902, the traceability information estimation unit 202 uses the learning model 402 to infer a part or product in a higher hierarchy than the determined part. In addition, in steps 704, 803, and 905, the traceability information estimation unit 202 determines whether the estimation has reached the final product.

[0092] FIG. 10 is a flowchart showing an example of processing by the supplier risk assessment device 103 according to the first embodiment of the present invention.

[0093] First, the traceability information acquisition unit 207 acquires the confirmed traceability information (that is, the information has been confirmed by the service user) and stores it as confirmed traceability information 210 (step 1001).

[0094] Next, the risk assessment unit 208 acquires information about each supplier included in the acquired traceability information from a database (step 1002). This database may be a database external to the supplier risk assessment device 103 in which information about each company is registered, or may be held within the supplier risk assessment device 103.

[0095] Next, the risk assessment unit 208 calculates a risk value for each supplier included in the acquired traceability information using the risk calculation information 211 (step 1003).

[0096] Next, the risk assessment unit 208 determines whether or not to notify the risk value calculated for each supplier based on the estimated accuracy of the supplier in the traceability information and the risk value (steps 1004, 1005). If it is determined that notification is necessary, the risk notification unit 209 notifies the service user of the suppliers who pose a risk (step 1006).

[0097] FIG. 11 is an explanatory diagram showing an example of processing of the supplier risk assessment device 103 according to the first embodiment of the present invention.

[0098] 11 shows an example of data acquired from a database regarding a certain supplier (Supplier A). This example includes information such as Supplier A's company name, company ID, defect rate, delivery date non-compliance rate, transaction unit price, annual production volume, sales ranking, production region, climate change risk, and executive composition. The risk assessment unit 208 assesses risk based on this information. The risks to be assessed are not limited, but may include, for example, at least one of QCD (quality, cost, delivery date) risks, financial risks, and ESG (environment, society, corporate governance) risks.

[0099] For example, the risk assessment unit 208 may assess QCD-related risk based on the defect rate, delivery non-compliance rate, transaction unit price, and annual production volume, assess financial risk based on sales ranking, and assess ESG-related risk based on manufacturing region, climate change risk, and executive composition. Specifically, a risk value may be calculated for each risk item so that the value falls within a range from 0 to 1, and a statistical value (e.g., maximum value or average value) of the risk values ​​for all items may be obtained as the overall risk value. Any method may be used to calculate the risk value for each item, and detailed explanations will be omitted.

[0100] Furthermore, the calculation of the risk value by the risk assessment unit 208 as described above is one example of a method for the risk assessment unit to acquire a risk value, and the risk assessment unit 208 may acquire the risk value by a method other than the above. For example, the risk assessment unit 208 may acquire a risk value calculated by an external system.

[0101] In the example of Figure 11, the risk value for QCD is calculated as "0.8," the risk value for finance as "0.4," and the risk value for ESG as "0.1." If the highest value is taken as the overall risk value, "0.8" is output.

[0102] However, in step 1004 of Fig. 10, whether to notify is determined based on both the risk value and the accuracy of the supplier's estimation. For example, thresholds may be set for the risk value and the accuracy, and it may be determined that the risk value should be notified if both are equal to or greater than the thresholds.

[0103] For example, if the risk value threshold is 0.8, the risk value for QCD in the above example meets the criteria. If the accuracy threshold is 80%, a QCD risk notification will be issued if the accuracy of Supplier A's estimate is 80% or higher. On the other hand, if the accuracy of Supplier A's estimate is less than 80%, no risk notification will be issued.

[0104] If a supplier's risk is high, it is desirable to notify the service user of that risk, but if the accuracy of the supplier's estimate is low, the need for notification is relatively low. By determining whether or not to notify based on the accuracy of the estimate as described above, it becomes possible to notify the service user of risks that require notification.

[0105] FIG. 12 is an explanatory diagram showing an example of a screen displayed by the information terminal 105 according to the first embodiment of the present invention.

[0106] 12 is an example of a screen displayed by the display unit 213 of the information terminal 105 when the traceability information determination unit 203 presents the estimated traceability information to the service user. Specifically, the display screen 1200 includes a traceability information display unit 1201, a supplier selection unit 1202, and a change input unit 1203.

[0107] The traceability information display unit 1201 displays the traceability information estimated by the traceability information estimation unit 202. For example, information similar to the estimation result 404 in Fig. 4 may be displayed. In the example of Fig. 12, information on confirmed child parts and their suppliers is displayed in thick solid line frames, information on estimated child parts and their suppliers is displayed in thin solid line frames, information on estimated child parts and their suppliers that includes multiple candidates is displayed in thin dashed line frames, and estimated parent-child relationships are displayed with thin solid lines connecting the frames.

[0108] When the traceability information estimation result includes multiple candidates, the supplier selection unit 1202 displays information to allow the service user to select one of them. For example, as shown in the estimation result 404 in Fig. 4, when two candidates, "part 5: S5" and "part 7: S7," are estimated as child parts of "part 2: S2," and the respective probabilities are "40%" and "60%,," the information is displayed in the supplier selection unit 1202. The service user can operate the input unit 212 to select either "part 5: S5" or "part 7: S7."

[0109] Information about child parts that have been estimated but not yet confirmed is displayed in change input section 1203. A service user can change or delete the estimation results, or add new information, by operating input section 212 and inputting necessary information into change input section 1203. [Example]

[0110] Next, a second embodiment of the present invention will be described. The configuration of the system according to the second embodiment is the same as that of the system according to the first embodiment shown in Figs. 1 to 12 except for the differences described below, and therefore the description will be omitted.

[0111] FIG. 13 is a block diagram illustrating an example of the internal configuration of a system according to the second embodiment of the present invention.

[0112] The system of the second embodiment is the same as the system of the first embodiment, except that the traceability information creation support device 102 is replaced by a traceability information creation support device 1301. The traceability information creation support device 1301 is the same as the traceability information creation support device 102 of the first embodiment, except that the traceability information creation support device 1301 does not hold the learning traceability information 206, the traceability information learning unit 204 is replaced by a prompt generation unit 1303, and the traceability information estimation unit 202 is replaced by a traceability information estimation unit 1302.

[0113] In the second embodiment, the traceability information estimation unit 1302 inputs the input partial traceability information to the prompt generation unit 1303 in step 702 of Fig. 7, step 802 of Fig. 8, and step 902 of Fig. 9. The prompt generation unit 1303 generates a prompt for estimating the missing traceability information based on the input partial traceability information, inputs the prompt to the large-scale language learning system 1304, and returns the output to the traceability information estimation unit 1302. The processing of the traceability information estimation unit 1302 other than the above is the same as the processing of the traceability information estimation unit 202 in the first embodiment.

[0114] In the second embodiment, step 705 in FIG. 7, step 806 in FIG. 8, and step 906 in FIG. 9 are not executed.

[0115] Large-scale language learning system 1304 may be a publicly available general-purpose system, such as a so-called generative AI. If the data learned to create large-scale language learning system 1304 contains information such as specifications, design information, and manufacturing companies of various products, it is possible to estimate traceability information by inputting appropriate prompts into large-scale language learning system 1304. This can assist in the creation of traceability information without creating a dedicated learning model.

[0116] Furthermore, the system according to the embodiment of the present invention may be configured as follows.

[0117] (1) A traceability information creation support system (for example, a traceability information creation support device 102, or an entire system including a supplier risk assessment device 103), comprising a processor (for example, a processor 301) and a storage device (for example, a memory 302 and a storage device 303), the storage device holds a model (for example, a learning model 402 created by a traceability information learning unit 204) that has learned the relationship between an item (for example, a product or a part) and items constituting the item and their suppliers, and the processor receives input of traceability information that indicates the relationship between an item and at least one item constituting the item and its supplier ( For example, steps 701, 801, 901), using the model to estimate one or more items and their suppliers that are missing from the input traceability information (e.g., steps 702, 802, 902), outputting the estimated one or more items and their suppliers for confirmation by a user (e.g., steps 703, 804, 903), storing the one or more items and their suppliers confirmed by the user as confirmed traceability information in the storage device (e.g., steps 703, 805, 904), and updating the model by learning the confirmed traceability information (e.g., steps 705, 806, 906).

[0118] This will help create traceability information that includes information on suppliers that a company cannot directly grasp.

[0119] (2) In the traceability information creation support system described in (1) above, the process of estimating one or more items and their suppliers that are missing in the input traceability information includes a process in which the processor uses the model to estimate one or more items and their suppliers that make up the item included in the input traceability information, and the processor repeats the process of estimating one or more items and their suppliers that make up the item included in the input traceability information until it reaches an item that cannot be divided any further (e.g., a loop of steps 802 to 803), outputs all of the estimated items and their suppliers for confirmation by the user (e.g., step 804), and stores the items and their suppliers confirmed by the user from all of the output items and their suppliers in the storage device as confirmed traceability information (e.g., step 805).

[0120] This allows the user to check the estimated traceability information all at once.

[0121] (3) In the traceability information creation support system described in (1) above, the process of estimating one or more items and their suppliers that are missing in the input traceability information includes a process in which the processor uses the model to estimate one or more items and their suppliers that make up the item included in the input traceability information, and the processor repeats the process of estimating the items and their suppliers that make up the item included in the input traceability information, outputting the estimated items and their suppliers for confirmation by the user, and storing the items and their suppliers confirmed by the user as confirmed traceability information in the storage device until it reaches an item that cannot be divided any further (for example, a loop of steps 902 to 905).

[0122] This allows the user to check the estimated traceability information each time.

[0123] (4) In the traceability information creation support system described in (1) above, the process of estimating one or more items and their suppliers that are missing in the input traceability information includes a process in which the processor uses the model to estimate one or more items and their suppliers that are composed of items included in the input traceability information, and the processor repeats the process of estimating items and their suppliers that are composed of items included in the input traceability information until it reaches an item corresponding to the final product (e.g., a loop of steps 802 to 803), outputs all of the estimated items and their suppliers for confirmation by the user (e.g., step 804), and stores the items and their suppliers confirmed by the user from all of the output items and their suppliers in the storage device as confirmed traceability information (e.g., step 805).

[0124] This allows the user to check the estimated traceability information all at once.

[0125] (5) In the traceability information creation support system described in (1) above, the process of estimating one or more items and their suppliers that are missing in the input traceability information includes a process in which the processor uses the model to estimate one or more items and their suppliers that are composed of items included in the input traceability information, and the processor repeats the process of estimating the items and their suppliers that are composed of items included in the input traceability information, outputting the estimated items and their suppliers for confirmation by the user, and storing the items and their suppliers confirmed by the user as confirmed traceability information in the storage device until it reaches an item corresponding to the final product (for example, a loop of steps 902 to 905).

[0126] This allows the user to check the estimated traceability information each time.

[0127] (6) A traceability information creation support system as described in (1) above, wherein the storage device holds learning traceability information (e.g., learning traceability information 206) including one or more relationships between an item and items that constitute the item and their suppliers, and the processor updates the learning traceability information by adding the determined traceability information, assigns a greater weight to the added traceability information among the updated learning traceability information than to other traceability information, and updates the model by learning the updated learning traceability information (e.g., steps 705, 806, 906).

[0128] This creates a more accurate model that more strongly reflects new information.

[0129] (7) In the traceability information creation support system described in (1) above, when the processor uses the model to estimate multiple combinations of items and their suppliers as candidates for one item and its supplier that is missing in the input traceability information, the processor outputs the accuracy of the estimation of each of the combinations for confirmation by the user, and the user's confirmation includes selecting one of the multiple candidates.

[0130] This helps the user to check the traceability information.

[0131] (8) In the traceability information creation support system described in (1) above, the processor acquires a risk value for each supplier included in the determined traceability information (e.g., step 1003), and determines whether to notify the user of the risk for each supplier based on the risk value for each supplier and the accuracy of the estimation for each supplier (e.g., steps 1004 to 1005).

[0132] This allows for the notification of risks that are most likely to require notification, taking into consideration not only the risk value but also the accuracy of the estimation.

[0133] (9) In the traceability information creation support system described in (8) above, the processor notifies the user of the risk of each supplier when the risk value of each supplier is higher than a predetermined standard and the accuracy of the estimation of each supplier is higher than a predetermined standard (e.g., steps 1004 to 1006).

[0134] This allows for the notification of risks that are most likely to require notification, taking into consideration not only the risk value but also the accuracy of the estimation.

[0135] (10) A traceability information creation support system (for example, an entire system including a traceability information creation support device 1301 and a supplier risk assessment device 103) includes a processor (for example, the processor 301) and a storage device (for example, the memory 302 and the storage device 303), and the processor receives input of traceability information indicating a relationship between an item and at least one item constituting the item and its supplier (for example, steps 701, 801, 901), and uses a large-scale language learning system to estimate one or more items and their suppliers that are missing from the input traceability information (for example, steps 702, 802, 902), output the estimated one or more items and their suppliers for confirmation by the user (e.g., steps 703, 804, 903), store the one or more items and their suppliers confirmed by the user in the storage device as confirmed traceability information (e.g., steps 703, 805, 904), obtain a risk value for each supplier included in the confirmed traceability information (e.g., step 1003), and determine whether or not to notify the user of the risk of each supplier based on the risk value of each supplier and the accuracy of the estimation of each supplier (e.g., steps 1004-1005).

[0136] This will help companies create traceability information that includes information on suppliers that they do not have direct access to, for example, using general-purpose models such as so-called generative AI.

[0137] The present invention is not limited to the above-described embodiments and includes various modifications. For example, the above-described embodiments have been described in detail to provide a better understanding of the present invention, and the present invention is not necessarily limited to those including all of the described configurations. Furthermore, it is possible to replace part of the configuration of one embodiment with the configuration of another embodiment, or to add the configuration of another embodiment to the configuration of one embodiment. Furthermore, it is possible to add, delete, or replace part of the configuration of each embodiment with other configurations.

[0138] Furthermore, the above-described configurations, functions, processing units, processing means, etc. may be partially or entirely implemented in hardware, for example, by designing them as integrated circuits. The above-described configurations, functions, etc. may also be implemented in software, with a processor interpreting and executing a program that implements each function. Information such as the programs, tables, and files that implement each function can be stored in storage devices such as nonvolatile semiconductor memory, hard disk drives, and solid-state drives (SSDs), or in computer-readable, non-transitory data storage media such as IC cards, SD cards, and DVDs.

[0139] In addition, the control lines and information lines shown are those that are considered necessary for the explanation, and not all control lines and information lines in the product are necessarily shown. In reality, it can be considered that almost all components are interconnected. [Explanation of symbols]

[0140] 101 Service Provider Server 102, 1301 Traceability information creation support device 103 Supplier Risk Assessment Device 104 Service user terminal 105 Information terminal 201 Traceability Information Acquisition Department 202, 1302 Traceability Information Estimation Department 203 Traceability Information Confirmation Department 204 Traceability Information Learning Department 205 Output section 206 Traceability Information for Education 207 Traceability Information Acquisition Department 208 Risk Assessment Department 209 Risk Notification Department 210 Confirmed Traceability Information 211 Risk Calculation Information 1303 Prompt Generation Unit 1304 Large-scale Language Learning System

Claims

1. A traceability information creation support system, a processor and a storage device, The storage device stores a model that has learned a relationship between an item, an item that constitutes the item, and a supplier of the item; The processor: Accepting input of traceability information indicating a relationship between an item, at least one item constituting the item, and its supplier; using the model to infer one or more missing items and their suppliers in the input traceability information; outputting the inferred one or more items and their suppliers for user confirmation; storing the one or more items and their suppliers identified by the user as established traceability information in the storage device; A traceability information creation support system characterized in that the model is updated by learning the determined traceability information.

2. The traceability information creation support system according to claim 1, The process of estimating one or more items and their suppliers that are missing in the input traceability information includes a process by the processor using the model to estimate one or more items and their suppliers that constitute the items included in the input traceability information; The processor: repeating the process of estimating one or more items and their suppliers that constitute the item included in the input traceability information until an item that cannot be divided any further is reached; outputting all of the inferred items and their suppliers for review by the user; A traceability information creation support system characterized in that the items and their suppliers confirmed by the user among all the output items and their suppliers are stored in the storage device as confirmed traceability information.

3. The traceability information creation support system according to claim 1, The process of estimating one or more items and their suppliers that are missing in the input traceability information includes a process by the processor using the model to estimate one or more items and their suppliers that constitute the items included in the input traceability information; The processor: A traceability information creation support system characterized by repeating the process of estimating the items and their suppliers that make up the item included in the input traceability information, outputting the estimated items and their suppliers for confirmation by the user, and storing the items and their suppliers confirmed by the user as confirmed traceability information in the storage device until it reaches an item that cannot be divided any further.

4. The traceability information creation support system according to claim 1, The process of estimating one or more items and their suppliers that are missing in the input traceability information includes a process by the processor using the model to estimate one or more items and their suppliers that are composed of items included in the input traceability information; The processor: repeating the process of estimating the products and their suppliers that are composed of the products included in the input traceability information until an product corresponding to the final product is reached; outputting all of the inferred items and their suppliers for review by the user; A traceability information creation support system characterized in that the items and their suppliers confirmed by the user among all the output items and their suppliers are stored in the storage device as confirmed traceability information.

5. The traceability information creation support system according to claim 1, The process of estimating one or more items and their suppliers that are missing in the input traceability information includes a process by the processor using the model to estimate one or more items and their suppliers that are composed of items included in the input traceability information; The processor: A traceability information creation support system characterized by repeating the process of estimating items and their suppliers that are composed of items included in the input traceability information, outputting the estimated items and their suppliers for confirmation by the user, and storing the items and their suppliers confirmed by the user as confirmed traceability information in the storage device until an item corresponding to the final product is reached.

6. The traceability information creation support system according to claim 1, The storage device stores learning traceability information including one or more relationships between an item and an item constituting the item and its supplier; The processor: updating the training traceability information by adding the determined traceability information; A traceability information creation support system characterized by updating the model by assigning a greater weight to the added traceability information among the updated learning traceability information than to other traceability information, and learning the updated learning traceability information.

7. The traceability information creation support system according to claim 1, The processor: When the model is used to estimate a plurality of combinations of an item and its supplier as candidates for one item and its supplier that is missing in the input traceability information, the accuracy of the estimation for each of the combinations is output for confirmation by the user; The traceability information creation support system, wherein the confirmation by the user includes selecting one of the plurality of candidates.

8. The traceability information creation support system according to claim 1, The processor: Obtain a risk value for each supplier included in the determined traceability information; A traceability information creation support system characterized in that it determines whether or not to notify the user of the risk of each supplier based on the risk value of each supplier and the accuracy of the estimation of each supplier.

9. The traceability information creation support system according to claim 8, A traceability information creation support system characterized in that the processor notifies the user of the risk of each supplier when the risk value of each supplier is higher than a predetermined standard and the accuracy of the estimation of each supplier is higher than a predetermined standard.

10. A traceability information creation support system, a processor and a storage device, The processor: Accepting input of traceability information indicating a relationship between an item, at least one item constituting the item, and its supplier; using a large scale language learning system to infer one or more missing items and their suppliers in the input traceability information; outputting the inferred one or more items and their suppliers for user confirmation; storing the one or more items and their suppliers identified by the user as established traceability information in the storage device; Obtain a risk value for each supplier included in the determined traceability information; A traceability information creation support system characterized in that it determines whether or not to notify the user of the risk of each supplier based on the risk value of each supplier and the accuracy of the estimation of each supplier.

11. A traceability information creation support method executed by a computer system, comprising: the computer system includes a processor and a storage device; The storage device stores a model that has learned a relationship between an item, an item that constitutes the item, and a supplier of the item; The traceability information creation support method includes: a step in which the processor receives input of traceability information indicating a relationship between an item, at least one item constituting the item, and its supplier; the processor using the model to infer one or more items and their suppliers that are missing from the input traceability information; the processor outputting the inferred one or more items and their suppliers for confirmation by a user; The processor stores the one or more items and their suppliers identified by the user as established traceability information in the storage device; and a step in which the processor updates the model by learning the determined traceability information.

Images