Method for manufacturing using method for supporting mixing design

Abstract

To provide a method for manufacturing using a method for supporting a mixing design which can adjust the mixing ratio of materials while grasping the whole image of mixing.SOLUTION: A method for manufacturing using a method for supporting a mixing design executed by a mixing design support device for an operator to adjust the mixing ratio of materials includes: a mixing display control step of displaying a mixing represented by a combination of drawings having an area according to the mixing ratio in a mixing adjustment screen; a mixing rate adjustment reception step of receiving adjustment of the mixing ratio by a variable operation that the operator does on the area of the drawings; a performance prediction value display control step of displaying the performance prediction value of the mixing; and a manufacturing step of manufacturing a product by using the mixing ratio of which adjustment was received.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present disclosure relates to a manufacturing method using a blend design support method. [Background technology]

[0002] In recent years, by constructing a machine learning model that predicts properties from a compounding formula using machine learning or statistical analysis, a graph showing the change in the predicted property value according to the change in the compounding formula has been displayed on a computer. For example, Patent Document 1 discloses a method for creating a graph depicting the predicted property value of a material as a technology for custom designing a product. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Special Publication No. 2020-535566 Summary of the Invention [Problem to be solved by the invention]

[0004] In a mix design support device that allows an operator to adjust the mix ratio of materials, it is necessary to adjust the mix ratio of materials (explanatory variables) to predict performance (objective variables) such as physical properties. Conventional mix design support devices have a problem in that it is not easy to adjust the mix ratio of materials while grasping the overall picture of the mix. Patent Document 1 does not solve the above problem.

[0005] The present disclosure aims to provide a manufacturing method using a formulation design support method that allows adjustment of the material formulation ratio while grasping the overall picture of the formulation. [Means for solving the problem]

[0006] The present disclosure has the following configuration.

[0007] [1] A mix design support device that allows an operator to adjust the mix ratio of materials, A combination display control unit configured to display a combination represented by a combination of figures having an area according to the combination ratio on a combination adjustment screen; a mixture ratio adjustment receiving unit configured to receive an adjustment of the mixture ratio based on a variable operation on the area of ​​the figure by the operator; A performance prediction value display control unit configured to display a performance prediction value of the composition; A mix design support device having the above.

[0008] [2] At least some of the materials are grouped; The mixture ratio adjustment receiving unit is configured to receive an adjustment of the ratio of the mixture ratios of the plurality of materials set in the same group while keeping the sum of the mixture ratios of the plurality of materials set in the same group fixed. [1] A mix design support device as described above.

[0009] [3] At least some of the materials are grouped; The mixture ratio adjustment receiving unit is configured to receive an adjustment to a sum of the mixture ratios of the materials set in the same group while keeping the ratio of the mixture ratios of the materials set in the same group fixed. The mix design support device according to [1] or [2].

[0010] [4] A correlation detection unit configured to detect a correlation of the blending ratio from data used in machine learning of a machine learning model that predicts performance from the blending; and The mixture ratio adjustment receiving unit is configured not to receive an adjustment of the mixture ratio that does not comply with the correlation. [1] A mix design support device as described above.

[0011] [5] A group setting unit configured to set a group based on the correlation for at least a portion of the material; The mix design support device according to [4], further comprising:

[0012] [6] A correlation detection unit configured to detect a correlation of the blending ratio from data used in machine learning of a machine learning model that predicts performance from the blending; Further comprising: The mixture ratio adjustment receiving unit is configured not to receive an adjustment of the mixture ratio that does not comply with the correlation. A mix design support device according to [2] or [3].

[0013] [7] A group setting unit configured to set a group based on the correlation for at least a portion of the material. The mix design support device according to [6], further comprising:

[0014] [8] At least some of the materials are grouped; The combination display control unit is configured to display a combination graph in a tree map format on the combination adjustment screen, which accepts adjustment of the mixture ratio of the materials, and accepts adjustment of the sum of the mixture ratios of the materials set in the same group by a variable operation in a first direction with respect to the area of ​​the figure while keeping the ratio of the materials set in the same group fixed, and accepts adjustment of the ratio of the mixture ratios of the materials set in the same group by a variable operation in a second direction with respect to the area of ​​the figure while keeping the sum of the mixture ratios of the materials set in the same group fixed. A mix design support device according to any one of [1] to [7].

[0015] [9] At least some of the materials are grouped; The combination display control unit is configured to display a combination graph in a sunburst chart format on the combination adjustment screen, accept adjustment of the sum of the combination rates of the materials set in the same group while keeping the ratio of the combination rates of the materials set in the same group fixed by a variable operation on the area of ​​the figure obtained by dividing a first circle, and accept adjustment of the ratio of the combination rates of the materials set in the same group while keeping the sum of the combination rates of the materials set in the same group fixed by a variable operation on the area of ​​the figure obtained by dividing a second circle. A mix design support device according to any one of [1] to [7].

[0016]

[10] A mix design support device that allows an operator to adjust the mix ratio of materials, A combination display control unit that displays a combination represented by a combination of figures having an area corresponding to the combination ratio on a combination adjustment screen; a mixture ratio adjustment receiving unit that receives an adjustment of the mixture ratio by a varying operation for the area of ​​the figure from the operator; A performance prediction value display control unit that displays the performance prediction value of the formulation; A program to function as a

[0017]

[11] A mix design support method implemented by a mix design support device that allows an operator to adjust the mix ratio of materials, comprising: A combination display control step of displaying a combination represented by a combination of figures having an area according to the combination ratio on a combination adjustment screen; a mixture ratio adjustment receiving step of receiving an adjustment of the mixture ratio from the operator by a variable operation on the area of ​​the figure; A performance prediction value display control step for displaying a performance prediction value of the blend; A method for supporting mix design. Effect of the Invention

[0018] According to the present disclosure, it is possible to provide a manufacturing method that uses a blending design support method for adjusting the blending ratio of materials while grasping the overall picture of the blending. [Brief description of the drawings]

[0019] [Figure 1] FIG. 1 is a configuration diagram of an example of a mix design support system according to this embodiment. [Diagram 2] FIG. 2 is a diagram illustrating an example of a hardware configuration of a computer according to the present embodiment. [Diagram 3] FIG. 1 is a functional configuration diagram of an example of a mix design support system according to an embodiment of the present invention. [Figure 4] FIG. 13 is an image diagram of an example of a blend adjustment screen according to the present embodiment. [Diagram 5] FIG. 13 is an explanatory diagram of an example of a variable operation for the area of ​​a mixture ratio figure. [Figure 6] FIG. 13 is an explanatory diagram of an example of a variable operation for the area of ​​a mixture ratio figure of a constant sum relationship. [Figure 7] FIG. 13 is an explanatory diagram of an example of a variable operation for the area of ​​a mixture ratio figure of a constant ratio relationship. [Figure 8] FIG. 13 is a diagram illustrating an example of adjustment of the mixture ratio when the mixture ratio has a constant sum relationship. [Figure 9] FIG. 13 is a diagram illustrating an example of adjustment of the mixture ratio when the mixture ratio has a constant ratio relationship. [Figure 10] FIG. 2 is an image diagram of an example of a combination graph according to the present embodiment. [Figure 11] FIG. 1 is a diagram for explaining an example of correlation formulation using the least absolute value method, RANSAC, Theil-Sen regression, or the like. [Figure 12] 1 is a flowchart of an example of processing of a mix design support system according to this embodiment. [Figure 13] FIG. 13 is an image diagram of an example of a formulation adjustment column on a formulation adjustment screen according to this embodiment. [Figure 14] FIG. 13 is a diagram showing an example of a profile obtained by plotting changes in predicted performance values ​​against changes in material blend ratios. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] Next, an embodiment of the present invention will be described in detail. Note that the present invention is not limited to the following embodiment. In this embodiment, the mixture ratio is the ratio of the mixture set for each material, for example, "material 1 is 30%". In this embodiment, the mixture is a set of multiple mixture ratios, the sum of which is 100%, for example, "material 1 is 30%, material 2 is 50%, and material 3 is 20%".

[0021] [First embodiment] <System configuration> Fig. 1 is a configuration diagram of an example of a mix design support system according to this embodiment. The mix design support system 1 in Fig. 1 has a mix design support device 10 and a user terminal 12. The mix design support device 10 and the user terminal 12 are connected to each other so that data can be communicated via a communication network 18 such as a local area network (LAN) or the Internet.

[0022] The user terminal 12 is an information processing device such as a PC, tablet terminal, or smartphone operated by an operator. The user terminal 12 displays a mix adjustment screen and accepts operations required to adjust the mix ratio of materials from the operator. The user terminal 12 transmits the contents of the operations accepted from the operator to the mix design support device 10. The user terminal 12 also displays performance prediction values ​​such as characteristics predicted from the mix by the mix design support device 10 on the mix adjustment screen for the operator to confirm.

[0023] The mix design support device 10 is an information processing device such as a PC or a workstation that executes a process of predicting the performance of a product from a mix based on an operation of adjusting the mix ratio of materials received from an operator and displaying it on a mix adjustment screen. The mix design support device 10 predicts the performance (objective variable) of a product generated by the mix from the mix ratio of materials (explanatory variable) using, for example, a machine learning model that machine-learns learning data. Hereinafter, the performance prediction value of a product generated by a mix may be referred to as the mix performance prediction value. The mix design support device 10 executes a process for displaying the predicted mix performance prediction value on the mix adjustment screen, such as sending the predicted mix performance prediction value to the user terminal 12.

[0024] In addition, image data for displaying the mix adjustment screen may be generated by the mix design support device 10 and transmitted to the user terminal 12, or data required to generate the mix adjustment screen may be transmitted from the mix design support device 10 and generated by the user terminal 12.

[0025] In the mix design support system 1 of this embodiment, a mix adjustment screen as described below is displayed to assist the worker in mix design, thereby helping the worker to easily adjust the mix ratios of materials while grasping the overall picture of the mix.

[0026] It goes without saying that the mix design support system 1 in Fig. 1 is just one example, and there are various system configurations depending on the application and purpose. For example, the mix design support device 10 may be realized by multiple computers, or may be realized as a cloud computing service. Also, for example, the mix design support system 1 in Fig. 1 may be realized by a stand-alone computer.

[0027] <Hardware configuration> The mix design support device 10 and the user terminal 12 in FIG. 1 are realized by, for example, a computer 500 having a hardware configuration shown in FIG.

[0028] Fig. 2 is a hardware configuration diagram of an example of a computer according to this embodiment. The computer 500 in Fig. 2 includes an input device 501, a display device 502, an external I / F 503, a RAM 504, a ROM 505, a CPU 506, a communication I / F 507, and a HDD 508, which are all connected to each other via a bus B. The input device 501 and the display device 502 may be connected to each other for use.

[0029] The input device 501 is a touch panel, operation keys or buttons, a keyboard, a mouse, etc., used by an operator to input various signals. The display device 502 is composed of a display such as a liquid crystal or organic electroluminescence display for displaying a screen, a speaker for outputting sound data such as voice and sound, etc. The communication I / F 507 is an interface for the computer 500 to perform data communication.

[0030] The HDD 508 is an example of a non-volatile storage device that stores programs and data. The stored programs and data include an OS, which is basic software that controls the entire computer 500, and applications that provide various functions on the OS. Note that the computer 500 may use a drive device (such as a solid-state drive (SSD)) that uses a flash memory as a storage medium instead of the HDD 508.

[0031] The external I / F 503 is an interface with an external device. The external device may include a recording medium 503a. This allows the computer 500 to read and / or write data from and to the recording medium 503a via the external I / F 503. The recording medium 503a may include a flexible disk, a CD, a DVD, an SD memory card, a USB memory, and the like.

[0032] The ROM 505 is an example of a non-volatile semiconductor memory (storage device) that can retain programs and data even when the power is turned off. The ROM 505 stores programs and data such as BIOS, OS settings, and network settings that are executed when the computer 500 is started. The RAM 504 is an example of a volatile semiconductor memory (storage device) that temporarily retains programs and data.

[0033] The CPU 506 is a calculation device that reads out programs and data from storage devices such as the ROM 505 and the HDD 508 onto the RAM 504 and executes processing to realize the overall control and functions of the computer 500. The computer 500 according to this embodiment can realize various functions of the mix design support device 10 and the user terminal 12, which will be described later, by executing programs. The CPU 506 may read the program from a recording medium 503a in which the program is stored via the external I / F 503 and execute the program.

[0034] <Functional configuration> The configuration of the mix design support system 1 according to this embodiment will be described. FIG. 3 is a functional block diagram of an example of the mix design support system according to this embodiment. Note that the block diagram of FIG. 3 omits parts that are not necessary for the description of this embodiment. Note that, in the following description, an example is described in which image data of the mix adjustment screen is generated by the mix design support device 10, but it may also be generated by a program such as an application installed in the user terminal 12.

[0035] The mix design support device 10 of the mix design support system 1 shown in Figure 3 has a request receiving unit 20, a response sending unit 22, a mix adjustment screen display control unit 24, a mix display control unit 26, a performance prediction value display control unit 28, a mix ratio adjustment receiving unit 30, a performance prediction unit 32, a correlation detection unit 34, a group setting unit 36, a learning data storage unit 40, a machine learning model storage unit 42, and a group storage unit 44. The user terminal 12 has an information display unit 50, an operation receiving unit 52, a request sending unit 54, and a response receiving unit 56.

[0036] The information display unit 50 of the user terminal 12 displays a mix adjustment screen, described below, on the display device 502 to support the mix design by the operator. The operation reception unit 52 receives various operations, such as operations on the mix adjustment screen, from the operator. The request transmission unit 54 transmits a processing request to the mix design support device 10 based on various operations, such as operations on the mix adjustment screen. The response reception unit 56 receives a response to the processing request transmitted by the request transmission unit 54 from the mix design support device 10.

[0037] The request receiving unit 20 of the mix design support device 10 receives a processing request based on an operation by an operator. The response sending unit 22 sends a response to the processing request received by the request receiving unit 20 to the user terminal 12.

[0038] The mixture adjustment screen display control unit 24 generates image data of the mixture adjustment screen described below, and controls the display on the user terminal 12. The mixture display control unit 26 controls the display of a mixture graph described below on the mixture adjustment screen. The performance prediction value display control unit 28 controls the display of the performance prediction value of the mixture adjusted by the operator on the mixture adjustment screen.

[0039] The mixture ratio adjustment receiving unit 30 receives an adjustment of the mixture ratio of materials from the operator based on the operator's operation on the mixture adjustment screen received at the user terminal 12. The performance prediction unit 32 performs machine learning on the learning data, and predicts a predicted performance value of the mixture using a machine learning model (hereinafter simply referred to as a machine learning model) that predicts the performance of a product from the mixture. For machine learning, for example, supervised learning can be used. The machine learning model that predicts the performance of a product from the mixture learns the relationship between the mixture and the performance of the product using learning data that corresponds to the mixture and the performance of the product generated by the mixture, and predicts the predicted performance value of the mixture according to that relationship.

[0040] The correlation detection unit 34 detects the correlation of the mixture ratio from the learning data used for the machine learning of the machine learning model. The correlation of the mixture ratio includes a relationship of a constant sum or a constant ratio of the mixture ratios present in the learning data. The constant sum of the mixture ratios represents a fixed relationship in which the sum of the mixture ratios of multiple materials set in the same group is fixed, as described later. Furthermore, the constant ratio of the mixture ratios represents a fixed relationship in which the ratio of the mixture ratios of multiple materials set in the same group is fixed.

[0041] The group setting unit 36 ​​sets groups based on the correlation of the mixture ratios. The group setting may be performed manually by an operator based on the intended functions of the filler and adhesive, or may be performed automatically based on the correlation of the mixture ratios detected by the correlation detection unit 34, or may be performed so as to inherit a recipe in which past mixtures are recorded (past recipe).

[0042] The learning data storage unit 40 stores learning data used in machine learning. The machine learning model storage unit 42 stores machine learning models. The group storage unit 44 stores set groups.

[0043] <Bixture adjustment screen> FIG. 4 is an image diagram of an example of a mixture adjustment screen according to this embodiment. The mixture adjustment screen 1000 in FIG. 4 includes a mixture adjustment field 1002 and a performance prediction value display field 1004. The mixture adjustment field 1002 includes a mixture graph 1006. The mixture graph 1006 is an example of a display that accepts adjustment of the mixture ratio of materials in a tree map format. The mixture graph 1006 represents the mixture by a combination of figures (hereinafter referred to as mixture ratio figures) having an area according to the mixture ratio. The area according to the mixture ratio is wider as the mixture ratio is larger, and narrower as the mixture ratio is smaller. The total of the mixture ratios of materials 1, 2, and 4 to 8 represented by the mixture graph 1006 is 100%. The mixture graph 1006 represents the overall image of the mixture. The mixture graph 1006 may be displayed by visually distinguishing each mixture ratio figure by color coding or hatching. When the mixture of materials is represented by the mixture amount, the mixture graph 1006 may be represented by a Marimekko chart (also called a mosaic plot).

[0044] Mark 1010 displayed on combination graph 1006 indicates that the mixture ratios of "material 4" and "material 5" are in a constant sum relationship. Mark 1012 displayed on combination graph 1006 indicates that the mixture ratios of "material 6", "material 7", and "material 8" are in a constant ratio relationship. For mixture ratios that are in a constant sum relationship, a check box is automatically placed in the check box for the set group. For mixture ratios that are in a constant ratio relationship, a check box is automatically placed in the check box for materials set in the same group.

[0045] Even if the mixture ratio of the ingredients is not a constant sum or constant ratio relationship, the mixture ratio of the ingredients that does not accept the operation to vary the area of ​​the mixture ratio figure can be specified by checking the check box.

[0046] The performance prediction value display field 1004 displays the performance prediction value of the formulation (current formulation) represented by the formulation graph 1006. "Comparison 1" and "Comparison 2" displayed in the performance prediction value display field 1004 of Fig. 4 are performance prediction values ​​of formulations different from the current formulation that have been saved as comparative examples by pressing the "Add to comparison" button 1008 in the formulation adjustment field 1002. In the performance prediction value display field 1004 of Fig. 4, the performance prediction values ​​of the current formulation, the formulation of "Comparison 1", and the formulation of "Comparison 2" are displayed in a bar graph for each performance.

[0047] As shown in Figs. 5 to 7, the mixture adjustment screen 1000 according to this embodiment accepts a varying operation for the area of ​​the mixture ratio graphic from the operator, and adjusts the mixture ratio based on the varying operation.

[0048] Fig. 5 is an explanatory diagram of an example of a variable operation for the area of ​​a mixture ratio diagram. In Fig. 5, an adjustment of the mixture ratio of "material 1" and "material 2" can be accepted based on a variable operation for the area of ​​the mixture ratio diagram of "material 1" and "material 2" in the mixture graph 1006. The operator can perform a variable operation for the area of ​​the mixture ratio diagram of "material 1" and "material 2" by dragging the line between the mixture ratio diagrams of "material 1" and "material 2" in the mixture graph 1006 in Fig. 5 to the left or right.

[0049] The area of ​​the mixture ratio figures can be changed by the operator specifying the mixture ratio figure to be enlarged and the mixture ratio figure to be reduced, or the operator can specify the mixture ratio figure to be enlarged and reduce the other mixture ratio figures by dividing them equally. Based on the change operation for the area of ​​the mixture ratio figures in the mixture graph 1006, the performance prediction value of the current mixture displayed in the performance prediction value display field 1004 changes.

[0050] Fig. 6 is an explanatory diagram of an example of a variable operation for the area of ​​a mixture ratio diagram of a constant sum relationship. In Fig. 6, it is possible to accept adjustments to the mixture ratios of "material 4" and "material 5" based on the variable operation for the area of ​​the mixture ratio diagrams of "material 4" and "material 5" in the mixture graph 1006. The operator can vary the area of ​​the mixture ratio diagrams of "material 4" and "material 5" by dragging the line between the mixture ratio diagrams of "material 4" and "material 5" set in the same "group 1" in Fig. 6 up and down.

[0051] In the mix graph 1006 of FIG. 6, the mix ratio diagrams of "material 4" and "material 5" have a constant sum relationship. FIG. 8 is a diagram for explaining an example of mix ratio adjustment when the mix ratios are in a constant sum relationship. FIG. 8 shows an example of "material A" and "material B" in a constant sum relationship. As shown in FIG. 8, the sum of "material A" and "material B" in the constant sum relationship remains fixed at "10%," while the mix ratio ratio (balance) can be adjusted from "material A: 3%" and "material B: 7%" to "material A: 5%" and "material B: 5%".

[0052] In the mixture graph 1006 of Fig. 6, the width of the mixture ratio diagram of "material 4" and "material 5" in the constant sum relationship is locked, so the sum of the mixture ratios of "material 4" and "material 5" in "group 1" remains fixed, and the ratio (balance) of the mixture ratios of "material 4" and "material 5" can be adjusted. Therefore, the operator can adjust the mixture ratios of materials within a group without changing the sum of the mixture ratios within the group for materials grouped based on the intended function.

[0053] Fig. 7 is an explanatory diagram of an example of a variable operation for the area of ​​a mixture ratio figure of a constant ratio relationship. In Fig. 7, adjustments to the mixture ratios of "material 6", "material 7", and "material 8" can be accepted based on the variable operation for the area of ​​the mixture ratio figures of "material 6", "material 7", and "material 8" in the mixture graph 1006. The worker can vary the area of ​​the mixture ratio figures of "material 6", "material 7", and "material 8" by dragging the width of "material 6", "material 7", and "material 8" that are set in the same "group 2".

[0054] In the mixture graph 1006 of FIG. 7, the mixture ratio diagrams of "material 6", "material 7" and "material 8" have a constant ratio relationship. FIG. 9 is a diagram for explaining an example of a mixture ratio adjustment when the mixture ratios are in a constant ratio relationship. FIG. 9 shows an example of "material A" and "material B" in a constant ratio relationship. As shown in FIG. 9, for "material A" and "material B" in a constant ratio relationship, the ratio (balance) of the mixture ratios within the group remains fixed, while the sum of the mixture ratios of "material A" and "material B" within the group can be adjusted from "10%" to "8%".

[0055] In the mixture ratio graph 1006 of FIG. 7, the mixture ratio diagram of "material 6", "material 7", and "material 8" which have a constant ratio relationship has the ratio (balance) of the mixture ratios of "material 6", "material 7", and "material 8" in "group 2" locked, so the sum of the mixture ratios of "material 6", "material 7", and "material 8" can be adjusted while keeping the ratio of the mixture ratios of "material 6", "material 7", and "material 8" in "group 2" fixed.

[0056] According to the combination graph 1006 of the combination adjustment screen 1000 of this embodiment, it is possible to adjust the combination ratio of ingredients while grasping the overall combination picture and in accordance with the grouping of ingredients based on the intended function.

[0057] The combination graph 1006 shown in FIG. 4 is an example, and may be a combination graph 1020 as shown in FIG. 10. FIG. 10 is an image diagram of an example of a combination graph according to this embodiment. The combination graph 1020 in FIG. 10 is also displayed in the combination adjustment field 1002 of the combination adjustment screen 1000 in FIG. 4, for example.

[0058] The combination graph 1020 is an example of a display that accepts adjustment of the mixture ratio of ingredients in the form of a sunburst chart. A sunburst chart is a hierarchical pie chart. The combination graph 1020 in FIG. 10 shows the combination by combining figures with areas according to the mixture ratios of "Ingredient 1", "Ingredient 2", "Group 1" and "Group 2" in the inner circle.

[0059] The outer circle of the composition graph 1020 in Fig. 10 shows the composition by combining figures (hereinafter referred to as composition ratio figures) having areas corresponding to the composition ratios of "ingredient 1", "ingredient 2", "ingredient 4", "ingredient 5", "ingredient 6", "ingredient 7" and "ingredient 8". The sum of the composition ratios of ingredients 1, 2, 4 to 8 shown in the outer circle of the composition graph 1020 is 100%. The composition graph 1020 shows the overall composition.

[0060] Mark 1022 displayed on the combination graph 1020 indicates that the combination ratios of "ingredients 4" and "ingredients 5" are in a constant sum relationship. Mark 1024 displayed on the combination graph 1020 indicates that the combination ratios of "ingredients 6", "ingredients 7", and "ingredients 8" are in a constant ratio relationship.

[0061] 5 to 7, the combination graph 1020 accepts a change operation for the area of ​​the combination ratio figure from the operator, and adjusts the combination ratio based on the change operation.

[0062] According to the combination graph 1006 of the combination adjustment screen 1000 of this embodiment, it is possible to adjust the combination ratio of ingredients while grasping the overall combination picture and in accordance with the grouping of ingredients based on the intended function.

[0063] <Avoid forbidden combinations> The mix design support device 10 according to this embodiment is preferably capable of prohibiting or automating changes in the area of ​​the mix ratio diagram so as to avoid mixes (prohibited mixes) that result in uncertain performance prediction values ​​in the performance prediction unit 32 on the mix adjustment screen 1000. Prohibited mixes are mixes that result in uncertain performance prediction values ​​due to inheritance of past recipes or a limited number of experiments (a number of experiments that make comprehensive experiments impossible in terms of man-hours).

[0064] Therefore, the mix design support device 10 of this embodiment detects correlations that exist in the mix ratios of ingredients from learning data and / or past recipes, and recognizes mixes that do not satisfy the detected correlations as prohibited mixes, allowing the operator to adjust the mix ratios of ingredients while avoiding prohibited mixes.

[0065] The correlation between the ingredient mix ratios from the learning data and / or past recipes can be detected as follows, for example, using the variance expansion factor (VIF). First, when detecting the correlation between the ingredient mix ratios from the learning data, the ingredient X to be examined is target Select.

[0066] Next, X target X using materials other than X as explanatory variables target The reproduction equation (prediction equation) is formulated as a correlation formula by using, for example, the least absolute value method, RANSAC (Random Sample Consensus), Theil-Sen regression, or the like.

[0067] FIG. 11 is a diagram for explaining an example of correlation formulation using the least absolute value method, RANSAC, Theil-Sen regression, or the like. As shown in FIG. 11, by inspecting the learning data, it is assumed that the correlation between the mix ratio of material A and the mix ratio of material B is "material A = 3 x material B". However, exceptional data is included. When the ordinary least square method is used, the formulated correlation may be disturbed due to the influence of exceptional data, for example, "material A = 2.9 x material B - 0.1". On the other hand, when the least absolute value method or RANSAC is used, the influence of exceptional data is suppressed by robustness, and the correlation between material A and material B, "material A = 3 x material B", is obtained. In this way, the mix design support device 10 according to this embodiment detects the correlation existing in the mix ratio of materials in the form of a constant sum and constant ratio of the mix ratios as simply as possible.

[0068] Next, X target Reproducibility (R 2 The VIF is calculated from the VIF (value). If the VIF is a threshold (e.g., 10 or more), it is considered that the correlation is established, and the reproduction formula is stored. The stored reproduction formula represents the correlation of intentional mixture ratios, such as constant sums and constant ratios of mixture ratios generated by learning data and / or past recipes.

[0069] Therefore, the mix design support device 10 of this embodiment displays the mix graph 1006 in a format that locks the horizontal width of the mix ratio figure when the mix ratios are in a constant sum relationship as shown in Figure 6, or in a format that locks the vertical width of the mix ratio figure when the mix ratios are in a constant ratio relationship as shown in Figure 7, in accordance with the stored reproduction formula so as to avoid prohibited mixes.

[0070] Among the material groups displayed on the mixture adjustment screen 1000, groups that are automatically set based on the detected correlation of the mixture ratio may not be released by the operator. The correlation of the mixture ratio that cannot be expressed by a simple formula such as the constant sum and constant ratio of the mixture ratio may not be displayed as a group on the mixture adjustment screen 1000, and the area change of the mixture ratio figure may be prohibited or automated so that prohibited mixtures can be avoided.

[0071] <Processing> Hereinafter, the process of the mix design support system 1 according to this embodiment for allowing the operator to adjust the mix ratio will be described. Fig. 12 is a flowchart of an example of the process of the mix design support system according to this embodiment.

[0072] In step S10, the mix design support device 10 detects correlations that exist in the mix ratios of ingredients from learning data, past recipes, etc., and classifies mixes that do not satisfy the detected correlations as prohibited mixes.

[0073] In step S12, the mix design support device 10 displays a mix adjustment screen 1000 for avoiding prohibited mixes, as shown in Figures 4 to 7. In step S14, the mix design support device 10 accepts an operation on the mix adjustment screen 1000 by an operator.

[0074] In step S16, the mix design support device 10 judges whether or not the area of ​​the mixture ratio figure has been changed by the operator. If the area of ​​the mixture ratio figure has not been changed, the mix design support device 10 performs the process of step S20. If the area of ​​the mixture ratio figure has been changed, the mix design support device 10 performs the process of step S18.

[0075] In step S18, the mix design support device 10 predicts the performance prediction value of the mix represented by the mix graph 1006 adjusted by the operator in step S14, and updates the performance prediction value of the current mix in the performance prediction value display field 1004 of the mix adjustment screen 1000. After the process of step S18, the mix design support device 10 performs the process of step S20.

[0076] In step S20, the mix design support device 10 judges whether the display of the mix adjustment screen 1000 has ended. If the display of the mix adjustment screen 1000 has not ended, the mix design support device 10 returns to the processing of step S14. If the display of the mix adjustment screen 1000 has ended, the mix design support device 10 ends the processing of the flowchart in FIG.

[0077] [Other embodiments] In this embodiment, an example has been described in which the change in area of ​​the mixture ratio diagram is prohibited or automated so as to avoid a mixture (prohibited mixture) in which the performance prediction value of the performance prediction unit 32 becomes indefinite. For example, in the process of inverse problem analysis, if exhaustive search points are generated randomly or at a predetermined interval and the performance of the product corresponding to each exhaustive search point is predicted using a trained machine learning model, the performance prediction value corresponding to the exhaustive search point that does not satisfy the detected correlation may be displayed visually differently as an indefinite performance prediction value.

[0078] The mixture adjustment screen 1000 shown in FIG. 4 to FIG. 7 may accept input of the mixture ratio from the operator by arranging at least one of the slider 1014 and the text box 1016 as described in the mixture adjustment field 1002 of FIG. 13. FIG. 13 is an image diagram of an example of the mixture adjustment field of the mixture adjustment screen according to this embodiment. Note that the mixture ratio that is fixed or linked with other mixture ratios, such as the mixture ratio that is a constant sum relationship or the mixture ratio that is a constant ratio relationship, may be grayed out to prevent input from the operator.

[0079] In addition, the mix adjustment screen 1000 shown in Figs. 4 to 7 may display a profile as shown in Fig. 14, for example, so that the worker can efficiently obtain the performance that he or she is aiming for. Fig. 14 is a diagram of an example of a profile plotting the change in the predicted performance value with respect to the change in the mix ratio of the materials. By referring to the profile in Fig. 14, the worker can know the change in the predicted performance value when the mix ratio of the materials is changed. For example, the worker who refers to the profile in Fig. 14 can know that the predicted performance value of "Performance 1" decreases and the predicted performance value of "Performance 2" increases when the mix ratio of "Material 2" is increased.

[0080] As described above, the mix design support system 1 according to this embodiment can support the operator to adjust the mix ratio of materials while grasping the overall picture of the mix. In addition, the mix design support system 1 according to this embodiment can prohibit or automate the area change of some mix ratio figures so that the operator does not adjust the mix whose performance prediction value becomes unstable.

[0081] The information on the composition adjusted by the composition design support system 1 according to this embodiment may be used as input information for a manufacturing device that produces a product according to the composition information. The manufacturing device can produce a product by blending multiple materials according to the composition information.

[0082] Although the present embodiment has been described above, it will be understood that various changes in form and details are possible without departing from the spirit and scope of the claims. Although the present invention has been described above based on the examples, the present invention is not limited to the above examples, and various modifications are possible within the scope of the claims. This application claims priority to basic application No. 2022-182597 filed on November 15, 2022 with the Japan Patent Office, the entire contents of which are incorporated herein by reference. [Explanation of symbols]

[0083] 1 Mix design support system 10 Mixture design support device 12 User terminal 18. Communications Networks 26 Mixture display control unit 28 Performance prediction value display control section 30 Mixture Ratio Adjustment Reception Department 34 Correlation detection unit 36 Group Settings 1000 Mixture adjustment screen 1002 Mixture adjustment column 1004 Performance prediction value display column 1006 Combination Graph

Claims

[Claim 1] A manufacturing method using a mix design support method executed by a mix design support device that allows an operator to adjust the mix ratio of materials, comprising: A combination display control step of displaying a combination represented by a combination of figures having an area according to the combination ratio on a combination adjustment screen; a mixture ratio adjustment receiving step of receiving an adjustment of the mixture ratio from the operator by a variable operation on the area of ​​the figure; A performance prediction value display control step for displaying a performance prediction value of the blend; a manufacturing step of manufacturing a product using the adjusted blending ratio; The manufacturing method comprising the steps of:

Images