Analysis method of manufacturing process, analyzer, analysis program, and computer-readable storage medium storing the analysis program

Abstract

To highly accurately and easily determine a manufacturing process that is considered to contribute to the quality of a product for each process group, for the product manufactured through a plurality of process groups.SOLUTION: An analysis method includes processes of: reading, from an SSD9, two or more types of machine learning models 69 generated in advance so as to associate each of a plurality of feature amounts P with one or more of a plurality of manufacturing processes Q and one or more of a plurality of process groups G; reading, from the plurality of feature amounts P, a specific feature amount Ps selected in advance; and estimating a series of one or more manufacturing processes Q estimated to contribute to quality determination for each type of machine learning model 69 on the basis of the specific feature amount Ps and the two or more types of machine learning models 69.SELECTED DRAWING: Figure 16A

Description

[Technical Field]

[0001] The present disclosure relates to a manufacturing process analysis method, an analysis device, an analysis program, and a computer-readable storage medium storing the analysis program. [Background technology]

[0002] For example, Patent Document 1 discloses a manufacturing process for a product that includes a characteristic adjustment step for providing characteristic control processing conditions and a characteristic inspection step that is reached after the specific adjustment step and via at least one other step. Here, the characteristic adjustment step is provided midway through multiple steps.

[0003] According to the description in Patent Document 1, in the characteristic adjustment process, a pre-generated learning model is used to search for optimal characteristic control processing conditions from intermediate characteristics obtained in a process prior to the characteristic adjustment process.

[0004] Furthermore, the learning model described in Patent Document 1 takes intermediate characteristics and characteristic control processing conditions as inputs and outputs product characteristics obtained in the characteristic inspection process. By using such a learning model, it is possible to automatically and appropriately search for characteristic control processing conditions for obtaining desired product characteristics from the intermediate characteristics obtained each time.

[0005] On the other hand, Patent Document 2 discloses that a first discrimination step and a classification step are performed for multiple product data. The first discrimination step is a process of discriminating between good and defective products based on a first learning model with training data. The classification step is a process of grouping product data that have been discriminated as defective based on cluster analysis without training data.

[0006] Furthermore, Patent Document 2 discloses that a second discrimination step is performed after the classification step. The second discrimination step is a process of discriminating between good and defective products for multiple product data and discriminating the type of defect for each defective product based on a second learning model.

[0007] Here, the second learning model is generated by performing machine learning using, as training data, the non-defective product data discriminated in the first discrimination step and the defective product data grouped by defect type in the classification step.

[0008] According to Patent Document 2, by successively performing classification using a first learning model and cluster analysis, it becomes possible to distinguish between good and defective products, as well as to distinguish the type of defect in each defective product. Furthermore, by generating a second learning model using the classification results from the first learning model and cluster analysis as training data, it becomes possible to use the second learning model to perform two types of discrimination. [Prior art documents] [Patent documents]

[0009] [Patent Document 1] Japanese Patent Application Laid-Open No. 2002-287803 [Patent Document 2] Japanese Patent Application Publication No. 2019-204232 Summary of the Invention [Problem to be solved by the invention]

[0010] However, the method described in Patent Document 1 can only be used in cases where a specific property adjustment process is assumed. In more general cases, there is a possibility that the desired product properties will not be achieved simply by changing the property control processing conditions in the specific property adjustment process.

[0011] When faced with such a problem, product manufacturers must consider which of the multiple processes should be adjusted. However, Patent Document 1 does not disclose or suggest such considerations.

[0012] On the other hand, Patent Document 2 discloses that the type of defect in each defective product is determined based on a machine learning model. However, even if the type of defect is determined, the process that caused the defect is not necessarily uniquely determined.

[0013] For example, in the case of a product that is manufactured through a pre-process in which raw materials are processed and a post-process in which the processed materials are assembled, even if a scratch is found on the product, it is not easy to determine whether the scratch occurred in the pre-process or the post-process. Even if it were possible to determine the scratch by inspecting the length, depth, etc. of the scratch, it would still be inconvenient because it would require skill to make such a determination.

[0014] This problem becomes more pronounced when a product is manufactured in multiple factories. In such cases, one or more manufacturing processes are performed in each factory, so it would be advantageous to identify the processes for each factory. However, this identification is more difficult than when analyzing a single factory.

[0015] Furthermore, the above problem is not limited to products manufactured through multiple factories, but is considered to be a common issue for products manufactured through multiple process groups in general.

[0016] The present disclosure has been made in consideration of these points, and its purpose is to enable accurate and easy determination of the manufacturing process that is thought to have contributed to the quality of a product manufactured through multiple process groups, for each process group. [Means for solving the problem]

[0017] A first aspect of the present disclosure relates to a method for analyzing a manufacturing process, which is performed using a computer having a memory unit and a calculation unit, and the results of judging the quality of a product manufactured through multiple process groups, each of which is composed of one or more manufacturing processes.

[0018] According to the first aspect, the multiple process groups are connected in the order in which the manufacturing stages of the product progress, and one or more of the multiple feature amounts acquired during the judgment are assumed to contribute to the judgment of the quality as specific feature amounts, and a sequence of one or more manufacturing processes selected for each process group from one or more manufacturing processes that constitute each of the multiple process groups and that are each assumed to contribute to the judgment of the quality is assumed to be a specific process group.The calculation unit reads from the memory unit two or more types of machine learning models that have been generated in advance to associate each of the multiple feature amounts with one or more of the multiple manufacturing processes across one or more of the multiple process groups, and the calculation unit estimates the specific process group for each type of machine learning model based on the specific feature amount and the two or more types of machine learning models.

[0019] According to the first aspect, the analysis method uses two or more machine learning models that associate each of a plurality of feature quantities with one or more of a plurality of manufacturing processes across one or more of a plurality of process groups. As an example of quality, consider the appearance of a product.

[0020] In this example, once the feature values ​​estimated to have contributed to the deterioration of appearance, such as the presence or absence of a scratch and its location, are determined, it is possible to use a machine learning model to estimate a specific manufacturing process in one process group that may have caused the scratch, as well as a specific manufacturing process in another process group connected to that process group. Estimation using a machine learning model can be performed easily and with high accuracy, even by non-expert workers.

[0021] Furthermore, for products made up of many parts, it is believed that there are countless feature values ​​that can be used for analysis. In such cases, even for an experienced worker, it is not easy to identify the process that is responsible for quality issues. However, estimation using machine learning models can be performed easily and with high accuracy, even when there are many feature values.

[0022] Furthermore, rather than simply using a machine learning model, specific process groups are estimated using each of two or more machine learning models, which allows for more multifaceted analysis and higher accuracy.

[0023] Furthermore, according to a second aspect of the present disclosure, when one of the plurality of process groups that is performed at the end of the order of progression is defined as a first process group, and another process group that is performed before the first process group is defined as a second process group, the two or more types of machine learning models are machine learning models that are generated in advance to associate the plurality of feature amounts with one or more manufacturing processes that make up the first process group, and to associate each manufacturing process that makes up the first process group with one or more manufacturing processes that make up the second process group, and the calculation unit may estimate the specific process group for each type of machine learning model based on the specific feature amounts and the two or more types of machine learning models.

[0024] By using the machine learning model of the second aspect, even when multiple process groups are connected, it is possible to estimate a specific manufacturing process in one process group and a specific manufacturing process in another process group connected to that process group, thereby realizing easy and highly accurate analysis.

[0025] Furthermore, according to a third aspect of the present disclosure, the two or more types of machine learning models may include a first machine learning model based on a decision tree algorithm and a second machine learning model based on a non-decision tree algorithm.

[0026] According to the third aspect, by using models constructed from different viewpoints, a more multifaceted analysis can be performed, thereby achieving a more accurate analysis.

[0027] Furthermore, according to a fourth aspect of the present disclosure, the calculation unit may calculate the importance of each manufacturing process included in the first process group that corresponds to each of the plurality of feature quantities in the first machine learning model, and the importance of each manufacturing process included in the second process group that is associated with each manufacturing process in the first process group, and may estimate the specific process group by determining the one or more manufacturing processes for each process group based on the importance of each manufacturing process.

[0028] According to the fourth aspect, the importance of each manufacturing process can be calculated automatically for each process group according to the machine learning model selected. For example, in the case of a random forest model, the importance specific to the model can be calculated. In this way, the user's discretion can be reduced when determining the specific process group. This allows for more accurate analysis.

[0029] Furthermore, according to a fifth aspect of the present disclosure, the first machine learning model may be a model constructed by sequentially generating a decision tree model constituted by a plurality of partial models by performing a process of generating partial models constituted by the plurality of explanatory variables and one of the plurality of objective variables using a random forest for all of the plurality of objective variables, when the plurality of explanatory variables and a plurality of objective variables are given, and another decision tree model constituted by newly generating the plurality of partial models when another variable linked to the plurality of explanatory variables is used as a new explanatory variable and when each of the plurality of explanatory variables is used as a new objective variable.

[0030] According to the fifth aspect, the specific process group is determined based on a chain of decision tree models. Since the importance of each model can be determined, the specific process group can be easily estimated.

[0031] Furthermore, according to a sixth aspect of the present disclosure, the second machine learning model is a Bayesian network in which each of the plurality of feature quantities is a child node, one or more manufacturing processes constituting the second process group are parent nodes, and one or more manufacturing processes constituting the first process group are intermediate nodes interposed between the child nodes and the parent nodes, and the calculation unit may, when visualizing the Bayesian network as a directed graph structure, include in the specific process group the manufacturing process corresponding to the intermediate node connected to the child node corresponding to the specific feature quantity via one or more edges, and the manufacturing process corresponding to the parent node connected to the intermediate node via one or more edges.

[0032] According to the sixth aspect, the manufacturing process is determined based on the probabilistic connections (dependencies). Since a more multifaceted analysis is performed, a more accurate analysis can be realized.

[0033] Furthermore, according to a seventh aspect of the present disclosure, when the determination result based on the first machine learning model differs from the determination result based on the second machine learning model, the calculation unit may notify the estimation results of each of the first and second machine learning models and the number of samples used in generating each of the first and second machine learning models.

[0034] According to the seventh aspect, the estimation results when using each model are notified to the user without overwriting or discarding the estimation results of one model with the other. At that time, by notifying the user of the number of samples, it becomes possible to perform a more comprehensive analysis, such as an analysis of the learning status of each model.

[0035] According to an eighth aspect of the present disclosure, the quality judgment results may be quantified as quality data that increases or decreases depending on whether the quality is good or bad, and the manufacturing processes constituting the first and second process groups may each be quantified to characterize the content of each manufacturing process, the memory unit may pre-store a first correlation indicating a correlation between each of the plurality of feature amounts and the quality data, a second correlation indicating a correlation between each manufacturing process in the first process group and each of the plurality of feature amounts, and a third correlation indicating a correlation between each manufacturing process in the second process group and each manufacturing process in the first process group, and after determining the specific process group, the calculation unit may notify the breakdown of the specific process group and a control mode for the specific process group to improve the quality based on the first, second, and third correlations.

[0036] According to the eighth aspect, it is possible not only to determine the specific process but also to propose a quality improvement plan, which makes it possible for even unskilled workers to easily improve the quality of products.

[0037] Furthermore, according to a ninth aspect of the present disclosure, the calculation unit may be configured to display, on a display unit capable of displaying the calculation results of the calculation unit, a breakdown of the specific processes determined for each process group and improvement methods for the specific processes for improving the quality, in order of their contribution to the quality.

[0038] According to the ninth aspect, when proposing a quality improvement plan, it is possible to provide a highly visible display, which makes it possible to easily improve the quality of products.

[0039] Furthermore, according to a tenth aspect of the present disclosure, the calculation unit may read, as the quality judgment result, a judgment result by a worker, or a judgment result by a rule-based judgment using the plurality of feature quantities as input, or an output from a pre-generated machine learning model.

[0040] As described in the tenth aspect, the quality assessment results used in the present disclosure are not limited to the output from a machine learning model. By having a worker perform the quality assessment and then having a computer perform the subsequent processes as described above, it is possible to achieve division of labor at the manufacturing site. This can improve the usability of the analysis method.

[0041] Furthermore, according to an eleventh aspect of the present disclosure, the calculation unit may acquire, as the plurality of features, at least one of image data, audio data, text data, and mechanical or electrical data that characterize the product.

[0042] As described in the eleventh aspect, the feature quantities used in the present disclosure are not limited to electrical data. For example, the analysis method according to the present disclosure can be executed based on text data written by a worker (e.g., text describing the condition of a scratch, etc.). This can improve the usability of the analysis method.

[0043] Furthermore, according to a twelfth aspect of the present disclosure, the product may be a metal product having an insulating layer on its surface, the quality of the product being the rust-preventing performance of the insulating layer, and the calculation unit may apply a voltage while a corrosion factor is in contact with the surface of the metal product, thereby acquiring a change over time in the current caused by the voltage, and acquiring, as the plurality of feature quantities, measurement values ​​characterizing the waveform of the change over time.

[0044] As in the twelfth aspect, the present disclosure is particularly effective for analyzing the rust prevention performance of metal products.

[0045] Furthermore, a thirteenth aspect of the present disclosure relates to a manufacturing process analysis device that is configured by a computer having a memory unit and a calculation unit, and that uses the quality judgment results of products manufactured through multiple process groups, each of which is configured by one or more manufacturing processes.

[0046] According to the thirteenth aspect, the plurality of process groups are connected in the order of progression of the manufacturing stages of the product, and one or more feature amounts, among the plurality of feature amounts acquired during the judgment, that are estimated to contribute to the judgment of the quality are defined as specific feature amounts, and a sequence of one or more manufacturing processes selected for each process group from one or more manufacturing processes that constitute each of the plurality of process groups and that are estimated to contribute to the judgment of the quality are defined as specific process groups. The analysis device includes a model reading means that reads from the memory unit two or more types of machine learning models that have been generated in advance so as to associate each of the plurality of feature amounts with one or more of the plurality of manufacturing processes across one or more of the plurality of process groups, and a specific process group estimation means that estimates the specific process groups for each type of machine learning model based on the specific feature amounts and the two or more types of machine learning models.

[0047] A fourteenth aspect of the present disclosure relates to a manufacturing process analysis program executed using a computer having a memory unit and a calculation unit, and quality assessment results of products manufactured through multiple process groups, each consisting of one or more manufacturing processes.

[0048] According to the fourteenth aspect, the plurality of process groups are connected in the order of progression of the product's manufacturing stages, and one or more feature amounts, among the plurality of feature amounts acquired during the judgment, that are estimated to contribute to the judgment of the quality are defined as specific feature amounts. A specific process group is a sequence of one or more manufacturing processes selected for each process group from one or more manufacturing processes that constitute each of the plurality of process groups and that are estimated to contribute to the judgment of the quality. The analysis program causes the computer to execute the following steps: the calculation unit reads from the memory unit two or more types of machine learning models that have been generated in advance so as to associate each of the plurality of feature amounts with one or more of the plurality of manufacturing processes across one or more of the plurality of process groups; and the calculation unit estimates the specific process group for each type of machine learning model based on the specific feature amount and the two or more types of machine learning models.

[0049] A fourteenth aspect of the present disclosure relates to a computer-readable storage medium, which stores the analysis program. [Effects of the Invention]

[0050] As described above, according to the present disclosure, for a product manufactured through multiple process groups, it is possible to accurately and easily determine, for each process group, the manufacturing process that is thought to have contributed to the quality of the product. [Brief explanation of the drawings]

[0051] [Figure 1] FIG. 1 is a system diagram illustrating the configuration of a manufacturing management system. [Figure 2] FIG. 2 is a diagram illustrating the configuration of a measurement device in a manufacturing system. [Figure 3] FIG. 3 is a diagram illustrating an example of a hardware configuration of the analysis device. [Figure 4] FIG. 4 is a diagram illustrating an example of the software configuration of the analysis device. [Figure 5] FIG. 5 is a flowchart illustrating the procedure of the analysis method. [Figure 6] FIG. 6 is a diagram illustrating an example of a change in current over time and a feature quantity obtained from the waveform. [Figure 7] FIG. 7 is a flowchart illustrating the measurement process. [Figure 8] FIG. 8 is a flow chart illustrating the measurement data analysis process. [Figure 9] FIG. 9 is a flow chart illustrating the performance evaluation process. [Figure 10] FIG. 10 is a flowchart illustrating the feature selection process. [Figure 11] FIG. 11 is a flowchart illustrating a specific process estimation process. [Figure 12] FIG. 12 is a flow chart illustrating the correlation analysis process. [Figure 13] FIG. 13 is a diagram for explaining the performance determination model. [Figure 14] FIG. 14 is a diagram illustrating the first machine learning model. [Figure 15] FIG. 15 is a diagram illustrating the second machine-based learning model. [Figure 16A] FIG. 16A is a diagram illustrating a specific process group. [Figure 16B] FIG. 16B is a diagram illustrating a specific process group. [Figure 17] FIG. 17 is a diagram illustrating the first correlation coefficient list. [Figure 18] FIG. 18 is a diagram illustrating the second correlation coefficient list. [Figure 19] FIG. 19 is a diagram illustrating an example of a display screen on a display. [Figure 20] FIG. 20 is a diagram illustrating an example of a display screen on a display. DETAILED DESCRIPTION OF THE INVENTION

[0052] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. Note that the following description is for illustrative purposes only.

[0053] <1. System configuration> 1 is a system diagram illustrating the configuration of a manufacturing management system 100 according to the present disclosure. This manufacturing management system 100 is made up of multiple systems. For example, in this embodiment, the manufacturing management system 100 includes an analysis system 101 and N manufacturing systems 102.

[0054] This analysis system 101 is configured to analyze a manufacturing process Q of a product W based on the quality assessment result of the product W manufactured through a plurality of process groups G. In this embodiment, the quality of the product W is assessed by a computer 1 of the analysis system 101 based on feature values ​​P acquired by the analysis system 101. Note that the plurality of process groups G referred to here are each composed of one or more manufacturing processes Q.

[0055] Hereinafter, the mth (1≦m≦N) manufacturing system 102 may be referred to as the mth system 102m. In this regard, the nth (1≦n≦N) process group G may be referred to as the nth process group G. n This is sometimes the case.

[0056] The multiple process groups G are connected in the order of the progress of the manufacturing stages of the product W (see the white arrows in FIG. 1). Each process group G is linked to one manufacturing system 102. That is, N process groups G are executed by N manufacturing systems 102. One manufacturing system 102 executes one process group G.

[0057] That is, the nth system 102n is the nth process group G n will be carried out.

[0058] Each manufacturing system 102 is, for example, a system installed by a different manufacturer or in a different factory. For example, if the product W is a metal product 201, the first process group G1 is a process group performed at a steel mill, the second process group G2 is a process group performed at a metal processing plant, and the Nth process group GN is, for example, a process group related to painting the product W, which is performed immediately before the metal product 201 is completed.

[0059] For example, the Nth process group GN can be regarded as a process group G that is carried out by a manufacturer called "Tier 1."

[0060] Each manufacturing system 102 includes a manufacturing device 121 and an information processing device 123. The manufacturing device 121 executes a process group G associated with each manufacturing system 102 to advance the manufacturing of a product W toward a finished product. The measuring device 122 inspects the product W manufactured through a plurality of manufacturing processes Q to acquire one or more feature quantities P that characterize the quality of the product W.

[0061] The analysis system 101 includes a computer 1 that functions as an analysis device, and a measurement device 122. The analysis system 101 analyzes a manufacturing process Q in a manufacturing system 102 by using the computer 1 and the judgment results of the quality of a product W.

[0062] The quality inspection of the product W (in other words, the process of acquiring one or more feature quantities P) may be performed by the manufacturing system 102 instead of the analysis system 101.

[0063] In this embodiment, the product W is a metal product. The product W as a metal product includes steel materials used in various parts that constitute a vehicle. When the product W is a metal product, the analysis system 101 may determine the rust prevention performance of the metal product as the quality. In this case, the metal product 201 may have an insulating layer 203 on its surface. Hereinafter, the rust prevention performance of the metal product 201 will be described as an example of quality.

[0064] In this case, the manufacturing management system 100 can be regarded as a rust prevention performance management system that manages the rust prevention performance of metal products manufactured through multiple process groups G. Note that even when a metal product is used for the product W, it is not essential that the product W be provided with the insulating layer 203.

[0065] Also, if the product W is a metal product, the Nth process group GN performed by the manufacturing equipment 121 of the Nth system 102 may include one or more processes related to the formation of the insulating layer 203, such as an electroplating process, a water washing process, a drying process, etc.

[0066] In the following, for the sake of comprehensive discussion, the mth process group Gm is m i It is assumed that the m-th process group G includes manufacturing processes Q (i≧2). m Among the multiple manufacturing processes that make up the i-th manufacturing process Q, we define it as the i-th process Q mi " is sometimes called.

[0067] Below, we will first explain an example of the configuration of the measuring device 122 when a metal product 201 having an insulating layer 203 on its surface is used as the product W. Then, we will explain a method for analyzing the results obtained by the measuring device 122 through an explanation of the computer 1 in the analysis system 101.

[0068] <2. Example of measurement device configuration> 2 is a diagram showing an example of the configuration of the measuring device 122 according to this embodiment. As described above, in this embodiment, the product W inspected by this measuring device 122 is a metal product 201. The metal product 201 has a base material 202 such as a steel plate, and an insulating layer 203 located on the surface of the base material 202.

[0069] Measurement by the measuring device 122 may be performed, for example, after the product W is completed.

[0070] Although not shown in the drawings, a chemical conversion coating may be formed on the surface of the substrate 202, and an insulating coating may be provided on the surface of the chemical conversion coating. In this case, the insulating layer 203 as in this embodiment is formed by the insulating coating.

[0071] The measuring device 122 is configured to apply a voltage to the surface of the metal product 201 as the product W, and acquire the change over time in the current caused by the voltage.

[0072] Specifically, the measuring device 122 applies a voltage to the surface of the metal product 201 while the corrosion factor 205 is in contact with the surface of the metal product 201. The corrosion factor 205 may be an electrolyte material containing a supporting electrolyte such as water or sodium chloride, and a clay mineral such as kaolinite.

[0073] Specifically, the measuring device 122 includes a container 220 , an electrode 221 , a power source 222 , a sealing material 223 , and wiring 224 .

[0074] Of these elements, the container 220 is placed on the surface of the metal product 201 (for example, the surface of the insulating layer 203) via a sealing material 223 to prevent liquid leakage. The corrosion factor 205 is contained in the container 220 and is in contact with the surface of the insulating layer 203. The shape and material of the container 220 are not particularly limited. The container 220 has a tubular shape and is made of a resin material such as acrylic resin or epoxy resin. The term "tubular shape" used here includes tubular shapes with any cross-sectional shape, such as a cylindrical shape or a polygonal cylindrical shape.

[0075] The sealing material 223 is a sheet-like sealing material made of, for example, silicone resin. When the container 220 is placed on the metal product 201, the sealing material 223 improves the adhesion between the container 220 and the insulating layer 203 and fills the gap between them. This effectively prevents the corrosion factor 205 from leaking from between the container 220 and the insulating layer 203.

[0076] The electrode 221 is used to apply a voltage between the substrate 202 and the insulating layer 203. The electrode 221 is configured so that at least its tip is embedded in the corrosion factor 205 in the container 220 and comes into contact with the corrosion factor 205. The electrode 221 may be, for example, an electrode that can be used for electrochemical measurement. The electrode 221 may be, for example, a carbon electrode or a platinum electrode.

[0077] The power supply 222 is connected to the electrode 221 and the substrate 202 via wiring 224, and applies a voltage between the electrode 221 and the substrate 202. At the same time, the power supply 222 measures the change over time in the current flowing between the electrode 221 and the substrate 202 as the voltage is applied. The application of the voltage by the power supply 222 and the measurement of the current are controlled by the measuring device 122.

[0078] The data measured by the power supply 222 (measurement data 49) is transmitted to the computer 1 of the analysis system 101 by wireless or wired communication.

[0079] The measurement data may be data in which the detected current value is plotted against time, or in the case of applying a gradually increasing voltage, data in which the detected current value is plotted against the applied voltage value. In addition to the detected current value, applied voltage value, and measurement time, the measurement data also includes information identifying the production lot of the metal product 201 on which the measurement was performed and the measurement position on the metal product 201.

[0080] 2 is merely an example of analyzing the rust prevention performance of a metal product (product W) having an insulating layer 203. The configuration of the measuring device 122 may vary depending on the type of product W and the type of quality to be analyzed.

[0081] Furthermore, the present disclosure does not necessarily require the measurement device 122. For example, the results of a craftsman's quality assessment may be input to the analysis system 101 together with the measurement data used by the craftsman during the inspection.

[0082] <3. Analysis System> Fig. 3 is a diagram illustrating an example of the hardware configuration of an analysis device (computer 1) according to the present disclosure, and Fig. 4 is a diagram illustrating an example of the software configuration of computer 1. This computer 1 is configured by a computer including a CPU 3 as a calculation unit, and a RAM 7 and an SSD 9 as storage units.

[0083] 3, the computer 1 includes a central processing unit (CPU) 3 that controls the entire computer 1, a read only memory (ROM) 5 that stores a boot program and the like, a random access memory (RAM) 7 that functions as a main memory, and a solid state drive (SSD) 9 that serves as a secondary storage device. Note that a hard disk drive (HDD) or the like can also be used as the secondary storage device instead of the SSD 9.

[0084] Of these elements, the CPU 3 executes various programs. The CPU 3 functions as a calculation unit in this embodiment. The RAM 7 and SSD 9 temporarily or continuously store the programs executed by the CPU 3. The RAM 7 and SSD 9 each function as a storage unit in this embodiment.

[0085] The analysis device 1 also includes a display 11, a graphics memory (Video RAM: VRAM) 13 that stores image data to be displayed on the display 11, and a keyboard 15 and a mouse 17 as man-machine interfaces. The keyboard 15 and the mouse 17 function as a reception unit that receives input from an operator. The display 11 functions as a display unit that displays a screen based on the results of calculations by the CPU 3. The analysis device 1 according to this embodiment can also send and receive data to and from external devices via a communication interface 21.

[0086] As shown in FIG. 4, the program memory of SSD 9 stores an operating system (OS) 19, a measurement data analysis program 291, a performance judgment program 292, a feature selection program 293, a model reading program 294, a specific process estimation program 295, a correlation analysis program 296, and an application program 39.

[0087] The analysis method of this embodiment is executed using a computer (analysis device 1) configured as described above and the quality judgment results of product W, and analyzes manufacturing process Q and process group G.

[0088] A measurement data analysis program 291, a performance judgment program 292, a feature selection program 293, a model reading program 294, a specific process group estimation program 295, and a correlation analysis program 296, which are coded to realize such analysis, constitute the analysis program 29 in this embodiment.

[0089] Here, the analysis program 29 is a program for executing the analysis method according to this embodiment, and is configured to cause a computer (analysis device 1) to execute each step constituting the analysis method. The analysis program 29 is pre-stored in a computer-readable storage medium 18. This storage medium 18 is a tangible storage medium such as a disk medium.

[0090] In the program memory of the SSD 9, each program constituting the analysis program 29 is started in response to a command input from the keyboard 15, mouse 17, etc. At that time, each program is loaded from the SSD 9 into the RAM 7 and executed by the CPU 3.

[0091] Meanwhile, measurement data 49 to be analyzed is stored in the data memory of SSD 9. The measurement data 49 is data measured by the measurement device 122 described above.

[0092] The data memory of the SSD 9 also stores one or more performance determination models 59 and two or more machine learning models 69. The performance determination models 59 and the machine learning models 69 are each machine learning models generated in advance. Details of these machine learning models will be described later.

[0093] The data memory of SSD9 also stores first, second, and third correlation coefficient lists 791, 792, and 793, which are used to provide specific suggestions for changes and adjustments to the manufacturing process Q. Furthermore, the data memory of SSD9 also stores quality data 89 generated by the performance evaluation program 292. These will be described in detail later.

[0094] In addition, various data generated by each program constituting the analysis program 29 and the execution results of the application program 39 are stored in the data memory of the SSD 9 or in the RAM 7 as the main memory, as necessary.

[0095] <4. Overview of analysis method> Fig. 5 is a flowchart illustrating the procedure of the analysis method. As shown in Fig. 5, the analysis method is implemented by sequentially executing a measurement data analysis process (step S1), a performance evaluation process (step S2), a feature selection process (step S3), a model loading process (step S4), a specific process group estimation process (step S5), and a correlation analysis process (step S6).

[0096] The analysis program 29 is configured to cause the computer 1 to execute these processes. That is, among these processes, the measurement data analysis process is carried out by the CPU 3 executing the measurement data analysis program 291 described above.

[0097] Similarly, the performance evaluation process is performed by the CPU3 executing a performance determination program 292. The feature selection process is performed by the CPU3 executing a feature selection program 293. The model loading process is performed by the CPU3 executing a model loading program 294. The specific process group estimation process is performed by the CPU3 executing a specific process group estimation program 295. The correlation analysis process is performed by the CPU3 executing a correlation analysis program 296.

[0098] When the CPU 3 executes the specific process group estimation program 295 etc., an analysis device is configured by the computer 1. That is, the computer 1 functions as an analysis device including a measurement data analysis means that executes a measurement data analysis process, a performance evaluation means that executes a performance evaluation process, a feature selection process, a model reading means that executes a model reading process, a specific process estimation means that executes a specific process estimation process, and a correlation analysis means that executes a correlation analysis process.

[0099] Below, the process (measurement process) performed prior to the analysis of the product 1 will be described with reference to FIGS. 6 and 7, and then the analysis method according to the present disclosure will be described in detail with reference to FIG. 5 again.

[0100] The measurement processing and analysis method described below are performed periodically, for example, for each production lot of product W. A production lot generally refers to a unit of production or order. In this embodiment, a production lot can be a unit separated by, for example, a shipping unit, a production unit, a production month / date, a raw material (e.g., paint) lot, a raw material change point, or a process change point (e.g., when replacing the cutting blade of the metal product 201). It is not necessary to perform the processing for each production lot. The measurement processing may be performed for each product one by one.

[0101] <5. Details of measurement process> FIG. 6 is a diagram illustrating an example of a change in current over time and feature quantities obtained from the waveform. FIG. 7 is a flowchart illustrating a measurement process according to the present disclosure. First, as shown in step S101 of FIG. 7, the measurement device 122 extracts one or more products W for each production lot. The number of products extracted is adjusted appropriately based on the size of the production lot, the expected probability of abnormality occurrence, etc. Note that extraction for each production lot is not essential. Measurement processing may be performed for each product individually.

[0102] In the following step S102, the measurement device 122 measures the measurement data 49 to be analyzed by the measurement device 122. This measurement is performed on each product W extracted in step S101.

[0103] Specifically, in step S102, the measurement device 122 applies a voltage between the insulating layer 203 and the substrate 202, for example, with the corrosion factor 205 in contact with the surface of the insulating layer 203. The manufacturing system 102 measures the change over time in the current caused by the applied voltage. The measurement locations include the main surfaces, edges, and welds of the metal product 201.

[0104] In the following step S103, the measurement device 122 converts the changes over time measured in step S103 into measurement data 49 and transmits it to the computer 1 of the analysis system 101. The transmitted measurement data 49 is stored in the RAM 7 or SSD 9 of the computer 1.

[0105] It should be noted that for a more general product W, the measurement data 49 is not limited to changes in current over time. The measurement data 49 may be at least one of image data, audio data, text data, and mechanical or electrical data that characterize the product W.

[0106] Image data is data obtained, for example, by capturing an image of the product W. Audio data is data obtained, for example, by recording sounds generated when the product W is in operation or when a strike is applied to the product W. Text data is data written related to the quality of the product W, such as inspection records by craftsmen. Mechanical or electrical data includes the weight and dimensions of the product W, as well as voltage and current values, as in this embodiment.

[0107] The following explanation will basically focus on the case where "measurement data 49 = change in current over time," but it can be replaced with other data as described above as appropriate.

[0108] <6. Details of analysis processing> Next, the analysis process will be described in detail. The following process is executed for each product W. It may also be executed for products W extracted for each production lot.

[0109] (6-1. Measurement data analysis process) Fig. 8 is a flowchart showing the details of the measurement data analysis process. When the control process proceeds to step S1 in Fig. 5, the CPU 3 executes the flow shown in Fig. 8 in order from step S201.

[0110] 8, the CPU 3 reads the measurement data 49 from the RAM 7, the SSD 9, etc. In the following step S202, the CPU 3 extracts a plurality of types of feature amounts P from the read measurement data 49.

[0111] Specifically, the CPU 3 acquires, as the plurality of feature quantities P, a plurality of parameters characterizing the change over time of the current (current resulting from the applied voltage) indicated by the measurement data 49. Here, in this embodiment, "change over time of the current" refers to "change over time of the current measured as a waveform." In other words, it refers to "current values ​​corresponding to voltage values ​​changed over time."

[0112] More specifically, the CPU 3 extracts a group of three or more parameters including the number Np of current peaks, the height Ip of each peak, and the gradient S of the current as the plurality of types of feature quantities P (see FIG. 6).

[0113] For example, if there is no abnormality or sign of an abnormality in the insulating layer 203 of the metal product 201, when a DC voltage that increases (gradually increases) over time is applied as described above, almost no current flows until the voltage (applied voltage) reaches the breakdown voltage, and once the applied voltage reaches the breakdown voltage, the current increases rapidly.

[0114] The insulating layer 203 maintains its blocking performance against corrosion factors 205 until the applied voltage reaches the breakdown voltage. As a result, almost no current flows. On the other hand, when the applied voltage reaches the breakdown voltage, the corrosion factors 205 are encouraged to penetrate the insulating layer 203, and the corrosion factors 205 may reach the surface of the substrate 202 at the weakest points of the insulating layer 203, such as points with relatively few cross-linked resin structures. As a result, the current may increase rapidly. In other words, a sudden increase in the detected current value indicates that the corrosion factors 205 have reached the surface of the substrate 202, causing the insulating layer 203 to lose its rust-preventing performance.

[0115] On the other hand, consider a case where there is a local defect in the insulating layer 203 (for example, foreign matter such as gas pins, welding spatter and slag, burrs, iron powder, or unevenness on the surface of the substrate 202), causing a localized area where the effective film thickness is small in the insulating layer 203. In this case, when a gradually increasing DC voltage is applied between the electrode 204 and the substrate 202, the current generated between them is considered to change over time as shown in FIG.

[0116] That is, if a local defect exists in the insulating layer 203, the corrosion factor 5 penetrates locally at the defect location. When the corrosion factor 205 penetrates the insulating layer 203 at a certain defect location and reaches the substrate 202, conduction occurs, causing an instantaneous increase in the current value. If a voltage equal to or greater than the voltage at which water electrolysis occurs is applied between the electrode and the substrate 202 at this point, electrochemical reactions such as water electrolysis proceed on the surface of the substrate 202 due to the conduction. As a result, the generated gas and electrolytic products accumulate in the defect, interrupting the conduction and causing a decrease in the current value. That is, if a local defect exists, the conduction and subsequent interruption at the defect location cause peaks in the waveform of the time-dependent change data of the detected current value, indicating an instantaneous increase and decrease in the current value.

[0117] When multiple defects exist, a number Np of peaks corresponding to the number of defects occurs, as shown in Figure 6. The current value at the peak, i.e., the peak height Ip, is related to the size of the conduction path through which the current flows, i.e., the size, type, and conductivity of the defect. Furthermore, since conduction occurs at a lower applied voltage in areas where the thickness of insulating layer 203 is thinner due to defects, the applied voltage that gives the peak is correlated with the film thickness of the area where the defect exists.

[0118] Furthermore, if the insulating layer 203 has a general defect (for example, a low cross-link density due to a lack of catalyst in the paint, or an oxide film on the surface of the substrate 202), corrosion factors 205 will penetrate the entire insulating layer 203 and reach the substrate 202. As a result, a gradual increase in the current value occurs in the waveform of the time-dependent change data of the detected current value. In other words, the slope S of the waveform of the time-dependent change data when the detected current value increases is related to the insulating quality (film quality) of the insulating layer 203.

[0119] The number, size, type, conductivity, film thickness, and film quality of the above defects are all factors that affect the rust-preventive performance of the insulating layer 203. Therefore, in this embodiment, the number Np of peaks in the current value, the height Ip of the peaks, and the slope S of the waveform of the change in current over time, which are related to these factors, are acquired as feature quantities P that characterize the rust-preventive performance of the insulating layer 203. Furthermore, since the likelihood of peaks occurring in the current varies depending on the position (including the main surface, edge, and weld of the metal product 201) at which the change in current over time is measured, position information indicating the position at which the change in current over time is measured may also be acquired from the measurement data as a parameter for evaluating the rust-preventive performance of the insulating layer 203.

[0120] Note that the parameters that can be used as the feature quantity P are not limited to the number of current peaks Np, the height of each peak Ip, and the current slope S. Any parameter can be used as long as it is related to the rust prevention performance of the metal product 201. When targeting other products W or other qualities (performances), more general parameters can be used.

[0121] Additionally, as described above, when image data, audio data, text data, and mechanical or electrical data are used as the measurement data 49, it is only necessary to extract a feature amount according to the type of data.

[0122] For comprehensive discussion, the plurality of feature quantities P are assumed to be N in total (N≧2). Accordingly, the Nth feature quantity P is referred to as the “Nth feature quantity P N In the case of the metal product 201, the first feature amount P1 corresponds to the number of peaks Np, the second feature amount P2 corresponds to the peak height Ip, and the third feature amount P3 corresponds to the slope S (see FIG. 6).

[0123] Thereafter, in step S203, the CPU 3 converts each of the plurality of feature quantities P into a dimensionless risk score between 0 and 1, and stores the risk score in the RAM 7, the SSD 9, or the like.

[0124] In this embodiment, the risk score indicates the level of risk of abnormalities occurring in the quality (rust prevention performance) of the metal product 201. In other words, it can be interpreted that the higher the risk score, the more likely abnormalities will occur in the metal product 201.

[0125] The inventors of the present application attempted to determine whether the quality of the metal product 201 is good or bad based on the level of the risk score corresponding to each of the multiple feature quantities P and the level of the average value of the multiple feature quantities P. Moreover, rather than simply dividing the quality of the metal product 201 into two categories, either "normal" or "abnormal," they attempted to achieve three or more classifications, including a gray area (an intermediate state) between white (normal) and black (abnormal), such as "there is a sign of an abnormality occurring."

[0126] However, such a judgment is not easy. Therefore, we decided to use a pre-generated machine learning model in the performance evaluation process that follows the measurement data analysis process. By using a machine learning model to make judgments, even those who are not skilled craftsmen can make judgments with a certain degree of accuracy.

[0127] For the sake of simplicity, the following description will be given assuming that each risk score is the same as the corresponding feature value P.

[0128] (6-2. Performance evaluation process) Fig. 9 is a flowchart showing the details of the performance evaluation process. When the control process proceeds to step S2 in Fig. 5, the CPU 3 executes the flow shown in Fig. 9 in order from step S301.

[0129] In the performance evaluation process, the CPU 3 determines and outputs the degree of quality (corrosion prevention performance) of the metal product 201 based on the multiple feature quantities P (more precisely, the risk scores corresponding to each feature quantity P) acquired in the measurement data analysis process. The term "degree of quality" here includes not only the two states of "normal" and "abnormal," but also the intermediate state of "there are signs of abnormality" as described above.

[0130] The quality level (quality) is quantified so that it increases or decreases depending on the quality. For example, in this embodiment, the quality exemplified as rust prevention performance is quantified as, for example, "normal = 2," "sign of abnormality present = 1," and "abnormal = 0." Hereinafter, the quantified quality level is also referred to as "quality data 89."

[0131] Specifically, first, in step S301, the CPU 3 reads a plurality of feature quantities P that have been scored as risk scores. In the specific example described above, the feature quantities P that are read here are the number of peaks Np as the first feature quantity P1, the peak height Ip as the second feature quantity P2, and the slope S as the third feature quantity P3.

[0132] In the next step S302, the CPU 3 reads the performance determination model 59 that has been generated in advance by machine learning. As shown in Fig. 13, the performance determination model 59 is a machine learning model that associates a plurality of feature quantities P with the degree of quality of the rust prevention performance (quality data 89).

[0133] In other words, the performance determination model 59 is a machine learning model that receives a plurality of feature quantities P as input and outputs an estimated value of the degree of quality of the rust prevention performance. The performance determination model 59 is generated in advance by supervised learning.

[0134] The performance assessment model 59 is trained using first training data. This first training data is composed of, for example, a plurality of feature quantities P obtained from each of the plurality of measurement data 49, and quality data 89 indicating the degree of quality of the rust prevention performance that is determined in advance when each measurement data 49 is acquired. Pairs each consisting of a plurality of feature quantities P and the associated quality data 89 are created in advance in the same number as the number of samples of the first training data.

[0135] Specifically, the performance evaluation model 59 may be a nonlinear regression model or a linear regression model. The performance evaluation model 59 as a nonlinear regression model may be a machine learning model based on a decision tree algorithm. The performance evaluation model 59 related to the decision tree may be a Random Forest (RF) model.

[0136] In the next step S303, the CPU 3 judges the pass / fail degree of the rust prevention performance based on the plurality of feature quantities P and the performance judgment model 59. Specifically, the CPU 3 inputs the plurality of feature quantities P read in step S301, i.e., the feature quantities P obtained from the product W to be analyzed, into the trained performance judgment model 59, and causes the model 59 to output an estimation result of the pass / fail degree. When quantified as described above, this estimation result is output as quality data 89.

[0137] In the following step S304, the CPU 3 stores the output result of step S303 in the RAM 7 or SSD 9. Thereafter, the control process returns from the flow in Fig. 9 to the flow in Fig. 5, and proceeds to step S3.

[0138] (6-3. Feature selection process) Fig. 10 is a flowchart showing the details of the feature selection process. When the control process proceeds to step S2 in Fig. 5, the CPU 3 executes the flow shown in Fig. 9 in order from step S301.

[0139] Hereinafter, among the multiple feature amounts P acquired during quality assessment, one or more feature amounts P estimated to contribute to the quality assessment will be referred to as specific feature amounts Ps. In the specific examples shown in Figures 2 and 6, the specific feature amounts Ps are one or more feature amounts P estimated to have contributed to the assessment of rust prevention performance.

[0140] Hereinafter, among the multiple feature amounts P acquired during quality assessment, one or more feature amounts P estimated to contribute to the quality assessment will be referred to as specific feature amounts Ps. In the specific examples shown in Figures 2 and 6, the specific feature amounts Ps are one or more feature amounts P estimated to have contributed to the assessment of rust prevention performance.

[0141] Specifically, in the feature selection process, the CPU 3 selects a specific feature Ps from among a plurality of feature P acquired from the measurement data 49. This selection can be made based on the importance (importance) of each of the plurality of feature P corresponding to the degree of pass / fail. In the case of the specific example described above, the CPU 3 selects which of the three parameters, the number of peaks Np as the first feature P1, the peak height Ip as the second feature P2, and the slope S as the third feature P3, contributed to the determination of the rust prevention performance, including abnormalities and their precursors.

[0142] In the example of FIG. 13, the specific feature amount Ps is composed of the number of peaks Np as the first feature amount P1 and the peak height Ip as the second feature amount P2.

[0143] This determination can be made by determining the importance of each of the multiple feature amounts P when the performance determination model 59 outputs the quality data 89, and setting one or more feature amounts P in descending order of importance as the specific feature amount Ps. When the RF model M1 is used for the performance determination model 59, the contribution of each feature amount P (so-called RF importance) can be used as the importance. The importance may be determined, for example, based on the Gini coefficient of the decision tree constituting the random forest model.

[0144] Alternatively, so-called SHAP (Shapley Additive exPlanations) processing may be used. In this case, based on a plurality of feature quantities P acquired from measurement data 59 and the performance evaluation model 59, the performance evaluation model 59 is locally approximated around the feature quantities P. The model (approximation model) generated by approximating the performance evaluation model 59 may be, for example, a simple model that easily explains the contribution of each feature quantity P. Thereafter, based on the approximation model, the contribution of each feature quantity P to the predicted value output from the approximation model is expressed by a so-called Shapley value used in cooperative game theory, etc. The contribution may be used as the importance for selecting a specific feature quantity Ps. In addition, various techniques in image processing for two-dimensional images, such as gradient processing, may be applied.

[0145] Specifically, in step S401, the CPU 3 determines the importance of each of the plurality of feature amounts P. In the following step S402, the CPU 3 selects one or more feature amounts P in descending order of importance as specific feature amounts Ps.

[0146] Here, the importance of each feature P may be scored as a dimensionless number between 0 and 1, and if the score is equal to or greater than a predetermined value (e.g., 0.5), it may be considered to be "highly important." Furthermore, if there are multiple feature Ps with scores thus defined that are equal to or greater than a predetermined value, only the top multiple feature Ps (e.g., three feature Ps) may be selected in descending order of importance.

[0147] Furthermore, when the quality of the product W is quantified into a dimensionless number between 0 and 2 as described above, a correlation coefficient between each feature amount P and the quality data 89 can be given in addition to the importance (see FIG. 16). In the case of this embodiment, such a correlation coefficient (hereinafter also referred to as a "first correlation coefficient") is calculated in advance and stored in the data memory of the SSD 9 as a first correlation coefficient list 791 shown in FIGS. 4 and 16.

[0148] In the following step S403, the CPU 3 stores the specific feature amount Ps selected in step S402 in the RAM 7, the SSD 9, etc. Thereafter, the control process returns from the flow in Fig. 10 to the flow in Fig. 5, and proceeds to step S4.

[0149] (6-4. Model loading process) In the model loading process, the CPU 3 loads two or more types (two in this embodiment) of machine learning models 691, 692 from the SSD 9 serving as a storage unit. Both of the two machine learning models 691, 692 associate each of a plurality of feature quantities P with the one or more manufacturing processes Q described above across one or more of the plurality of process groups G. As will be described in detail later, the manufacturing processes Q constituting each process group G are each quantified (digitized) so as to characterize the content of each manufacturing process Q.

[0150] Here, among the multiple process groups G, the Nth process group G is the last one to be performed in the order of progression. N and an (Nn)th process group G which is performed before the Nth process group. N-n The following explanation will be given using the following (n≧1): The former is an example of the "first process group", and the latter is an example of the "second process group".

[0151] -First machine learning model- The first machine learning model 691 is a model of a plurality of features P and an N-th process group G N and associates the Nth process group G N A plurality of manufacturing processes Q constituting the (Nn)th process group G N-n and a plurality of manufacturing processes Q that constitute the manufacturing process.

[0152] More specifically, the first machine learning model 691 is a model of a plurality of features P and an N-th process group G N a first partial model 6911 that associates a plurality of manufacturing processes Q that constitute the Nth process group G N The (N-1)th process group G N-1 a second partial model 6912 that associates a plurality of manufacturing processes Q that constitute the N-1th process group G 1N-1 The multiple manufacturing processes Q and the (N-2)th process group G N-2 The model is composed of N-1 partial models arranged in the order of the manufacturing progress of the product W, such as a third partial model 6913 that associates the first partial model with a plurality of manufacturing processes Q that constitute the first partial model.

[0153] For example, if you start from one feature P, that feature P has the Nth process group G N The manufacturing process Q that constitutes the Nth process group G is linked (see FIG. 14). N Each manufacturing process Q that constitutes the N-1 process group G N-1 By linking in this way, it is possible to trace back from the analysis system 101 located at the end of the manufacturing process to each manufacturing process Q that constitutes the first process group G1.

[0154] The first machine learning model 691 is generated in advance by supervised learning.

[0155] The first machine learning model 691 is trained using second training data. This second training data is composed of numerical data measured in correspondence with each manufacturing process Q for each process group G and a plurality of feature quantities P. The number of samples of the second training data is stored in the SDD 9 or the like.

[0156] Specifically, the first machine learning model 691 according to this embodiment is a model based on a decision tree algorithm. An example of a machine learning model based on a decision tree is an RF model. When an RF model is used, the contribution of each manufacturing process Q (so-called RF importance) acquired for each process group G can be used as the importance. The example in FIG. 14 can be interpreted as a model in which multiple RF models are nested.

[0157] In this way, the method for constructing the first machine learning model 691 in which multiple RF models are nested has already been systematized by the present inventors. Details of this systematization are disclosed in, for example, Japanese Patent Application Laid-Open No. 2023-57729.

[0158] The model described in the publication has multiple explanatory variables (Nth process group G N A process of generating partial models using a random forest, each of which is composed of a manufacturing process Q constituting the product and one of a plurality of objective variables (e.g., feature P), is performed for all of the plurality of objective variables, thereby generating a decision tree model 6911 composed of a plurality of partial models.

[0159] After that, after generating the first decision tree model 6911, the model is configured to generate a plurality of explanatory variables (Nth process group G N Another variable (the N-1th process group G N-1 The manufacturing process Q that constitutes the Nth process group G is used as a new explanatory variable, and multiple explanatory variables ( N When each of the manufacturing processes Q constituting the first decision tree model 691 is set as a new objective variable, another decision tree model 6912 is further generated by newly generating a plurality of partial models. In this way, the first machine learning model 691 in this embodiment is a model constructed by generating a plurality of decision tree models in a chain reaction.

[0160] Note that each decision tree model does not connect all variables. Instead, it connects those with high importance (for example, the top three) based on the impurity of the partial model. The impurity of a partial model represents the degree of error improvement before and after branching of each decision tree. The impurity may be, for example, the amount of reduction in the Gini coefficient (Gini impurity).

[0161] In the case of Figure 14, the Nth process group G N Among the manufacturing processes Q belonging to the above, the manufacturing process Q with the highest importance for one feature P is the NAth process Q N and NB process Q NB and NC process Q NC N-1 process group G N-1 Among the manufacturing processes Q belonging to N The manufacturing process Q that is most important to this is the (N-1)-I process Q. (N-1)-I and the (N-1)-J process Q (N-1)-J is.

[0162] Similarly, models in which multiple RF models are nested are generated in advance for the other feature quantities P. By performing this process for all feature quantities P, it is possible to generate a first machine learning model 691 in which each of the multiple feature quantities P is associated with one or more of the multiple manufacturing processes Q across one or more of the multiple process groups G, as shown in FIG.

[0163] By following the arrows in Fig. 14, paths (see Fig. 16) connecting the manufacturing processes Q can be formed. As a result, for example, when focusing on one specific feature amount Ps, manufacturing processes Q that have a high degree of importance with respect to that specific feature amount Ps can be extracted from each process group G. In this embodiment, each manufacturing process Q connected by such a path is included in a specific process group G, which will be described later.

[0164] When a feature value P is input, the first machine learning model 691 outputs, for each process group G, one or more manufacturing processes Q that are connected to the feature value P based on the RF model. This output is nothing other than the specific process group Gs described below. The output from the first machine learning model 691 corresponds to a series of manufacturing processes Q that are highly important to the feature value P input to the model 691.

[0165] Furthermore, when the quality of product W is quantified as a dimensionless number between 0 and 2 as described above, a correlation coefficient can be given between each manufacturing process Q and each feature quantity P, in addition to the importance (see Figure 18). Note that for manufacturing processes Q that can only be quantified as categorical data, the correlation coefficient is left undefined, as in the case of the third process Q3 in Figure 18.

[0166] In this embodiment, such correlation coefficients (hereinafter also referred to as "second correlation coefficients") are calculated in advance and stored in the data memory of the SSD 9 as a second correlation coefficient list 792 shown in FIGS.

[0167] Furthermore, although only shown in FIG. 4, a correlation coefficient can be given between each manufacturing process Q in one process group G and each manufacturing process Q in another process group G connected thereto.

[0168] In the case of this embodiment, such correlation coefficients (hereinafter also referred to as "third correlation coefficients") are calculated in advance and stored in the data memory of the SSD 9 as the third correlation coefficient list 793 shown in FIG. 4. The third correlation coefficient list 793 can be set for each process group G. For example, the third correlation coefficient list 793 according to this embodiment is made up of (N-1) lists, each of which is stored in the data memory of the SSD 9.

[0169] -Second machine learning model- As before, the second machine learning model 692 is a set of multiple feature quantities P and an N-th process group G N and associates the Nth process group G NA plurality of manufacturing processes Q constituting the (Nn)th process group G N-n and a plurality of manufacturing processes Q that constitute the manufacturing process.

[0170] In detail, the second machine learning model 692 is a Bayesian network in which each of the plurality of feature quantities P is a child node and each of the plurality of manufacturing processes Q is a parent node.

[0171] More specifically, the second machine learning model 692 has each of the plurality of feature quantities P as a child node, and a second process group G N-n One or more manufacturing processes Q that make up the Nth process group G are used as parent nodes, and the Nth process group G is used as the first process group. N This is a Bayesian network in which one or more manufacturing processes Q constituting the above are intermediate nodes interposed between the child node and the parent node.

[0172] This Bayesian network can be visualized as, for example, a directed graph structure as shown in Fig. 15. The second machine learning model 692 is generated in advance by unsupervised learning.

[0173] The second machine learning model 692 is trained using third training data. This third training data is composed of waveforms of current changes over time corresponding to each feature value P, which are measured in advance for each of the L pieces of sample data (measurement data for learning, not for inspection), and the contents (quantified numerical data) of each of the multiple manufacturing processes Q in each of the multiple process groups G.

[0174] Specifically, the second machine learning model 692 may be a model based on a non-decision tree algorithm. The second machine learning model 692 related to the non-decision tree may have a graph structure determined by so-called graph structured analysis (hereinafter referred to as "GSA").

[0175] GSA is a big data analysis method proposed by the inventors of the present application that combines probability theory (Bayesian estimation) and graph theory. In this embodiment, GSA is used to determine the graph structure. However, it is not necessary to use GSA to determine the graph structure. For details of GSA, please refer to JP 2021-111063 A.

[0176] In FIG. 15, boxes with text entered indicate nodes. Arrows connecting boxes indicate edges. In the case of FIG. 15, the Nth process group G N One or more manufacturing processes Q belonging to are connected to the feature P via one edge.

[0177] The probability distribution function for a Bayesian network is usually expressed by multiplying multiple conditional probabilities. Two nodes connected via an edge mean that the probability distribution function contains a conditional probability where one of the nodes is a condition and the other is a variable. This suggests that the two nodes connected via the edge have a relatively strong dependency compared to other nodes.

[0178] Also, in FIG. 15, the N-1 step group G N-1 One or more manufacturing processes Q belonging to the Nth process group G N By connecting edges in this way, the feature quantity P can be used to determine the Nth process group G. N After passing through one or more manufacturing processes Q belonging to the first process group G1, the process can be traced back to one or more manufacturing processes Q belonging to the first process group G2.

[0179] 15, paths connecting manufacturing processes Q can be formed. As a result, when focusing on one specific feature amount Ps, for example, manufacturing processes Q that are highly dependent on that specific feature amount Ps can be extracted from each process group G. In this embodiment, each manufacturing process Q connected by such a path is included in a specific process group G, which will be described later.

[0180] When a feature quantity P is input, the second machine learning model 692 outputs, for each process group G, one or more manufacturing processes Q that are connected to the feature quantity P via one or more edges. This output is nothing other than the specific process group Gs described below. The output from the second machine learning model 692 corresponds to a series of manufacturing processes Q that have a strong dependency on the feature quantity P input to the model 692.

[0181] Once the first and second machine learning models 691, 692 are loaded, the control process proceeds from step S4 to step S5 of FIG.

[0182] (6-5. Specific process estimation process) Fig. 11 is a flowchart showing the details of the specific process estimation process. When the control process proceeds to step S5 in Fig. 5, the CPU 3 executes the flow shown in Fig. 11 in order from step S501.

[0183] Hereinafter, one or more manufacturing processes Q among the multiple manufacturing processes Q that are estimated to contribute to quality determination will be referred to as a specific process Qs. At the same time, a series of one or more manufacturing processes Q selected for each multiple process group G from one or more manufacturing processes Q that make up each of the multiple process groups G and that are estimated to contribute to quality determination will be referred to as a specific process group Gs. The specific processes Qs and specific process groups Gs are illustrated in Figures 16A and 16B, which will be described later.

[0184] In the specific examples shown in Figures 2 and 4, the specific process Qs is presumed to have contributed to the determination of the rust prevention performance, N The specific process Qs is one or more manufacturing processes Q in the Nth process group G, which is presumed to have contributed to the determination of the rust-preventive performance. N The specific process group Gs is one or more manufacturing processes Q in the Nth process group G N The manufacturing process Q is highly important and / or highly dependent on the specific process Qs selected from the first process group G1 to the Nth process group G N It is a series of manufacturing processes Q selected from each of the above.

[0185] For example, as shown in FIG. 16A, the specific process Qs is the NC process Q NC In this case, the specific process group Gs is the NC process Q NC and the (N-1)-J process Q (N-1)J and (N-2)-M process Q (N-2)M It is a sequence that includes and.

[0186] More generally, the CPU 3 generates a first process group (Nth process group G N ) and the importance of each manufacturing process Q included in the first process group (Nth process group G N ) the second process group (N-1 process group G N-1 The specific process group Gs can be estimated by calculating the importance of each manufacturing process W included in the process group G and determining one or more manufacturing processes Q for each process group G based on each importance.

[0187] 16A shows an example based on the first machine learning model 691, but the specific process group Gs can also be given in the same way when based on the second machine learning model 692 (see FIG. 16B). In that case, a manufacturing process Q corresponding to a node connected by an edge may be selected for each process group G. Furthermore, similar to the specific process Ps, multiple specific process groups Gs may be output.

[0188] Returning to FIG. 11, in step S501, the CPU 3 reads a specific feature quantity Ps that has been selected in advance from among a plurality of feature quantities P in a feature quantity selection process.

[0189] In the following steps S502 to S504, the CPU 3 determines specific processes Qs and specific process groups Gs from among the multiple manufacturing processes Q. This determination is performed for each type of machine learning model 69 based on the specific feature amount Ps and two or more types of machine learning models 69.

[0190] Specifically, in step S502, the CPU 3 inputs the specific feature Ps to the first machine learning model 691. The first machine learning model 691 outputs a manufacturing process Q that has a high contribution to the specific feature Ps, that is, a specific process Qs, and a specific process group Gs.

[0191] In the following step S503, the CPU 3 inputs the specific feature Ps to the second machine learning model 692. The second machine learning model 692 outputs a manufacturing process Q having a strong dependency on the specific feature Ps, that is, a specific process Qs, and a specific process group Gs.

[0192] In the following step S504, the CPU 3 stores the specific process Qs and the specific feature Ps output for each model in the RAM 7 or the SSD 9. Thereafter, the control process returns from the flow in Fig. 10 to the flow in Fig. 5, and proceeds to step S5.

[0193] (6-6. Correlation analysis process) Fig. 12 is a flowchart showing the details of the correlation analysis process. When the control process proceeds to step S6 in Fig. 5, the CPU 3 executes the flow shown in Fig. 12 in order from step S601.

[0194] In this correlation analysis process, the CPU 3 notifies the breakdown of the specific process Qs and the specific process group Gs, and the control mode of the specific process group Gs to improve quality, based on the first, second, and third correlation coefficient lists 791, 792, and 793.

[0195] Specifically, the CPU 3 displays the breakdown of the specific processes Qs and the control modes of the specific process groups Gs for improving quality in order of the specific process groups Gs that contribute to the quality on the display 11. The processing related to the display order of the specific process groups Gs is performed when there are multiple specific process groups Gs.

[0196] Specifically, in step S601, the CPU 3 reads one or more specific processes Qs and specific feature quantities Ps associated with each specific process Qs.

[0197] In the following step S602, the CPU 3 refers to the second correlation coefficient list 792 to read out the correlation coefficient (second correlation coefficient) of the specific feature Ps corresponding to each specific process group Gs, particularly the specific process Ps that constitutes the process group Gs.

[0198] Also in step S602, the CPU 3 selects, for each specific process group Gs, specific feature quantities Ps whose absolute value of the second correlation coefficient is equal to or greater than a second predetermined value (e.g., 0.2). If there are no specific feature quantities Ps that satisfy this condition, the CPU 3 displays a message to that effect on the display 11.

[0199] In the next step S603, the CPU 3 generates a union of the specific feature amounts Ps selected for each specific process group Gs, and regards the specific feature amounts Ps included in the union as the feature amounts P to be controlled.

[0200] In the following step S604, the CPU 3 reads out the correlation coefficients (first correlation coefficients) of the quality data 89 corresponding to the specific feature quantities Ps constituting the union by referring to the first correlation coefficient list 791. This reading is performed for each specific process group Gs.

[0201] In the same step S604, the CPU 3 selects, from among the specific feature quantities Ps constituting the union, a specific feature quantity Ps whose absolute value of the correlation coefficient (first correlation coefficient) read immediately before is equal to or greater than a third predetermined value (e.g., 0.2). If there is no specific feature quantity Ps that satisfies this condition, the CPU 3 displays that fact on the display 11.

[0202] In the next step S605, the CPU 3 lists the specific feature amounts Ps selected in step S604 in association with the corresponding specific process groups Gs. As a result, one or more specific feature amounts Ps that are strongly correlated with the quality data 89 are listed for each specific process group Gs.

[0203] In the next step S606, the CPU 3 multiplies, for each specific process group Gs, the specific feature amount Ps listed in step S607 by the first correlation coefficient and the second correlation coefficient. Hereinafter, this multiplied value will be referred to as an "effect index."

[0204] In this example, the Nth process group G N This is an example of controlling a manufacturing process Q belonging to the N-1 process group G. N-1 When controlling a manufacturing process Q belonging to the process group G, the product of the first correlation coefficient and the second correlation coefficient is multiplied by the third correlation coefficient read from the third correlation coefficient list 793. Different effect indicators may be used depending on the process group G to be controlled.

[0205] For one specific process group Gs, an effect index is calculated for each specific feature amount Ps associated with the process. For example, if there are two specific feature amounts Ps associated with one specific process group Gs, two effect indexes are also calculated. Furthermore, if there are two specific process groups Gs based on the first machine learning model 691, an effect index is calculated for each of the two specific process groups Gs.

[0206] The larger the effect index, the larger the change in the quality data 89 when the corresponding specific process Qs is changed. By paying attention to the positive or negative sign of the multiplied value, it is possible to determine the control mode of the specific process Qs in order to improve the quality data 89.

[0207] As described above, the effect index is calculated for each specific process group Gs and for each specific feature amount Ps. Therefore, in step S607, CPU 3 adds up the effect indexes calculated for each specific feature amount Ps. A total value of the effect indexes is calculated for each specific process group Gs. This total value is calculated for each manufacturing process Q that constitutes the specific process group Gs.

[0208] Thereafter, in the following step S608, the CPU 3 displays the specific process groups Gs and their control modes on the display 11 in descending order of the absolute value of the effect index.

[0209] Note that there is a possibility that the estimation result of the specific process group Gs by the first machine learning model 691 and the estimation result of the specific process Gs by the second machine learning model 692 may differ from each other.

[0210] 12, the CPU 3 individually notifies the estimation results of the first and second machine learning models 691 and 692 and the number of samples used in generating the first and second machine learning models 691 and 692. This notification may be performed, for example, by displaying them on the display 11.

[0211] Here, the details of the control mode are set in advance for each manufacturing process Q of each process group G. Each manufacturing process Q may include one or more control factors, and when displaying the information such as in step S608, the adjustment direction of the control factors may be displayed.

[0212] As an example, the Nth process group G N Assume that a specific manufacturing process Q belonging to the above is an electrodeposition coating process. In this case, manufacturing process Q includes, as control factors, the electrical conductivity (μS / cm) during electrodeposition coating, the paint temperature, the voltage value during electrodeposition, the ion concentration of the electrodeposition solution (MEQ), the acid concentration of the electrodeposition solution, the distribution of the electrodeposition solution, the amount and type of solvent in the electrodeposition solution, the presence or absence of foreign matter in the electrodeposition solution, and the takt time.

[0213] The above-mentioned numerical data quantified corresponding to each manufacturing process Q may be the control target of these control factors. When the control process Q includes multiple control factors, the numerical data obtained by quantifying each control process Q may be numerical data corresponding to each control factor.

[0214] Another example is the Nth process group G NSuppose a specific manufacturing process Q belonging to the above is a drying process using a drying oven. In this case, manufacturing process Q includes the moisture content in the drying oven, the set temperature of the drying oven, the baking time in the drying oven, the capacity of the drying oven, etc. as control factors.

[0215] The RAM 7 serving as a storage unit stores in advance the association between increasing or decreasing each of the control factors described above and increasing or decreasing the numerical data corresponding to each manufacturing process Q. By increasing or decreasing one or more control factors based on the stored contents, it becomes possible to control each manufacturing process Q constituting the specific process group Gs in a direction that improves quality.

[0216] Each control factor constituting each manufacturing process Q may be regarded as an independent manufacturing process Q. For example, in an electrodeposition coating process, the electrical conductivity may be regarded as one manufacturing process Q, and the paint temperature may be regarded as another manufacturing process Q.

[0217] In other words, instead of directly estimating each manufacturing process Q that constitutes the specific process group Gs, the CPU 3 may estimate the control factors that constitute each manufacturing process Q based on the specific feature Ps and two or more types of machine learning models, by type of machine learning model.

[0218] (6-7. Display example) 19 and 20 are diagrams illustrating display screens Sc and Sc′ on the display 11. For example, when the processing shown in FIG. 11 is completed, the CPU 3 displays on the display 11 the specific feature values ​​Ps that contributed to the selection of the specific process groups Gs, the specific process groups Gs selected by the first machine learning model 691, and the specific process groups Gs selected by the second machine learning model 692. See boxed section C1 in FIG. 19 for the specific feature values ​​Ps, boxed section C2 for the specific process groups Gs related to the first machine learning model 691, and boxed section C3 for the specific process groups Gs related to the second machine learning model 692. In the illustrated example, there are two specific feature values ​​Ps, but only one corresponding specific process group Gs. In more general cases, multiple specific process groups Gs may be selected for each machine learning model or multiple specific feature values ​​Ps.

[0219] In the illustrated example, the specific process group Gs for the first machine learning model 691 is the Cth manufacturing process (3-C process Q) in the third process group G3. 3C ) and the Jth manufacturing process in the second process group G2 (2-J process Q 2J ) and the Mth manufacturing process in the first process group G1 (1-M process Q 1M In this example, "N=3" is set as shown in Fig. 1, Fig. 15, etc., and analysis by analysis system 101 is performed after third process group G3.

[0220] In the illustrated example, the specific process group Gs for the second machine learning model 692 is the C-th manufacturing process (the 3-C process Q) in the third process group G3. 3C ) and the Gth manufacturing process in the second process group G2 (2-G process Q 2G ) and is composed of.

[0221] Next, the CPU 3 generates a union of the specific process group Gs associated with the first machine learning model 691 and the specific process group Gs associated with the second machine learning model 692 (see boxed part C4). As shown in Fig. 20 described below, the CPU 3 determines whether the union is an empty set.

[0222] The CPU 3 also calculates the above-mentioned effect index for each of the manufacturing processes Q that make up the specific process group Gs corresponding to the first machine learning model 691.

[0223] In the example shown, the 3-C process Q that constitutes one specific process group Gs 3C , 2nd-G process Q 2G and 1-M Process Q 1M For each of the specific process groups Gs, an effect index is calculated for each specific feature amount Ps that is the end of the specific process group Gs. If there are multiple specific process groups Gs, an effect index is further calculated for each specific process group Gs.

[0224] As shown in the figure, when one specific process group Gs is specified, the 3-C process Q that constitutes the specific process group Gs 3C For the second-G step Q, the first correlation coefficient and the second correlation coefficient are multiplied for each specific feature Ps. 2G Regarding the 3-C process Q 3C The multiplied value calculated for the second step Q 2G and 3-C Process Q 3C The correlation coefficient between the first and second steps (third correlation coefficient) is further multiplied. 1M Regarding the 2nd-G process Q 2G The multiplied value calculated for the 1st-Mth step Q 1M and 2nd-G process Q 2G Then, the CPU 3 adds up the calculated effect indexes for each specific process group Gs and for each manufacturing process Q.

[0225] Thereafter, the CPU 3 displays on the display 11, in descending order of the calculated effect index, each manufacturing process Q constituting each specific process group Gs and its control mode (e.g., "increase," "decrease") (see boxed section C5). Here, if each manufacturing process Q includes multiple control factors, a display may be made for each control factor (e.g., a display indicating "increase paint temperature"). Furthermore, if the specific feature quantity Ps at the end of each specific process Gs is, for example, a categorical value and it is not possible to set a correlation coefficient, and therefore "increase," "decrease," etc., then instead of displaying "increase," "decrease," etc., "estimation not possible" may be displayed (not shown).

[0226] Furthermore, as shown in boxed areas C12 and C13 in Figure 20, if the specific process Qs differs between the two models 691, 692, a message indicating this (e.g., "Mismatch") may be displayed on screen Sc', and the number of samples for each model 691, 692 may also be displayed (see boxed area C14).

[0227] 7. Significance of the analysis method As described above, the analysis method according to the embodiment uses two or more machine learning models 691, 692 that associate each of a plurality of feature quantities P with one or more of a plurality of manufacturing processes Q across one or more of a plurality of process groups G (see FIGS. 14 and 15). Here, consider the appearance of a product W as an example of quality.

[0228] In this example, once the feature value P estimated to have contributed to the deterioration of appearance, such as the presence or absence of a scratch and its location, is determined, it is possible to use the machine learning model 69 to estimate a specific manufacturing process Q in one process group G that may have caused the scratch, and a specific manufacturing process Q in another process group G connected to that process group G. The estimation using the machine learning model 69 can be performed easily and with high accuracy, even by non-expert workers.

[0229] Furthermore, in the case of a product W consisting of many parts, such as a car body, it is believed that there are an infinite number of feature values ​​P that can be used as candidates for analysis. In such cases, even if a skilled worker is involved, it is not easy to identify the process that is responsible for the quality. Estimation using the machine learning model 69 can be performed easily and with high accuracy, even when there are many feature values.

[0230] Furthermore, the specific process group Gs is estimated not only by using the machine learning model 69 but also by using each of two or more types of machine learning models 691 and 692. This allows for more multifaceted analysis and more accurate analysis.

[0231] 14, even when multiple process groups G are connected, it is possible to estimate a specific manufacturing process Q in one process group G and a specific manufacturing process Q in another process group G connected to the first process group G. This makes it possible to realize an easy and highly accurate analysis.

[0232] Furthermore, as illustrated in Figures 14 and 15, the two or more types of machine learning models 69 include a first machine learning model 691 based on a decision tree algorithm and a second machine learning model 692 based on a non-decision tree algorithm.

[0233] In this way, by using models constructed from different perspectives, a more multifaceted analysis can be performed, which makes it possible to achieve a more accurate analysis.

[0234] Furthermore, the importance of each manufacturing process Q can be calculated automatically for each process group G according to the selection of the machine learning model 69. For example, in the case of a random forest model as in this embodiment, it is sufficient to calculate the importance specific to that model. In this way, it is possible to reduce the possibility of the user's intervention in determining the specific process group Gs. This allows for more accurate analysis.

[0235] Furthermore, the specific process group Gs can be determined based on a chain of decision tree models, as shown in Fig. 14. Since the importance can be determined for each model, the specific process group Gs can be easily estimated.

[0236] 15 is used to select the second machine learning model 692, the manufacturing process Q is determined based on probabilistic connections (dependencies). This results in a more multifaceted analysis, making it possible to achieve a more accurate analysis.

[0237] Furthermore, even if the estimation results from two types of machine learning models 69 differ from each other, one estimation result is not overwritten or discarded by the other, and the estimation results when using each machine learning model 69 are notified to the user. At that time, by notifying the number of samples as well, it becomes possible to perform a more comprehensive analysis, such as an analysis of the learning status of each model.

[0238] 12, the specific process group Gs is not simply determined, but a quality improvement plan is also proposed, which makes it possible for even an unskilled worker to easily improve the quality of the product W.

[0239] 12, when proposing a quality improvement plan, a highly visible display can be provided, which makes it possible to easily improve the quality of the product W.

[0240] Furthermore, the quality determination results used in the present disclosure are not limited to the output from the machine learning model 69. By having a worker perform the quality determination and then having the computer 1 perform the subsequent processes as described above, it is possible to achieve division of labor at the manufacturing site. This can improve the ease of use of the analysis method.

[0241] Furthermore, the feature P used in the present disclosure is not limited to electrical data. For example, the analysis method according to the present disclosure can be executed based on text data written by a worker (e.g., text describing the condition of a scratch, etc.). This can improve the usability of the analysis method.

[0242] 5, the specific feature Ps is determined by the performance evaluation model 59, separately from the machine learning model 69 for determining the manufacturing process Q. This reduces the possibility of the user's intervention in the analysis of the manufacturing process Q. This allows for a more accurate analysis to be achieved.

[0243] As described with reference to FIG. 2 etc., the present disclosure is particularly effective for analyzing the rust prevention performance of the metal product 201.

[0244] <8. Other embodiments> In the above embodiment, the output data (quality data 89) of the performance determining program 292 is used as the quality determination result, but the present disclosure is not limited to such a configuration.

[0245] That is, in the flow shown in FIG. 5, at least the measurement data analysis process (step S1) and the performance evaluation process (step S2) may be omitted.

[0246] The "quality judgment result" in the present disclosure may be, instead of or in addition to the output from a pre-generated machine learning model, a judgment result by a factory worker or a judgment result by rule-based judgment using multiple feature quantities P as input. For example, when using the judgment result by a factory worker, as described above, the configuration of machine learning model 69 can be reconfigured so that the text data recorded by the factory worker is used as input.

[0247] The feature value P used in this process may be, for example, the index (color, shape, texture, etc. of the product) used by a worker when he or she judges the quality based on experience. By quantifying these indexes, the process similar to that of the above embodiment can be performed.

[0248] Furthermore, although the above embodiment has shown an example in which the analysis device is configured by one computer 1, the present disclosure is not limited to this example. The analysis method and analysis program 29 according to the present disclosure may be executed using multiple computers 1, such as by having a first computer execute processing related to the specific process estimation process and a second computer execute processing related to the correlation analysis process. Furthermore, the computer 1 in the present disclosure also includes parallel computers such as supercomputers and PC clusters.

[0249] Furthermore, the screen on which various information can be displayed is not limited to the display screen on the display 11 of the computer 1. The graph structure may be displayed on a screen prepared separately from the computer 1. [Explanation of symbols]

[0250] 1. Computer 3 CPU (arithmetic unit) 7 RAM (memory section) 9 SSD (storage unit) 11 Display (display unit) 15 Keyboard (reception area) 17 Mouse (Reception) 18 Storage medium 29 Analysis Program 291 Measurement Data Analysis Program 292 Performance Evaluation Program 293 Feature Selection Program 294 Model Loading Program 295 Specific Process Group Estimation Program 296 Correlation Analysis Program 49 Measurement Data 59 Performance Evaluation Model 69 Machine Learning Models 691 First Machine Learning Model 692 Second Machine Learning Model 89 Quality Data P feature Ps specific feature Q Manufacturing process G process group Gs Specific process group W Products 201 Metal products 203 Insulating layer

Claims

1. A manufacturing process analysis method executed using a computer including a storage unit and a calculation unit, and a quality assessment result of a product manufactured through a plurality of process groups each consisting of one or more manufacturing processes, comprising: The plurality of process groups are connected in order of progression of the manufacturing stages of the product, Among the plurality of feature amounts acquired during the assessment, one or more feature amounts estimated to contribute to the assessment of quality are defined as specific feature amounts, and a series of one or more manufacturing processes selected for each process group from one or more manufacturing processes constituting each of the plurality of process groups and estimated to contribute to the assessment of quality are defined as specific process groups: the calculation unit reads from the storage unit two or more types of machine learning models that have been generated in advance so as to associate each of the plurality of feature quantities with one or more of the plurality of manufacturing processes across one or more of the plurality of process groups; The calculation unit estimates the specific process group for each type of machine learning model based on the specific feature amount and the two or more types of machine learning models. A manufacturing process analysis method comprising:

2. 2. The manufacturing process analysis method according to claim 1, Among the plurality of process groups, one process group performed at the end of the progression order is designated as a first process group, and another process group performed before the first process group is designated as a second process group. The two or more machine learning models each include: a machine learning model that is generated in advance to associate the plurality of feature amounts with one or more manufacturing processes that constitute the first process group, and to associate each manufacturing process that constitutes the first process group with one or more manufacturing processes that constitute the second process group, The calculation unit estimates the specific process group for each type of machine learning model based on the specific feature amount and the two or more types of machine learning models. A manufacturing process analysis method comprising:

3. 3. The manufacturing process analysis method according to claim 2, The two or more machine learning models are a first machine learning model based on a decision tree algorithm; a second machine learning model based on a non-decision tree algorithm. A manufacturing process analysis method comprising:

4. 4. The manufacturing process analysis method according to claim 3, The calculation unit calculates the importance of each manufacturing process included in the first process group that corresponds to each of the plurality of feature quantities in the first machine learning model, and the importance of each manufacturing process included in the second process group that is associated with each manufacturing process in the first process group, and determines the one or more manufacturing processes for each process group based on the importance, thereby estimating the specific process group. A manufacturing process analysis method comprising:

5. 4. The manufacturing process analysis method according to claim 3, The first machine learning model is: A process of generating partial models using a random forest, each of which is composed of a plurality of explanatory variables and one of a plurality of objective variables, is performed for all of the plurality of objective variables to generate a decision tree model composed of the plurality of partial models; and another decision tree model constructed by newly generating the plurality of partial models when another variable linked to the plurality of explanatory variables is set as a new explanatory variable and each of the plurality of explanatory variables is set as a new objective variable. A manufacturing process analysis method comprising:

6. 4. The manufacturing process analysis method according to claim 3, the second machine learning model is a Bayesian network in which each of the plurality of feature quantities is a child node, one or more manufacturing processes constituting the second process group is a parent node, and one or more manufacturing processes constituting the first process group is an intermediate node interposed between the child node and the parent node; When the Bayesian network is visualized as a directed graph structure, the calculation unit includes, in the specified process group, a manufacturing process corresponding to the intermediate node connected to a child node corresponding to the specific feature via one or more edges, and a manufacturing process corresponding to the parent node connected to the intermediate node via one or more edges. A manufacturing process analysis method comprising:

7. 3. The manufacturing process analysis method according to claim 2, When the determination result by the first machine learning model and the determination result by the second machine learning model differ, Estimation results of the first and second machine learning models; and and the number of samples used in generating the first and second machine learning models, respectively. A manufacturing process analysis method comprising:

8. 3. The manufacturing process analysis method according to claim 2, The quality judgment result is quantified as quality data that increases or decreases depending on whether the quality is good or bad, The manufacturing processes constituting the first and second process groups are quantified so as to characterize the contents of each manufacturing process, The storage unit a first correlation indicating a correlation between each of the plurality of feature amounts and the quality data; a second correlation indicating a correlation between each manufacturing process in the first process group and each of the plurality of feature amounts; a third correlation indicating a correlation between each manufacturing process in the second process group and each manufacturing process in the first process group is stored in advance; After determining the specific process group, the calculation unit calculates, based on the first, second, and third correlations, A breakdown of the specific step group; and notifying the specific process group of control modes for improving the quality. A manufacturing process analysis method comprising:

9. 9. The manufacturing process analysis method according to claim 8, The calculation unit includes a display unit capable of displaying the calculation results of the calculation unit. A breakdown of the specific steps determined for each step group; and displaying improvement methods for the specific process for improving the quality in order of contribution to the quality. A manufacturing process analysis method comprising:

10. 2. The manufacturing process analysis method according to claim 1, The calculation unit reads, as the quality judgment result, a judgment result by a worker, or a judgment result by a rule-based judgment using the plurality of feature amounts as input, or an output from a pre-generated machine learning model. A manufacturing process analysis method comprising:

11. 2. The manufacturing process analysis method according to claim 1, The calculation unit acquires, as the plurality of feature quantities, at least one of image data, audio data, text data, and mechanical or electrical data that characterize the product. A manufacturing process analysis method comprising:

12. 2. The manufacturing process analysis method according to claim 1, The product is a metal product having an insulating layer on a surface of the product, The quality of the product is the rust prevention performance of the insulating layer, The calculation unit A voltage is applied to the surface of the metal product with a corrosion factor in contact with the surface, and a change over time in the current caused by the voltage is obtained; As the plurality of feature quantities, measurement values ​​that characterize the waveform of the change over time are acquired. A manufacturing process analysis method comprising:

13. A manufacturing process analysis device that uses a quality assessment result of a product manufactured through a plurality of process groups, each of which is made up of one or more manufacturing processes, and is configured by a computer having a storage unit and a calculation unit, comprising: The plurality of process groups are connected in order of progression of the manufacturing stages of the product, Among the plurality of feature amounts acquired during the assessment, one or more feature amounts estimated to contribute to the assessment of quality are defined as specific feature amounts, and a series of one or more manufacturing processes selected for each process group from one or more manufacturing processes constituting each of the plurality of process groups and estimated to contribute to the assessment of quality are defined as specific process groups: a model reading means for reading from the storage unit two or more types of machine learning models that have been generated in advance so as to associate each of the plurality of feature quantities with one or more of the plurality of manufacturing processes across one or more of the plurality of process groups; and a specific process group estimation means for estimating the specific process group for each type of machine learning model based on the specific feature amount and the two or more types of machine learning models. A manufacturing process analysis device characterized by:

14. A manufacturing process analysis program executed using a computer having a storage unit and a calculation unit, and a quality assessment result of a product manufactured through a plurality of process groups each consisting of one or more manufacturing processes, The plurality of process groups are connected in order of progression of the manufacturing stages of the product, Among the plurality of feature amounts acquired during the assessment, one or more feature amounts estimated to contribute to the assessment of quality are defined as specific feature amounts, and a series of one or more manufacturing processes selected for each process group from one or more manufacturing processes constituting each of the plurality of process groups and estimated to contribute to the assessment of quality are defined as specific process groups: The computer, the calculation unit reading from the storage unit two or more types of machine learning models that have been generated in advance so as to associate each of the plurality of feature quantities with one or more of the plurality of manufacturing processes across one or more of the plurality of process groups; the calculation unit estimates the specific process group for each type of machine learning model based on the specific feature amount and the two or more types of machine learning models. A manufacturing process analysis program characterized by:

15. The analysis program according to claim 14 is stored. A computer-readable storage medium comprising:

Images