Manufacturing control device, manufacturing control method, and program

Abstract

This manufacturing control device (100) comprises: a first condition determination unit (111) that uses a model to determine, as a first manufacturing condition, a manufacturing condition under which a target characteristic value, which is a target characteristic value of a product, is obtained; a manufacturing control unit (112) that causes a manufacturing device (200) to execute synthesis according to the determined first manufacturing condition; an evaluation characteristic acquisition unit (113) that acquires, as an evaluation characteristic value, the measured characteristic value from a characteristic evaluation unit (230) that measures a characteristic value of a manufactured product; and a second condition determination unit (114) that determines a second manufacturing condition for bringing the evaluation characteristic value close to the target characteristic value on the basis of the model. The manufacturing control unit (112) changes the manufacturing condition used for the manufacturing device (200) from the first manufacturing condition to the second manufacturing condition.

Description

Manufacturing control device, manufacturing control method, and program

[0001] The present disclosure relates to a manufacturing control device that controls a manufacturing device.

[0002] Conventionally, a production system has been proposed that produces or manufactures a product from raw materials through multiple processes (see, for example, Patent Document 1). In the production system of Patent Document 1, cluster analysis is performed on the characteristics of the raw materials, and cluster analysis is also performed on the manufacturing conditions of each process. Then, a route for producing the product, which connects the clusters of each process, is determined based on the characteristics of the raw materials. Then, in each process, target values ​​of the manufacturing conditions (i.e., control factors) are set so as to follow the determined route. As a result, the target values ​​of the manufacturing conditions for each process are dynamically set based on the characteristics of the raw materials, thereby stabilizing the quality of the product.

[0003] Patent No. 7259179

[0004] In the production system of Patent Document 1, when a product is continuously manufactured, for example, it may be difficult to maintain constant quality of the product. In other words, in the production system of Patent Document 1, in order to maintain constant quality of the continuously manufactured product, it is necessary to evaluate the characteristics of the raw materials, and it is difficult to evaluate changes in the raw materials over time.

[0005] Therefore, the present disclosure provides a manufacturing control device and the like that can easily increase the likelihood of maintaining constant quality of continuously manufactured products.

[0006] A manufacturing control device according to a first aspect of the present disclosure includes a first condition determination unit that determines, as first manufacturing conditions, manufacturing conditions under which a target characteristic value, which is a target characteristic value of the product, is obtained, using a model that indicates the relationship between the characteristic value of the product and the manufacturing conditions for manufacturing the product; a manufacturing control unit that causes a manufacturing device that manufactures the product by synthesizing raw materials flowing through each of a plurality of flow paths to perform synthesis in accordance with the determined first manufacturing conditions; an evaluation characteristic acquisition unit that acquires, from a characteristic evaluation unit that measures the characteristic value of the manufactured product, the measured characteristic value as an evaluation characteristic value; and a second condition determination unit that determines, based on the model, second manufacturing conditions for bringing the evaluation characteristic value closer to the target characteristic value, and the manufacturing control unit changes the manufacturing conditions used by the manufacturing device from the first manufacturing conditions to the second manufacturing conditions.

[0007] The comprehensive or specific aspects may be realized as a system, a method, an integrated circuit, a computer program, or a computer-readable recording medium such as a CD-ROM, or may be realized as any combination of a system, a method, an integrated circuit, a computer program, and a recording medium. The recording medium may also be a non-transitory recording medium.

[0008] The manufacturing control device of the present disclosure can easily increase the likelihood of maintaining constant quality of continuously manufactured products.

[0009] Further advantages and effects of one aspect of the present disclosure will become apparent from the specification and drawings. Such advantages and / or effects are provided by some of the embodiments and configurations described in the specification and drawings, but not all of the configurations are necessarily required.

[0010] FIG. 1 is a diagram illustrating an example of a manufacturing system according to an embodiment. FIG. 2 is a block diagram illustrating an example of the configuration of a manufacturing control device according to an embodiment. FIG. 3 is a diagram illustrating an example of a first processing operation performed by the manufacturing control device according to an embodiment. FIG. 4 is a flowchart illustrating an example of the first processing operation performed by the manufacturing control device according to an embodiment. FIG. 5 is a diagram illustrating an example of a second processing operation performed by the manufacturing control device according to an embodiment. FIG. 6 is a flowchart illustrating an example of the second processing operation performed by the manufacturing control device according to an embodiment. FIG. 7 is a diagram illustrating a third processing operation performed by the manufacturing control device according to an embodiment. FIG. 8 is a flowchart illustrating an example of the third processing operation performed by the manufacturing control device according to an embodiment.

[0011] (Findings that Form the Basis of the Present Invention) The present inventors have found that the above-mentioned Patent Document 1, described in the "Background Art" section, has the following problems.

[0012] In a manufacturing device that continuously produces a functional material product by flowing two or more raw materials through a flow path and synthesizing them, various characteristics of the product, such as molecular weight and particle size, must be maintained constant. The quality of the product is determined by its characteristics. In other words, it is important that the quality of the continuously produced product be maintained within an acceptable range. Because the characteristics of the product change depending on the processing performed in each process in the manufacturing device, manufacturing conditions such as control variables and upper and lower limits for those manufacturing conditions are set for each process. Furthermore, the characteristics of the product in each process are affected by the state or characteristics of the raw materials. Therefore, the production system described in Patent Document 1 dynamically adjusts the manufacturing conditions for each process according to the characteristics of the raw materials, rather than uniquely defining them, in order to more strictly maintain the quality of the product within a certain range.

[0013] However, the production system of Patent Document 1 cannot continuously and dynamically adjust the production conditions of each process in response to changes in the raw materials over time. To control the production conditions in response to changes in the raw materials over time, it is necessary to measure the changes in the raw materials over time. However, it is extremely difficult to quantitatively measure changes in the raw materials over time, such as changes in the complex ion state and the oxidation state, in a short period of time. Therefore, the production system of Patent Document 1 has a problem in that, when the raw materials change over time, it is difficult to continuously maintain constant product quality even if the changes over time are measured.

[0014] To solve this problem, a manufacturing control device according to a first aspect of the present disclosure includes a first condition determination unit that determines, as first manufacturing conditions, manufacturing conditions under which a target characteristic value, which is a target characteristic value of the product, is obtained using a model that indicates the relationship between a characteristic value of the product and the manufacturing conditions for manufacturing the product, a manufacturing control unit that causes a manufacturing device that manufactures the product by synthesizing raw materials flowing through each of a plurality of flow paths to synthesize the product according to the determined first manufacturing conditions, an evaluation characteristic acquisition unit that acquires, from a characteristic evaluation unit that measures the characteristic value of the manufactured product, the measured characteristic value as an evaluation characteristic value, and a second condition determination unit that determines, based on the model, second manufacturing conditions for bringing the evaluation characteristic value closer to the target characteristic value, and the manufacturing control unit changes the manufacturing conditions used by the manufacturing device from the first manufacturing conditions to the second manufacturing conditions. For example, the manufacturing device continuously manufactures products.

[0015] As a result, even if at least one of the raw materials changes over time after production of the product under the first production conditions is initiated, the characteristic values ​​of the product produced using those raw materials are measured. Then, second production conditions are determined to bring the measured characteristic value, i.e., the evaluated characteristic value, closer to the target characteristic value, and the production conditions used by the production equipment are changed from the first production conditions to the second production conditions. Therefore, even if at least one of the raw materials changes over time, the likelihood of maintaining the characteristic values ​​of the continuously produced product at the target characteristic value is increased. Therefore, the production conditions can be continuously and dynamically controlled without measuring the characteristics of the raw materials or using the measurement results of the characteristics of the raw materials. As a result, the likelihood of easily maintaining the quality of the continuously produced product, obtained from the characteristic values ​​of the product, is increased. The characteristic or characteristic value of the product can also be considered a characteristic linked to the quality of the product.

[0016] In the manufacturing control device according to a second aspect, the second condition determination unit may determine the second manufacturing conditions by calculating a difference between the evaluation characteristic value and the target characteristic value and changing the first manufacturing conditions in accordance with the difference between the model and the first manufacturing conditions. Note that the second aspect may be subordinate to the first aspect.

[0017] As a result, for example, when the first manufacturing conditions are expressed as values ​​of explanatory variables, the amount of change in the first manufacturing conditions is determined based on the model and the difference, and the value obtained by changing the first manufacturing conditions by the amount of change is determined as the second manufacturing conditions. Therefore, it is possible to prevent a sudden change in the characteristic values ​​of the products being continuously manufactured due to a change from the first manufacturing conditions to the second manufacturing conditions, and to make it easier to maintain the characteristic values ​​at the target characteristic values.

[0018] In a third aspect of the manufacturing control device, the manufacturing conditions for manufacturing the product may be expressed by explanatory variables, and the model may represent the relationship between the characteristic values ​​of the product and the explanatory variables by a continuous function. Note that the third aspect may be dependent on the first or second aspect.

[0019] Since the model is a continuous function, the slope (i.e., gradient) of the equation can be calculated and used to determine the second manufacturing conditions, thereby determining appropriate second manufacturing conditions.

[0020] In a fourth aspect of the manufacturing control device, the first manufacturing condition and the second manufacturing condition may be expressed by values ​​of the explanatory variables, and the second condition determination unit may identify a slope of the continuous function for the first manufacturing condition and determine, based on the slope, whether to determine the second manufacturing condition greater than the first manufacturing condition or to determine the second manufacturing condition smaller than the first manufacturing condition. Note that the fourth aspect may be dependent on the third aspect.

[0021] For example, if the value indicating the slope is positive, the model can predict that an increase in the manufacturing conditions will increase the evaluation characteristic value, and that a decrease in the manufacturing conditions will decrease the evaluation characteristic value. Furthermore, if the value indicating the slope is negative, the model can predict that an increase in the manufacturing conditions will decrease the evaluation characteristic value, and that a decrease in the manufacturing conditions will increase the evaluation characteristic value. In the fourth aspect, in order to bring the evaluation characteristic value closer to the target characteristic value, it is possible to determine, based on such predictions, whether to increase or decrease the manufacturing conditions from the first manufacturing conditions. As a result, it is possible to appropriately determine second manufacturing conditions for bringing the evaluation characteristic value closer to the target characteristic value by increasing or decreasing from the first manufacturing conditions.

[0022] In a fifth aspect of the manufacturing control device, the second condition determination unit may further determine, based on the slope, whether or not the second manufacturing conditions can be determined based on the model, and may determine the second manufacturing conditions when it is determined that the second manufacturing conditions can be determined. Note that the fifth aspect may be dependent on the fourth aspect.

[0023] This allows the second manufacturing conditions to be determined only when they can be determined, and eliminates wasteful processing such as forcibly determining the second manufacturing conditions when they cannot be determined.

[0024] In a sixth aspect of the manufacturing control device, the second condition determination unit may determine that the second manufacturing conditions can be determined based on the model when the numerical value indicating the slope is not within a predetermined range including 0, and when determining the second manufacturing conditions, may determine whether to determine the second manufacturing conditions greater than the first manufacturing conditions or to determine the second manufacturing conditions smaller than the first manufacturing conditions based on whether the numerical value indicating the slope is positive or negative and the difference between the evaluation characteristic value and the target characteristic value. Note that the sixth aspect may be dependent on the fourth or fifth aspect.

[0025] For example, if the numerical value indicating the slope is 0, it is determined that it is impossible to determine the second manufacturing conditions based on the model. Furthermore, if the numerical value indicating the slope is close to 0, whether the numerical value indicating the slope is positive or negative, the reliability of the slope is low, and therefore it is determined that it is impossible to determine the second manufacturing conditions based on the model. On the other hand, if the numerical value indicating the slope is neither 0 nor close to 0, i.e., if the numerical value indicating the slope is not within a predetermined range including 0, the reliability of the slope is high. Therefore, depending on whether the numerical value indicating the slope is positive or negative, it is possible to appropriately determine the second manufacturing conditions for bringing the evaluation characteristic value closer to the target characteristic value.

[0026] In a seventh aspect of the present invention, when the second condition determination unit determines that it is impossible to determine the second manufacturing conditions based on the model, the manufacturing control unit causes the manufacturing apparatus to manufacture the product according to each of a plurality of values ​​of the explanatory variables within a survey range that includes the first manufacturing conditions, the evaluation property acquisition unit acquires, for each of the plurality of values, the measured property values ​​of the product manufactured according to the respective values ​​as survey property values ​​from the property evaluation unit, and the manufacturing control apparatus further includes a model processing unit that updates the model based on the plurality of values ​​and the survey property values ​​acquired for each of the plurality of values, and the second condition determination unit determines the second manufacturing conditions based on the updated model. Note that the seventh aspect may be subordinate to any one of the fourth to sixth aspects.

[0027] As a result, if it is determined that the second manufacturing conditions cannot be determined based on the model, an investigation is performed. In the investigation, the manufacturing conditions used to manufacture the product by the manufacturing equipment are set to each manufacturing condition within the investigation range (i.e., each numerical value of the explanatory variables), and investigation characteristic values ​​corresponding to each manufacturing condition are obtained. Then, the model is updated based on each manufacturing condition and each investigation characteristic value, and the second manufacturing conditions are determined based on the updated model. Therefore, even if it is determined that the second manufacturing conditions cannot be determined based on the model, the second manufacturing conditions can be appropriately determined by using the updated model.

[0028] Hereinafter, the embodiments will be specifically described with reference to the drawings.

[0029] The embodiments described below are all comprehensive or specific examples. The numerical values, shapes, materials, components, component placement and connection configurations, steps, and step order shown in the following embodiments are merely examples and are not intended to limit the present disclosure. Furthermore, among the components in the following embodiments, components that are not described in the independent claims that represent the highest concepts are described as optional components. Furthermore, each figure is a schematic diagram and is not necessarily an exact illustration. Furthermore, the same components are designated by the same reference numerals in each figure.

[0030] First Embodiment FIG. 1 is a diagram showing an example of the configuration of a manufacturing system according to this embodiment.

[0031] The manufacturing system 1 in this embodiment includes a manufacturing control device 100 and a manufacturing device 200 .

[0032] The manufacturing apparatus 200 produces a product by continuously flowing multiple raw materials through flow paths to synthesize them. For example, the manufacturing apparatus 200 performs flow synthesis by flowing multiple types of liquids, each of which is a raw material, through the flow paths and sequentially reacting the liquids within the flow paths. The manufacturing apparatus 200 includes a transport section 210, a manufacturing process section 220, a property evaluation section 230, and a recovery section 240, which are connected in series by flow paths.

[0033] The transport unit 210 transports the above-mentioned multiple raw materials to the manufacturing process unit 220. For example, the transport unit 210 may transport the multiple raw materials using a pressure-transfer means such as a plunger pump, a diaphragm pump, a syringe pump, a mono pump, a tube pump, or a pressure tank. Note that the transport of the multiple raw materials by the transport unit 210 is a manufacturing process, and specifically, may be a process of adjusting the flow rate or flow rate of each of the multiple raw materials.

[0034] The manufacturing process unit 220 synthesizes the multiple raw materials transported by the transport unit 210. Furthermore, when synthesizing the multiple raw materials, the manufacturing process unit 220 performs one or more manufacturing processes on each of the multiple raw materials or on a compound obtained by synthesizing the multiple raw materials. The one or more manufacturing processes may include, for example, a process for adjusting the flow rate or flow rate of each of the multiple raw materials and the compound, a process for heating or cooling, a process for introducing or mixing a solvent or additive, a process for a photoreaction, a process for a catalytic reaction, a pressurization process, and the like, for each of the multiple raw materials and the compound. In other words, the manufacturing process unit 220 may include, as production equipment for performing these manufacturing processes, a heater, a cooler, a photoreaction reactor, a catalytic reaction device using a catalyst-packed column, a backpressure valve, and the like.

[0035] The characteristic evaluation unit 230 evaluates the characteristics of the product, which is a composite product that is finally discharged from the manufacturing process unit 220. The characteristics are evaluated as values. That is, the characteristic evaluation unit 230 measures the characteristic values ​​of the product.

[0036] Specifically, the characteristic evaluation unit 230 may include an optical measuring instrument that measures, as characteristic values, values ​​indicating the luminescence characteristics, absorbance, optical density, turbidity, etc., of the product. Furthermore, the characteristic evaluation unit 230 may include a Raman spectrometer or FT-IR (Fourier Transform Infrared Spectroscopy) and measure values ​​indicating the analysis results thereof as characteristic values. Furthermore, the characteristic evaluation unit 230 may include a particle size distribution measuring instrument that measures, as a characteristic value, a value indicating the particle size distribution of the product, or a viscometer that measures, as a characteristic value, the kinematic viscosity of the product.

[0037] The characteristic evaluation unit 230 repeatedly measures the characteristic values ​​of the product at predetermined time intervals. The predetermined time intervals may be, for example, 60 seconds, 10 seconds, or 0.1 seconds. The characteristic evaluation unit 230 continuously measures the characteristic values ​​of the product at such time intervals, and outputs an output signal indicating the characteristic value to the manufacturing control device 100 each time a characteristic value is measured. In other words, the output signal is a signal indicating a plurality of characteristic values ​​arranged in chronological order.

[0038] The product characteristics evaluated by the characteristic evaluation unit 230 can also be considered characteristics linked to the product's quality. In the manufacturing system 1 according to the present embodiment, in order to maintain the product's quality at a required level, the characteristic evaluation unit 230 may evaluate the product's quality itself as a characteristic, or may evaluate another characteristic that affects the product's quality. For example, if the required quality of a product that is a polymeric fluid is a predetermined molecular weight, the characteristic evaluation unit 230 may evaluate (i.e., measure) the kinematic viscosity as a characteristic linked to the predetermined molecular weight. Furthermore, for example, if the required quality of a product that is a microparticle suspension is particle size, the characteristic evaluation unit 230 may evaluate (i.e., measure) turbidity, optical density, or the like as a characteristic linked to the particle size.

[0039] The collection unit 240 collects the product evaluated by the characteristic evaluation unit 230. Such a collection unit 240 may be a container such as a beaker or a tank. The collection unit 240 may also use a fraction collector, a flow path switching valve, or the like to control switching of the collection location or container from which the product is collected.

[0040] The manufacturing control device 100 includes, for example, a personal computer 110 and a mutation state output unit 120. The personal computer 110 controls the manufacturing apparatus 200 and the mutation state output unit 120. For example, the personal computer 110 controls the transport unit 210 and the manufacturing process unit 220 of the manufacturing apparatus 200 by outputting control signals to each of them. Specifically, the personal computer 110 controls explanatory variables related to the transport of multiple raw materials by the transport unit 210 and explanatory variables related to the manufacturing process of multiple raw materials or compounds by the manufacturing process unit 220. These explanatory variables are control variables or control factors, and can also be considered manufacturing conditions. In other words, the control signal indicates the manufacturing conditions. The personal computer 110 also acquires an output signal from the characteristic evaluation unit 230 and changes the control signal (i.e., the manufacturing conditions) sent to the manufacturing apparatus 200 in accordance with the output signal. Furthermore, when the personal computer 110 sends a control signal to the manufacturing apparatus 200, it outputs the control signal to the mutation state output unit 120. The personal computer 110 may further control at least one of the characteristic evaluation unit 230 and the collection unit 240 .

[0041] The mutation state output unit 120 acquires a control signal from the personal computer 110 every time the personal computer 110 transmits the control signal to the manufacturing apparatus 200. Then, every time the mutation state output unit 120 acquires the control signal, the mutation state output unit 120 evaluates a change in the manufacturing conditions indicated by the control signal by identifying the change as a mutation state. Furthermore, the mutation state output unit 120 outputs the identified mutation state. In other words, the mutation state output unit 120 evaluates and outputs the mutation state of the manufacturing conditions used in the manufacturing apparatus 200.

[0042] The mutation state output unit 120 may include, for example, a signal light and change the illumination of the signal light depending on the mutation state. Furthermore, the mutation state output unit 120 may include, for example, a display and display the mutation state as an image on the display. Alternatively, the mutation state output unit 120 may display the mutation state as an image on the display of the personal computer 110. In this case, the mutation state output unit 120 may be provided in the personal computer 110. Alternatively, the mutation state output unit 120 may display the mutation state on a user interface for operating the personal computer 110. The mutation state may be the amount of change in the manufacturing conditions from the start of manufacturing the product by the manufacturing apparatus 200, or the degree of change, which is the amount of change in the manufacturing conditions per unit time, or a numerical value calculated using at least one of the amount of change and the degree of change.

[0043] Although the manufacturing control device 100 in this embodiment includes a personal computer 110, it may include a programmable controller, a microcontroller, or the like instead of the personal computer 110.

[0044] FIG. 2 is a block diagram showing an example of the configuration of the manufacturing control device 100 according to this embodiment.

[0045] The manufacturing control device 100 includes a first condition determination unit 111, a manufacturing control unit 112, an evaluation characteristic acquisition unit 113, a second condition determination unit 114, a mutation state output unit 120, a model processing unit 115, and a model storage unit 116. Note that each component included in the manufacturing control device 100, except for the mutation state output unit 120, is provided in the personal computer 110, but does not necessarily have to be provided in the personal computer 110.

[0046] The model storage unit 116 is a recording medium for storing a model showing the relationship between the characteristic values ​​of a product and the manufacturing conditions for manufacturing the product. The model can also be considered a model for predicting the characteristic values ​​of a product from the manufacturing conditions. Furthermore, if the manufacturing apparatus 200 performs multiple manufacturing processes to manufacture a product, the model storage unit 116 may store a model corresponding to each of the manufacturing conditions for the multiple manufacturing processes. The model storage unit 116 may be a hard disk drive, a random access memory (RAM), a read-only memory (ROM), or a semiconductor memory. Such a model storage unit 116 may be volatile or non-volatile.

[0047] The first condition determination unit 111 uses the model stored in the model storage unit 116 to determine, as the first manufacturing conditions, manufacturing conditions that will result in target characteristic values, which are characteristic values ​​that are targets for the product.

[0048] The manufacturing control unit 112 causes the manufacturing apparatus 200, which manufactures a product by synthesizing raw materials flowed through each of a plurality of flow paths, to execute synthesis in accordance with the first manufacturing conditions determined by the first condition determination unit 111. Furthermore, when the second manufacturing conditions are determined by the second condition determination unit 114, the manufacturing control unit 112 changes the manufacturing conditions used by the manufacturing apparatus 200 from the first manufacturing conditions to the second manufacturing conditions. For example, the manufacturing control unit 112 outputs a control signal to the manufacturing apparatus 200 to cause the manufacturing apparatus 200 to execute synthesis in accordance with the first manufacturing conditions or change the first manufacturing conditions to the second manufacturing conditions.

[0049] The evaluation characteristic acquisition unit 113 acquires an output signal from the characteristic evaluation unit 230 that measures the characteristic value of the manufactured product, and thereby acquires the measured characteristic value as an evaluation characteristic value.

[0050] For example, when the evaluation characteristic value differs from the target characteristic value, the second condition determination unit 114 determines the second manufacturing conditions for bringing the evaluation characteristic value closer to the target characteristic value based on the model stored in the model storage unit 116.

[0051] The model processing unit 115, for example, constructs the above-mentioned model and stores it in the model storage unit 116. For example, the model processing unit 115 acquires a past data set for each manufacturing process, and generates, as a model, a mathematical formula that associates manufacturing conditions and characteristic values ​​in the manufacturing process based on the past data set. The past data set includes, for each of a plurality of manufacturing conditions used to manufacture a product in the past, the manufacturing conditions and characteristic values ​​of the product manufactured by the manufacturing process under those manufacturing conditions.

[0052] The following describes three processing operations of the manufacturing control apparatus 100 in this embodiment: a first processing operation, a second processing operation, and a third processing operation.

[0053] [First Processing Operation] FIG. 3 is a diagram for explaining an example of a first processing operation by the manufacturing control device 100. In FIG.

[0054] The model processing unit 115 constructs a model as a continuous function, as shown in FIG. 3 . The continuous function is expressed, for example, as Y = f(X). Here, Y is a characteristic value of the product, and X is a manufacturing condition (i.e., explanatory variable). In other words, in this embodiment, the manufacturing conditions for manufacturing the product are expressed by the explanatory variable X. The model then shows the relationship between the characteristic value Y of the product and the explanatory variable X using a continuous function. Such a model can predict the characteristic value Y from the explanatory variable X, and conversely, can predict the explanatory variable X from the characteristic value Y. The explanatory variable X may be the flow rate of the raw material transported by the transport unit 210. Furthermore, if the manufacturing process unit 220 has a heater, the explanatory variable X may be the set temperature of the heater. Furthermore, the explanatory variable X is not limited to the flow rate or set temperature, and may be other variables depending on the configuration of the manufacturing apparatus 200, the type of product, etc. The model processing unit 115 may also generate a model using multiple regression analysis. Alternatively, the model processing unit 115 may generate a model using Gaussian process regression, or may generate a model using machine learning such as a neural network.

[0055] Using this model, the first condition determination unit 111 determines, as the first manufacturing condition, the explanatory variable X = X1 that yields the target characteristic value Yt of the product. Then, the manufacturing control unit 112 causes the manufacturing equipment 200 to start a manufacturing process in accordance with the first manufacturing conditions, i.e., to manufacture the product in accordance with the first manufacturing conditions. When the product is manufactured, the characteristic evaluation unit 230 measures the characteristic value Y of the manufactured product as an evaluated characteristic value Ya and outputs an output signal indicating the evaluated characteristic value Ya to the manufacturing control equipment 100. The evaluated characteristic acquisition unit 113 of the manufacturing control equipment 100 acquires the output signal to obtain the evaluated characteristic value Ya.

[0056] Here, if at least one of the multiple raw materials used in the production of the product changes over time, the evaluation characteristic value Ya may deviate from the target characteristic value Yt, even if the product is produced according to the first production conditions (i.e., explanatory variable X = X1). In other words, the relational expression (relational expression shown by the dashed line in FIG. 3 ) showing the actual relationship between the product characteristic value Y and explanatory variable X deviates from the model over time (or continuously) due to the above-mentioned change over time. As a result, the evaluation characteristic value Ya deviates from the target characteristic value Yt by ΔY (= Ya - Yt). In other words, the evaluation characteristic value Ya changes from the target characteristic value Yt by the change amount ΔY.

[0057] Therefore, the second condition determination unit 114 determines a second manufacturing condition for bringing the amount of change ΔY closer to 0. Specifically, since the model is a continuous function expressed as Y = f(X), the second condition determination unit 114 first calculates the slope (i.e., gradient) of the continuous function at explanatory variable X = X1 by differentiation.

[0058] Here, if the slope is positive, in the vicinity of explanatory variable X = X1 in the continuous function, as explanatory variable X increases from value X1, the characteristic value Y also increases, and as explanatory variable X decreases from value X1, the characteristic value Y also decreases. Conversely, if the slope is negative, in the vicinity of explanatory variable X = X1 in the continuous function, as explanatory variable X increases from value X1, the characteristic value Y decreases, and as explanatory variable X decreases from value X1, the characteristic value Y increases. Furthermore, due to changes in raw materials over time, the general form of the relational equation showing the actual relationship between characteristic value Y of the product and explanatory variable X rarely changes drastically from the model. Therefore, by continuously changing explanatory variable X based on the above-mentioned slope and change amount ΔY so that future change amount ΔY approaches 0, the characteristic value Y of the product can be maintained constant.

[0059] 3, the above-mentioned slope is positive, and the amount of change ΔY is also positive. Therefore, the second condition determination unit 114 determines the second manufacturing conditions by decreasing the explanatory variable X from the value X1 so that the future amount of change ΔY approaches 0, that is, so that the evaluation characteristic value Ya to be measured in the future decreases. For example, the second condition determination unit 114 determines a negative value as the amount of change ΔX of the explanatory variable X, thereby determining the explanatory variable X = X1 + ΔX as the second manufacturing conditions.

[0060] If the slope is positive and the change ΔY is negative, the second condition determination unit 114 determines a positive value as the change ΔX of the explanatory variable X, thereby determining the explanatory variable X = X1 + ΔX as the second manufacturing condition. If the slope is negative and the change ΔY is positive, the second condition determination unit 114 determines a positive value as the change ΔX of the explanatory variable X, thereby determining the explanatory variable X = X1 + ΔX as the second manufacturing condition. If the slope is negative and the change ΔY is negative, the second condition determination unit 114 determines a negative value as the change ΔX of the explanatory variable X, thereby determining the explanatory variable X = X1 + ΔX as the second manufacturing condition.

[0061] The absolute value of the change amount ΔX may be predetermined, or may be smaller as the magnitude of the above-mentioned slope (i.e., absolute value) increases. The greater the slope of the continuous function, the greater the change in the characteristic value Y relative to the change amount ΔX. Therefore, by reducing the absolute value of the change amount ΔX as the slope increases, it becomes easier to bring the evaluation characteristic value Ya of the product closer to the target characteristic value Yt.

[0062] 3, there is one explanatory variable, but even if there are multiple explanatory variables, the second manufacturing conditions can be determined in the same way as when there is one explanatory variable. In other words, when there are multiple explanatory variables X, the second manufacturing conditions can be determined by expanding the derivative of the explanatory variable X to a partial derivative with respect to the explanatory variable X.

[0063] FIG. 4 is a flowchart showing an example of the first processing operation by the manufacturing control device 100.

[0064] First, the model processing unit 115 of the manufacturing control device 100 constructs a model (step S1). Next, the first condition determination unit 111 uses the model to determine, as the first manufacturing conditions, manufacturing conditions that will obtain the target characteristic value Yt of the product (step S2). The first manufacturing conditions can also be considered to be manufacturing conditions at the start of manufacturing the product. Note that, if the evaluation characteristic value Ya of a previously manufactured product is the target characteristic value Yt, the first condition determination unit 111 may determine, as the first manufacturing conditions, the manufacturing conditions used in the previously manufactured product without using the model.

[0065] Then, the manufacturing control unit 112 transmits a control signal to the manufacturing apparatus 200, thereby causing the manufacturing apparatus 200 to execute the manufacturing process in accordance with the first manufacturing conditions determined in step S2 (step S3). That is, the manufacturing control unit 112 causes the manufacturing apparatus 200 to manufacture the product in accordance with the first manufacturing conditions.

[0066] Next, the evaluation characteristic acquisition unit 113 acquires an evaluation characteristic value Ya of the product manufactured in step S3 by acquiring an output signal from the characteristic evaluation unit 230 (step S4). Note that this evaluation characteristic value Ya is a value measured by the characteristic evaluation unit 230. Here, if the multiple raw materials used in manufacturing the product change over time, the evaluation characteristic value Ya may deviate from the target characteristic value Yt. Therefore, the second condition determination unit 114 determines a change amount ΔX of the first manufacturing conditions based on the model constructed in step S1, the first manufacturing conditions determined in step S2, and the evaluation characteristic value Ya acquired in step S4 (step S5). The conditions obtained by changing the first manufacturing conditions by the change amount ΔX are the above-mentioned second manufacturing conditions. That is, in step S5, the second condition determination unit 114 determines second manufacturing conditions for bringing the evaluation characteristic value Ya of the product to be manufactured in the future closer to the target characteristic value Yt.

[0067] The production control unit 112 then determines whether the production volume, which is the volume of products recovered by the recovery unit 240, has reached the target volume (step S6). If the production control unit 112 determines that the production volume has reached the target volume (Yes in step S6), it causes the production apparatus 200 to stop production of the products and terminates the process related to the production of the products by the production apparatus 200. On the other hand, if the production control unit 112 determines that the production volume has not reached the target volume (No in step S6), it changes the first production conditions determined in step S2 and used in the production of step S3 by the change amount ΔX determined in step S5 (step S7). That is, the production control unit 112 changes the production conditions used by the production apparatus 200 from the first production conditions to the second production conditions by sending a control signal to the production apparatus 200. The production control apparatus 100 then repeatedly executes the process from step S3. That is, the production control unit 112 causes the production apparatus 200 to produce the products according to the changed production conditions.

[0068] Thus, in the first processing operation of this embodiment, even if at least one of the raw materials changes over time after production of the product according to the first manufacturing conditions is initiated, the characteristic values ​​of the product manufactured using those raw materials are measured. Then, second manufacturing conditions are determined to bring the measured characteristic value, the evaluated characteristic value Ya, closer to the target characteristic value Yt, and the manufacturing conditions used by the manufacturing apparatus 200 are changed from the first manufacturing conditions to the second manufacturing conditions. Therefore, even if at least one of the raw materials changes over time, the likelihood of maintaining the characteristic value of the continuously manufactured product at the target characteristic value Yt is increased. Therefore, the manufacturing conditions can be continuously and dynamically controlled without measuring the characteristics of the raw materials or using the measurement results of the characteristics of the raw materials. As a result, the likelihood of easily maintaining the quality of the continuously manufactured product, obtained from the characteristic values ​​of the product, is increased. The characteristic or characteristic value of the product can also be considered a characteristic associated with the quality of the product.

[0069] In addition, in this embodiment, the second condition determination unit 114 calculates the change amount ΔY, which is the difference between the evaluation characteristic value Ya and the target characteristic value Yt, and determines the second manufacturing conditions by changing the first manufacturing conditions according to the model and the change amount ΔY.

[0070] As a result, for example, when the first manufacturing condition is expressed as the value of the explanatory variable X, the change amount ΔX of the first manufacturing condition is determined according to the model and the change amount ΔY, and the value obtained by changing the first manufacturing condition by the change amount ΔX is determined as the second manufacturing condition. Therefore, it is possible to suppress a sudden change in the characteristic value of the continuously manufactured products due to a change from the first manufacturing condition to the second manufacturing condition, and to make it easier to maintain the characteristic value at the target characteristic value Yt.

[0071] In addition, in this embodiment, since the model is a continuous function, the slope (i.e., gradient) of the equation can be calculated and used to determine the second manufacturing conditions, thereby enabling appropriate second manufacturing conditions to be determined.

[0072] Furthermore, in this embodiment, the first manufacturing condition and the second manufacturing condition are each expressed by the value of the explanatory variable X. Then, the second condition determination unit 114 specifies the first manufacturing condition by calculating the slope of the continuous function, which is the model. Furthermore, the second condition determination unit 114 determines, based on the slope, whether to determine a second manufacturing condition that is greater than the first manufacturing condition or a second manufacturing condition that is smaller than the first manufacturing condition. In other words, the second condition determination unit 114 determines whether the change amount ΔX from the explanatory variable X = X1, which is the first manufacturing condition, is a positive value or a negative value.

[0073] For example, if the slope is positive, i.e., the numerical value indicating the slope is positive, the model can predict that an increase in the manufacturing condition will increase the evaluation characteristic value Ya, and a decrease in the manufacturing condition will decrease the evaluation characteristic value Ya. Furthermore, if the slope is negative, i.e., the numerical value indicating the slope is negative, the model can predict that an increase in the manufacturing condition will decrease the evaluation characteristic value Ya, and a decrease in the manufacturing condition will increase the evaluation characteristic value Ya. Therefore, in the first processing operation, in order to bring the evaluation characteristic value Ya closer to the target characteristic value Yt, it is possible to determine, based on such prediction, whether to increase or decrease the manufacturing condition from the first manufacturing condition. As a result, it is possible to appropriately determine second manufacturing conditions for bringing the evaluation characteristic value Ya closer to the target characteristic value Yt by increasing or decreasing from the first manufacturing condition.

[0074] 5 is a diagram for explaining an example of a second processing operation by the manufacturing control device 100. Note that in the second processing operation, another process is further added to the first processing operation.

[0075] 3 and 4, the sign (i.e., positive or negative) of the change amount ΔX of the first manufacturing condition is determined depending on whether the slope of the continuous function for the first manufacturing condition is positive or negative. However, if the slope is close to 0, it is difficult to determine the sign of the change amount ΔX. Therefore, the manufacturing control device 100 determines whether the change amount ΔX can be determined from the model, i.e., whether the second manufacturing condition can be determined based on the model, by determining whether the slope is close to 0.

[0076] 5A, when the evaluation characteristic value Ya deviates from the target characteristic value Yt by ΔY, the second condition determination unit 114 calculates the slope of the explanatory variable X=X1 of the model expressed by Y=f(X). Then, the second condition determination unit 114 determines whether the slope is close to 0, that is, whether the slope is within a predetermined range of 0±Δa.

[0077] Note that Δa may be, for example, a numerical value greater than 0 and less than 0.1, or may be any other numerical value. Alternatively, Δa may be set according to the explanatory variable X=X1 or the evaluation characteristic value Ya. For example, Δa may be a numerical value that is 10% of the value X1, or may be a numerical value that is a percentage other than 10%.

[0078] If the slope is within a predetermined range of 0±Δa, it is difficult to predict whether the evaluation characteristic value Ya will increase or decrease in the future if the explanatory variable X, which is the first manufacturing condition, is made larger than X1. Similarly, it is difficult to predict whether the evaluation characteristic value Ya will increase or decrease if the explanatory variable X, which is the first manufacturing condition, is made smaller than X1. Therefore, if the slope is within a predetermined range of 0±Δa, the second condition determination unit 114 determines that it is impossible to determine the change amount ΔX from the model.

[0079] If it is determined that the change amount ΔX cannot be determined, the manufacturing control device 100 performs an investigation of the evaluation characteristic value Ya by changing the manufacturing conditions within the investigation range. Such an investigation may be performed when ΔY is not within the allowable range. The allowable range may be any predetermined range. Specifically, the manufacturing control unit 112 continuously or intermittently changes the value of the explanatory variable X, which is a manufacturing condition, within the investigation range defined by X1 ± Δb. Δb may be any predetermined numerical value or may be a value derived from a predetermined ratio to X1. The evaluation characteristic acquisition unit 113 then acquires, from the characteristic evaluation unit 230, an evaluation characteristic value Ya of the product manufactured using each value of the explanatory variable X changed within the investigation range. The evaluation characteristic value Ya acquired for each value of the explanatory variable X changed within the investigation range is also referred to as an investigation characteristic value.

[0080] The model processing unit 115 updates the model using the survey characteristic values ​​for each value of the explanatory variable X that has been changed within the survey range, which are obtained by the survey described above. That is, the model processing unit 115 updates the model expressed by Y = f(X) shown in FIG. 5A to the model expressed by Y = f'(X) shown in FIG. 5B. The model processing unit 115 may update the model by multiple regression analysis, Gaussian process regression, machine learning, or the like. As a result, the model stored in the model storage unit 116 is updated.

[0081] It should be noted that the surveyed characteristic value of the product manufactured as a result of the survey may deviate from the target characteristic value Yt. In other words, the difference between the surveyed characteristic value and the target characteristic value Yt may be large. Therefore, the production control unit 112 may cause the collection unit 240 to discard the product manufactured as a result of the survey. Alternatively, the production control unit 112 may control the collection unit 240 so that the product manufactured as a result of the survey is collected in an unauthorized container. The unauthorized container is a container different from the container used to collect products manufactured other than those for the survey.

[0082] When the model is updated, the second condition determination unit 114 determines the change amount ΔX of the first manufacturing condition in the same manner as described above, using the updated model Y=f'(X) instead of the model before the update. That is, the second condition determination unit 114 calculates the slope of the explanatory variable X=X1 of Y=f'(X). Then, if the slope is negative and ΔY (=Ya-Yt) is positive, the second condition determination unit 114 determines a positive ΔX as the change amount ΔX for bringing ΔY closer to 0, that is, as the change amount ΔX for bringing the evaluation characteristic value Ya to be acquired in the future closer to the target characteristic value Yt. As a result, the explanatory variable X=X1+ΔX is determined as the second manufacturing condition.

[0083] 5, the manufacturing condition is represented by one explanatory variable, but the number of explanatory variables is not limited to one and may be two or more. When the manufacturing condition is represented by multiple explanatory variables, the same processing as in FIG. 5 is performed for each of the multiple explanatory variables.

[0084] Furthermore, even if an investigation is performed and the model is updated, if the slope at explanatory variable X = X1 is close to 0, the second condition determination unit 114 cannot determine the change amount ΔX of the first manufacturing condition. In such a case, the raw material may be undergoing unacceptable changes over time. Therefore, in such a case, the manufacturing control device 100 may suspend or stop the manufacturing of the product by the manufacturing device 200.

[0085] In this way, in the second processing operation, if the first manufacturing condition is an extreme condition of the continuous function (i.e., Y=f(X)), and therefore the second manufacturing condition for bringing the evaluation characteristic value Ya closer to the target characteristic value Yt cannot be determined, an investigation is performed to update the continuous function, and the second manufacturing condition can be determined using the updated continuous function.

[0086] FIG. 6 is a flowchart showing an example of the second processing operation by the manufacturing control device 100.

[0087] First, the manufacturing control device 100 executes steps S1 to S4, similar to the first processing operation shown in FIG. 4 , and then executes step S6 after step S4. That is, in step S6, the manufacturing control unit 112 determines whether the production volume, which is the volume of products collected by the collection unit 240, has reached the target volume (step S6). If the manufacturing control unit 112 determines that the production volume has reached the target volume (Yes in step S6), it stops the manufacturing of the products by the manufacturing apparatus 200 and terminates the manufacturing process related to the products by the manufacturing apparatus 200. On the other hand, if the manufacturing control unit 112 determines that the production volume has not reached the target volume (No in step S6), the second condition determination unit 114 determines whether it is possible to determine the change amount ΔX of the first manufacturing conditions (step S11). Specifically, the second condition determination unit 114 determines whether it is possible to determine the change amount ΔX based on the slope of the continuous function, which is the model constructed in step S1, under the first manufacturing conditions.

[0088] If the second condition determination unit 114 determines that the change amount ΔX of the first manufacturing conditions can be determined (Yes in step S11), it performs the processes of steps S5 and S7, as in the example of FIG. 4 . Then, the manufacturing control device 100 repeatedly executes the processes from step S3. On the other hand, if the second condition determination unit 114 determines that the change amount ΔX of the first manufacturing conditions cannot be determined (No in step S11), the model processing unit 115 determines whether the change amount ΔY, which is the difference between the evaluation characteristic value Ya acquired in step S4 and the target characteristic value Yt, is within the allowable range (step S12). If the model processing unit 115 determines that the change amount ΔY is within the allowable range (Yes in step S12), the manufacturing control device 100 repeatedly executes the processes from step S3. In other words, the first manufacturing conditions are not changed, and manufacturing continues under the first manufacturing conditions. On the other hand, if the model processing unit 115 determines that the change ΔY is not within the allowable range (No in step S12), it causes the manufacturing control unit 112 and the evaluation characteristic acquisition unit 113 to investigate the evaluation characteristic value Ya. Furthermore, the model processing unit 115 updates the model based on the investigation results (step S13). This updates the model stored in the model storage unit 116. Then, the manufacturing control device 100 executes the processes of steps S5 and S7, similar to the example in FIG. 4, and then repeatedly executes the processes from step S3. The updated model is used in steps S5 and S7.

[0089] Thus, in the second processing operation, the second condition determination unit 114 further determines whether it is possible to determine the second manufacturing conditions based on the model, based on the slope of the continuous function under the first manufacturing conditions, and if it determines that it is possible to determine the second manufacturing conditions, it determines the second manufacturing conditions.

[0090] This allows the second manufacturing conditions to be determined only when they can be determined, and eliminates wasteful processing such as forcibly determining the second manufacturing conditions when they cannot be determined.

[0091] Furthermore, in this embodiment, the second condition determination unit 114 determines that it is possible to determine the second manufacturing conditions based on the model if the slope, i.e., the numerical value indicating the slope, is not within a predetermined range that includes 0. When determining the second manufacturing conditions, the second condition determination unit 114 determines whether to determine second manufacturing conditions that are greater than the first manufacturing conditions or smaller than the first manufacturing conditions, based on whether the numerical value indicating the slope is positive or negative and the amount of change ΔY that is the difference between the evaluation characteristic value Ya and the target characteristic value Yt.

[0092] For example, if the numerical value indicating the slope is 0, it is determined that it is impossible to determine the second manufacturing conditions based on the model. Furthermore, if the numerical value indicating the slope is close to 0, whether the numerical value indicating the slope is positive or negative, the reliability of the slope is low, and therefore it is determined that it is impossible to determine the second manufacturing conditions based on the model. On the other hand, if the numerical value indicating the slope is neither 0 nor close to 0, i.e., if the numerical value indicating the slope is not within a predetermined range including 0, the reliability of the slope is high. Therefore, depending on whether the numerical value indicating the slope is positive or negative, it is possible to appropriately determine the second manufacturing conditions for bringing the evaluation characteristic value Ya closer to the target characteristic value Yt.

[0093] Furthermore, in this embodiment, if the second condition determination unit 114 determines that it is impossible to determine second manufacturing conditions based on the model, the manufacturing control unit 112 causes the manufacturing apparatus 200 to manufacture a product according to each of multiple values ​​of the explanatory variable X within a survey range that includes the first manufacturing conditions. Then, the evaluation property acquisition unit 113 acquires, for each of the multiple values, the measured property value Y as a survey property value from the property evaluation unit 230, which measures the property value Y of the product manufactured according to that value. The model processing unit 115 updates the model based on the multiple values ​​and the survey property values ​​acquired for each of the multiple values. Then, the second condition determination unit 114 determines second manufacturing conditions based on the updated model.

[0094] As a result, if it is determined that the second manufacturing conditions cannot be determined based on the model, an investigation is performed. In the investigation, the manufacturing conditions used in manufacturing the product by the manufacturing apparatus 200 are set to each manufacturing condition within the investigation range (i.e., each numerical value of the explanatory variable X), and investigation characteristic values ​​corresponding to each manufacturing condition are obtained. Then, the model is updated based on each manufacturing condition and each investigation characteristic value, and the second manufacturing conditions are determined based on the updated model. Therefore, even if it is determined that the second manufacturing conditions cannot be determined based on the model, the second manufacturing conditions can be appropriately determined by using the updated model.

[0095] [Third Processing Operation] In the third processing operation, another process is added to the first processing operation or the second processing operation. Specifically, in the third processing operation, the manufacturing control device 100 stops manufacturing of the product by the manufacturing device 200 in accordance with the mutation state of the manufacturing conditions identified by the mutation state output unit 120. In other words, the mutation state output unit 120 evaluates the mutation state of the manufacturing conditions used by the manufacturing device 200 by identifying the mutation state from the first manufacturing conditions to the second manufacturing conditions. Furthermore, the manufacturing control unit 112 stops manufacturing of the product by the manufacturing device 200 in accordance with the mutation state evaluated by the mutation state output unit 120.

[0096] FIG. 7 is a diagram for explaining the third processing operation by the manufacturing control device 100.

[0097] For example, one or more manufacturing processes of the manufacturing apparatus 200 are controlled by explanatory variables Xa and Xb, respectively. The explanatory variables Xa and Xb are manufacturing conditions. That is, the manufacturing control unit 112 of the manufacturing control apparatus 100 transmits a control signal to the manufacturing apparatus 200, thereby causing the manufacturing apparatus 200 to manufacture a product in accordance with the explanatory variables Xa and Xb.

[0098] In this case, as shown in FIG. 7A, the first condition determination unit 111 determines the explanatory variables (Xa, Xb) = (Xa1, Xb1) as the first manufacturing conditions. Then, the manufacturing control unit 112 causes the manufacturing apparatus 200 to manufacture the product in accordance with the first manufacturing conditions. Next, the second condition determination unit 114 determines the explanatory variables (Xa, Xb) = (Xa2, Xb2) as the second manufacturing conditions. Then, the manufacturing control unit 112 causes the manufacturing apparatus 200 to manufacture the product in accordance with the second manufacturing conditions.

[0099] As a result, the variation state output unit 120 identifies a variation state of the explanatory variables (Xa, Xb) from (Xa1, Xb1) to (Xa2, Xb2) as a variation state from the first manufacturing condition to the second manufacturing condition. More specifically, the manufacturing control unit 112 identifies, as a variation state, the Euclidean distance between the position indicated by the explanatory variables (Xa, Xb) = (Xa1, Xb1) and the position indicated by the explanatory variables (Xa, Xb) = (Xa2, Xb2) in the coordinate space. Note that this coordinate space is a space configured by the explanatory variables Xa and Xb.

[0100] In other words, the mutation state output unit 120 specifies the length of the resultant vector V determined by the vectors va and vb as a mutation state from the first manufacturing condition to the second manufacturing condition. The vector va is a vector indicating a change in the explanatory variable Xa from the value Xa1 to the value Xa2, and the vector vb is a vector indicating a change in the explanatory variable Xb from the value Xb1 to the value Xb2.

[0101] The production control unit 112 stops the production of the product by the manufacturing apparatus 200 when the Euclidean distance or the length of the resultant vector V specified as described above is longer than a predetermined threshold value.

[0102] This allows an appropriate mutation state to be identified, allowing the manufacturing apparatus 200 to stop production of the product at an effective timing based on the mutation state. For example, to identify the mutation state, it is conceivable to individually identify the change amounts of the explanatory variables Xa and Xb, or to identify the Manhattan distance, which is the sum of the absolute values ​​of these change amounts. However, as typified by response surface methodology and response surface models, the region formed by multiple explanatory variables for a product to satisfy certain quality or characteristics is almost always curved. On the other hand, the individual change amounts of multiple explanatory variables or the Manhattan distance described above are effective when the region formed by multiple explanatory variables is rectangular, but not when the region is curved. Therefore, for a region formed by multiple explanatory variables for a product to satisfy certain quality or characteristics, the mutation state can be appropriately or effectively identified by using the Euclidean distance or the length of the resultant vector V.

[0103] In the above example, manufacturing conditions such as the first manufacturing condition and the second manufacturing condition are represented by two explanatory variables, but the number of explanatory variables may be, for example, three. In a specific example, the transport unit 210 and the manufacturing process unit 220 of the manufacturing equipment 200 perform three manufacturing processes. The three manufacturing processes are controlled by explanatory variables Xa, Xb, and Xc, respectively. The explanatory variables Xa, Xb, and Xc are manufacturing conditions. In other words, the manufacturing control unit 112 of the manufacturing control device 100 sends a control signal to the manufacturing equipment 200, causing the manufacturing equipment 200 to manufacture a product in accordance with the explanatory variables Xa, Xb, and Xc.

[0104] In this case, as shown in FIG. 7B, the first condition determination unit 111 determines the explanatory variables (Xa, Xb, Xc) = (Xa1, Xb1, Xc1) as the first manufacturing conditions. Then, the manufacturing control unit 112 causes the manufacturing apparatus 200 to manufacture the product in accordance with the first manufacturing conditions. Next, the second condition determination unit 114 determines the explanatory variables (Xa, Xb, Xc) = (Xa2, Xb2, Xc2) as the second manufacturing conditions. Then, the manufacturing control unit 112 causes the manufacturing apparatus 200 to manufacture the product in accordance with the second manufacturing conditions.

[0105] As a result, the variation state output unit 120 identifies a variation state of the explanatory variables (Xa, Xb, Xc) from (Xa1, Xb1, Xc1) to (Xa2, Xb2, Xc2) as a variation state from the first manufacturing condition to the second manufacturing condition. More specifically, the manufacturing control unit 112 identifies, as a variation state, the Euclidean distance between a position indicated by the explanatory variables (Xa, Xb, Xc) = (Xa1, Xb1, Xc1) in the coordinate space and a position indicated by the explanatory variables (Xa, Xb, Xc) = (Xa2, Xb2, Xc2). Note that this coordinate space is a space configured by the explanatory variables Xa, Xb, and Xc.

[0106] In other words, the mutation state output unit 120 specifies the length of a resultant vector V determined by vectors va, vb, and vc as a mutation state from the first manufacturing condition to the second manufacturing condition. Vector va is a vector indicating a change in explanatory variable Xa from value Xa1 to value Xa2, vector vb is a vector indicating a change in explanatory variable Xb from value Xb1 to value Xb2, and vector vc is a vector indicating a change in explanatory variable Xc from value Xc1 to value Xc2.

[0107] Then, the manufacturing control unit 112 stops the manufacturing of the product by the manufacturing apparatus 200 when the Euclidean distance or the length of the resultant vector V identified as described above is longer than a predetermined threshold value.

[0108] In the above example, the manufacturing conditions such as the first manufacturing conditions and the second manufacturing conditions are represented by two or three explanatory variables, but the number of explanatory variables is not limited to these numbers and may be four or more.

[0109] In this way, the first manufacturing condition and the second manufacturing condition are each expressed as a position in a coordinate space configured by a plurality of explanatory variables. When the manufacturing condition used in the manufacturing apparatus 200 is changed from the first manufacturing condition to the second manufacturing condition, the mutation state output unit 120 identifies, as a mutation state, the Euclidean distance between the position corresponding to the first manufacturing condition in the coordinate space and the position corresponding to the second manufacturing condition in the coordinate space.

[0110] As a result, even if the values ​​of the multiple explanatory variables change when the manufacturing conditions are changed from the first manufacturing conditions to the second manufacturing conditions, the Euclidean distance is used to identify the mutation state, so that the mutation state can be appropriately identified. In other words, a more appropriate mutation state can be identified than with the Manhattan distance. As a result, the possibility of maintaining consistent product quality can be further increased.

[0111] In other words, the first manufacturing condition and the second manufacturing condition are each represented by a plurality of explanatory variables. When the manufacturing condition used in the manufacturing apparatus 200 is changed from the first manufacturing condition to the second manufacturing condition, the mutation state output unit 120 identifies, as the mutation state, the length of a resultant vector V determined from vectors indicating the amount and direction of change of each of the plurality of explanatory variables.

[0112] As a result, even if the values ​​of the multiple explanatory variables change when the manufacturing conditions are changed from the first manufacturing conditions to the second manufacturing conditions, the length of the resultant vector V is identified as the mutation state, so that the mutation state can be appropriately identified. In other words, a more appropriate mutation state can be identified than the simple sum of the amounts of change of the multiple explanatory variables. As a result, the possibility of maintaining consistent product quality can be further increased.

[0113] Furthermore, in the manufacturing system 1 according to the present embodiment, the manufacturing conditions are continuously and dynamically controlled to maintain a constant quality of the product. Therefore, because the quality of the product is stable within a certain range, even if the raw materials significantly change over time, it is difficult to detect the change in the raw materials from the quality (or characteristics) of the product. Therefore, it may be impossible to perform raw material replacement or maintenance of the manufacturing apparatus 200 at an appropriate time based on the quality of the product. However, in the third processing operation, if the raw materials continuously change significantly, the change is identified as a mutation state. Therefore, the extent to which the raw materials have changed over time can be determined from the identified and output mutation state. As a result, the manufacturing of the product by the manufacturing apparatus 200 can be stopped before the raw material changes over time to an unacceptable extent. This further increases the likelihood of maintaining a constant quality of the product.

[0114] It should be noted that the multiple explanatory variables representing the manufacturing conditions may or may not be independent of each other. For example, if the explanatory variable Xc is not independent of the explanatory variables Xa and Xb, the explanatory variable Xc may be expressed as Xc = j Xa + k Xb, where j and k are each an arbitrary value. Furthermore, it is possible to determine from the model whether the effect of the explanatory variable Xc on the quality or characteristics of the product is independent of the respective effects of the explanatory variables Xa and Xb. Therefore, when there are multiple explanatory variables that interact with each other, i.e., multiple explanatory factors that are not independent of each other, among the multiple explanatory variables for the quality or characteristics of the product, the mutation state can be more effectively identified by using the length of the composite vector of these multiple explanatory variables.

[0115] FIG. 8 is a flowchart showing an example of the third processing operation by the manufacturing control device 100.

[0116] First, the manufacturing control device 100 executes steps S1 to S7, similar to the first processing operation shown in FIG. 4 . Then, when the manufacturing conditions of the manufacturing device 200 are changed from the first manufacturing conditions to the second manufacturing conditions in step S7, the mutation state output unit 120 evaluates the mutation state of the manufacturing conditions (step S21). For example, the mutation state output unit 120 evaluates the mutation state by determining the Euclidean distance or the length of the resultant vector shown in the example of FIG. 7 . The evaluated mutation state is then output. Next, the manufacturing control unit 112 determines whether the mutation state evaluated in step S21 is within an acceptable range (step S22). That is, the manufacturing control unit 112 determines whether the Euclidean distance or the length of the resultant vector is within an acceptable range equal to or less than a threshold value.

[0117] If the manufacturing control unit 112 determines that the mutation state is within the tolerance range (Yes in step S22), the manufacturing control device 100 repeats the process from step S3. That is, if the Euclidean distance or the length of the resultant vector is equal to or less than the threshold, manufacturing of the product continues. On the other hand, if the manufacturing control unit 112 determines that the mutation state is not within the tolerance range (No in step S22), it causes the manufacturing equipment 200 to stop manufacturing the product and terminates the manufacturing process of the product by the manufacturing equipment 200. That is, the manufacturing control unit 112 causes the manufacturing equipment 200 to stop manufacturing the product before the characteristics of the raw material change to an unacceptable extent due to changes in the raw material over time.

[0118] In the above example, if the production control unit 112 determines that the mutation state is not within the tolerance range, it causes the production apparatus 200 to stop production of the product. In other words, if the mutation state is not within the tolerance range, production of the product by the production apparatus 200 is automatically stopped. However, if the evaluated mutation state is output and displayed, for example, on a display, the production control unit 112 may accept an operation by a user who has confirmed the mutation state displayed on the display, and cause the production apparatus 200 to stop production of the product in accordance with the operation.

[0119] In this way, in the third processing operation, the variation state of the manufacturing conditions used in the manufacturing apparatus 200 is evaluated. In other words, when the manufacturing conditions are changed from the first manufacturing conditions to the second manufacturing conditions, the variation state from the first manufacturing conditions to the second manufacturing conditions is identified and output. Therefore, the output variation state allows the user to understand the extent to which the raw material has changed over time. As a result, the user can stop the manufacturing of the product by the manufacturing apparatus 200 before the characteristics of the raw material change to an unacceptable extent due to the change over time of the raw material. As a result, the possibility of maintaining a consistent quality of the product can be further increased.

[0120] Furthermore, in the third processing operation, the production control unit 112 causes the manufacturing apparatus 200 to stop manufacturing the product in accordance with the evaluated mutation state, so that the manufacturing of the product can be automatically stopped. In other words, even without the user checking the output mutation state, the manufacturing of the product by the manufacturing apparatus 200 can be automatically stopped before the characteristics of the raw material change to an unacceptable extent due to changes in the raw material over time. As a result, the workload on the user can be reduced, and the possibility of maintaining consistent product quality can be increased.

[0121] (Other Modifications) While the manufacturing control device 100 according to one aspect of the present disclosure has been described above based on the above embodiment, the present disclosure is not limited to the embodiment. As long as the modifications do not deviate from the spirit of the present disclosure, various modifications that a person skilled in the art can conceive of to the above embodiment may also be included in the present disclosure.

[0122] For example, in the above embodiment, the second condition determination unit 114 determines the second manufacturing conditions by determining the amount of change ΔX from the first manufacturing conditions, but the second manufacturing conditions may be determined directly without determining the amount of change ΔX. Specifically, the second condition determination unit 114 may determine, as the second manufacturing conditions, manufacturing conditions that change the evaluation characteristic value Ya by −ΔY based on the model and the amount of change ΔY.

[0123] In the above-described embodiments, each component may be configured with dedicated hardware, or may be realized by executing a software program suitable for each component. Each component may be realized by a program execution unit such as a CPU or processor reading and executing a software program recorded on a recording medium such as a hard disk or semiconductor memory. Here, software for realizing the manufacturing control device 100 of the above-described embodiments is, for example, a program that causes a computer to execute each step of the flowchart shown in FIG. 4, FIG. 6, or FIG. 8.

[0124] The following cases are also included in this disclosure:

[0125] (1) The at least one device is specifically a computer system comprising a microprocessor, ROM (Read Only Memory), RAM (Random Access Memory), a hard disk unit, a display unit, a keyboard, a mouse, etc. A computer program is stored in the RAM or hard disk unit. The at least one device achieves its function when the microprocessor operates in accordance with the computer program. Here, the computer program is composed of a combination of multiple instruction codes that indicate commands to the computer to achieve a predetermined function.

[0126] (2) Some or all of the components constituting at least one of the above devices may be configured as a single system LSI (Large Scale Integration). A system LSI is an ultra-multifunctional LSI manufactured by integrating multiple components on a single chip, and specifically, is a computer system configured to include a microprocessor, ROM, RAM, etc. A computer program is stored in the RAM. The system LSI achieves its functions by the microprocessor operating in accordance with the computer program.

[0127] (3) Some or all of the components constituting at least one of the above devices may be configured as an IC card or a standalone module that can be attached to or detached from the device. The IC card or module is a computer system configured with a microprocessor, ROM, RAM, etc. The IC card or module may include the above-mentioned ultra-multifunctional LSI. The IC card or module achieves its functions when the microprocessor operates in accordance with a computer program. This IC card or module may be tamper-resistant.

[0128] (4) The present disclosure may be embodied as the methods described above, a computer program for implementing these methods on a computer, or a digital signal comprising the computer program.

[0129] The present disclosure may also be a computer program or a digital signal recorded on a computer-readable recording medium, such as a flexible disk, a hard disk, a CD (Compact Disc)-ROM, a DVD, a DVD-ROM, a DVD-RAM, a BD (Blu-ray (registered trademark) Disc), a semiconductor memory, etc. Alternatively, the present disclosure may be a digital signal recorded on such a recording medium.

[0130] The present disclosure may also be applied to transmitting a computer program or digital signal via a telecommunications line, a wireless or wired communication line, a network such as the Internet, data broadcasting, or the like.

[0131] Furthermore, the program or digital signal may be recorded on a recording medium and transferred, or the program or digital signal may be transferred via a network or the like, so that the program or digital signal may be implemented by another independent computer system.

[0132] The manufacturing control device of the present disclosure can, for example, increase the likelihood of easily maintaining constant the quality of continuously manufactured products, and can be applied to control devices or control systems of manufacturing devices that manufacture products by continuously synthesizing multiple raw materials.

[0133] 1 Manufacturing system 100 Manufacturing control device 110 Personal computer 111 First condition determination unit 112 Manufacturing control unit 113 Evaluation characteristic acquisition unit 114 Second condition determination unit 115 Model processing unit 116 Model storage unit 120 Mutation state output unit 200 Manufacturing device 210 Transportation unit 220 Manufacturing process unit 230 Characteristic evaluation unit 240 Recovery unit X, Xa, Xb, Xc Explanatory variable (manufacturing condition) Y Characteristic value Ya Evaluation characteristic value Yt Target characteristic value ΔX Change amount ΔY Change amount

Claims

1. A manufacturing control device comprising: a first condition determination unit that determines, as first manufacturing conditions, manufacturing conditions under which a target characteristic value, which is a desired characteristic value of the product, is obtained, using a model that indicates the relationship between the characteristic value of the product and the manufacturing conditions for manufacturing the product; a manufacturing control unit that causes a manufacturing device that manufactures the product by synthesizing raw materials flowing through each of a plurality of flow paths to execute synthesis in accordance with the determined first manufacturing conditions; an evaluation characteristic acquisition unit that acquires, from a characteristic evaluation unit that measures the characteristic value of the manufactured product, the measured characteristic value as an evaluation characteristic value; and a second condition determination unit that determines, based on the model, second manufacturing conditions for bringing the evaluation characteristic value closer to the target characteristic value, wherein the manufacturing control unit changes the manufacturing conditions used by the manufacturing device from the first manufacturing conditions to the second manufacturing conditions.

2. The manufacturing control device described in claim 1, wherein the second condition determination unit determines the second manufacturing conditions by calculating the difference between the evaluation characteristic value and the target characteristic value and changing the first manufacturing conditions according to the model and the difference.

3. The manufacturing control device according to claim 2, wherein the manufacturing conditions for manufacturing the product are expressed by explanatory variables, and the model indicates the relationship between the characteristic values ​​of the product and the explanatory variables by a continuous function.

4. The manufacturing control device described in claim 3, wherein the first manufacturing conditions and the second manufacturing conditions are each expressed by the values ​​of the explanatory variables, and the second condition determination unit identifies the slope of the continuous function for the first manufacturing conditions, and determines, based on the slope, whether to determine the second manufacturing conditions that are greater than the first manufacturing conditions or smaller than the first manufacturing conditions.

5. The manufacturing control device described in claim 4, wherein the second condition determination unit further determines whether it is possible to determine the second manufacturing conditions based on the model based on the slope, and determines the second manufacturing conditions if it determines that it is possible to determine the second manufacturing conditions.

6. The manufacturing control device described in claim 5, wherein the second condition determination unit determines that the second manufacturing conditions can be determined based on the model when the numerical value indicating the slope is not within a predetermined range including 0, and when determining the second manufacturing conditions, determines whether to determine the second manufacturing conditions that are greater than the first manufacturing conditions or smaller than the first manufacturing conditions based on whether the numerical value indicating the slope is positive or negative and the difference between the evaluation characteristic value and the target characteristic value.

7. The manufacturing control device described in claim 6, wherein, when the second condition determination unit determines that it is impossible to determine the second manufacturing conditions based on the model, the manufacturing control unit causes the manufacturing equipment to manufacture the product according to each of a plurality of values ​​of the explanatory variable within an investigation range including the first manufacturing conditions, the evaluation characteristic acquisition unit acquires, for each of the plurality of values, the measured characteristic values ​​from the characteristic evaluation unit that measures the characteristic values ​​of the product manufactured according to the value as investigation characteristic values, the manufacturing control device further includes a model processing unit that updates the model based on the plurality of values ​​and the investigation characteristic values ​​acquired for each of the plurality of values, and the second condition determination unit determines the second manufacturing conditions based on the updated model.

8. A manufacturing control method performed by a computer, comprising: using a model showing the relationship between a characteristic value of a product and the manufacturing conditions for manufacturing the product, determining, as first manufacturing conditions, manufacturing conditions under which a target characteristic value, which is a desired characteristic value of the product, is obtained; causing a manufacturing device that manufactures the product by synthesizing raw materials flowing through each of a plurality of flow paths to execute synthesis in accordance with the determined first manufacturing conditions; obtaining, from a characteristic evaluation unit that measures the characteristic value of the manufactured product, the measured characteristic value as an evaluated characteristic value; determining, based on the model, second manufacturing conditions for bringing the evaluated characteristic value closer to the target characteristic value; and changing the manufacturing conditions used by the manufacturing device from the first manufacturing conditions to the second manufacturing conditions.

9. A program that causes a computer to execute the following steps: determine, as first manufacturing conditions, manufacturing conditions that will result in a target characteristic value, which is a characteristic value that is the target of the product, using a model that shows the relationship between the characteristic value of a product and the manufacturing conditions for manufacturing the product; cause a manufacturing device that manufactures the product by synthesizing raw materials flowing through each of a plurality of flow paths to execute synthesis in accordance with the determined first manufacturing conditions; obtain, from a characteristic evaluation unit that measures the characteristic value of the manufactured product, the measured characteristic value as an evaluated characteristic value; determine, based on the model, second manufacturing conditions that will bring the evaluated characteristic value closer to the target characteristic value; and change the manufacturing conditions used by the manufacturing device from the first manufacturing conditions to the second manufacturing conditions.

Images