Recommended data generation device, recommended data generation method, and program

Abstract

A recommendation data generation device (2000) acquires a plurality of pieces of material specification information (10) indicating material specifications and acquires, for each piece of the material specification information (10), physical property information (20) indicating physical quantities for each of a plurality of physical properties of a product (70) that is possibly generated on the bases of the material specifications indicated by the material specification information (10). The recommendation data generation device (2000) uses the physical property information (20) to generate a self-organizing map (30) in which a material property vector indicating values related to the physical quantity of each of the plurality of types of physical properties of the product (70) is assigned to each node in a map space. The recommendation data generation device (2000) selects one or more target nodes from among the nodes of the self-organizing map (30) on the basis of an arrangement of nodes corresponding to respective physical property information (20) in the map space. The recommendation data generation device (2000) generates recommendation data (80) indicating material specifications corresponding to the target node(s).

Description

Technical Field

[0004] , , , ,

[0005] , , , , ,

[0003] , , ,

[0001] The present disclosure relates to a technology for providing information related to product development.

Background Art

[0002] In product development, it is useful to grasp the relationship between materials and products. Therefore, a system for assisting in grasping the relationship between materials and products has been developed. For example, Patent Document 1 discloses a system that uses a self-organizing map to assist in grasping the causal relationship between the design values and physical property values of tires.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In Patent Document 1, a self-organizing map is used to identify which of the plurality of design variables of a tire is an important factor. Therefore, it is not assumed to use the self-organizing map for other purposes. The present disclosure is in view of such problems, and the object of the present disclosure is to provide a new technology that provides information useful for product development.

Means for Solving the Problems

[0005] The recommendation data generation device of this disclosure includes: an acquisition means that acquires a plurality of material specification information indicating material specifications, and for each of the material specification information, an acquisition means that acquires material property information indicating the quantity of each of a plurality of physical properties of an output that can be produced with the material specification information indicated in the material specification information; a first generation means that uses the material property information to generate a self-organizing map in which a material property vector indicating the value of the quantity of each of a plurality of types of physical properties of the output is assigned to each node in the map space; a selection means that selects one or more target nodes from the nodes of the self-organizing map based on the arrangement of the nodes corresponding to each of the material property information in the map space; and a second generation means that generates recommendation data indicating the material specifications corresponding to the target node.

[0006] The recommended data generation method of this disclosure is performed by a computer. The recommended data generation method includes: an acquisition step of acquiring a plurality of material specification information indicating material specifications, and for each of the material specification information, acquiring material property information indicating the quantity of each of a plurality of physical properties of an output that can be produced with the material specification information indicated in that material specification information; a first generation step of using the material property information to generate a self-organizing map in which a material property vector indicating the value of the quantity of each of the plurality of physical properties of the output is assigned to each node in the map space; a selection step of selecting one or more target nodes from the nodes of the self-organizing map based on the arrangement of the nodes corresponding to each of the material property information in the map space; and a second generation step of generating recommended data indicating the material specifications corresponding to the target node.

[0007] The non-temporary computer-readable media of this disclosure contains a program that causes a computer to execute the data generation method recommended in this disclosure. [Effects of the Invention]

[0008] According to this disclosure, a new technology is provided that offers information useful for product development. [Brief explanation of the drawing]

[0009] [Figure 1] This figure illustrates an overview of the operation of the recommended data generation device in Embodiment 1. [Figure 2] This is a block diagram illustrating the functional configuration of the recommended data generation device of Embodiment 1. [Figure 3] This is a block diagram illustrating the hardware configuration of a computer that realizes the recommended data generation device of Embodiment 1. [Figure 4] This flowchart illustrates the processing flow performed by the recommended data generation device of Embodiment 1. [Figure 5] This diagram illustrates material specifications in a table format. [Figure 6] This diagram illustrates material properties in a table format. [Figure 7] This diagram illustrates the structure of a self-organizing map in a table format. [Figure 8A] This flowchart illustrates the process for selecting a target node. [Figure 8B] This flowchart illustrates the process for selecting a target node. [Figure 9] This diagram illustrates a map space divided into multiple sub-regions. [Figure 10] This diagram illustrates the structure of a self-organizing map, where each node is assigned a parameter vector, in a table format. [Figure 11] This is an example diagram illustrating a map image. [Figure 12] This figure illustrates an overview of the operation of the recommended data generation device in Embodiment 2. [Figure 13] This is a block diagram illustrating the functional configuration of the recommended data generation device in Embodiment 2. [Figure 14] This flowchart illustrates the processing flow performed by the recommended data generation device of Embodiment 2. [Modes for carrying out the invention]

[0010] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. In each drawing, the same or corresponding elements are denoted by the same reference numerals, and redundant descriptions are omitted as necessary for clarity of explanation. Also, unless otherwise specified, information such as predetermined values and threshold values that are predetermined is stored in advance in a storage device or the like accessible from the device that uses the information. Furthermore, unless otherwise specified, the storage unit is constituted by one or more storage devices of any number.

[0011] [Embodiment 1] [Overview] FIG. 1 is a diagram illustrating an overview of the operation of the recommended data generation device 2000 according to Embodiment 1. Here, FIG. 1 is a diagram for facilitating understanding of the overview of the recommended data generation device 2000, and the operation of the recommended data generation device 2000 is not limited to that shown in FIG. 1.

[0012] The recommended data generation device 2000 generates a self-organizing map 30 representing the distribution of the physical properties of the various products 70 that can be generated in a specific process of product development (hereinafter, the target process). The products 70 are those predicted to be generated by processing the material 60 in the generation process of the target process or those actually generated. The material 60 is the material used for generating the products 70. In the target process, various patterns of the material 60 can be used. The physical properties of the products 70 may vary depending on the material 60 used.

[0013] One pattern of the material 60 is specified by the material specifications. In other words, materials 60 with different material specifications are treated as different patterns of the material 60. On the other hand, materials 60 with the same material specifications are treated as the same pattern of the material 60.

[0014] The material specifications are represented by, for example, the type of material, the type of substances constituting the material, the blending ratio of each substance, and the type of processing performed to create the material. Examples of the type of material include carbon fiber reinforced plastic and stainless steel. For example, assume that material 60 is carbon fiber reinforced plastic. In this case, the material specifications of material 60 include the type of each one or more carbon fibers constituting material 60 (such as polyacrylonitrile fiber and cellulose carbonized fiber), the type of each one or more resins constituting material 60 (such as epoxy and polyether terephthalate), and the blending ratio of these substances. Further, the material specifications may further include the type of fiber orientation polymerization method, the type of crimping method, and the resin composition, etc.

[0015] Note that the target process may be one process or a combination of a plurality of consecutive processes. In the latter case, the product 70 is a product that can be generated by processing material 60 through these plurality of consecutive processes. For example, assume that the target process is a combination of process P1 and process P2. In this case, the product 70 can be obtained by processing material 60 in the production process of process P1 and then processing the product of process P1 in the production process of process P2.

[0016] The self-organizing map 30 has a plurality of nodes arranged on an m-dimensional map space. Here, in order to be able to visually represent (for example, in an image) the map space, m is set to 2 or 3. In the visually represented map space, each node is represented by, for example, a cell in a grid or a lattice on a grid.

[0017] Each node in the self-organizing map 30 is assigned multidimensional data (hereinafter referred to as "property vector") that represents the magnitude of each of several types of material properties. For example, suppose we use four types of material properties: flame retardancy, heat resistance, elastic modulus, and toughness. In this case, the property vector is four-dimensional data that represents the magnitude of each of these four types of material properties. Hereinafter, let the number of dimensions of the property vector be n, where n > m. That is, in the self-organizing map 30, the space of the property vectors is a high-dimensional space, and the map space is a low-dimensional space.

[0018] To generate such a self-organizing map 30, the recommendation data generation device 2000 acquires material specification information 10 representing the material specifications of each of the multiple material patterns 60 (in other words, for each material 60 specified by each of the multiple material specifications), and physical property information 20 corresponding to that material specification information 10. That is, the recommendation data generation device 2000 acquires multiple pairs of material specification information 10 and physical property information 20. The physical property information 20 corresponding to the material specification information 10 indicates the quantity of each of multiple types of physical properties for the output product 70 that can be produced in the target process using the material 60 of the material specifications represented by the material specification information 10. The types of physical properties are, for example, flame retardancy, heat resistance, elastic modulus, or toughness, as mentioned above. The number of types of physical properties indicated by the physical property information 20 is greater than or equal to the number of dimensions n of the physical property vector.

[0019] The recommended data generation device 2000 identifies the corresponding node in the self-organizing map 30 for each piece of material property information 20 (hereinafter referred to as the corresponding node). Here, the corresponding node for material property information 20 is the node in the self-organizing map 30 to which the material property vector most similar to the material property vector obtained from that material property information 20 is associated. In other words, the corresponding node for material property information 20 is the node in the self-organizing map 30 that represents the material property that is closest to the material property represented by the material property information 20.

[0020] The recommended data generation device 2000 selects one or more nodes as target nodes from among the nodes that do not correspond to any of the physical property information 20 (in other words, from among the nodes that are not corresponding nodes) based on the arrangement of corresponding nodes for each physical property information 20 in the map space. For example, a node located far from each corresponding node in the map space may be selected as a target node.

[0021] The recommendation data generation device 2000 uses the self-organizing map 30 to estimate the material parameters corresponding to the target node. The recommendation data generation device 2000 then generates recommendation data 80 showing the estimated material parameters.

[0022] <An example of effects> The pairs of material specification information 10 and physical property information 20 acquired by the recommended data generation device 2000 are generated based on the results of simulations and experimental generation of the output product 70 that have already been carried out. Therefore, the pairs of material specification information 10 and physical property information 20 can be said to represent knowledge about the correspondence between material specifications and physical properties obtained through simulations and other methods performed to date.

[0023] When conducting further simulations to increase knowledge, it is useful to efficiently increase knowledge by appropriately selecting the material parameters used in the simulations. One way to efficiently increase knowledge is to conduct simulations in a way that produces output 70 whose physical properties are as different as possible from those obtained so far. In this way, knowledge can be obtained about the correspondence between various physical properties that are not very similar to each other and the material parameters. In other words, it is possible to avoid obtaining knowledge only about similar physical properties.

[0024] In this regard, according to the recommended data generation device 2000, nodes (corresponding nodes) of the self-organizing map 30 corresponding to each physical property information 20 are identified, and one or more target nodes are selected based on the arrangement of corresponding nodes in the map space of the self-organizing map 30. Here, the arrangement of corresponding nodes in the map space represents the distribution of physical properties for which knowledge has already been obtained. Therefore, by selecting target nodes based on the arrangement of corresponding nodes in the map space, it is possible to select nodes that represent physical properties that do not have a high degree of similarity to physical properties for which knowledge has already been obtained as target nodes.

[0025] Then, according to the recommended data generation device 2000, material parameters corresponding to the target node are estimated, and recommended data 80 showing those material parameters is generated. Here, the material parameters corresponding to the target node are the material parameters that are estimated to produce an output 70 having the physical properties corresponding to that target node. Therefore, by performing simulations etc. using these material parameters, there is a high probability that an output 70 with physical properties that are not highly similar to those for which knowledge has already been obtained can be obtained through simulations etc. From this, it can be seen that the recommended data generation device 2000 makes it easy to obtain material parameters that enable efficient simulations etc. for the generation of output 70.

[0026] The recommended data generation device 2000 of this embodiment will be described in more detail below.

[0027] <Example of functional configuration> Figure 2 is a block diagram illustrating the functional configuration of the recommended data generation device 2000 of Embodiment 1. The recommended data generation device 2000 includes an acquisition unit 2020, a first generation unit 2040, a selection unit 2060, and a second generation unit 2080. The acquisition unit 2020 acquires material specification information 10 and physical property information 20 for each of the multiple material patterns 60. The first generation unit 2040 generates a self-organizing map 30 using physical property vectors obtained from each physical property information 20. The selection unit 2060 identifies the corresponding node for each physical property information 20. Based on the arrangement of each corresponding node in the map space, the selection unit 2060 selects one or more nodes that are not corresponding nodes as target nodes. The second generation unit 2080 generates recommended data 80 indicating the material specifications corresponding to the target node.

[0028] <Example of hardware configuration> Each functional component of the recommended data generation device 2000 may be implemented by hardware (e.g., hardwired electronic circuits) or by a combination of hardware and software (e.g., a combination of electronic circuits and programs that control them). The following will further explain the case where each functional component of the recommended data generation device 2000 is implemented by a combination of hardware and software.

[0029] Figure 3 is a block diagram illustrating the hardware configuration of a computer 1000 that realizes the recommended data generation device 2000 of Embodiment 1. Computer 1000 is any computer. For example, computer 1000 is a stationary computer such as a PC (Personal Computer) or a server machine. Alternatively, computer 1000 is a portable computer such as a smartphone or a tablet terminal. Computer 1000 may be a dedicated computer designed to realize the recommended data generation device 2000, or it may be a general-purpose computer.

[0030] For example, by installing a predetermined application on computer 1000, the various functions of the recommended data generation device 2000 are realized on computer 1000. The above application consists of programs for realizing each functional component of the recommended data generation device 2000. The method of obtaining the above program is arbitrary. For example, the program can be obtained from a storage medium (such as a DVD disc or USB memory) on which it is stored. Alternatively, the program can be obtained by downloading it from a server device that manages the storage device on which it is stored.

[0031] Computer 1000 includes a bus 1020, a processor 1040, memory 1060, a storage device 1080, an input / output interface 1100, and a network interface 1120. The bus 1020 is a data transmission path for the processor 1040, memory 1060, storage device 1080, input / output interface 1100, and network interface 1120 to send and receive data from each other. However, the method of connecting the processor 1040 and other components is not limited to bus connection.

[0032] The processor 1040 is a variety of processors such as a CPU (Central Processing Unit), GPU (Graphics Processing Unit), or FPGA (Field-Programmable Gate Array). The memory 1060 is the main memory, implemented using RAM (Random Access Memory), etc. The storage device 1080 is the auxiliary storage, implemented using a hard disk, SSD (Solid State Drive), memory card, or ROM (Read Only Memory), etc.

[0033] The input / output interface 1100 is an interface for connecting the computer 1000 with input / output devices. For example, input devices such as keyboards and output devices such as display devices are connected to the input / output interface 1100.

[0034] The network interface 1120 is an interface for connecting computer 1000 to a network. This network may be a LAN (Local Area Network) or a WAN (Wide Area Network).

[0035] The storage device 1080 stores programs that implement each functional component of the recommended data generation device 2000 (programs that implement the aforementioned applications). The processor 1040 reads these programs into memory 1060 and executes them to implement each functional component of the recommended data generation device 2000.

[0036] The recommended data generation device 2000 may be implemented using one computer 1000 or multiple computers 1000. In the latter case, the configuration of each computer 1000 does not need to be identical and can be different.

[0037] <Processing flow> Figure 4 is a flowchart illustrating the processing flow performed by the recommended data generation device 2000 of Embodiment 1. The acquisition unit 2020 acquires material specification information 10 and physical property information 20 for each of the multiple material patterns 60 (S102). The first generation unit 2040 generates a self-organizing map 30 using the physical property vectors obtained from each physical property information 20 (S104). The selection unit 2060 identifies the corresponding node for each physical property information 20 (S106). The selection unit 2060 selects a target node based on the arrangement of the corresponding nodes in the map space (S108). The second generation unit 2080 generates recommended data 80 indicating the material specifications corresponding to the target node (S110).

[0038] <Acquisition of material specifications information 10 and physical properties information 20: S102> The acquisition unit 2020 acquires, for each of the multiple patterns of material 60, material specification information 10 representing the material specifications of that material 60, and physical property information 20 about the deliverable 70 that can be produced using that material 60 (S102). Figure 5 is a diagram illustrating the material specification information 10 in table format. The table 100 in Figure 5 has columns for material identification information 102 and material specifications 104. Material identification information 102 shows the identification information assigned to material 60. Material specifications 104 shows the specifications of material 60.

[0039] In Figure 5, the material specification information 10 is represented by a single record in table 100. That is, the material specification information 10 associates the identification information of material 60 with the material specifications of material 60 that possesses that identification information.

[0040] Figure 6 is a diagram illustrating the physical property information 20 in table format. Table 200 in Figure 6 has columns for deliverable identification information 202 and physical properties 204. Deliverable identification information 202 shows the identification information of deliverable 70. Physical properties 204 shows the physical properties of deliverable 70. In Table 200, the physical properties of deliverable 70 are represented by showing a correspondence for each physical property in the form of "label representing the type of physical property: physical quantity of that physical property".

[0041] In Figure 6, the physical property information 20 is represented by a single record in Table 200. That is, the physical property information 20 associates the identification information of the deliverable 70 with the physical properties of the deliverable 70 that possesses that identification information.

[0042] The acquisition unit 2020 acquires multiple pairs of material specification information 10 and physical property information 20. There are various ways in which the acquisition unit 2020 acquires pairs of material specification information 10 and physical property information 20. For example, pairs of material specification information 10 and physical property information 20 are stored in advance in any storage unit accessible from the recommended data generation device 2000. The acquisition unit 2020 acquires pairs of material specification information 10 and physical property information 20 by accessing this storage unit. Alternatively, for example, the acquisition unit 2020 may acquire pairs of material specification information 10 and physical property information 20 by accepting user input to input such pairs. Alternatively, for example, the acquisition unit 2020 may acquire pairs of material specification information 10 and physical property information 20 by receiving pairs of material specification information 10 and physical property information 20 transmitted from other devices.

[0043] There are various methods for generating pairs of material specification information 10 and physical property information 20. For example, a pair of material specification information 10 and physical property information 20 can be generated by simulating the generation of the output product 70. Specifically, by providing specific material specifications as input and running the simulation, physical property information 20 is generated that shows the predicted values ​​of the physical properties of each physical property for the output product 70. Then, a pair is obtained of the generated physical property information 20 and material specification information 10 that shows the material specifications provided as input. Here, existing technologies can be used to realize a simulation that takes material specifications as input and outputs predicted data of the physical properties of an output product generated in a specific process using the material specified by those material specifications.

[0044] Alternatively, for example, the pair of material specification information 10 and physical property information 20 may be generated by actually producing the output product 70. Specifically, the output product 70 is experimentally produced by using a material 60 represented by specific material specifications in the target process. Furthermore, physical property information 20 is generated by measuring the physical property quantities of each property of the generated output product 70. As a result, a pair of the generated physical property information 20 and the material specification information 10 representing the material 60 used is obtained.

[0045] Furthermore, the physical property information 20 acquired by the acquisition unit 2020 may include data that is expressed in different ways. For example, it is possible that different labels are used for essentially the same physical properties. Also, it is possible that physical property quantities of the same physical property are expressed in different units. In such cases, it is preferable for the acquisition unit 2020 to unify the way the data is expressed by unifying labels or converting units. This situation in which the way the data is expressed differs among the physical property information 20 can occur, for example, when acquiring both physical property information 20 generated using simulation and physical property information 20 generated by actually producing the output product 70. It is also preferable that the same unification of the way the data is expressed be performed for the material specification information 10.

[0046] <Generating Self-Organizing Map 30: S104> The first generation unit 2040 generates a self-organizing map 30 using each of the physical property information 20 (S104). The self-organizing map 30 has multiple nodes arranged on an m-dimensional map space (m=2 or m=3). Whether to adopt 2 dimensions or 3 dimensions as the number of dimensions of the map space may be predetermined or specified by the user. Each node of the self-organizing map 30 is assigned an n-dimensional physical property vector.

[0047] Figure 7 illustrates the structure of the self-organizing map 30 in table format. Table 300 has two columns: node 302 and material property vector 304. Each record in Table 300 represents that the material property vector shown in the material property vector 304 of that record is assigned to the node identified by node 302 of that record. In Figure 7, node 302 indicates the coordinates of the node in the map space.

[0048] The assignment of physical property vectors to each node is performed by training the self-organizing map 30. Training the self-organizing map 30 can be done by inputting n-dimensional training data into the self-organizing map 30. Existing methods can be used to specifically train the self-organizing map using the training data.

[0049] For example, the first generation unit 2040 initializes the self-organizing map 30 in an arbitrary way. One possible initialization method is to initialize the physical property vectors of each node to random values. The first generation unit 2040 obtains multiple physical property vectors by obtaining physical property vectors from each of the multiple physical property information 20. The first generation unit 2040 generates the self-organizing map 30 by training the self-organizing map 30 using each of these multiple physical property vectors as training data. As a result, the physical property vectors corresponding to each node of the self-organizing map 30 become n-dimensional data representing values ​​for each of the n types of physical property quantities.

[0050] <Identifying the corresponding node: S106> The selection unit 2060 identifies the corresponding node for each piece of material property information 20 (S106). The corresponding node for the material property information 20 is the node in the self-organizing map 30 that has the material property vector most similar to the material property vector obtained from that piece of material property information 20.

[0051] The degree of similarity between physical property vectors can be determined, for example, based on the distance between them. For example, the selection unit 2060 performs the following processing for each piece of physical property information 20. First, the selection unit 2060 calculates the distance between the physical property vector obtained from the physical property information 20 and the physical property vector of each node in the self-organizing map 30. Then, the selection unit 2060 identifies the node with the smallest calculated distance as the node with the physical property vector most similar to the physical property vector obtained from that piece of physical property information 20. Therefore, the node identified here is identified as the corresponding node for that piece of physical property information 20.

[0052] <Select target node: S108> The selection unit 2060 selects a target node from among the nodes other than the corresponding node, based on the arrangement of the corresponding node in the map space (S108). Conceptually, the selection unit 2060 selects a node located far from the corresponding node in the map space as the target node. The following is a specific example of how the target node is selected.

[0053] Figures 8A and 8B are flowcharts illustrating the process of selecting a target node. Hereafter, Figures 8A and 8B will be collectively referred to as Figure 8.

[0054] The selection unit 2060 divides the map space into multiple sub-regions (S202). Each sub-region contains a predetermined number of nodes. Figure 9 is an example of a map space 32 divided into multiple sub-regions. In the map space 32 of Figure 9, each sub-region 36 is a rectangular area enclosed by a thick border, and contains four nodes 34. Nodes 34 marked with circles are corresponding nodes.

[0055] S204 to S212 constitute a loop process L1 that is executed on each of the multiple subregions. In S204, the selection unit 2060 determines whether or not the loop process L1 has been executed on all subregions. If the loop process L1 has already been executed on all subregions, the process in Figure 8 proceeds to S214. On the other hand, if there are subregions that have not yet been targeted by the loop process L1, the selection unit 2060 selects one of them. The subregion selected here is denoted as "subregion R". After that, the process in Figure 8 proceeds to S206.

[0056] The selection unit 2060 calculates an evaluation score for a representative point of the subregion R based on the positional relationship between that representative point and each corresponding node (S206). The representative point of the subregion R is, for example, one of the vertices of the subregion R or the center of the subregion R.

[0057] The evaluation score for representative points is defined, for example, by the following formula (1).

number

[0058] According to equation (1), in the distribution of corresponding nodes in the map space, representative points located in areas with low density will have lower evaluation scores.

[0059] Here, if the representative point is located at the center of a node, the coordinates of that representative point are expressed by the coordinates of that node. On the other hand, if the representative point is not located at the center of a node, for example, the coordinates of the representative point are determined based on the coordinates of each node adjacent to the representative point. For example, in the example in Figure 9, suppose the representative point of a subregion is one of the vertices of that subregion. In this case, the coordinates of the representative point are calculated, for example, as the coordinates of the centers of the four nodes that have that representative point as a vertex (the nodes to the upper left, upper right, lower right, and lower left of that representative point).

[0060] The distance between a representative point and its corresponding node may be expressed as a distance in map space, or as a distance in the higher-dimensional space (space of material property vectors) of the self-organizing map 30. In the latter case, the distance between the representative point and its corresponding node is expressed as the distance between the material property vectors corresponding to these two points.

[0061] The definition of the evaluation score for a representative point is not limited to the definition in equation (1). For example, the evaluation score for a representative point may be defined by the following equation (2).

number

[0062] According to equation (2), the further a representative point is from the nearest corresponding node, the lower its evaluation score will be.

[0063] The selection unit 2060 determines whether the evaluation score of the representative point of subregion R is lower than the evaluation score of any other representative point adjacent to that representative point (S208). If the evaluation score of the representative point of subregion R is lower than the evaluation score of any other representative point adjacent to that representative point (S208: YES), the selection unit 2060 selects the representative point of subregion R as a reference point to be used for subsequent evaluations and stores it in the list of reference points Lp (S210).

[0064] On the other hand, if there is another representative point adjacent to the representative point of subregion R whose evaluation score is less than or equal to the evaluation score of the representative point of subregion R (S208:NO), the process in Figure 8 proceeds to S212. In this case, the representative point of subregion R is not selected as the reference point.

[0065] For example, in the example in Figure 9, there are eight adjacent representative points to the representative point. The representative point with an evaluation score lower than any of the eight adjacent representative points is selected as the reference point.

[0066] S212 marks the end of loop processing L1. Therefore, the process in Figure 8 proceeds to S204.

[0067] When loop processing L1 is completed, S214 is executed. S214 to S226 constitute loop processing L2, which is executed repeatedly until a predetermined termination condition is met. For example, the predetermined termination condition is that loop processing L2 is executed a predetermined number of times. Another example of a predetermined termination condition is that the size of the evaluation area, which will be described later, falls below a predetermined threshold. The predetermined number of times and the predetermined threshold that define the termination condition are set in advance by, for example, the user of the recommended data generation device 2000.

[0068] S216 to S224 constitute a loop process L3 that is executed for each reference point included in list Lp. In S216, the selection unit 2060 determines whether loop process L3 has been executed for all reference points included in Lp. If loop process L3 has already been executed for all reference points, the process in Figure 8 proceeds to S226. Since S226 is the end of loop process L2, the process in Figure 8 proceeds to S214.

[0069] On the other hand, if there are reference points that have not yet been targeted by loop processing L3, the selection unit 2060 selects one of them. The reference point selected here is denoted as reference point q. After that, the process in Figure 8 proceeds to S218.

[0070] The selection unit 2060 determines an evaluation region that is an area in the map space and is centered on the reference point q, and divides the determined evaluation region into multiple sub-regions (S218). Here, if the current S218 is included in the first iteration of the loop process L2, for example, the size of the evaluation region is calculated by multiplying the size of the map space by a predetermined ratio α less than 1. On the other hand, if the current S218 is included in the second or subsequent iterations of the loop process L2, for example, the size of the evaluation region is calculated by multiplying the size of the previous evaluation region by a ratio α less than 1. In this way, the size of the evaluation region decreases each time the loop process L2 is executed.

[0071] The method for determining the ratio α is arbitrary. For example, the ratio α may be fixed and set by the administrator of the recommended data generation device 2000. Alternatively, for example, the ratio α may be specified by the user of the recommended data generation device 2000. Here, in the repeated execution of the loop process L2, the value of the ratio α may be the same each time, or different values ​​may be used each time. In the latter case, for example, the value of the ratio α is set individually for the first iteration of the loop process L2 and for subsequent iterations.

[0072] The selection unit 2060 calculates the aforementioned evaluation score for the representative points of each sub-region obtained from the evaluation region (S220). The selection unit 2060 replaces the reference point q stored in the list Lp with the representative point that has the minimum evaluation score calculated in S220 (S222).

[0073] According to the processing in S222, the reference point q included in Lp is replaced with a point with a smaller evaluation score that is included in the evaluation region surrounding the reference point q. Here, as mentioned above, the size of the evaluation region decreases with each execution of the loop process L2. As a result, by repeatedly executing the loop process L2, the position of each reference point is repeatedly corrected, while gradually narrowing the correction range.

[0074] S224 marks the end of loop processing L3. Therefore, the process in Figure 8 proceeds to S216.

[0075] After the completion of loop processing L3, the selection unit 2060 selects the node corresponding to each reference point in the list Lp as the target node. The node corresponding to a reference point is, for example, a node that contains the reference point within a region when the map space is divided into multiple nodes as shown in Figure 9. Alternatively, for example, the node corresponding to a reference point is the node with the coordinates closest to the coordinates of the reference point.

[0076] According to the method shown in Figure 8, a so-called grid search is used to find nodes with the aforementioned low evaluation scores in the map space. When the evaluation score defined by equation (1) is used, nodes located in areas with low density in the distribution of corresponding nodes in the map space are selected as target nodes. On the other hand, when the evaluation score defined by equation (2) is used, nodes that are farther away from the nearest corresponding node are selected as target nodes.

[0077] Here, the method for selecting the target node is not limited to the method shown in Figure 8. For example, the selection unit 2060 may select the node corresponding to each reference point in the list Lp obtained in loop processing L1 as the target node, without performing loop processing L2 and L3 in Figure 8B. Alternatively, for example, the selection unit 2060 may calculate an evaluation score for each node other than the corresponding node and select the node with the smallest evaluation score, or the first predetermined number of nodes in ascending order of evaluation scores, as the target node.

[0078] <Generating recommended data 80: S110> The second generation unit 2080 estimates the material parameters corresponding to the target node and generates recommended data 80 showing the estimated material parameters (S110). To this end, the second generation unit 2080 uses the material parameter information 10 to assign multidimensional data (hereinafter referred to as parameter vectors) representing the values ​​of each of the multiple types of parameters (hereinafter referred to as parameter parameters) of the material parameters to each node of the self-organizing map 30. For example, parameter vectors are data that represent the information shown in the material parameters 104 in Figure 5 as vectors. The second generation unit 2080 then estimates the parameter vectors assigned to the target node as the material parameters corresponding to the target node.

[0079] The following describes how to assign a parameter vector to each node.

[0080] First, the second generation unit 2080 assigns the parameter vectors obtained from each material parameter information 10 to the node corresponding to that material parameter information 10. Here, the node corresponding to the material parameter information 10 is the corresponding node of the physical property information 20 corresponding to that material parameter information 10.

[0081] The parameter vector obtained from the material parameter information 10 may represent the values ​​for all parameter parameters indicated by the material parameter information 10, or it may represent the values ​​for some of those parameter parameters. That is, if the dimension of the parameter vector is k, the value of k may be the same as the number of parameter parameters indicated by the material parameter information 10, or it may be less than the number of parameter parameters indicated by the material parameter information 10.

[0082] For example, suppose the material specification information 10 includes both parameters that take continuous values ​​(e.g., the mixing ratio of materials) and parameters that do not take continuous values ​​(e.g., the type of processing). In this case, for example, the specification vector is generated by the parameters that take continuous values.

[0083] Furthermore, the parameters used to generate the specification vector from the material specification information 10 may be predetermined or specified by the user. In addition, the specification vector may represent the parameter values ​​shown in the material specification information 10 as they are, or it may represent values ​​obtained by transforming the values ​​of each parameter in a predetermined way (for example, by performing normalization or standardization).

[0084] Furthermore, the second generation unit 2080 estimates the parameter vectors to be assigned to nodes to which the material parameter information 10 is not associated. Specifically, the second generation unit 2080 estimates the distribution of parameter vectors in the map space based on the parameter vectors of each node to which the material parameter information 10 is associated, and the arrangement of those nodes in the map space. Then, the second generation unit 2080 uses the estimated distribution to assign parameter vectors even to nodes to which the material parameter information 10 is not associated.

[0085] There are various methods for estimating the distribution of parameter vectors in map space. For example, the second generation unit 2080 estimates the distribution of parameter vectors in map space from the parameter vectors of each node to which the material parameter information 10 is associated, and the arrangement of those nodes in map space, using arbitrary interpolation processes such as linear interpolation or spline interpolation. Alternatively, the second generation unit 2080 may estimate the distribution of parameter vectors in map space using sparse estimation. Furthermore, when estimating the distribution of parameter vectors, the estimation accuracy may be improved by applying Bayesian estimation.

[0086] Figure 10 is a diagram illustrating the structure of a self-organizing map 30 in table format, where each node is assigned a parameter vector. The table 300 in Figure 10 has four columns: node 302, material property vector 304, material identification information 306, and parameter vector 308. The table 300 has one record for each node.

[0087] Node 302 indicates the coordinates of the node in map space. The material property vector 304 represents the material property vector assigned to the node. The material identification information 306 indicates the identification information of material 60 indicated by the material specification information 10 assigned to the node for nodes to which material specification information 10 is assigned. For records of nodes to which material specification information 10 is not assigned, the material identification information 306 shows "-". The specification vector 308 represents the specification vector assigned to the node.

[0088] <Output of results> The recommended data generation device 2000 outputs information representing the processing results (hereinafter referred to as "output information") in any manner. Hereinafter, the functional component of the recommended data generation device 2000 that outputs the output information will be referred to as the output unit. For example, the output information shows recommended data 80 representing the material specifications corresponding to each of one or more target nodes, associated with the coordinates of that target node.

[0089] For example, the output unit stores the output information in an arbitrary memory unit. Alternatively, the output unit can output the output information to a display device, causing the information to be displayed on the display device. Furthermore, the output unit can transmit the output information to any other device (for example, the simulator mentioned above).

[0090] Furthermore, the recommended data generation device 2000 may generate a map image that visually represents the map space of the self-organizing map 30, and include the map image in the output information. The map image is an image that represents the correspondence between the distribution of material parameters and the distribution of physical properties.

[0091] Figure 11 is an example of a map image. In Figure 11, the map image 40 visually represents the map space of the self-organizing map 30. Nodes to which material specification information 10 is assigned are superimposed with a material specification display 42 that represents some or all of the material specification information 10.

[0092] The nodes in map image 40 are divided into clusters based on their physical property vectors. The thick borders in map image 40 represent the cluster boundaries. To divide the nodes into clusters, the recommendation data generator 2000 performs clustering on the physical property vectors corresponding to each node in the self-organizing map 30. By dividing the physical property vectors into clusters in this way, the nodes corresponding to the physical property vectors can also be divided into clusters. For example, the recommendation data generator 2000 performs clustering of physical property vectors using various clustering algorithms such as the k-means method.

[0093] Each node in the map image 40 may be colored based on its physical property vectors. Here, various existing methods can be used to color each node of the self-organizing map according to the data associated with that node.

[0094] <Examples of how to use the recommended data 80> For example, the recommended data 80 is used for simulating the generation of the deliverable 70 or for experimentally generating the deliverable 70. Here, the recommended data 80 may be referenced by the operator performing the simulation or experimental generation, or it may be referenced by the simulator performing the simulation. In the former case, the operator performs the simulation by inputting the recommended data 80 into the simulator, or experimentally generates the deliverable 70 by preparing the material 60 specified by the material specifications shown in the recommended data 80. The latter case will be described below as Embodiment 2.

[0095] [Embodiment 2] Figure 12 illustrates an overview of the operation of the recommended data generation device 2000 according to Embodiment 2. Here, Figure 12 is intended to facilitate understanding of the overview of the recommended data generation device 2000, and the operation of the recommended data generation device 2000 is not limited to what is shown in Figure 12.

[0096] The recommendation data generation device 2000 inputs the recommendation data 80 into the simulator 400, causing the simulator 400 to perform a simulation. The simulator 400 simulates the production process of the target process. Specifically, the simulator 400 acquires input data representing material specifications and generates prediction data 410 representing the predicted physical properties of the output product 70, which is produced using the material 60 specified by the input data. Here, the simulator 400 can be any existing simulator that acquires input data representing material specifications, performs a simulation using that input data, and outputs prediction data for the physical properties of the output product. Furthermore, the simulator 400 may be implemented on the computer that implements the recommendation data generation device 2000, or it may be implemented on another computer.

[0097] The recommendation data generation device 2000 acquires the prediction data 410 generated by the simulator 400 and outputs a pair of recommendation data 80 and prediction data 410. As a result, a pair of recommendation data 80 and prediction data 410 obtained by inputting the recommendation data 80 into the simulator 400 is output.

[0098] The above pair can be said to correspond to the pair of material specification information 10 and physical property information 20 mentioned above. Therefore, by using the recommended data generation device 2000, it is possible to improve the efficiency of subsequent simulations based on the results of simulations and experiments that have already been conducted to generate the output product 70.

[0099] For example, for the first certain number of iterations, the material parameters to be simulated or experimentally generated are determined either according to past knowledge or randomly. Subsequently, by inputting these results into the recommended data generation device 2000, material parameters that are predicted to produce a material with properties less similar to those of each output 70 obtained from previously performed simulations are estimated and output as recommended data 80. Then, by inputting the material parameters shown in the recommended data 80 into the simulator 400, new pairs of material parameters and properties are obtained.

[0100] By using the recommended data generation device 2000 in this way, the similarity between the physical properties of the output 70 obtained from a new simulation and the physical properties of each output 70 obtained from previously performed simulations can be reduced. Therefore, it is possible to avoid a situation where only output 70 with similar physical properties are obtained from simulations, and to increase the variety of physical properties of the output 70 with fewer simulations.

[0101] The output method of the pair of recommended data 80 and predicted data 410 is arbitrary. For example, the recommended data generation device 2000 stores the pair in any memory accessible from the recommended data generation device 2000. Alternatively, for example, the recommended data generation device 2000 displays the pair on any display device controllable from the recommended data generation device 2000. Alternatively, for example, the recommended data generation device 2000 transmits the pair to any device that is communicatively connected to the recommended data generation device 2000.

[0102] Furthermore, the recommendation data generator 2000 may update the self-organizing map 30 by further training the self-organizing map 30 using the prediction data 410 as new training data. In addition, the recommendation data generator 2000 may generate a map image 40 for the updated self-organizing map 30 and output it in any manner.

[0103] The recommended data generation device 2000 may, each time it generates recommended data 80, have the simulator 400 run a simulation using that recommended data 80, or it may, after generating multiple sets of recommended data 80, have the simulator 400 run simulations sequentially for each of those sets of recommended data 80.

[0104] <Example of functional configuration> Figure 13 is a block diagram illustrating the functional configuration of the recommendation data generation device 2000 of Embodiment 2. The recommendation data generation device 2000 of Embodiment 2 further comprises a simulator control unit 2120 and a simulation result output unit 2140. The simulator control unit 2120 inputs the recommendation data 80 to the simulator 400, causing the simulator 400 to perform a simulation of generating an output product 70 using the material 60 specified by the material specifications represented by the recommendation data 80. The simulation result output unit 2140 acquires the prediction data 410 output from the simulator 400 and outputs a pair of the recommendation data 80 and the prediction data 410.

[0105] <Example of hardware configuration> The hardware configuration of the recommended data generation device 2000 in Embodiment 2 is similar to that of the recommended data generation device 2000 in Embodiment 1, as shown in Figure 3, for example. However, the storage device 1080 in Embodiment 2 stores programs that implement each functional component of the recommended data generation device 2000 in Embodiment 2.

[0106] <Processing flow> Figure 14 is a flowchart illustrating the processing flow performed by the recommended data generation device 2000 of Embodiment 2. Here, steps S102 to S110 are the same as those shown in Figure 4 and are therefore not shown in Figure 14. In S112, the simulator control unit 2120 inputs the recommended data 80 to the simulator 400 and causes the simulator 400 to perform a simulation. The simulation result output unit 2140 acquires the predicted data 410 from the simulator 400 (S114). The simulation result output unit 2140 outputs a pair of recommended data 80 and predicted data 410 (S116). The first generation unit 2040 updates the self-organizing map 30 by training using the predicted data 410 as training data (S118).

[0107] Although the present invention has been described above with reference to the embodiments, the present invention is not limited to the above embodiments. Various modifications to the structure and details of the present invention can be made, which can be understood by those skilled in the art within the scope of the present invention.

[0108] In the above examples, the program includes a set of instructions (or software code) that, when loaded into a computer, cause the computer to perform one or more of the functions described in the embodiments. The program may be stored on a non-temporary computer-readable medium or a physical storage medium. Examples, but not limited to, include random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technologies, CD-ROM, digital versatile disc (DVD), Blu-ray® disc or other optical disc storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices. The program may be transmitted over a temporary computer-readable medium or a communication medium. Examples, but not limited to, include temporary computer-readable medium or a communication medium that includes electrical, optical, acoustic or other forms of propagating signals.

[0109] Some or all of the above embodiments may also be described as follows, but are not limited to the following: (Note 1) Acquisition means for acquiring multiple material specification information indicating material specifications, and for each of the said material specification information, acquiring physical property information indicating the quantity of each of the multiple physical properties of the output product that can be produced with the material specifications indicated in that material specification information, A first generation means generates a self-organizing map in which, using the aforementioned physical property information, physical property vectors representing the values ​​of the physical property quantities for each of the multiple types of physical properties of the deliverable are assigned to each node in the map space. Selection means for selecting one or more target nodes from the nodes of the self-organizing map based on the arrangement of the nodes corresponding to each of the physical property information in the map space, A recommendation data generation device having a second generation means for generating recommendation data indicating the material specifications corresponding to the target node. (Note 2) The selection means calculates an evaluation score for each of the multiple locations in the map space based on the distance between that location and the node corresponding to each of the physical property information, selects one or more of the multiple locations based on the evaluation score, and selects the node corresponding to the selected location as the target node, as described in Appendix 1. (Note 3) The recommended data generation device as described in Appendix 2, wherein the evaluation score is the sum of the reciprocals of the distances between the position and the node corresponding to each of the physical property pieces, or the reciprocal of the minimum distance between the position and the node corresponding to each of the physical property pieces. (Note 4) The second generating means is Using the material specification information for each of the aforementioned materials, a specification vector representing a value related to the material specification is assigned to each of the aforementioned nodes. A recommendation data generation device according to any one of the appendices 1 to 3, which generates the recommendation data indicating the specification vector assigned to the target node. (Note 5) A simulator control means causes a simulator that generates predictive data on the physical properties of an output product that can be produced using a material specified by the input material specifications to perform a simulation using the material specifications represented by the recommended data as input. A recommendation data generation device according to any one of the appendices 1 to 3, comprising: a simulation result output means for acquiring the prediction data generated by the simulator and outputting the prediction data and the recommendation data. (Note 6) Acquisition step: Acquire multiple material specification information indicating material specifications, and for each of the said material specification information, acquire physical property information indicating the quantity of each of the multiple physical properties of the output product that can be produced with the material specifications indicated in that material specification information. A first generation step involves generating a self-organizing map in which, using the aforementioned physical property information, physical property vectors representing the values ​​of the physical property quantities for each of the multiple types of physical properties of the deliverable are assigned to each node in the map space. A selection step of selecting one or more target nodes from the nodes of the self-organizing map based on the arrangement of the nodes corresponding to each of the physical property information in the map space, A method for generating recommended data, performed by a computer, comprising a second generation step of generating recommended data indicating the material specifications corresponding to the target node. (Note 7) The recommended data generation method described in Appendix 6, wherein in the selection step, for each of the multiple locations in the map space, an evaluation score is calculated based on the distance between that location and the node corresponding to each of the physical property information, one or more of the multiple locations are selected based on the evaluation score, and the node corresponding to the selected location is selected as the target node. (Note 8) The recommended data generation method described in Appendix 7, wherein the evaluation score is the sum of the reciprocals of the distances between the position and the node corresponding to each of the physical property pieces, or the reciprocal of the minimum distance between the position and the node corresponding to each of the physical property pieces. (Note 9) In the second generation step, Using the material specification information for each of the aforementioned materials, a specification vector representing a value related to the material specification is assigned to each of the aforementioned nodes. A method for generating recommended data according to any one of the appendices 6 to 8, which generates the recommended data indicating the specification vector assigned to the target node. (Note 10) A simulator control step involves causing a simulator that generates predictive data on the physical properties of an output product that can be produced using the material specified by the input material specifications to perform a simulation using the material specifications represented by the recommended data as input. A method for generating recommended data according to any one of the appendices 6 to 8, comprising: a simulation result output step of acquiring the prediction data generated by the simulator and outputting the prediction data and the recommended data. (Note 11) Acquisition step: Acquire multiple material specification information indicating material specifications, and for each of the said material specification information, acquire physical property information indicating the quantity of each of the multiple physical properties of the output product that can be produced with the material specifications indicated in that material specification information. A first generation step involves generating a self-organizing map in which, using the aforementioned physical property information, physical property vectors representing the values ​​of the physical property quantities for each of the multiple types of physical properties of the deliverable are assigned to each node in the map space. A selection step of selecting one or more target nodes from the nodes of the self-organizing map based on the arrangement of the nodes corresponding to each of the physical property information in the map space, A non-temporary computer-readable medium containing a program that causes a computer to perform a second generation step, which generates recommended data indicating the material specifications corresponding to the target node. (Note 12) The computer-readable medium according to Appendix 11, wherein in the selection step, for each of the multiple locations in the map space, an evaluation score is calculated based on the distance between that location and the node corresponding to each of the physical property information, one or more of the multiple locations are selected based on the evaluation score, and the node corresponding to the selected location is selected as the target node. (Note 13) The computer-readable medium described in Appendix 12, wherein the evaluation score is the sum of the reciprocals of the distances between the position and the node corresponding to each of the physical property pieces, or the reciprocal of the minimum distance between the position and the node corresponding to each of the physical property pieces. (Note 14) In the second generation step, Using the material specification information for each of the aforementioned materials, a specification vector representing a value related to the material specification is assigned to each of the aforementioned nodes. A computer-readable medium according to any one of the appendices 11 to 13, which generates the recommended data showing the parameter vectors assigned to the target node. (Note 15) A simulator control step involves causing a simulator that generates predictive data on the physical properties of an output product that can be produced using the material specified by the input material specifications to perform a simulation using the material specifications represented by the recommended data as input. A computer-readable medium according to any one of the appendices 11 to 13, comprising a simulation result output step of acquiring the prediction data generated by the simulator and outputting the prediction data and the recommended data. [Explanation of symbols]

[0110] 10. Material Specifications 20. Physical Properties Information 30 Self-organizing maps 32 Map Space 34 nodes 36 subregion 40 Map Images 42 Material specification display 60 materials 70 deliverables 80 Recommended Data 100 tables 102 Material Identification Information 104 Material specifications 200 tables 202 Deliverable Identification Information 204 Physical Properties 300 tables 302 nodes 304 Material property vectors 306 Material Identification Information 308 Parameter Vectors 400 Simulators 410 Prediction Data 1000 computers 1020 Bus 1040 processor 1060 memory 1080 Storage Devices 1100 Input / Output Interface 1120 Network Interface 2000 Recommended Data Generator 2020 Acquisition Department 2040 1st generation part 2060 Selection Section 2080 Second generation part 2120 Simulator Control Unit 2140 Simulation Result Output Unit

Claims

1. Acquisition means for acquiring multiple material specification information indicating material specifications, and for each of the said material specification information, acquiring physical property information indicating the quantity of each of the multiple physical properties of the output product that can be produced with the material specifications indicated in that material specification information, A first generation means generates a self-organizing map in which, using the aforementioned physical property information, physical property vectors representing the values ​​of the physical property quantities for each of the multiple types of physical properties of the deliverable are assigned to each node in the map space. Selection means for selecting one or more target nodes from the nodes of the self-organizing map based on the arrangement of the nodes corresponding to each of the physical property information in the map space, It includes a second generation means for generating recommended data showing the material specifications corresponding to the target node, The selection means is a recommendation data generation device that calculates an evaluation score for each of the multiple locations in the map space based on the distance between that location and the node corresponding to each of the physical property information, selects one or more of the multiple locations based on the evaluation score, and selects the node corresponding to the selected location as the target node.

2. The recommendation data generation device according to claim 1, wherein the evaluation score is the sum of the reciprocals of the distances between the position and the node corresponding to each of the physical property information, or the reciprocal of the minimum distance between the position and the node corresponding to each of the physical property information.

3. The second generating means is, Using the material specification information for each of the aforementioned materials, a specification vector representing a value related to the material specification is assigned to each of the aforementioned nodes. A recommendation data generation device according to claim 1 or 2, which generates the recommendation data indicating the specification vector assigned to the target node.

4. A simulator control means causes a simulator that generates predictive data on the physical properties of an output product that can be produced using a material specified by the input material specifications to perform a simulation using the material specifications represented by the recommended data as input. The recommendation data generation device according to claim 1 or 2, further comprising: a simulation result output means for acquiring the prediction data generated by the simulator and outputting the prediction data and the recommendation data.

5. Acquisition step: Acquire multiple material specification information indicating material specifications, and for each of the said material specification information, acquire physical property information indicating the quantity of each of the multiple physical properties of the output product that can be produced with the material specifications indicated in that material specification information. A first generation step involves generating a self-organizing map in which, using the aforementioned physical property information, physical property vectors representing the values ​​of the physical property quantities for each of the multiple types of physical properties of the deliverable are assigned to each node in the map space. A selection step of selecting one or more target nodes from the nodes of the self-organizing map based on the arrangement of the nodes corresponding to each of the physical property information in the map space, The system includes a second generation step of generating recommended data showing the material specifications corresponding to the target node, A recommended data generation method comprising the selection step, which involves calculating an evaluation score for each of the multiple locations in the map space based on the distance between that location and the node corresponding to each of the physical property information, selecting one or more of the multiple locations based on the evaluation score, and selecting the node corresponding to the selected location as the target node.

6. The recommended data generation method according to claim 5, wherein the evaluation score is the sum of the reciprocals of the distances between the position and the node corresponding to each of the physical property information, or the reciprocal of the minimum distance between the position and the node corresponding to each of the physical property information.

7. Acquisition step: Acquire multiple material specification information indicating material specifications, and for each of the said material specification information, acquire physical property information indicating the quantity of each of the multiple physical properties of the output product that can be produced with the material specifications indicated in that material specification information. A first generation step involves generating a self-organizing map in which, using the aforementioned physical property information, physical property vectors representing the values ​​of the physical property quantities for each of the multiple types of physical properties of the deliverable are assigned to each node in the map space. A selection step of selecting one or more target nodes from the nodes of the self-organizing map based on the arrangement of the nodes corresponding to each of the physical property information in the map space, A second generation step is performed by having a computer generate recommended data showing the material specifications corresponding to the target node. A program that, in the selection step, calculates an evaluation score for each of the multiple locations in the map space based on the distance between that location and the node corresponding to each of the physical property information, selects one or more of the multiple locations based on the evaluation score, and selects the node corresponding to the selected location as the target node.

Images