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

Abstract

The aim is to group multiple products in a manner suitable for analysis, compared to using all of the product's manufacturing data. [Solution] The information processing device includes a processor, which acquires manufacturing data for each of the multiple products, selects partial data from the manufacturing data to be used for grouping by characteristic, uses the partial data to group the multiple products into product groups according to their characteristics, and outputs information corresponding to the product groups.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 describes, "An abnormal factor identification method is an abnormal factor identification method for identifying factors causing defective products in a production process that executes one or more manufacturing processes for a manufactured product and an inspection process including a quality inspection for the manufactured product that has undergone the manufacturing process. In the production process, for each manufactured product, one or more types of manufacturing data including manufacturing conditions for that manufactured product and inspection data representing the quality obtained in the inspection process are acquired as a data set, and it is determined whether the manufactured product is a good product or a defective product based on the inspection data. From the manufacturing data for the manufactured products determined to be good products among the data sets, multivariate analysis is performed to extract at least one feature quantity that best represents the manufacturing conditions of the manufactured product. Based on the extracted feature quantity, the significant difference between the average manufacturing data during good product manufacturing and each manufacturing data during defective product manufacturing is calculated respectively, and the contribution rate of each manufacturing data to the significant difference is calculated as the influence degree respectively. From the manufacturing data, manufacturing data for which the influence degree has substantially increased during defective product manufacturing compared to good product manufacturing is extracted as an abnormal factor candidate. An extraction step for abnormal factor candidates; extracting the manufacturing data as the abnormal factor candidate from the data set, and using the extracted manufacturing data to perform a statistical test on whether the manufacturing data affects the quality. A test step; and a test result display step of displaying the result obtained in the test step based on an input or based on a predetermined criterion."

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

[0004] Patent Document 1 describes a method for statistically analyzing inspection results of manufactured goods to determine whether they are good or defective. Such quality inspections are very effective when mass-producing the same product. However, in small-batch, high-mix manufacturing where multiple products of different types are produced in small quantities, attempting to statistically analyze inspection results for a single product presents the problem of insufficient sample size to obtain the desired results. This problem can occur not only in product quality inspections but also in various processes that statistically analyze products. To address this, one could consider adding other products to the statistical population. In this case, it is necessary to group multiple products into product groups based on their characteristics. However, product manufacturing data tends to be increasingly fragmented and numerous, making it difficult to group multiple products while considering all of this data.

[0005] This disclosure is made in view of these circumstances and aims to provide an apparatus, method, and program that can group multiple products in a manner suitable for analysis, compared to using all of the product's manufacturing data. [Means for solving the problem]

[0006] An information processing device according to a first aspect of this disclosure includes a processor, which acquires manufacturing data for each of a plurality of products, selects partial data from the manufacturing data to be used for grouping by characteristic, uses the partial data to group the plurality of products into product groups according to their characteristics, and outputs information corresponding to the product groups.

[0007] An information processing device according to a second aspect of this disclosure, in an information processing device according to a first aspect, the processor uses the partial data to identify at least one similar product from among the plurality of products that is similar to the target product, and outputs information corresponding to the similar product.

[0008] An information processing apparatus according to a third aspect of this disclosure, in an information processing apparatus according to a second aspect, wherein the processor adds the similar products to the population, statistically analyzes the inspection results obtained by inspecting the quality of the target product, and outputs the results of the analysis.

[0009] An information processing device according to a fourth aspect of this disclosure, in an information processing device according to any one of the first to third aspects, the processor stores a plurality of test patterns for inspecting the quality of a product and factors that contribute to the inspection results of the plurality of test patterns in association with each other, and in response to one of the plurality of test patterns being specified, selects data relating to the factors associated with that one test pattern from the manufacturing data as the partial data.

[0010] An information processing apparatus according to a fifth aspect of this disclosure, in an information processing apparatus according to a fourth aspect, the processor determines the similarity between an unknown test pattern other than the plurality of test patterns and the plurality of test patterns, and if the unknown test pattern is similar to any of the plurality of test patterns, stores the unknown test pattern in association with the factor associated with the test pattern similar to the unknown test pattern.

[0011] An information processing device according to a sixth aspect of this disclosure, in an information processing device according to a fourth or fifth aspect, the processor compares the inspection results obtained by inspecting the quality of two or more products using unknown test patterns other than the plurality of test patterns, estimates factors that contribute to the inspection results using the unknown test patterns based on the results of the comparison, and stores the unknown test patterns and the estimated factors in association.

[0012] An information processing device according to the seventh aspect of this disclosure, in an information processing device according to any one of the first to sixth aspects, the processor groups the plurality of products based on the model output output by the machine learning model in response to inputting the partial data into the machine learning model.

[0013] The information processing device according to the eighth aspect of this disclosure is an information processing device according to any one of the first to seventh aspects, wherein the processor acquires at least one of product specification information and environmental information in which the product was manufactured as manufacturing data.

[0014] An information processing method relating to the ninth aspect of this disclosure comprises a computer acquiring manufacturing data for each of several products, selecting partial data from the manufacturing data to be used for grouping by characteristic, using the partial data to group the several products into product groups according to their characteristics, and outputting information corresponding to the product groups.

[0015] An information processing program according to the tenth aspect of this disclosure is an information processing program that causes a computer to perform the following: a process of acquiring manufacturing data for each of a plurality of products; a process of selecting partial data from the manufacturing data to be used for grouping by characteristic; a process of grouping the plurality of products into product groups according to their characteristics using the partial data; and a process of outputting information corresponding to the product groups.

[0016] According to the information processing device, information processing method, and information processing program relating to this disclosure, multiple products can be grouped in a manner suitable for analysis, compared to the case where all of the product manufacturing data is used. [Brief explanation of the drawing]

[0017] [Figure 1] This figure shows an example of the schematic configuration of the information processing system 10 according to this embodiment. [Figure 2] This figure shows an example of data stored in the storage unit 120 by the information processing device 100 according to this embodiment. [Figure 3] This is a diagram showing an example of the functional configuration of the information processing apparatus 100 according to the present embodiment. [Figure 4] This is a diagram showing an example of data stored in the storage unit 120 included in the information processing apparatus 100 according to the present embodiment. [Figure 5] This is a diagram showing an example of grouping of a plurality of products by the information processing apparatus 100 according to the present embodiment. [Figure 6] This is a diagram showing an example of the flow of information processing executed by the information processing apparatus 100 according to the present embodiment. [Figure 7] An example of inspection results of different products with an unknown test pattern is shown.

Embodiments for Carrying Out the Invention

[0018] Hereinafter, an example of an embodiment of the present disclosure will be described while referring to the drawings. In each of the drawings, the same or equivalent components and parts are given the same reference numerals. Also, the dimensional ratios in the drawings are exaggerated for convenience of explanation and may be different from the actual ratios.

[0019] FIG. 1 is a diagram showing an example of the schematic configuration of the information processing system 10 according to the present embodiment. The information processing system 10 includes an information processing apparatus 100. In the information processing system 10, the information processing apparatus 100 acquires manufacturing data of a plurality of products and groups the plurality of products into product groups for each feature using part of the manufacturing data.

[0020] FIG. 2 is a diagram showing an example of the hardware configuration of the information processing apparatus 100 according to the present embodiment. The information processing apparatus 1000 includes a processor 101, a ROM (Read Only Memory) 102, a RAM (Random Access Memory) 103, a storage 104, a communication interface 105, and a user interface 106. These components are connected to each other via a bus 109 so as to be communicable.

[0021] Processor 101 executes various programs and controls each component. Here, it is assumed that processor 101 is a CPU (Central Processing Unit). ROM 102 stores various programs and various data. RAM 103 temporarily stores programs or data as a working area. Storage 104 is composed of an HDD (Hard Disk Drive) or an SSD (Solid State Drive) and stores various programs and various data including an operating system.

[0022] In the information processing apparatus 100 according to the present embodiment, an information processing program 107 is stored in ROM 102 or storage 104. In this figure, a case where the information processing program 107 is stored in storage 104 is shown as an example. Then, the processor 101 reads the information processing program from ROM 102 or storage 104 and executes it using RAM 103 as a working area, thereby controlling each component and performing various arithmetic processes according to the information processing program.

[0023] Also, a machine learning model 108 is stored in ROM 102 or storage 104. In this figure, a case where the machine learning model 108 is stored in storage 104 is shown as an example. However, it is not limited to this. The machine learning model 108 may be stored outside the information processing apparatus 100 (for example, in the cloud). The machine learning model 108 may be, for example, a clustering model. Details thereof will be described later.

[0024] Communication interface 105 is an interface for the information processing apparatus 100 to communicate with other devices. User interface 106 is an input / output interface for the information processing apparatus 100 to exchange information with a user. User interface 106 may include a mouse, a keyboard, a touch panel, a microphone, etc. as input devices and may include a monitor, a speaker, etc. as output devices.

[0025] Figure 3 shows an example of the functional configuration of the information processing device 100 according to this embodiment. The information processing device 100 includes an acquisition unit 110, a storage unit 120, a selection unit 130, a group unit 140, an analysis unit 150, and an output unit 160. These functional configurations may be implemented in the information processing device 100 by having the processor 101 read an information processing program from the ROM 102 or storage 104, expand it in the RAM 103, and execute it.

[0026] The acquisition unit 110 acquires manufacturing data for each of the multiple products. In this case, the acquisition unit 110 may acquire at least one of the product specification information and the environmental information in which the product was manufactured as manufacturing data.

[0027] The memory unit 120 stores multiple test patterns for inspecting product quality and factors that contribute to the inspection results of each of the multiple test patterns, in an associated manner.

[0028] The selection unit 130 selects partial data from the manufacturing data to be used for grouping by characteristic. In this case, the selection unit 130 may select data from the manufacturing data relating to the factors associated with one of the multiple test patterns that has been specified as partial data.

[0029] The grouping unit 140 uses partial data to group multiple products into product groups based on their characteristics. The grouping unit 140 may also use partial data to identify at least one similar product from among the multiple products that is similar to the target product. In this case, the grouping unit 140 may group multiple products based on the model output output by the machine learning model 108 in response to inputting partial data into the machine learning model.

[0030] The analysis unit 150 adds similar products to the population and statistically analyzes the inspection results obtained by inspecting the quality of the target product.

[0031] The output unit 160 outputs information corresponding to the product group. In this case, the output unit 160 may output the results of statistical analysis by adding similar products to the population.

[0032] Figure 4 shows an example of data stored in the storage unit 120 by the information processing device 100 according to this embodiment. Hereafter, we will explain using the case where the product is a storage battery as an example. However, we are not limited to this. The product may be any manufactured or processed product, such as a car, smartphone, or processed food. Furthermore, the product is not limited to finished products, but may also include work-in-progress and semi-finished products.

[0033] The memory unit 120 stores the test pattern, the inspection summary, and the factors in association. When the product is a rechargeable battery, it is necessary to inspect multiple quality aspects, such as charge acceptance performance, low-temperature discharge performance, charge / discharge cycle durability performance, and temperature characteristic performance.

[0034] The "Test Patterns" section defines the test patterns used to inspect the quality of such products. In this diagram, only a simplified identifier for each test pattern is listed in the "Test Patterns" column. However, the conditions and procedures for executing each test pattern may be predefined. The "Inspection Overview" section provides an overview of the inspection content for each test pattern.

[0035] In the Factor section, factors that contribute to the test results for each test pattern, that is, important factors that influence the test results, are defined. For example, in charge acceptance performance testing, it is theoretically or empirically known that the test results change significantly depending on the conversion temperature. In this case, conversion temperature may be defined in the Factor section for test pattern = 001. Similarly, in low-temperature discharge performance testing, it is theoretically or empirically known that the test results change significantly depending on the plate weight. In this case, plate weight may be defined in the Factor section for test pattern = 002.

[0036] In this figure, only one factor is associated with one test pattern as an example. However, if multiple factors contribute to the test results from a single test pattern, then multiple factors may be associated with that single test pattern.

[0037] Figure 5 shows an example of grouping multiple products by the information processing device 100 according to this embodiment. In this figure, one example is shown in which nine products, from product 1 to product 9, are grouped into three product groups, from product group A to product group C.

[0038] In this diagram, the horizontal axis represents the distance between manufacturing data (e.g., Euclidean distance), with the distance increasing as you move to the right. The vertical axis shows multiple products arranged so that products that are close together are placed closer together. This type of visualization is called a dendrogram. In this case, by setting a threshold Th, multiple products can be grouped into any number of product groups.

[0039] The information processing device 100 may group multiple products by obtaining such a dendrogram using, for example, a hierarchical clustering model. Examples of algorithms for this include Ward's method, centroid method, shortest distance method, or group average method. However, it is not limited to these. The information processing device 100 may also group multiple products using non-hierarchical clustering. Examples of algorithms for this include the k-means method or mixture of normal distributions.

[0040] Figure 6 shows an example of the information processing flow executed by the information processing device 100 according to this embodiment. This flow may be started when the processor 101 reads an information processing program from the ROM 102 or storage 104, loads it into the RAM 103, and executes it.

[0041] In this diagram, an example is given of the information processing device 100 statistically analyzing the results of a product quality inspection. Therefore, it is assumed that the information processing device 100 has already stored the data shown in Figure 4 in the storage unit 120 prior to this flow. However, the information processing device 100 may apply or modify this flow to various processes for statistically analyzing a product.

[0042] In step S210, the processor 101, acting as an acquisition unit 110, acquires manufacturing data for each of the multiple products. As an example, the processor 101 may acquire the manufacturing data for each of the multiple storage batteries via a network in cooperation with the communication interface 105.

[0043] In this case, the processor 101 may acquire at least one of the product specification information and the environmental information in which the product was manufactured as manufacturing data. The specification information may include, for example, information on the components that make up the product, such as electrode material, electrode weight, electrolyte material, and separator material. The product specification information may also include information on rated performance that indicates the performance of the product, such as the applied power, nominal voltage, and operating temperature range.

[0044] Environmental information may include manufacturing conditions such as the date, location, and process conditions for each process. Environmental information may also include external conditions such as ambient temperature, room temperature, and weather on the day each process is implemented.

[0045] In step S220, the processor 101 accepts the specification of a test pattern. Once a test pattern is specified, the processor 101 proceeds to step S230. Here, it is assumed that a user who wishes to test the charge acceptance performance of the target battery has specified test pattern = 001 via the user interface 106.

[0046] In step S230, the processor 101, as the selection unit 130, selects partial data from the manufacturing data to be used for grouping by characteristic. For example, the processor 101 may select data related to a factor associated with a test pattern specified in step S220 from the manufacturing data acquired in step S210 as partial data. Here, since test pattern = 001 is specified, the processor 101 may select data related to the chemical conversion temperature from the manufacturing data as partial data.

[0047] In step S240, the processor 101, as the grouping unit 140, groups multiple products into product groups based on their characteristics using partial data. For example, the processor 101 may group multiple products based on the model output of the machine learning model 108 in response to inputting partial data into the machine learning model 108. Here, it is assumed that data related to chemical temperature is input into a hierarchical clustering model, and as a result, the clustering model outputs the dendrogram shown in Figure 5.

[0048] In this case, the processor 101 may group products 3, 9, 2, 6, and 8 into product group A, product 7 into product group B, and products 4, 5, and 1 into product group C, according to the threshold Th. As mentioned above, other methods besides hierarchical clustering may be used to group multiple products.

[0049] In step S250, the processor 101 accepts the designation of the target product. Once the target product is designated, the processor 101 proceeds to step S260. Here, since the product that the user wishes to test for charge acceptance performance is product 9, it is assumed that the user designated the target product = product 9 via the user interface 106.

[0050] In step S260, the processor 101, as a group unit 140, uses partial data to identify at least one similar product from among multiple products that is similar to the target product. For example, the processor 101 may identify products other than the target product that are included in the product group that includes the target product specified in step S250, from among the product group that was grouped in step S240, as similar products. Here, since the target product = product 9 is specified, products 3, 2, 6, and 8 included in product group A may be identified as similar products.

[0051] In step S270, the processor 101, acting as an analysis unit 150, statistically analyzes the inspection results of the quality tests performed on the target product, adding similar products to the population. Here, since products 3, 2, 6, and 8 were identified as similar products in step S260, the processor 101 adds the inspection results of the charge acceptance performance tests performed on products 3, 2, 6, and 8 to the population when analyzing the inspection results of the charge acceptance performance tests performed on product 9, which is the target product.

[0052] Next, the processor 101 adds products 3, 2, 6, and 8 to the denominator and calculates the average value and standard deviation of charging speed, etc. Then, the processor 101 analyzes whether the target product is good or defective, for example, by determining whether the inspection result value of the target product is within a predetermined multiple of the standard deviation value.

[0053] In step S280, the processor 101 outputs information corresponding to the product group as an output unit 160. For example, the processor 101 may display the results of the analysis performed in step S270 on the monitor, which is the output device of the user interface 106.

[0054] In the above explanation, the case in which the processor 101 statistically analyzes the inspection results and outputs the analysis results as information corresponding to the product group was shown as an example. However, it is not limited to this. The processor 101 may output information on similar products identified in step S260 as information corresponding to the product group, or it may output the information of the grouped product groups in step S240 as information corresponding to the product group. Then, the processor 101 terminates this flow.

[0055] As described above, the information processing device 100 according to this embodiment acquires manufacturing data for multiple products, selects partial data from the manufacturing data to be used for grouping by characteristic, uses the partial data to group the multiple products into product groups according to their characteristics, and outputs information corresponding to the product groups.

[0056] Generally, when the manufacturing data to be analyzed is extensive, the process of grouping multiple products becomes computationally intensive, and data that does not correlate with the output becomes noise. In contrast, the information processing device 100 according to this embodiment uses only partial data, rather than all, of the product's manufacturing data when grouping multiple products. As a result, the information processing device 100 according to this embodiment can group multiple products in a manner suitable for analysis, compared to the case where all of the product's manufacturing data is used. Therefore, the information processing device 100 according to this embodiment makes it possible to obtain optimal output even if the amount of product manufacturing data increases significantly in the future.

[0057] In this case, the information processing device 100 according to this embodiment can also use partial data to identify at least one similar product from among multiple products that is similar to the target product, and output information corresponding to the similar product. As a result, the information processing device 100 according to this embodiment can output information limited to only similar products that are similar to the target product.

[0058] Furthermore, the information processing device 100 according to this embodiment can also statistically analyze the inspection results obtained by inspecting the quality of the target product by adding similar products to the population, and output the results of the analysis. However, for products that have just started production or products that are produced in small quantities, uncertainty arises in quality analysis due to insufficient data. In response to this, the information processing device 100 according to this embodiment statistically analyzes the data of other similar products by adding it to the population. As a result, the information processing device 100 according to this embodiment can suppress uncertainty in quality analysis even when there is insufficient data for the same product as the target product.

[0059] Furthermore, the information processing device 100 according to this embodiment stores multiple test patterns for inspecting product quality and factors that contribute to the inspection results of each of the multiple test patterns in association with each other. When one of the multiple test patterns is specified, it can also select data related to the factors associated with that test pattern as partial data from the manufacturing data. Thus, according to the information processing device 100 according to this embodiment, when the important factors that influence the inspection results are known, it is possible to group multiple products by focusing on manufacturing data that reflects the known knowledge.

[0060] This embodiment is not limited to the above, and can be modified or applied in various other forms. For example, the above description showed a case where the test pattern and factors are known in advance. That is, it was assumed that the information processing device 100 had already stored the data shown in Figure 4 in the storage unit 120. However, the information processing device 100 can also add and store new data in the storage unit 120. This will be explained in detail.

[0061] (Application Example 1) Suppose the user wanted to test an unknown test pattern for the target product, other than the multiple test patterns stored in the memory unit 120. Furthermore, it is unknown which factors would influence the test results for the unknown test pattern.

[0062] Here, the unknown test pattern is assumed to be test pattern = 006, and the test summary is capacity recovery rate. In such a case, the processor 101 may determine the similarity between the unknown test pattern and the multiple test patterns. As an example, suppose that five test patterns, test patterns = 001 to 005, and their factors are associated and stored in the memory unit 120. In this case, the processor 101 may determine the similarity between test pattern = 006 and test patterns = 001 to 005. In this case, the processor 101 may determine the similarity of the test patterns based on the similarity of the test conditions and procedures.

[0063] Here, suppose that the similarity between test pattern = 006 and test pattern = 001 is the highest, and that the similarity is above a predetermined threshold. In this case, the processor 101 may determine that test pattern = 006 is similar to test pattern = 001.

[0064] Accordingly, the processor 101 may infer that, for test pattern = 006, just as with test pattern = 001, the chemical reaction temperature will affect the test results. The processor 101 may then store the relationship between test pattern = 006 and the chemical reaction temperature.

[0065] The processor 101 can, for example, determine the similarity between an unknown test pattern other than the multiple test patterns and the multiple test patterns, and if the unknown test pattern is similar to any of the multiple test patterns, it can associate and store the factors associated with the test pattern that is similar to the unknown test pattern. Thus, the information processing device 100 according to this embodiment may be applied to determine which past test pattern a new test pattern is similar to when it is introduced, and to automatically identify important factors based on the determination result.

[0066] (Application Example 2) Application Example 1 illustrates a case where unknown test patterns and factors are associated and memorized based on the similarity of test patterns. However, it is also possible to associate unknown test patterns and factors based on the test results obtained by testing multiple products using the same test pattern.

[0067] Here, the unknown test pattern is assumed to be test pattern = 007, and the inspection summary is 5-hour rate capacity. In this case, the processor 101 may compare the inspection results obtained by inspecting the quality of two or more products using the unknown test pattern. As an example, the inspection results obtained by inspecting the quality of products 1 to 9 using test pattern = 007 may be compared. In this case, suppose the following comparison results are obtained.

[0068] Figure 7 shows an example of the inspection results of different products using an unknown test pattern. Here, we assume that when 24 samples of product 6 were inspected using test pattern = 007, the data shown on the left of this figure was obtained. In this data, the mean is 71.55 [Ah], the standard deviation is 0.25565, the upper limit of 95% is 72.07 [Ah], and the lower limit of 95% is 71.03 [Ah].

[0069] Similarly, suppose that when 18 samples of product 8 were inspected using test pattern = 007, the data shown on the right of this figure was obtained. In this data, the mean is 71.42 [Ah], the standard deviation is 0.29519, the upper limit of 95% is 72.02 [Ah], and the lower limit of 95% is 70.83 [Ah].

[0070] Comparing these, it can be seen that the two sets of data are extremely similar, meeting predetermined criteria. In this case, processor 101 compares the manufacturing data of product 6 and product 8. Here, it is assumed that the plate weight, separator material, applied charge, nominal voltage, and operating temperature range of product 6 and product 8 were not similar among the manufacturing data. On the other hand, it is assumed that the plate material and electrolyte material of product 6 and product 8 were identical or similar among the manufacturing data.

[0071] In this case, the processor 101 may infer that the electrode material and electrolyte material are influencing the test results for test pattern = 007. That is, the processor 101 may infer that the fact that the electrode material and electrolyte material are the same or similar is the reason why the test results for product 6 and product 8 using test pattern = 007 are similar. The processor 101 may then store the relationship between test pattern = 007 and the electrode material and electrolyte material.

[0072] The processor 101 may, for example, compare the inspection results obtained by testing the quality of two or more products using an unknown test pattern other than the multiple test patterns, estimate factors contributing to the inspection results using the unknown test pattern based on the comparison results, and store the unknown test pattern and the estimated factors in association. In this way, the information processing device 100 according to this embodiment may be applied to automatically identify important factors based on the comparison of inspection results obtained by testing multiple products using an unknown test pattern.

[0073] The processes described above can also be implemented using dedicated hardware circuits. In this case, the process may be executed on a single piece of hardware or on multiple pieces of hardware.

[0074] Furthermore, in the above explanation, the term "processor" refers to a broad type of processor, including general-purpose processors (e.g., CPU: Central Processing Unit, etc.) and specialized processors (e.g., GPU: Graphics Processing Unit, ASIC: Application Specific Integrated Circuit, FPGA: Field Programmable Gate Array, programmable logic device, etc.).

[0075] Furthermore, the processor operations described above may not be performed by a single processor, but may also be performed by multiple processors located in physically separate locations working together. Also, the order of the processor operations is not limited to the order described above and may be changed as appropriate.

[0076] Furthermore, the aforementioned program may be provided on a computer-readable non-temporary recording medium such as a USB (Universal Serial Bus) memory, flexible disk, or CD-ROM (Compact Disc Read Only Memory), or it may be provided online via a network such as the Internet. In this case, the program recorded on the computer-readable non-temporary recording medium is usually transferred to and stored in memory or storage. This program may also be provided, for example, as a standalone application software, or it may be incorporated into the software of each device as a function of that device.

[0077] Furthermore, the aforementioned program can be provided as a program product. A program product includes any form of product for providing a program. For example, a program product includes a program provided via a network such as the Internet, and non-temporary computer-readable recording media such as CD-ROMs and DVDs on which the program is stored.

[0078] This disclosure is not limited to the foregoing, and it goes without saying that it can be implemented in various modified forms without departing from its intent. [Explanation of Symbols]

[0079] 10 Information Processing Systems 100 Information Processing Devices 101 Processors 102 ROM 103 RAM 104 storage 105 Communication Interface 106 User Interface 107 Information Processing Programs 108 Machine Learning Models 109 Bus 110 Acquisition Department 120 Storage section 130 Selection Section 140 Group Department 150 Analysis Department 160 Output section

Claims

1. The processor comprises, Manufacturing data was obtained for each of the multiple products, From the aforementioned manufacturing data, select partial data to be used for grouping by characteristic, Using the aforementioned partial data, the multiple products are grouped into product groups according to their characteristics. Outputs information corresponding to the aforementioned product group. Information processing device.

2. The aforementioned processor, Using the aforementioned partial data, at least one similar product that is similar to the target product is identified from among the multiple products. Outputs information corresponding to the aforementioned similar product. The information processing apparatus according to claim 1.

3. The aforementioned processor, The aforementioned similar products were added to the population, and the inspection results of the quality tests of the target product were statistically analyzed. Output the results of the analysis described above. The information processing apparatus according to claim 2.

4. The aforementioned processor, Multiple test patterns for inspecting product quality and factors contributing to the inspection results of each of the multiple test patterns are stored in association with each other. In response to the selection of one of the aforementioned multiple test patterns, data relating to the factor associated with that test pattern is selected from the manufacturing data as the partial data. The information processing apparatus according to any one of claims 1 to 3.

5. The aforementioned processor, The similarity between unknown test patterns other than the aforementioned plurality of test patterns and the aforementioned plurality of test patterns is determined, If the unknown test pattern is similar to any of the multiple test patterns, the unknown test pattern and the factor associated with the similar test pattern are stored in association. The information processing apparatus according to claim 4.

6. The aforementioned processor, The test results obtained by inspecting the quality of two or more products using unknown test patterns other than the aforementioned multiple test patterns are compared. Based on the results of the comparison described above, we estimate the factors that contribute to the test results using the unknown test pattern. The unknown test pattern and the estimated factor are stored in association. The information processing apparatus according to claim 4.

7. The processor groups the multiple products based on the model output that the machine learning model outputs in response to inputting the partial data into the machine learning model. The information processing apparatus according to any one of claims 1 to 3.

8. The processor acquires at least one of the product specification information and the environmental information in which the product was manufactured as manufacturing data. The information processing apparatus according to any one of claims 1 to 3.

9. Computers Obtaining manufacturing data for multiple products, Selecting partial data from the aforementioned manufacturing data to be used for grouping by characteristic, Using the aforementioned partial data, the multiple products are grouped into product groups according to their characteristics, The output of information corresponding to the aforementioned product group, Information processing methods.

10. On the computer, The process of acquiring manufacturing data for multiple products, A process to select partial data to be used for grouping by characteristic from the aforementioned manufacturing data, Using the aforementioned partial data, a process is performed to group the multiple products into product groups according to their characteristics. The process of outputting information corresponding to the aforementioned product group is executed. Information processing program.

Images