Information processor, information processing method, and information processing program

Abstract

To improve flexibility and adaptability in model management, while improving accuracy of quality prediction of a resin molding.SOLUTION: An information processor (2) includes: a state prediction part (202) for predicting the state of a target mold, using a state prediction model (211) for predicting the state of a mold from manufacture data; and a quality prediction part (203) for predicting the quality of the resin molding using a quality prediction model (212) including an explanatory variable indicating the state of the mold.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, and an information processing program. [Background technology]

[0002] Patent Document 1 discloses a method for determining the quality of a resin molded product molded by an injection molding machine, in which a group of monitoring data, such as the temperature of the mold and the filling pressure when filling the mold with resin, is combined with waveform data in the same dimension to determine the quality. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2023-172022 Summary of the Invention [Problem to be solved by the invention]

[0004] Although this is not limited to injection molding, the contamination state of the mold may have a negative impact on the quality of the resin molded product. Therefore, in order to determine the quality of the resin molded product, it is necessary to take the contamination state of the mold into consideration, but this point is not disclosed in Patent Document 1, and there is room for improvement.

[0005] One possible method for taking mold contamination into account is to include data indicating the mold contamination state in the explanatory variables of a quality prediction model that predicts the quality of plastic molded products based on actual measured values ​​such as heater temperature and injection pressure. However, because the mold contamination state is prone to change over time, quality prediction models that use data indicating the mold contamination state as explanatory variables have problems with their flexibility and adaptability in management. For example, if data indicating the mold contamination state is included in the explanatory variables of a quality prediction model, the frequency of maintenance and tuning of the quality prediction model will have to be increased, but it is difficult to tune a quality prediction model that is used continuously in production in a timely manner.

[0006] An aspect of the present invention aims to improve the accuracy of quality prediction of resin molded products while improving flexibility and adaptability in model management. [Means for solving the problem]

[0007] In order to solve the above problems, an information processing device according to one embodiment of the present invention includes a condition prediction unit that predicts the condition of a target mold using a condition prediction model that predicts the condition of the mold used to manufacture the resin molded product from manufacturing data of the resin molded product, and a quality prediction unit that predicts the quality of a resin molded product manufactured with the target mold from the prediction result of the condition prediction unit using a quality prediction model for the resin molded product that includes explanatory variables that indicate the condition of the mold.

[0008] In order to solve the above-mentioned problems, an information processing method according to one embodiment of the present invention is an information processing method executed by one or more information processing devices, and includes: a state prediction step of predicting the state of a target mold using a state prediction model that predicts the state of the mold used to manufacture the resin molded product from manufacturing data of the resin molded product; and a quality prediction step of predicting the quality of a resin molded product manufactured with the target mold from the prediction result of the state prediction step using a quality prediction model of the resin molded product that includes explanatory variables that indicate the state of the mold.

[0009] The information processing device according to each aspect of the present invention may be realized by a computer. In this case, the information processing program that causes the computer to operate as each part (software element) of the information processing device to realize the information processing device on the computer, and the computer-readable recording medium on which the program is recorded, also fall within the scope of the present invention. [Effects of the Invention]

[0010] According to one aspect of the present invention, it is possible to improve the accuracy of quality prediction of resin molded products while improving the flexibility and adaptability of model management. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram showing an example of a main configuration of an information processing apparatus according to an embodiment of the present invention; [Figure 2] 1 is a diagram showing an overview of an information processing system according to an embodiment of the present invention. [Figure 3] FIG. 1 is a diagram illustrating an example of the configuration of an injection molding machine and its flow paths. [Figure 4] FIG. 2 is a schematic diagram showing an example of the flow of data such as manufacturing data during the manufacture of a resin molded product. [Figure 5] 10 is a flowchart illustrating an example of processing executed by an information processing apparatus according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0012] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the description of the drawings, the same parts are designated by the same reference numerals and the description thereof will be omitted.

[0013] (Outline of information processing system) An overview of the information processing system 5 according to this embodiment will be described with reference to Fig. 2. Fig. 2 is a diagram showing an overview of the information processing system 5. Fig. 2 also shows an overview of an injection molding machine. The information processing system 5 is a system that can be used to support injection molding, and includes information processing devices 1 and 2. Below, the injection molding to be supported will be described, followed by a description of the information processing device 1 and the information processing device 2.

[0014] Injection molding uses an injection molding machine that includes an injection unit, a mold, and a mold clamping unit. In injection molding, resin pellets are first fed from a hopper into a cylinder, and the resin in the cylinder is heated by a heater to form molten resin. Next, a screw is advanced by a motor, and the molten resin is injected into the mold. The injected molten resin is cooled and solidified inside the mold. After the molten resin has solidified, the crosshead is moved to open the mold along the tie bars, and the molded resin product is removed from the mold by an ejector mechanism.

[0015] The removed resin molded product is inspected for quality issues before being used as a finished product. On the other hand, resin molded products that do not meet quality standards are deemed defective. Figure 2 shows an example in which a spoon is used as a resin molded product, and its quality is determined using a visual inspection machine. The information processing system 5 can support any injection molding process. While the configuration of the injection molding machine used and the resin molded product to be injection-molded are not particularly limited, the information processing system 5 is particularly suitable for molding resins with low decomposition temperatures. For example, the information processing system 5 can be used effectively for molding biodegradable plastics. Specifically, the resin may be a poly(3-hydroxyalkanoate)-based resin or a polyhydroxyalkanoate resin such as poly(3-hydroxybutyrate-co-3-hydroxyhexanoate). Polyhydroxyalkanoate resins are biodegradable aliphatic polyesters.

[0016] The information processing device 1 generates a condition prediction model and a quality prediction model through machine learning using training data, and the information processing device 2 uses the condition prediction model and the quality prediction model to predict the quality of the resin molded product from the prediction data.

[0017] (Configuration examples of information processing device 1 and information processing device 2) 1 is a block diagram showing an example of the main configuration of the information processing device 1 and the information processing device 2. Below, each configuration of the information processing device 1 and the information processing device 2 will be described in order based on FIG.

[0018] 1, the information processing device 1 includes a control unit 10 that controls each unit of the information processing device 1, and a storage unit 11 that stores various data used by the information processing device 1. The information processing device 1 also includes a communication unit 12 that enables the information processing device 1 to communicate with other devices, an input unit 13 that accepts various data input to the information processing device 1, and an output unit 14 that enables the information processing device 1 to output various data. The control unit 10 also includes a training data acquisition unit 101 and a learning unit 102. The storage unit 11 stores training data sets 111 and 112.

[0019] The information processing device 1 may be configured as a general-purpose computer. When the information processing device 1 is configured as a general-purpose computer, the control unit 10 may be configured to include a processor, a read-only memory (ROM), a random access memory (RAM), etc. The processor can read programs from the ROM and execute various programs using the RAM as a working area. The storage unit 11 may be configured as a hard disk drive (HDD), a solid state drive (SSD), etc.

[0020] The input unit 13 is typically a keyboard or a mouse, but is not limited to these and may be any configuration that allows various data to be input. For example, the input unit 13 may be a microphone, a touch panel, etc. The output unit 14 is typically a display, but is not limited to these and may be any configuration that allows various data to be output. For example, the output unit 14 may be a speaker, a warning light, etc.

[0021] The training data sets 111 and 112 stored in the storage unit 11 will be described. The training data sets 111 and 112 are data sets including multiple pieces of training data (labeled data used in machine learning). Machine learning is a technique for learning features included in input data and generating a "model" that predicts results corresponding to newly input data. A state prediction model 211 and a quality prediction model 212, which will be described later, are trained models generated by machine learning using the training data sets 111 and 112, respectively.

[0022] As will be described in detail later, the condition prediction model 211 uses manufacturing data of a resin molded product as input data and predicts the condition of the mold used to manufacture the resin molded product. The predicted "mold condition" may be expressed, for example, as an evaluation value that evaluates the degree of mold contamination. As described above, when a biodegradable plastic with a low decomposition temperature is used for a resin molded product, the mold is more likely to be contaminated by thermal decomposition products. Therefore, by predicting the degree of mold contamination and using the prediction results to predict the quality of the resin molded product, the accuracy of the quality prediction of the resin molded product can be improved.

[0023] The "mold condition" to be predicted is not limited to the degree of mold contamination. The "mold condition" to be predicted may also be whether or not the mold needs to be cleaned, the number of injections remaining until cleaning is required, or the next time the mold will be cleaned. In this case, the "mold condition" may be expressed as an index value indicating whether or not the mold needs to be cleaned. In the following, as an example, unless otherwise specified, the condition prediction model 211 will be described as predicting the "degree of mold contamination" as the mold condition.

[0024] The quality prediction model 212 receives as input the predicted value (evaluation value for evaluating the degree of contamination of the mold) predicted by the condition prediction model 211 and manufacturing data of the resin molded product, and predicts the quality of the resin molded product manufactured when the manufacturing data is applied to the degree of contamination of the mold. In this way, the explanatory variables of the quality prediction model 212 include the degree of contamination of the mold.

[0025] The training data set 111 includes a plurality of pieces of training data used for machine learning of the state prediction model 211. The training data included in the training data set 111 is obtained by associating input data input to the state prediction model 211 with values ​​to be output by the state prediction model 211 as correct answer data. This input data can also be referred to as explanatory variables, and the correct answer data as objective variables.

[0026] As described above, the predicted value predicted by the condition prediction model 211 is used to predict quality by the quality prediction model 212. The condition prediction model 211 is a model that predicts the degree of mold contamination from manufacturing data of a resin molded product. Therefore, the training data included in the training dataset 111 is data in which the degree of mold contamination when the manufacturing data is applied to "input data," which is the manufacturing data of a resin molded product, is associated as "correct data."

[0027] An example of the "input data" is various process values ​​(PV) measured during the production of a resin molded product, such as the temperature of the resin being kneaded inside the screw, the temperature around the heaters installed in each part of the cylinder, the injection pressure at the injection port, screw position data, etc. These process values ​​are measured by temperature sensors, pressure sensors, position sensors, etc. installed in the injection molding machine.

[0028] Next, an example of "correct answer data" will be described. Correct answer data is data that indicates the objective variable of the condition prediction model 211, and the degree of mold contamination is the correct answer data. More specifically, when certain "input data" is input to the condition prediction model 211, the degree of mold contamination that the condition prediction model 211 should output is the "correct answer data" that corresponds to that "input data."

[0029] The training data is generated by associating the above-described input data with the correct answer data, and is stored as a training data set 111 in the storage unit 11. The process of generating the training data set 111 may be performed by the information processing device 1 or another device.

[0030] When generating the state prediction model 211, the training data acquisition unit 101 refers to the storage unit 11 and acquires the training data set 111. The training data acquisition unit 101 outputs the acquired training data set 111 to the learning unit 102.

[0031] The learning unit 102 generates a condition prediction model 211 for predicting the degree of mold contamination through machine learning using the training data set 111 acquired from the training data acquisition unit 101. The condition prediction model 211 is a prediction model that uses manufacturing data of the resin molded product as an explanatory variable and the degree of mold contamination as a target variable. The prediction algorithm is not particularly limited. For example, the condition prediction model 211 may be a neural network model, a multiple regression model, a random forest model, or the like.

[0032] Furthermore, the learning unit 102 generates a quality prediction model 212 in addition to the state prediction model 211. Like the state prediction model 211, the quality prediction model 212 is also a model generated by machine learning. However, in generating the quality prediction model 212, training data (data in which input data is associated with correct answer data) used for machine learning is different from the training data used for machine learning of the state prediction model 211. Specifically, the training dataset 112 is used when generating the quality prediction model 212. The training data included in the training dataset 112 will be described below.

[0033] The training data included in the training dataset 112 is a correspondence between "input data (which can also be referred to as explanatory variables)" and "ground truth data (which can also be referred to as objective variables)." This "input data" includes various data related to the quality to be predicted. Specifically, the "input data" includes the degree of contamination of the mold output from the condition prediction model 211. In addition to this, the "input data" also includes manufacturing data for resin molded products.

[0034] The "correct answer data" in the training data included in the training data set 112 may be data indicating the quality of the resin molded product. The "quality of the resin molded product" can be expressed by evaluation values ​​for evaluating the quality of the resin molded product, such as the defect rate, strength, dimensions, etc.

[0035] A data set generated by associating such input data with the correct answer data becomes training data for generating the quality prediction model 212. The training data generated in this manner is stored in the storage unit 11 as a training data set 112. The process of generating the training data set 112 may be performed by the information processing device 1 or another device.

[0036] When generating the quality prediction model 212, the training data acquisition unit 101 acquires the training data set 112 and outputs it to the learning unit .

[0037] The state prediction model 211 and the quality prediction model 212 generated by the learning unit 102 are transmitted to the information processing device 2 and stored in the storage unit 21 of the information processing device 2. Note that the state prediction model 211 and the quality prediction model 212 generated by the learning unit 102 may be stored in the storage unit 11.

[0038] On the other hand, the information processing device 2 includes a control unit 20 that controls each unit of the information processing device 2, and a storage unit 21 that stores various data used by the information processing device 2. The information processing device 2 also includes a communication unit 22 that enables the information processing device 2 to communicate with other devices, an input unit 23 that accepts various data input to the information processing device 2, and an output unit 24 that enables the information processing device 2 to output various data. The control unit 20 also includes a data acquisition unit 201, a state prediction unit 202, a quality prediction unit 203, and an output control unit 204. The storage unit 21 stores a state prediction model 211 and a quality prediction model 212.

[0039] The data acquisition unit 201 acquires input data to be input to the condition prediction model 211 and input data to be input to the quality prediction model 212. For example, when actual measurement values ​​such as resin temperature and injection pressure, which are manufacturing data of a resin molded product, are input data to the condition prediction model 211 and the quality prediction model 212, the data acquisition unit 201 acquires these actual measurement values. There are no particular limitations on the method for acquiring the manufacturing data, and for example, the data acquisition unit 201 may acquire these actual measurement values ​​from various sensors provided in an injection molding machine. The data acquisition unit 201 outputs the acquired manufacturing data to the condition prediction unit 202. Note that the manufacturing data acquired by the data acquisition unit 201 may include data for predicting the degree of contamination of a mold and data for predicting the quality of a resin molded product.

[0040] The condition prediction unit 202 predicts the value of an explanatory variable (an evaluation value that evaluates the degree of mold contamination) in a quality prediction model 212 that predicts the quality of a resin molded product from the manufacturing data of the resin molded product. Specifically, the condition prediction unit 202 predicts the degree of mold contamination from the manufacturing data of the resin molded product acquired by the data acquisition unit 201, using a condition prediction model 211 that is generated by learning the relationship between the manufacturing data of the resin molded product and the degree of mold contamination. The degree of mold contamination predicted by the condition prediction unit 202 is output to the quality prediction unit 203.

[0041] The quality prediction unit 203 predicts the quality of the resin molded product using the degree of contamination of the mold predicted by the condition prediction unit 202. Specifically, the quality prediction unit 203 predicts the quality of the resin molded product by inputting the degree of contamination of the mold predicted by the condition prediction unit 202 and other data (manufacturing data of the resin molded product) that are explanatory variables of the quality prediction model 212 into the quality prediction model 212. The quality of the resin molded product predicted by the quality prediction unit 203 is output to the output control unit 204.

[0042] The output control unit 204 causes the output unit 24 to output the quality of the resin molded product predicted by the quality prediction unit 203. The output may be in any form, and for example, the output control unit 204 may cause the predicted quality of the resin molded product to be output in at least one form of display output, audio output, or print output. Furthermore, the device that outputs the predicted quality of the resin molded product may be a device external to the information processing device 2. Such an output allows the operator to grasp the accurately predicted quality of the resin molded product.

[0043] Furthermore, the condition prediction unit 202 may output the predicted degree of mold contamination to the output control unit 204. Then, the output control unit 204 may output the degree of mold contamination acquired from the condition prediction unit 202 to the output unit 24. This output allows the operator to understand the degree of mold contamination and to make plans necessary for production management, such as when to clean the mold.

[0044] As described above, the information processing device 2 includes a condition prediction unit 202 that predicts the condition of a target mold using a condition prediction model 211 that predicts the condition of the mold used to manufacture the resin molded product from manufacturing data of the resin molded product, and a quality prediction unit 203 that predicts the quality of a resin molded product manufactured with the target mold from the prediction result of the condition prediction unit 202 using a quality prediction model 212 for the resin molded product that includes explanatory variables that indicate the condition of the mold.

[0045] (Regarding manufacturing data for resin molded products) In addition to the resin temperature and injection pressure described above, the manufacturing data for resin molded products may also include the difference between the decomposition temperature at which the molten resin thermally decomposes during injection molding and the temperature of the flow path that guides the molten resin to the mold, the time the molten resin remains in the flow path, etc. Here, data that can be included in the manufacturing data for resin molded products will be explained with reference to Fig. 3. Fig. 3 is a diagram showing an example of the configuration of an injection molding machine and its flow path.

[0046] A1 in Figure 3 shows an injection molding machine similar to that in Figure 2. As shown in Figure 3, the cylinder is provided with heaters h1 to h3 that heat the resin inside. In addition, a heater h4 is also provided in the flow path that guides the molten resin injected from the injection port e at the end of the cylinder to the cavity c in the mold. This flow path can also be called a hot runner. Note that the rear part of the cylinder (near the injection port e) can also be considered part of this flow path.

[0047] The temperature of the flow path can be measured by installing a temperature sensor near heater h4. Therefore, using the measurement value of the temperature sensor, it is possible to calculate the difference between the decomposition temperature at which the molten resin thermally decomposes and the temperature of the flow path that leads the molten resin to the mold. Furthermore, since the measurement value of the temperature sensor changes significantly depending on whether or not the molten resin is flowing in the flow path, it is also possible to determine the time that the molten resin remains in the flow path.

[0048] The above-mentioned flow paths may also be branched. A2 in Figure 3 shows an example of a branched flow path configuration. The flow paths include a first-stage flow path r1, a second-stage flow path r2, and a third-stage flow path r3. The molten resin injected from the cylinder first flows into the first-stage flow path r1. Flow path r1 branches into two at its end, each of which leads to the second-stage flow path r2. This results in two flow paths r2. Each of the two flow paths r2 then branches into four, each of which leads to the third-stage flow path r3. This results in eight flow paths r3. These eight flow paths r3 are connected to cavities c1 to c8, respectively. Using a mold with such branched flow paths allows for the simultaneous production of eight molded products in a single injection molding run. In a hot runner, each of the branched flow paths is controlled by a heater. To eliminate unevenness between cavities, the lengths of each flow path are generally designed to be equal.

[0049] When a mold includes multiple cavities, parameters related to multiple flow paths connected to each of the multiple cavities may be included in the manufacturing data for the resin molded product. In this case, the condition prediction unit 202 may predict the degree of contamination for each of the multiple cavities. This makes it possible to manage the need for mold cleaning on a cavity-by-cavity basis. To predict the degree of contamination for each of the multiple cavities, a model for predicting the degree of contamination for each of the multiple cavities may be generated. In other words, a condition prediction model 211 may be prepared for each cavity. In this case, the explanatory variables of the quality prediction model 212 may include data indicating the degree of contamination for each cavity, or a representative value (e.g., average, maximum, median, etc.) of the degree of contamination for each cavity may be included in the explanatory variables of the quality prediction model 212.

[0050] Furthermore, for example, by providing a temperature sensor in each of the flow paths connected to each cavity, it is possible to measure the temperature of the molten resin flowing through each path. This makes it possible to calculate, for each cavity, the difference between the decomposition temperature of the molten resin and the temperature of the path that guides the molten resin to the mold, as well as the residence time of the molten resin in the path. This also makes it possible to predict, for example, the quality of the resin molded product for each cavity.

[0051] The manufacturing data of a resin molded product is not limited to the various actual measurement values ​​measured during the manufacturing of the resin molded product. The manufacturing data of a resin molded product may be various setting values ​​(SV: Setting Value) set during the manufacturing of the resin molded product. The setting values ​​include the cylinder temperature, mold temperature, injection speed, injection time, pressure holding time, cooling time, etc., which are to be controlled.

[0052] Furthermore, the manufacturing data for the resin molded product may be an index value calculated from a set value and an actual measurement value corresponding to the set value. For example, the manufacturing data for the resin molded product may include a parameter indicating the flow state of the molten resin calculated using the output value of a heater provided along a flow path that guides the molten resin to a mold and that maintains the molten resin in the flow path at a predetermined temperature. This makes it possible to obtain reasonable prediction results that reflect changes in the flow state that are a sign of molding defects caused by mold contamination using an easily obtainable value, namely the heater output value.

[0053] For example, in the example of A1 in Figure 3, the output value of heater h4 can be used to calculate a parameter indicating the flow state of the molten resin. For example, when controlling the output of heater h4 so that the temperature of the molten resin flowing through the flow path falls within a predetermined range, the parameter may be the ratio or difference of the output of heater h4 relative to the normal state. This is because when the flow state of the molten resin deteriorates and the flow rate decreases, the amount of heat required per given time to keep the temperature of the molten resin within the predetermined range decreases, and the output of heater h4 drops significantly compared to the normal state.

[0054] In addition, when predicting the state of a mold when the manufacturing data of a resin molded product is the above-mentioned set value or index value, a state prediction model that has been trained using the set value or index value as input data can be used.

[0055] (Data flow during manufacturing) FIG. 4 is a schematic diagram showing an example of the flow of data such as manufacturing data during the manufacture of a resin molded product.

[0056] The data acquisition unit 201 acquires a plurality of pieces of manufacturing data 40, 41, 42, etc., such as resin temperature and injection pressure, when a resin molded product is being manufactured. The manufacturing data 40, 41, 42, etc. acquired by the data acquisition unit 201 are input to a state prediction model 211.

[0057] The state prediction model 211 predicts the degree of contamination of the mold based on the input manufacturing data 40, 41, 42, etc. In Fig. 4, the degree of contamination of the mold is shown as "mold state 50".

[0058] The degree of contamination of the mold predicted by the state prediction model 211 and the manufacturing data 40, 41, 42, . . . acquired by the data acquisition unit 201 are input to the quality prediction model 212.

[0059] The quality prediction model 212 predicts the quality of a resin molded product based on the input degree of contamination of the mold and the manufacturing data 40, 41, 42, etc. The prediction results are output as values ​​51 such as the defect rate, strength, dimensions, etc.

[0060] (Processing flow) Next, the flow of processing executed by the information processing device 2 will be described with reference to Fig. 5. Fig. 5 is a flowchart showing an example of processing executed by the information processing device 2.

[0061] In step S101, the data acquisition unit 201 acquires prediction data. The prediction data is manufacturing data indicating actual values ​​measured by various sensors when manufacturing a resin molded product, specifically, resin temperature, injection pressure, etc. The prediction data is used as input data for a state prediction model 211, and is also used as input data for a quality prediction model 212 (see FIG. 4).

[0062] The data acquiring unit 201 may acquire a set value instead of or in addition to the actual measurement value. There is no particular limitation on the method of acquiring the set value, but for example, the data acquiring unit 201 may acquire data input by an operator to the input unit 23 as the set value. Furthermore, if the information processing device 2 is configured with a PLC (Programmable Logic Controller), the data acquiring unit 201 may acquire data set by the PLC as the set value.

[0063] The process proceeds to step S102, where the state prediction unit 202 inputs the prediction data acquired in the process of step S101 into the state prediction model 211 to predict the degree of contamination of the mold. As an example, the state prediction model 211 outputs an evaluation value that evaluates the degree of contamination of the mold.

[0064] The process proceeds to step S103, where the quality prediction unit 203 predicts the quality of the resin molded product by inputting the prediction data acquired in the process of step S101 and the degree of mold contamination predicted in the process of step S102 into the quality prediction model 212. The quality prediction model 212 outputs evaluation values ​​for evaluating the quality of the resin molded product, such as the defect rate, strength, dimensions, etc., for example.

[0065] In step S104, the output control unit 204 outputs the quality of the resin molded product predicted in the processing of step S103 to the output unit 24. This allows the operator to grasp the quality of the resin molded product predicted with high accuracy.

[0066] As described above, the information processing method according to this embodiment includes a condition prediction step (S102) of predicting the condition of a target mold using a condition prediction model 211 that predicts the condition of the mold used to manufacture the resin molded product from manufacturing data of the resin molded product, and a quality prediction step (S103) of predicting the quality of a resin molded product manufactured with the target mold from the prediction result of the condition prediction step using a quality prediction model 212 for the resin molded product that includes explanatory variables that indicate the condition of the mold.

[0067] (Action and effect) As described above, according to this embodiment, the following advantageous effects can be obtained.

[0068] In this embodiment, the state of the target mold is predicted using a state prediction model 211 that predicts the state of the mold used to manufacture the resin molded product from the manufacturing data of the resin molded product, and the quality of the resin molded product manufactured with the target mold is predicted from the predicted state of the mold using a quality prediction model 212 for the resin molded product that includes explanatory variables that indicate the state of the mold.

[0069] As described above, this embodiment employs a two-stage prediction process in which output from the condition prediction model 211 is input to the quality prediction model 212. According to the two-stage prediction process, quality is predicted in two stages using two models, improving the accuracy of quality prediction. More specifically, because the quality prediction model 212 includes explanatory variables that indicate the mold condition predicted by the condition prediction model 211, a so-called "cascade effect" occurs, improving the accuracy of quality prediction of the resin molded product.

[0070] Furthermore, by configuring the condition prediction model 211 to predict the condition (degree of contamination) of the mold, which is prone to change over time, it is possible to maintain prediction accuracy by updating the condition prediction model 211 relatively frequently, while reducing the update frequency of the quality prediction model 212. This enables model management with increased flexibility and adaptability.

[0071] Furthermore, in machine learning, while a large number of variables are required to generate a predictive model that performs quality prediction as in this embodiment, by adopting a two-stage prediction process, fewer variables are required for each model, contributing to a reduction in the amount of required data. Furthermore, generating models with fewer variables also contributes to improved flexibility and adaptability in model management.

[0072] Furthermore, in hierarchical processing in which the output from the state prediction model 211 is input to the quality prediction model 212, by selecting or designing appropriate features for each layer, it becomes possible to automatically tune these models, thereby reducing the maintenance of the models.

[0073] Furthermore, the prediction results of the quality prediction model 212 may be used to optimize the manufacturing conditions (e.g., various temperatures and pressures) of the resin molded product. A well-known optimization method may be used for the optimization calculation. Well-known optimization methods include grid search, random search, genetic algorithm, and Bayesian optimization. The optimized manufacturing conditions may be output to the output unit 24. This allows the operator to understand the optimized manufacturing conditions, making it possible to manufacture products with the desired quality. Furthermore, when a PLC is used to control the injection molding machine, the optimized manufacturing conditions are automatically applied, making it possible to automatically manufacture products with the desired quality.

[0074] The resin molded product may also be a molded product containing a biodegradable resin. Generally, biodegradable resins tend to have large viscosity variations in raw material pellets. When such raw material pellets are used for molding, the viscosity variations can easily cause mold contamination, and this contamination can affect the quality of the resin molded product. According to the above configuration, quality prediction is performed taking into account the state of the mold, so that quality can be predicted with high accuracy, and manufacturing conditions can be adjusted based on the prediction results to efficiently mold the biodegradable resin.

[0075] The biodegradable resin may also be a polyhydroxyalkanoate-based resin. Because polyhydroxyalkanoate-based resins have excellent degradability, they can solve the environmental problems caused by discarded plastics. Particularly preferably, the polyhydroxyalkanoate-based resin is a poly(3-hydroxyalkanoate)-based resin (hereinafter sometimes referred to as a P3HA-based resin). More specifically, the P3HA-based resin preferably contains 3-hydroxybutyrate (3HB) units. The P3HA-based resin containing 3HB units is preferably selected from the group consisting of poly(3-hydroxybutyrate) (P3HB), poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (P3HB3HV), poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) (P3HB3HH), poly(3-hydroxybutyrate-co-3-hydroxyvalerate-co-3-hydroxyhexanoate) (P3HB3HV3HH), poly(3-hydroxybutyrate-co-4-hydroxybutyrate) (P3HB4HB), poly(3-hydroxybutyrate-co-3-hydroxyoctanoate), and poly(3-hydroxybutyrate-co-3-hydroxydecanoate). The P3HA-based resin may contain only one type, or two or more types.

[0076] Furthermore, the operator can also simulate the production of resin molded products by inputting various manufacturing conditions and having the state prediction unit 202 and the quality prediction unit 203 make predictions. In this way, the information processing device 2 can also be used as an auxiliary tool for determining the manufacturing conditions of resin molded products.

[0077] Although the present embodiment has been described with reference to injection molding as a method for processing resin, the present invention is not limited to this and can be applied to various methods for processing resin, such as extrusion molding, vacuum molding, and pressure molding.

[0078] The execution entity of each process described in the above embodiment is arbitrary and is not limited to the above example. In other words, functions similar to those of information processing devices 1 and 2 can be realized by a plurality of information processing devices that can communicate with each other. For example, the processing of each step shown in FIG. 5 may be shared and executed by a plurality of information processing devices. Furthermore, an information processing device that combines the functions of information processing device 1 and information processing device 2 is also included in the scope of the present invention.

[0079] [Software implementation example] The functions of the information processing devices 1 to 2 (hereinafter simply referred to as "devices") can be realized by a program (information processing program) for causing a computer to function as the device, and a program for causing a computer to function as each control block of the device (particularly each part included in the control unit 10 and control unit 20).

[0080] In this case, the device includes a computer having at least one control device (e.g., a processor) and at least one storage device (e.g., a memory) as hardware for executing the program. The computer executes the program to realize each function described in the above embodiment.

[0081] The program may be stored non-transitory on one or more computer-readable storage media. The storage media may or may not be included in the device. In the latter case, the program may be supplied to the device via any wired or wireless transmission medium.

[0082] In addition, some or all of the functions of each control block can be realized by a logic circuit. For example, an integrated circuit in which a logic circuit that functions as each control block is formed is also included in the scope of the present invention. In addition, the functions of each control block can also be realized by, for example, a quantum computer.

[0083] The present invention is not limited to the above-described embodiment, and various modifications are possible within the scope of the claims. [Explanation of symbols]

[0084] 1, 2 Information processing device 202 State Prediction Unit 203 Quality Prediction Department 211 State Prediction Model 212 Quality Prediction Model 40~42 Manufacturing data

Claims

1. a state prediction unit that predicts the state of a target mold using a state prediction model that predicts the state of the mold used in manufacturing the resin molded product based on manufacturing data of the resin molded product; and a quality prediction unit that predicts the quality of a resin molded product manufactured with the target mold from a prediction result of the state prediction unit using a quality prediction model of the resin molded product including explanatory variables that indicate the state of the mold. Information processing device.

2. The resin molded article is a molded article containing a biodegradable resin. The information processing device according to claim 1 .

3. The biodegradable resin is a polyhydroxyalkanoate-based resin. The information processing device according to claim 2 .

4. An information processing method executed by one or more information processing devices, a state prediction step of predicting a state of a target mold using a state prediction model that predicts a state of the mold used in manufacturing the resin molded product from manufacturing data of the resin molded product; a quality prediction step of predicting the quality of a resin molded product manufactured with the target mold from the prediction result of the state prediction step using a quality prediction model of the resin molded product including explanatory variables indicating the state of the mold, Information processing methods.

5. 2. An information processing program for causing a computer to function as the information processing device according to claim 1, the information processing program causing a computer to function as the state prediction unit and the quality prediction unit.

Images