System, method and program for determining blend ratio

Abstract

To provide a system capable of determining a more suitable blending ratio of a flowability modifier that compensates a difference between flowabilities of material manufacturing lots even when using a general-purpose injection molding machine, to stabilize quality.SOLUTION: A system has one or more processors, and one or more storage media. The processors learn information obtained from a sensor value of a sensor installed on a manufacturing facility of a product using a material, and a blending ratio of a flowability modifier contained in the material corresponding to the information. Thereby, the system generates a learned prediction model that predicts a blending ratio of the flowability modifier capable of achieving a desired flowability of a material with the same type number as the material.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a system, a mixing ratio determination method, and a program. [Background technology]

[0002] Patent Document 1, in relation to an injection molding machine, describes a method for automatically adjusting an injection molding process to compensate for variations in the flowability of a molten plastic material, the method including the steps of: providing an injection molding machine having at least one mold cavity; and providing an injection molding controller including a pressure control output, the pressure control output configured to generate a control signal that at least partially determines an injection molding pressure for the injection molding process of the injection molding machine, the method further including the steps of: measuring a first control signal generated from the pressure control output at a first time in an injection molding cycle; measuring a second control signal generated from the pressure control output at a second time in the injection molding cycle subsequent to the first time; comparing the first control signal generated from the pressure control output with the second control signal generated from the pressure control output to obtain a comparison result; and determining a third control signal for the pressure control output at a third time subsequent to the second time based at least in part on the comparison result. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6144840 Summary of the Invention [Problem to be solved by the invention]

[0004] Patent Document 1 discloses an injection molding machine that takes into account changes in the fluidity of the material during injection molding operation. In addition, in the method disclosed in the document, a pressure control signal is determined during a shot, which is a cycle for obtaining one molded product, or for the next shot, based on a control signal generated from a pressure control output. Therefore, the technology described in the document requires an injection molding machine equipped with an injection molding controller that can measure the control signal and determine a new control signal.

[0005] On the other hand, for manufacturers who own multiple injection molding machines, updating each of their injection molding machines to one equipped with such a controller would be a large cost burden, and updating an injection molding machine that is still within its useful life is not desirable from the perspective of effective use of resources.

[0006] In addition, with the social trend of shifting from the conventional linear economy of mass consumption and mass waste to a circular economy, there is an increasing social demand for the use of recycled materials that are recovered and regenerated from waste. In addition, countries around the world are setting carbon neutral targets to achieve the 1.5℃ scenario set out in the Paris Agreement, and the use of recycled materials, which emit less CO2 during production compared to virgin materials, has become an urgent issue for manufacturers.

[0007] On the other hand, since the physical properties of recycled materials vary depending on the history of the collected waste materials, the fluidity may vary significantly for each material production lot. Therefore, manufacturers need to manufacture products while taking into consideration the fluidity that differs for each material production lot of recycled materials, while also utilizing existing injection molding machines. Therefore, there is a problem that it is difficult to meet the need for utilizing recycled materials just by improving the performance of the injection molding machine itself, as in the technology described in Patent Document 1.

[0008] The present invention has been made in consideration of the above problems, and aims to determine a more appropriate blending ratio of a flowability modifier that can compensate for differences in flowability between material production lots and stabilize quality, even when a general-purpose injection molding machine is used. [Means for solving the problem]

[0009] The present application includes a plurality of means for solving at least part of the above problems, examples of which are as follows: A system according to one aspect of the present invention for solving the above problems is a system having at least one or more processors and one or more storage media, wherein the processors: resin material Model number and production lot Flow improver Multiple combinations of model numbers The mixture ratio, The resin Using materials By injection molding product Manufacture Installed in manufacturing facilities and acquiring information correlating with the fluidity of the resin material at the blending ratio. The sensor value is obtained from Ta By learning information, resin Material and same model number resin Materials are the goal of liquidity So that A trained predictive model is generated to predict the blend ratio of the flow modifier. Effect of the Invention

[0010] According to the present invention, even when a general-purpose injection molding machine is used, it is possible to determine a more appropriate compounding ratio of a flowability modifier that can compensate for differences in flowability between material production lots and stabilize quality. [Brief description of the drawings]

[0011] [Figure 1] FIG. 2 is a schematic configuration diagram showing an example of a functional configuration of the system according to the first embodiment. [Diagram 2] FIG. 1 is a flow diagram illustrating an example of a manufacturing process. [Diagram 3] FIG. 11 is a flow diagram illustrating an example of a process for generating a trained prediction model. [Figure 4]FIG. 11 is a flow chart showing an example of a blending ratio determination process. [Diagram 5] FIG. 13 is a diagram showing an outline of an experimental example for verifying the effects of the embodiment. [Figure 6] FIG. 2 is a diagram illustrating an example of a hardware configuration of a system according to a first embodiment. [Figure 7] FIG. 11 is a diagram illustrating an example of a hardware configuration of a system according to a second embodiment. [Figure 8] FIG. 13 is a diagram illustrating an example of a hardware configuration of a system according to a third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0013] First Embodiment 1 is a schematic diagram showing an example of the functional configuration of a system 1000 according to this embodiment. As shown in the figure, this system 1000 has multiple subsystems, and the system 1000 as a whole determines the blending ratio of a flowability modifier to be added to a material (e.g., a recycled material or a raw material partially containing a recycled material), and manufactures a product using the material to which the flowability modifier has been added in the determined blending ratio.

[0014] Specifically, the system 1000 has the following subsystems: a production management system 100, a manufacturing supervision system 200, a blending ratio determination system 300, and a manufacturing execution system 400. The subsystems cooperate with each other to execute various processes.

[0015] In addition to the subsystems, the system 1000 also has functional units (not shown). Specifically, the system 1000 has an input receiving unit that receives instructions and information input from a user via an input device included in the system 1000. The system 1000 also has an output display unit that generates display information (screen information) to be displayed on an output device (display) included in the system 1000.

[0016] The system 1000 also has a storage unit that stores various information. Specifically, the storage unit has a mixture ratio information DB320, a fluidity information DB340, and a prediction model DB360 used in the mixture ratio determination system 300. The storage unit also has an injection molding process DB420 that stores a learning injection molding process 421, a trial injection molding process 422, and a mass production injection molding process 423 used in the manufacturing execution system 400.

[0017] The system 1000 also has a communication unit that communicates information with an external device (e.g., the manufacturing facility 500). Specifically, the communication unit is connected to the external device so as to be able to communicate with each other via a communication path (CN2 shown in FIG. 8) such as a public network such as the Internet, a LAN (Local Area Network), or a WAN (Wide Area Network).

[0018] The input reception unit, output display unit, storage unit, and communication unit cooperate with the functional units of each subsystem to execute various processes.

[0019] Next, the functional configuration of the production management system 100 will be described.

[0020] The production management system 100 functions as a processing unit that executes various processes related to the management of production plans. Specifically, the production management system 100 has a production plan management unit 110. The production plan management unit 110 is a functional unit that generates a production plan including production specifications, quantities, and timing in accordance with the order status and inventory status.

[0021] Next, the functional configuration of the manufacturing control system 200 will be described.

[0022] The manufacturing control system 200 functions as a processing unit that performs overall processing related to manufacturing, such as issuing a predetermined type of manufacturing instruction to the manufacturing execution system 400. Specifically, the manufacturing control system 200 has a manufacturing condition determination unit 210, a blending ratio acquisition unit 220, and a manufacturing instruction unit 230.

[0023] The manufacturing condition determination unit 210 is a functional unit that determines manufacturing conditions using a production plan generated by the production plan management unit 110. The manufacturing conditions include, for example, information that identifies the manufacturing equipment 500 of the product (in this example, including manufacturing equipment 510 such as an injection molding machine used in product manufacturing), information that identifies the model number of the material used in manufacturing (hereinafter, may be referred to as material model number) and its manufacturing lot (hereinafter, may be referred to as material manufacturing lot), the model number of the flowability modifier, the desired flowability value, and the quantity and production time of the products to be manufactured.

[0024] Here, the material model number is identification information (e.g., a number) that is determined by the material manufacturer to identify the characteristics of the material. The material production lot is the minimum unit of production or delivery determined by the material manufacturer for materials with the same model number. The model number of the flowability modifier is identification information (e.g., a number) that is determined by the material manufacturer to identify the characteristics of the flowability modifier provided by the flowability modifier manufacturer. The target flowability value is a value that indicates the flowability of the material during injection molding, and is a flowability value obtained by adding the flowability modifier to the material at a specified blend ratio. The blend ratio of the flowability modifier indicates the ratio of the amount of the flowability modifier to the amount of the material, and is managed, for example, by weight ratio (weight percent) or parts by weight (PHR: Part Hundred Resin).

[0025] The blending ratio acquisition unit 220 is a functional unit that acquires information on the blending ratio of the fluidity modifier to the material from the blending ratio determination system 300. Specifically, the blending ratio acquisition unit 220 acquires the blending ratio of the fluidity modifier from the blending ratio determination system 300 by sending basic information necessary for determining the blending ratio of the fluidity modifier to the blending ratio determination system 300. Note that such basic information includes, for example, the material model number, the material production lot, the model number of the fluidity modifier, and the desired fluidity value. Note that, as described below, when the blending ratio corresponding to the material model number, the material production lot, and the model number of the fluidity modifier is not yet determined, the desired fluidity value is used by the blending ratio determination system 300 when determining such blending ratio using the trained prediction model.

[0026] In addition, when the mixing ratio acquisition unit 220 acquires the mixing ratio for the combination of the material production lot to be used and the model number of the fluidity modifier from the mixing ratio determination system 300, it requests the production instruction unit 230 to mass-produce the product based on the aforementioned production conditions determined by the production condition determination unit 210 and the mixing ratio of the fluidity modifier acquired from the mixing ratio determination system 300.

[0027] In addition, when the mixing ratio acquisition unit 220 acquires information from the mixing ratio determination system 300 indicating that the mixing ratio for the combination of the material production lot of the material model number to be used and the model number of the fluidity modifier has not been determined, i.e., when a learned predictive model corresponding to the combination of the material model number and the model number of the fluidity modifier exists but the mixing ratio of the fluidity modifier for that material production lot has not been determined, the mixing ratio acquisition unit 220 requests the production instruction unit 230 to perform trial molding to determine the mixing ratio.

[0028] In addition, when the mixing ratio acquisition unit 220 acquires information from the mixing ratio determination system 300 indicating that there is no learned predictive model corresponding to the combination of the material model number and the fluidity modifier model number used, it requests the manufacturing instruction unit 230 to perform learning molding to generate such a predictive model.

[0029] The trial molding refers to a process of injection molding a production lot of the material to be used without adding any flowability modifier in order to determine the compounding ratio of the flowability modifier.

[0030] Moreover, the learning molding is an injection molding process for generating a learned prediction model corresponding to a combination of the material model number and the flowability modifier model number used, and is an injection molding process performed at multiple blending ratios for each combination of multiple material manufacturing lots for the material model number used and the flowability modifier model number used. Note that the multiple blending ratios also include the case where the blending ratio of the flowability modifier is "0 (zero)", that is, the flowability modifier is not added.

[0031] The manufacturing instruction unit 230 is a functional unit that instructs the manufacturing execution system 400 to execute manufacturing. Specifically, the manufacturing instruction unit 230 outputs a manufacturing (molding) instruction for mass production, trial molding, or learning molding to the manufacturing execution unit 410 based on a request from the compounding ratio acquisition unit 220.

[0032] Specifically, when the manufacturing instruction unit 230 instructs mass production, it outputs to the manufacturing execution unit 410 instructions for executing the injection molding process for mass production based on information identifying the manufacturing equipment 500 for the product determined by the manufacturing condition determination unit 210, information identifying the material model number and the material manufacturing lot used in production, information identifying the model number of the fluidity modifier, the desired fluidity value, the quantity and production time of the product to be manufactured, and the blending ratio of the fluidity modifier obtained from the blending ratio determination system 300.

[0033] Furthermore, when the manufacturing instruction unit 230 instructs trial molding, it instructs trial molding based on information specifying the manufacturing equipment 500 for the product determined by the manufacturing condition determination unit 210 and information specifying the material model number and the material production lot used in the manufacturing. Specifically, the manufacturing instruction unit 230 outputs to the manufacturing execution unit 410 an instruction to execute an injection molding process for trial molding for the material production lot of the material model number to be used in a state where no fluidity modifier has been added.

[0034] In addition, when the manufacturing instruction unit 230 instructs the learning molding, it outputs to the manufacturing execution unit 410 an instruction to execute the injection molding process of the learning molding based on information specifying the manufacturing equipment 500 of the product determined by the manufacturing condition determination unit 210, information specifying the material model number used in the manufacturing, information specifying multiple material manufacturing lots for the material model number, information specifying the model number of the fluidity modifier, and multiple compounding ratios in each combination of the material manufacturing lot and the model number of the fluidity modifier. Note that the multiple compounding ratios in each combination of the material manufacturing lot and the model number of the fluidity modifier refer to multiple arbitrary compounding ratios (e.g., compounding ratio of the fluidity modifier = 0%, 1%, 2%, 3%) set for each combination in each combination (a / M, b / M, c / M, d / M) of each of multiple material manufacturing lots (e.g., a, b, c, d) corresponding to the material model number used and the model number M of the fluidity modifier used.

[0035] Next, the functional configuration of the blend ratio determination system 300 will be described.

[0036] The blending ratio determination system 300 functions as a processing unit that performs processing for determining the blending ratio of the fluidity modifier. Specifically, the blending ratio determination system 300 has a blending ratio determination unit 310, a fluidity information generation unit 330, and a prediction model generation unit 350.

[0037] The compounding ratio determination unit 310 is a functional unit that determines the compounding ratio between the material production lot and the fluidity modifier. Specifically, when the compounding ratio determination unit 310 receives a compounding ratio acquisition request including the material model number, the material production lot, the model number of the fluidity modifier, and the target fluidity value from the compounding ratio acquisition unit 220, the compounding ratio determination unit 310 refers to the compounding ratio information DB 320. In addition, when a compounding ratio corresponding to the combination of the material model number, the material production lot, and the model number of the fluidity modifier related to the acquisition request is stored in the compounding ratio information DB 320, the compounding ratio determination unit 310 outputs the compounding ratio to the compounding ratio acquisition unit 220.

[0038] In addition, when the mixture ratio corresponding to the combination of the material model number, the material production lot, and the model number of the fluidity modifier related to the acquisition request is not stored in the mixture ratio information DB 320, the mixture ratio determination unit 310 refers to the prediction model DB 360. In addition, when a learned prediction model corresponding to the combination of the material model number and the model number of the fluidity modifier related to the acquisition request exists, the mixture ratio determination unit 310 outputs information indicating that the mixture ratio is undecided to the mixture ratio acquisition unit 220.

[0039] In addition, if there is no learned prediction model corresponding to the combination of the material model number related to the acquisition request and the model number of the fluidity modifier, the mixing ratio determination unit 310 outputs information indicating that such a prediction model does not exist to the mixing ratio acquisition unit 220.

[0040] In addition, the compounding ratio determination unit 310 inputs the fluidity information generated by the trial molding (fluidity information corresponding to the material production lot related to the compounding ratio acquisition request) and the value indicating the desired fluidity included in the acquisition request into the learned prediction model corresponding to the combination of the material model number related to the acquisition request and the model number of the fluidity modifier, thereby determining the compounding ratio in the combination of the material production lot to be used and the model number of the fluidity modifier (i.e., the compounding ratio determination unit 310 determines the value output from the prediction model as the compounding ratio of the fluidity modifier). In addition, the compounding ratio determination unit 310 stores the material model number, material production lot, and model number of the fluidity modifier related to the acquisition request in the compounding ratio information DB320 in association with the determined compounding ratio.

[0041] The fluidity information generating unit 330 is a functional unit that acquires the value output from the sensor 520 included in the manufacturing equipment 500 (for example, a mold in this case) and the product quality (for example, product weight) as measured values ​​in the trial molding or learning molding, and generates fluidity information. The fluidity information generating unit 330 also associates the generated fluidity information with the material model number, material production lot, and model number of the fluidity modifier used in the trial molding or learning molding, and stores them in the fluidity information DB 340.

[0042] Here, the fluidity information refers to a feature quantity correlating with the physical properties of the material obtained from the measurement value of the sensor 520. Specifically, one type of fluidity information is a feature quantity correlating with the fluidity of a material production lot (e.g., peak pressure, maximum differential value of pressure, integral value of pressure, etc.). Such a feature quantity is a feature quantity calculated from a measurement value during trial molding, i.e., a measurement value by the sensor 520 when a material production lot to which no fluidity modifier has been added is used.

[0043] Moreover, one of the pieces of fluidity information is a feature value (e.g., peak pressure, maximum differential value of pressure, integral value of pressure, etc.) that correlates with the fluidity of a material production lot to which a fluidity modifier is added at a predetermined blending ratio. Such feature values ​​are feature values ​​calculated from the measurement values ​​by the sensor 520 when using the measurement values ​​during learning molding, i.e., a material production lot to which no fluidity modifier is added (when the blending ratio of the fluidity modifier is zero), and feature values ​​calculated from the measurement values ​​by the sensor 520 when using a material production lot to which a fluidity modifier is added at a predetermined blending ratio.

[0044] In addition, such a feature correlated with fluidity may be a single feature obtained from a single sensor, or may be a dimension-reduced feature obtained by reducing the dimensions of multiple features obtained from multiple sensors. In addition, when performing dimension reduction, a method such as UMAP (Uniform Manifold Approximation and Projection) or t-SNE (t-distributed Stochastic Neighbor Embedding) may be used.

[0045] The prediction model generating unit 350 is a functional unit that generates a trained prediction model for the material. Specifically, the prediction model generating unit 350 generates a trained prediction model that predicts the blending ratio of the fluidity modifier so that the new material production lot to be used has the desired fluidity during production by learning the feature amount obtained in the training molding and the material model number and the blending ratio of the fluidity modifier for the material production lot at that time. The trained prediction model is model information generated for a combination of the material model number and the model number of the fluidity modifier. Therefore, a new material production lot of the same material model number to be used in production (meaning that the blending ratio is not yet determined) and the value of the desired fluidity are input to the trained prediction model corresponding to the combination of the material model number and the model number of the fluidity modifier, and the blending ratio of the fluidity modifier that will obtain the fluidity is output.

[0046] The predictive model generation unit 350 may construct a trained predictive model by learning, for a material whose fluidity is known in advance by measurement using another measuring means, feature quantities that correlate with the fluidity when no fluidity modifier is added (i.e., the fluidity when the fluidity modifier has a known fluidity and the blending ratio of the fluidity modifier is zero), as well as feature quantities that correlate with the fluidity when the fluidity modifier is added in any of a plurality of blending ratios and the blending ratios at that time.

[0047] Furthermore, the prediction model generating unit 350 stores the generated trained prediction model in association with the material model number and the model number of the fluidity modifier in the prediction model DB 360. Note that as the prediction model, a supervised prediction model such as ridge regression or a support vector machine (SVM) may be used.

[0048] In addition, the predictive model generation unit 350 performs active learning to re-learn the trained predictive model by associating features correlated with the fluidity of the material production lot obtained by trial molding and mass production with the compounding ratio of the fluidity modifier.

[0049] Next, the functional configuration of the manufacturing execution system 400 will be described.

[0050] The manufacturing execution system 400 functions as a processing unit that manufactures products by executing a predetermined manufacturing process based on instructions from the manufacturing control system 200. Specifically, the manufacturing execution system 400 has a manufacturing execution unit 410 and a quality inspection unit 430.

[0051] The manufacturing execution unit 410 is a functional unit that manufactures products by executing a predetermined manufacturing process based on instructions from the manufacturing supervision system 200. Specifically, when the manufacturing execution unit 410 receives an instruction for learning molding from the manufacturing instruction unit 230, it acquires a learning injection molding process 421 from the injection molding process DB 420. In addition, the manufacturing execution unit 410 outputs an instruction to execute learning molding according to the learning molding process at multiple compounding ratios for each combination of multiple material manufacturing lots and flowability modifiers included in the instruction to the manufacturing equipment 510.

[0052] Furthermore, when the manufacturing execution unit 410 is instructed to perform trial molding by the manufacturing instruction unit 230, it acquires a trial injection molding process 422 from the injection molding process DB 420. Furthermore, the manufacturing execution unit 410 outputs an instruction to the manufacturing equipment 510 to perform trial molding according to the trial injection molding process 422 using the material manufacturing lot included in the instruction, without adding a flowability modifier.

[0053] Furthermore, when the manufacturing execution unit 410 is instructed to carry out mass production by the manufacturing instruction unit 230, it acquires a mass production injection molding process 423 from the injection molding process DB 420. Furthermore, the manufacturing execution unit 410 outputs to the manufacturing equipment 510 an execution instruction for mass production in accordance with the mass production injection molding process 423 based on the manufacturing conditions and compounding ratio included in the instruction.

[0054] The quality inspection unit 430 is a functional unit that inspects the quality of the product. Specifically, the quality inspection unit 430 inspects the quality of the product based on, for example, shape characteristics, mechanical and optical characteristics, and the like.

[0055] In addition, the quality inspection unit 430 outputs the material model number for which the quality inspection was performed, the material production lot, the model number of the fluidity modifier, and the quality inspection results to the fluidity information generation unit 330 of the blending ratio determination system 300 in association with each other.

[0056] The functional configuration of each subsystem has been described above.

[0057] Next, various types of information stored in the storage unit of the present system 1000 will be described.

[0058] The mixing ratio information DB320 used in the mixing ratio determination system 300 stores, in correspondence with one another, information identifying a material model number, information identifying a material production lot, information identifying a model number of a fluidity modifier, and the mixing ratio of the fluidity modifier.

[0059] In addition, the fluidity information DB340 used in the blending ratio determination system 300 stores information indicating features correlated with fluidity, calculated from measured values, for example, information identifying the material model number, information identifying the material production lot, and information identifying the model number of the fluidity modifier, all in correspondence with each other.

[0060] Further, the prediction model DB 360 used in the blending ratio determination system 300 stores a learned prediction model corresponding to each combination of, for example, a material model number and a flowability modifier model number.

[0061] Furthermore, the injection molding process DB 420 used in the manufacturing execution system 400 stores various manufacturing (molding) processes for executing learning molding, trial molding, and mass production.

[0062] The storage unit also stores predetermined information such as the order status and inventory status of the product, and the operating status of the manufacturing equipment 500 in the manufacturing factory.

[0063] The communication unit of the system 1000 communicates information between the blending ratio determining system 300 and the manufacturing equipment 500. The communication unit also communicates information between the manufacturing execution system 400 and the manufacturing equipment 500.

[0064] The manufacturing facility 500 includes a manufacturing device 510 that manufactures products based on the manufacturing process output from the manufacturing execution unit 410, and a plurality of various sensors 520 that acquire measured values ​​by measuring physical quantities during the manufacturing process. The sensor 520 also outputs the measured values ​​to the fluidity information generation unit 330 of the blending ratio determination system 300.

[0065] The schematic configuration of the system 1000 according to this embodiment has been described above.

[0066] [Explanation of operation] 2 is a flow diagram showing an example of a manufacturing process executed by the present system 1000. This process is started when the input receiving unit receives an instruction to execute the manufacturing process from a user via an input device, for example.

[0067] When the process starts, the production management system 100 determines a production plan (step S10). Specifically, the production plan management unit 110 of the production management system 100 acquires predetermined information such as order status and inventory status from the storage unit. In addition, the production plan management unit 110 generates screen information for displaying the information acquired from the storage unit via the output display unit, and displays this on an output device (display).

[0068] The user refers to the displayed order status and inventory status, and inputs the appropriate production specifications, quantities, and timing via the input device. When the production plan management unit 110 receives the input information via the input reception unit, it determines the production plan based on a predetermined algorithm.

[0069] The production plan management unit 110 may automatically determine the production plan without relying on input from the user by using a mathematical planning model and its algorithm for optimizing the entire logistics. The mathematical planning model and algorithm may be stored in the storage unit in advance.

[0070] Next, the manufacturing condition determination unit 210 of the manufacturing control system 200 determines the manufacturing conditions (step S20). Specifically, the manufacturing condition determination unit 210 acquires the production plan and the operating status of the manufacturing equipment 500 from the storage unit. In addition, the manufacturing condition determination unit 210 generates screen information for displaying the information acquired from the storage unit via the output display unit, and displays this on the output device (display).

[0071] The user refers to the displayed production plan and the operating status of the manufacturing equipment 500, and inputs manufacturing conditions related to an appropriate combination of the manufacturing equipment 500 and materials via the input device. When the manufacturing condition determination unit 210 receives the input information via the input reception unit, it determines the manufacturing conditions based on a predetermined algorithm. In addition, the manufacturing condition determination unit 210 outputs the manufacturing conditions including the model number of the material used in the manufacture of the product, the material manufacturing lot, the model number of the flowability modifier, etc. to the blending ratio acquisition unit 220.

[0072] The manufacturing condition determination unit 210 may automatically determine the manufacturing conditions without relying on input from the user by using a mathematical programming model and its algorithm for optimizing production efficiency. Moreover, such a mathematical programming model and algorithm may be stored in the storage unit in advance.

[0073] Next, the compounding ratio acquisition unit 220 outputs a compounding ratio acquisition request for the combination of the material model number, the material production lot, and the model number of the fluidity modifier included in the manufacturing conditions (step S30). Specifically, the compounding ratio acquisition unit 220 outputs a compounding ratio acquisition request including the material model number, the material production lot, the model number of the fluidity modifier, and the target fluidity value to the compounding ratio determination unit 310.

[0074] Next, the compounding ratio determination unit 310 judges whether or not there is a compounding ratio related to the acquisition request (step S40). Specifically, the compounding ratio determination unit 310 refers to the compounding ratio information DB 320, and when a compounding ratio in which the material model number related to the acquisition request, the material production lot, and the model number of the flowability modifier are associated is stored, it judges that there is a compounding ratio related to the acquisition request.

[0075] If it is determined that such a mixture ratio does not exist (No in step S40), the mixture ratio determination unit 310 shifts the process to step S70. On the other hand, if it is determined that such a mixture ratio exists (Yes in step S40), the mixture ratio determination unit 310 acquires the mixture ratio related to the acquisition request from the mixture ratio information DB 320, outputs it to the mixture ratio acquisition unit 220, and shifts the process to step S50.

[0076] In step S50, when the mixture ratio acquisition unit 220 acquires the mixture ratio from the mixture ratio determination unit 310, it requests the manufacturing instruction unit 230 to mass-produce the product based on the manufacturing conditions determined by the manufacturing condition determination unit 210 and the acquired mixture ratio. In addition, the manufacturing instruction unit 230 outputs an instruction for mass production to the manufacturing execution unit 410 based on the request from the mixture ratio acquisition unit 220.

[0077] Next, the manufacturing execution unit 410 outputs an execution instruction for mass production according to the mass production injection molding process 423 based on the manufacturing conditions and compounding ratio included in the instruction to the manufacturing equipment 510 via the communication unit, thereby executing mass production of products according to the production plan and manufacturing conditions (step S60). Furthermore, when the manufacturing execution unit 410 outputs the mass production injection molding process 423 to the manufacturing equipment 500, this flow ends.

[0078] The predictive model generation unit 350 performs active learning to re-learn the trained predictive model by associating the fluidity information of the material production lot for the material model number obtained in mass production (fluidity information in a state in which the fluidity modifier has been added in the blending ratio determined by the blending ratio determination unit 310) with the blending ratio of the fluidity modifier at that time.

[0079] In addition, in step S70, which is performed when it is determined in step S40 that there is no compounding ratio related to the acquisition request (No in step S40), the compounding ratio determination unit 310 determines whether or not there is a trained prediction model corresponding to the combination of the material model number and the model number of the fluidity modifier related to the acquisition request. Specifically, the compounding ratio determination unit 310 refers to the prediction model DB 360 and determines whether or not such a trained prediction model is stored.

[0080] If the learned prediction model is not stored, the compounding ratio determination unit 310 determines that such a learned prediction model does not exist (No in step S70), outputs information indicating that there is no learned prediction model corresponding to the combination of the material model number and the model number of the fluidity modifier related to the acquisition request to the compounding ratio acquisition unit 220, and proceeds to step S90. In step S90, a process for generating a learned prediction model is executed. The details of the process for generating a learned prediction model will be described later.

[0081] On the other hand, when the learned prediction model is stored, the mixture ratio determination unit 310 determines that the learned prediction model exists (Yes in step S70), outputs information indicating that the mixture ratio related to the acquisition request is undetermined to the mixture ratio acquisition unit 220, and proceeds to step S80. In step S80, a mixture ratio determination process is executed. The details of the mixture ratio determination process will be described later.

[0082] Next, the process of generating a trained prediction model (step S90) will be described in detail.

[0083] FIG. 3 is a flow diagram illustrating an example of a process for generating a trained prediction model.

[0084] In step S91, the manufacturing supervision system 200, the manufacturing execution system 400, and the blending ratio determination system 300 cooperate to execute a learning molding to generate a learned prediction model. Specifically, when the blending ratio acquisition unit 220 acquires information indicating that there is no learned prediction model corresponding to the combination of the material model number related to the blending ratio acquisition request and the model number of the fluidity modifier, the blending ratio acquisition unit 220 requests the manufacturing instruction unit 230 to execute a learning molding to generate such a prediction model.

[0085] Furthermore, the manufacturing instruction unit 230 outputs to the manufacturing execution unit 410 an execution instruction for the learning molding, including a plurality of material manufacturing lots for the material model number used in the learning molding, a model number of the fluidity modifier to be used, and a plurality of compounding ratios for each combination of the material manufacturing lot and the model number of the fluidity modifier, based on a request from the compounding ratio acquisition unit 220. The manufacturing instruction unit 230 may also output an execution instruction for the learning molding using a material whose fluidity is known in advance.

[0086] Furthermore, when the manufacturing execution unit 410 acquires an execution instruction for the learning molding from the manufacturing instruction unit 230, it acquires the learning injection molding process 421 from the injection molding process DB 420. Furthermore, the manufacturing execution unit 410 performs the learning molding by outputting an execution instruction for the learning injection molding process 421 according to the content of the execution instruction acquired from the manufacturing instruction unit 230 to the manufacturing equipment 510.

[0087] Next, the fluidity information generating unit 330 generates fluidity information using the measured values ​​in the learning molding (step S92). Specifically, the fluidity information generating unit 330 acquires the measured values ​​(sensor values ​​and product quality) output from the sensor 520 and the quality inspection unit 430 included in the manufacturing equipment 500 (for example, a mold in this case) in the learning molding. The fluidity information generating unit 330 also calculates a feature value correlating with the material property (fluidity) from the acquired measured value to generate fluidity information. The fluidity information generating unit 330 also associates the generated fluidity information with the material model number, the material production lot, the model number of the fluidity modifier, and each compounding ratio in each combination of the material production lot and the model number of the fluidity modifier, and stores them in the fluidity information DB 340. Note that, when the learning molding is performed using a material whose fluidity is known in advance, the fluidity information generating unit 330 also associates the fluidity with the fluidity and stores it in the fluidity information DB 340.

[0088] Next, the prediction model generating unit 350 generates a trained prediction model (step S93). Specifically, the prediction model generating unit 350 uses a predetermined supervised prediction model to learn the feature amount of the fluidity information, thereby generating a trained prediction model corresponding to the combination of the material model number in the learning molding and the model number of the fluidity modifier. Note that, when a material with known fluidity is used, the prediction model generating unit 350 may construct (generate) a trained prediction model that directly predicts fluidity from measured values ​​by having the prediction model learn the feature amount correlated with fluidity and the correlation with the fluidity.

[0089] Next, the prediction model generation unit 350 registers the generated trained prediction model in the prediction model DB 360 (step S94), and ends this flow. After ending this flow, the prediction model generation unit 350 returns the process to step S30 in FIG.

[0090] The process of generating a trained prediction model has been described above.

[0091] Next, the blending ratio determination process (step S80) will be described in detail.

[0092] FIG. 4 is a flow diagram showing an example of the blending ratio determination process.

[0093] In step S81, the manufacturing supervision system 200, the manufacturing execution system 400, and the blending ratio determination system 300 cooperate to execute trial molding and determine the blending ratio of the flowability modifier. Specifically, when the blending ratio acquisition unit 220 acquires information indicating that the blending ratio related to the acquisition request has not been determined, it requests the manufacturing instruction unit 230 to perform trial molding to determine the blending ratio.

[0094] Furthermore, based on a request from the compounding ratio acquisition unit 220, the manufacturing instruction unit 230 outputs to the manufacturing execution unit 410 an instruction to execute trial molding, including the material manufacturing lot of the material model number to be used in the trial molding.

[0095] Furthermore, when the manufacturing execution unit 410 acquires an instruction to execute trial molding from the manufacturing instruction unit 230, it acquires a trial injection molding process 422 from the injection molding process DB 420. Furthermore, the manufacturing execution unit 410 performs trial molding by outputting an instruction to execute the trial injection molding process 422 according to the content of the execution instruction acquired from the manufacturing instruction unit 230 to the manufacturing equipment 510.

[0096] Next, the fluidity information generating unit 330 generates fluidity information using the measured values ​​in the trial molding (step S82). Specifically, the fluidity information generating unit 330 acquires the measured values ​​(sensor values ​​and product quality) output from the sensor 520 and the quality inspection unit 430 in the manufacturing equipment 500 (for example, a mold in this case) in the trial molding. The fluidity information generating unit 330 also calculates a feature value correlating with the material characteristic (fluidity) from the acquired measured values ​​to generate fluidity information. The fluidity information generating unit 330 also associates the generated fluidity information with the material model number and the material manufacturing lot, and stores them in the fluidity information DB 340.

[0097] Next, the compounding ratio determination unit 310 inputs the generated fluidity information and the desired fluidity value into the trained prediction model corresponding to the combination of the material model number of the material production lot and the model number of the fluidity modifier for which the compounding ratio is to be obtained (related to the request for obtaining the compounding ratio), thereby determining the compounding ratio for the combination of the material production lot and the model number of the fluidity modifier for that material model number (step S83). Specifically, the compounding ratio determination unit 310 identifies a trained prediction model corresponding to the combination of the material model number of the material production lot and the model number of the fluidity modifier for which the compounding ratio is to be obtained from the prediction model DB360. In addition, the compounding ratio determination unit 310 identifies the fluidity information generated by trial molding using the material production lot from the fluidity information DB340. Then, the compounding ratio determination unit 310 inputs the identified fluidity information and the desired fluidity value included in the request for obtaining the compounding ratio into the identified trained prediction model, thereby acquiring the compounding ratio output from the trained prediction model, and determines this as the compounding ratio.

[0098] Next, the compounding ratio determination unit 310 associates the material model number, the material production lot, and the model number of the flowability modifier with the determined compounding ratio, and stores (registers) them in the compounding ratio information DB 320 (step S84).

[0099] Next, the prediction model generation unit 350 performs active learning to re-learn the learned prediction model by associating the fluidity information (fluidity information when no fluidity modifier is added) of the material production lot for the material model number obtained by trial molding with the blending ratio of the fluidity modifier determined by the blending ratio determination unit 310 using the learned prediction model to obtain the target fluidity (step S85). After performing the re-learning, the prediction model generation unit 350 ends this flow. After the end of this flow, the prediction model generation unit 350 returns the process to step S30 in FIG. 2.

[0100] The blend ratio determination process has been described above.

[0101] Next, an outline of an experimental example for verifying the effects obtained when the system 1000 according to this embodiment is implemented will be described.

[0102] 5 is a diagram showing an outline of an experimental example for verifying the effect of this embodiment. As the experimental example, an experimental example in an injection molding process for manufacturing a resin product (for example, a product using a plastic resin) is shown. The illustrated mold 600 has a structure in which resin flows into a two-cavity dumbbell test piece-shaped product forming section 603 by a four-point side gate method via a sprue 601 and a runner 602. In the experimental example, a plurality of pressure sensors and resin temperature sensors are arranged in each of a sensor arrangement section 604 provided on the runner 602 and a sensor arrangement section 605 provided on the product forming section 603. Then, measured values ​​of the pressure and temperature changes over time were obtained from these sensors.

[0103] Among the measured values ​​obtained in the experiment, multiple feature quantities correlating with fluidity were obtained. In this experimental example, the peak pressure, the pressure integral value during the injection process, the pressure integral value from the start to the end of injection, and the maximum differential value of the pressure were obtained. In addition, the product weight was measured as an indicator of product quality. Note that here, the pressure integral value during the injection process was used as the feature quantity correlating with the fluidity of the material.

[0104] In addition, one model number of polypropylene derived from waste home appliances was used as the material used in the production, and three different production lots of that model number were used. In addition, a flowability modifier manufactured by Company A was used as the flowability modifier.

[0105] In addition, the predicted and measured values ​​were compared for the pressure integral value during the injection process, which indicates the fluidity of the material after the addition of the fluidity modifier, which is the objective variable. For the learning data, the predicted and measured values ​​matched well. Similarly, for the validation data, the predicted and measured values ​​matched well. This confirmed the generalization performance of the trained prediction model that was generated.

[0106] As described above, prediction accuracy can be continuously improved by re-learning and updating the prediction model using feature quantities obtained from measurements in trial molding and mass production.

[0107] The above describes the system 1000 according to the present embodiment. With such a system, even when a general-purpose injection molding machine is used, it is possible to determine a more appropriate blending ratio of a flowability modifier that can compensate for differences in flowability between material production lots and stabilize quality.

[0108] In particular, this system can determine the compounding ratio of the flow modifier that will provide the desired flowability even in a general-purpose injection molding machine based on the measured values ​​obtained from the sensor during the injection molding process. In addition, by determining the compounding ratio of the flow modifier from the measured values ​​obtained from the sensor during the injection molding process, the difference in flowability for each material (material production lot) can be compensated for, and the flowability during the injection molding process can be automatically made uniform for mass production. Therefore, according to this system, even when materials with large variations, such as recycled materials, are used in a general-purpose injection molding machine, the quality can be stabilized without relying on skilled workers. As a result, this system makes it possible to utilize recycled materials even in products that require higher precision, and can contribute to the realization of a resource-circulating society and carbon neutrality.

[0109] In addition, by associating the fluidity information (feature of fluidity) obtained from trial molding and mass production with the compounding ratio of the fluidity modifier and re-learning the trained prediction model, the accuracy of the trained prediction model can be improved each time it is used.

[0110] For example, when multiple users use the blending ratio determination system 300, the feature quantities of the fluidity of the material (material production lot) differ slightly depending on the conditions of the injection molding machine, mold, sensor, etc. used by each user as manufacturing equipment. Therefore, the learned prediction model may perform machine learning associated with the conditions of these manufacturing equipment for each user, and when determining the blending ratio of the fluidity modifier, a prediction model corresponding to these conditions may be used to obtain a more accurate blending ratio of the fluidity modifier.

[0111] Alternatively, a learning molding may be performed on a material whose fluidity is known in advance, and a learned prediction model that directly predicts fluidity based on information obtained from the sensor 520 may be constructed by learning the correlation between the obtained feature amount and fluidity. In this case, by using the fluidity as the objective variable, a learned prediction model constructed without relying on an injection molding machine or a mold can be generated. Note that any parameter such as melt flow rate or melt viscosity can be used as an index of fluidity.

[0112] Note that this embodiment includes both cases where the same business operator generates and re-trains the trained prediction model and mass-produces products using the mixing ratios output from the trained prediction model, and where the business operator who generates and re-trains the trained prediction model is different from the business operator who mass-produces products using the mixing ratios output from the trained prediction model.

[0113] Next, the hardware configuration of the system 1000 according to this embodiment will be described.

[0114] 6 is a diagram showing an example of a hardware configuration of the present system 1000. The present system 1000 can be realized, for example, by using one computer 700 (information processing device).

[0115] Specifically, the computer 700 includes an arithmetic unit 710, a memory 720, a storage device 730, an input device 740, an output device 750, a communication device 760, and a media interface 770, and these devices are electrically connected to each other via a communication path CN1. The communication path CN1 is, for example, an internal bus or a LAN (Local Area Network).

[0116] The computing device 710 is configured, for example, by a processor (including a microprocessor, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or other semiconductor devices).

[0117] The memory 720 is a non-volatile or volatile storage medium such as a random access memory (RAM) or a read only memory (ROM).

[0118] The storage device 730 is a device that stores computer programs 731 that realize the functional units of each subsystem, various information used in the processing of the entire system 1000 or each subsystem, etc. The storage device 730 is a rewritable storage medium such as a flash memory, a hard disk, or an SSD (Solid State Drive).

[0119] The input device 740 is a device through which a user (e.g., an operator using the system) inputs information and instructions to the computer 700. Examples of the input device 740 include a keyboard, a touch panel, a pointing device such as a mouse, and a voice input device such as a microphone. The output device 750 (display device) is a device that outputs (displays) information generated by the computer 700 or information obtained from an external device. Examples of the output device 750 include a display, a printer, and a voice synthesizer.

[0120] The communication device 760 is a device that allows an external information processing device and the computer 700 to communicate with each other via the communication path CN2. The external information processing device may be another computer 700 (not shown) or an external storage device 800. The external storage device 800 may be, for example, a Universal Serial Bus (USB) memory, a memory card, or a hard disk.

[0121] The computer 700 can read information (such as unique information and production performance information used by the computer 700 of the present system 1000) and computer programs stored in the external storage device 800. The computer 700 can also transmit all or part of the computer program 731 and information stored in the storage device 730 to the external storage device 800 for storage therein.

[0122] The medium interface 770 is a device for reading and writing data from and to an external recording medium 810 (such as a USB memory, a memory card, or a DVD). The computer 700 can transfer the computer program 731 and information from the external recording medium 810 to the storage device 730, and can also transfer all or part of the computer program 731 and information stored in the storage device 730 to the external recording medium 810 for storage therein.

[0123] The hardware configuration of the system 1000 according to this embodiment has been described above.

[0124] The processing units of the system 1000, i.e., the functional units of each subsystem, are realized by a program 731 that causes the arithmetic device 710 to perform processing. This program is stored in, for example, the storage device 730 or the external storage device 800, and is loaded onto, for example, the memory 720 when the program 731 is executed, and is executed by the arithmetic device 710.

[0125] Moreover, the input reception unit is realized by the input device 740. Moreover, the output display unit is realized by the output device 750. Moreover, the storage unit is realized by the memory 720, the storage device 730, or the external storage device 800, or a combination of these. Moreover, the communication unit is realized by the communication device 760.

[0126] Furthermore, the configurations, functions, processing units, processing means, etc. of the present system 1000 may be partly or entirely realized in hardware, for example by designing them as integrated circuits. Furthermore, the present system 1000 may be partly or entirely realized in software, or may be realized by a combination of software and hardware. Furthermore, the present system 1000 may use hardware having fixed circuits, or may use hardware having at least some of its circuits that are changeable.

[0127] As described above, each subsystem can also be realized by a user (operator) performing some or all of the functions without using dedicated software or hardware.

[0128] In this embodiment, the case where the system 1000 is realized from one computer 700 has been described as an example, but the present invention is not limited to this, and one or more systems can be constructed by linking multiple computers 700. The configuration of such a system will be described in detail in the second and third embodiments below.

[0129] Second Embodiment In the above-described first embodiment, a case has been described in which the present system 1000 is realized by one computer 700, but the present invention is not limited to such a configuration.

[0130] FIG. 7 is a diagram showing an example of the hardware configuration of the system 1000 according to the second embodiment. As shown in the figure, the system 1000 according to the second embodiment includes a computer 700a (hereinafter, sometimes referred to as the compounding ratio determination computer 700a) that realizes the compounding ratio determination system 300, and a plurality of computers 700b (hereinafter, sometimes referred to as the user side computers 700b) that realize the production management system 100, the manufacturing supervision system 200, and the manufacturing execution system 400, which are other subsystems, each of which is connected to be able to communicate with each other via a communication path CN2. The user side computer 700b may be a single computer. The same components are denoted by the same reference numerals, and detailed description thereof will be omitted.

[0131] With such a system configuration, even when a general-purpose injection molding machine is used, it is possible to determine a more appropriate blending ratio of the flowability modifier that can compensate for the difference in flowability for each material production lot and stabilize the quality. In particular, in this system, the blending ratio determination side computer is placed on the cloud, and each user side computer requests the blending ratio determination side computer to determine the blending ratio via the communication path CN2, thereby obtaining the same effect as in the first embodiment. Note that, when the user side computer obtains the blending ratio from the blending ratio determination side computer, it is sufficient to transmit the material model number, the material production lot, the modifier of the flowability modifier, and the desired flowability value to the blending ratio determination side computer, in the same manner as the blending ratio acquisition unit in the above-mentioned manufacturing supervision system.

[0132] In addition, according to this configuration, the mixing ratio determination side computer (mixing ratio determination system) receives requests for mixing ratio determination from a plurality of user side computers, and therefore, the learned prediction model can be subjected to more re-learning, thereby improving the accuracy of the prediction model. In addition, the user side computer can carry out production based on the mixing ratio output using the learned prediction model with improved accuracy, so that the manufacturer can carry out production with higher reliability for the materials used.

[0133] It is to be noted that such a mixing ratio determination computer 700a may be owned, for example, by a business that generates and re-learns a trained predictive model, and the user computer 700b may be owned by a business that requests the mixing ratio determination computer 700a to obtain a mixing ratio using such a trained predictive model, and that mass-produces products based on the obtained mixing ratio.

[0134] <Third embodiment> In the second embodiment described above, the computer 700a that realizes the blending ratio determination system 300 and the multiple computers 700b that realize the other subsystems are each connected to each other so that they can communicate with each other via communication path CN2, but the present invention is not limited to such a configuration.

[0135] 8 is a diagram showing an example of a hardware configuration of a system 1000 according to the third embodiment. As shown in the figure, in the system 1000 according to the third embodiment, each subsystem is realized by a separate computer 700o to 700r, which are communicatively connected to each other via a communication path CN2. As shown in the figure, in this embodiment, there may be multiple computers 700r that realize the manufacturing execution system 400. Note that the same components are denoted by the same reference numerals, and detailed description thereof will be omitted.

[0136] With this configuration, the same effect as in the first embodiment can be obtained. In particular, when there are multiple computers 700r that realize the manufacturing execution system 400, even if each is located in a distributed manufacturing plant, the processes executed by each computer 700 of the manufacturing execution system 400 can be consolidated and managed by the common production management system 100, manufacturing integration system 200, and blending ratio determination system 300.

[0137] The computer 700 according to the system of the present invention can also be provided as a computer (distribution server) that distributes a program 731 for realizing an arbitrary subsystem among subsystems such as the production management system 100, the manufacturing supervision system 200, the blending ratio determination system 300, and the manufacturing execution system 400 to other computers 700. For example, when the present system 1000 receives an acquisition request for the program 731 for realizing an arbitrary subsystem from an external computer 700 (not shown), the present system 1000 can also transmit the computer program 731 for realizing each subsystem stored in the storage unit to the external computer 700 via the communication unit.

[0138] In addition, it is conceivable that the calculator 700q relating to such a mixing ratio determination system is owned, for example, by a business that generates and re-learns a trained predictive model, and some or all of the other calculators 700p to 700r are owned by a business that requests the calculator 700q to obtain a mixing ratio using such a trained predictive model and mass-produces products based on the obtained mixing ratio.

[0139] Furthermore, the present invention is not limited to the above-mentioned embodiment and modified examples, and various modified examples are included within the scope of the same technical idea. For example, the above-mentioned embodiment has been described in detail to explain the present invention in an easy-to-understand manner, and is not necessarily limited to those having all of the configurations described. Furthermore, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. Furthermore, it is possible to add, delete, or replace a part of the configuration of each embodiment with another configuration.

[0140] In addition, in the above explanation, the control lines and information lines are those that are considered necessary for the explanation, and not all control lines and information lines in the product are necessarily shown. In reality, it can be considered that almost all components are connected to each other. [Explanation of symbols]

[0141] 1000 ··· system, 100 ··· production management system, 110 ··· production planning management section, 200 ··· manufacturing supervision system, 210 ··· manufacturing condition determination section, 220 ··· mix ratio acquisition section, 230 ··· manufacturing instruction section, 300 ··· mix ratio determination system, 310 ··· mix ratio determination section, 320 ··· mix ratio information DB, 330 ··· fluidity information generation section, 340 ··· fluidity information DB, 350 ··· prediction model generation section, 360 ··· prediction model DB, 400 ··· manufacturing execution system, 410 ··· manufacturing execution section, 42 0···Injection molding process DB, 430···Quality inspection section, 500···Manufacturing equipment, 510···Manufacturing devices, 520···Sensors, 700···Calculator, 710···Calculation unit, 720···Memory, 730···Storage device, 731···Computer program for implementing each functional unit of the subsystem, 740···Input device, 750···Output device, 760···Communication device, 770···Media interface, 800···External storage device, 810···External recording medium, CN1, CN2···Communication path

Claims

1. A system having at least one or more processors and one or more storage media, The processor, A trained prediction model is generated that learns a plurality of blending ratios for each combination of the model number and production lot of a resin material and the model number of a fluidity modifier, and information obtained from a sensor value of a sensor that is installed in a manufacturing facility that manufactures products by injection molding using the resin material and obtains information correlating with the fluidity of the resin material at the blending ratio, so that a resin material having the same model number as the resin material has a desired fluidity. A system characterized by:

2. 2. The system of claim 1, The processor, For a new resin material having the same model number as the resin material, the blending ratio obtained by inputting the information obtained from the sensor and the target fluidity into the trained prediction model is determined as the blending ratio of the fluidity modifier in the new resin material. A system characterized by:

3. 2. The system of claim 1, The processor, calculating a feature value correlated with the fluidity of the resin material from each of the sensor values ​​when the fluidity modifier is not added and when the fluidity modifier is added at a predetermined blending ratio; The calculated feature amount is input to a predetermined prediction model as information obtained from the sensor value, thereby generating the trained prediction model. A system characterized by:

4. 4. The system of claim 3, The processor, The information correlating with the fluidity corresponding to the blending ratio of the fluidity modifier not added, obtained by trial molding, for a new resin material having the same model number as the resin material; The blending ratio of the fluidity modifier corresponding to the target fluidity obtained by inputting the information correlating with the fluidity at a blending ratio when the fluidity modifier is not added and the target fluidity related to the requirements of the trial molding into the trained prediction model; and inputting the trained prediction model into the trained prediction model, thereby retraining the trained prediction model. A system characterized by:

5. 4. The system of claim 3, The processor, The information obtained from the sensor value by mass production in a state in which the fluidity modifier is added at a predetermined blending ratio for a resin material of the same model number as the resin material is input to the trained prediction model, and the predetermined blending ratio of the fluidity modifier obtained by inputting the information and the target fluidity into the trained prediction model is associated with each other and input to the trained prediction model, thereby retraining the trained prediction model. A system characterized by:

6. 2. The system of claim 1, The resin material is a material containing recycled material at least in part. A system characterized by:

7. 4. The system of claim 3, The feature amount includes: At least one of the following is included: peak pressure, pressure integral value during the injection process, pressure integral value from the start to the end of injection, and maximum differential value A system characterized by:

8. A method for determining a blending ratio performed by a system having at least one processor and one or more storage media, comprising: The processor, A step of generating a learned prediction model that predicts the blending ratio of the fluidity modifier so that a resin material having the same model number as the resin material has a desired fluidity by learning a plurality of blending ratios for each combination of the model number and production lot of the resin material and the model number of the fluidity modifier, and information obtained from a sensor value of a sensor that is installed in a manufacturing facility that manufactures products by injection molding using the resin material and obtains information correlating with the fluidity of the resin material at the blending ratio, so as to obtain a learned prediction model that predicts the blending ratio of the fluidity modifier so that a resin material having the same model number as the resin material has a desired fluidity. A method for determining a blend ratio.

9. The method for determining a blend ratio according to claim 8, The processor, A step of inputting information obtained from the sensor about a new resin material having the same model number as the resin material and the target fluidity into the trained prediction model to obtain a compounding ratio is performed as the compounding ratio of the fluidity modifier in the new resin material. A method for determining a blend ratio.

10. A program executed on a system having at least one or more processors and one or more storage media, The processor, A trained prediction model is generated that learns a plurality of blending ratios for each combination of the model number and production lot of the resin material and the model number of the fluidity modifier, and information obtained from a sensor value of a sensor that is installed in a manufacturing facility that manufactures products by injection molding using the resin material and obtains information correlating with the fluidity of the resin material at the blending ratio, to thereby predict the blending ratio of the fluidity modifier so that a resin material having the same model number as the resin material has a desired fluidity. A program for executing the process.

11. The program according to claim 10, The processor, The compounding ratio obtained by inputting the information obtained from the sensor about a new resin material having the same model number as the resin material and the target fluidity into the trained prediction model is determined as the compounding ratio of the fluidity modifier in the new resin material. A program for carrying out the process.

Images