Prediction system and prediction method

Abstract

To predict with greater accuracy the rate of bitterness transfer in the manufacturing process of beer-flavored beverages. [Solution] The prediction system includes an acquisition unit that acquires at least one target data value related to manufacturing conditions in a process that constitutes at least a part of the manufacturing process of a beer-flavored beverage as an input value set, and a prediction unit that inputs the input value set to a prediction model that has been trained to accept input data values ​​related to manufacturing conditions and calculate the bitterness transfer rate, which is the transfer rate of bitterness in the process, and predicts the bitterness transfer rate in the process.

Description

Technical Field

[0001] One aspect of the present disclosure relates to a prediction system and a prediction method.

Background Art

[0002] Patent Document 1 describes a method for predicting beverage characteristics by a computer. The computer inputs brewing samples with setting items in the brewing conditions of the beverage as explanatory variables into a plurality of prediction models, and obtains estimated values of the component values of the beverage for each prediction model; inputs the brewing samples into an error determination model corresponding to the prediction model, and calculates the probability within the allowable error for each prediction model; and inputs the explanatory variables related to the brewing samples and a plurality of probabilities within the allowable error into a method selection model, and selects an estimated value presented as a component value predicted under the brewing conditions of the beverage based on the output value of the method selection model.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] A method for predicting the transfer rate of the bitterness amount in the manufacturing process of a beer-taste beverage with higher accuracy is desired. A

Means for Solving the Problems

[0005] A prediction system relating to one aspect of this disclosure includes an acquisition unit that acquires at least one target data value related to manufacturing conditions in a process that constitutes at least a part of the manufacturing process of a beer-flavored beverage as an input value set, and a prediction unit that inputs the input value set to a prediction model that has been trained to accept input data values ​​related to manufacturing conditions and calculate a bitterness transfer rate, which is the transfer rate of bitterness in the process, and predicts the bitterness transfer rate in the process.

[0006] In this regard, a predictive model pre-trained to calculate the bitterness transfer rate from data values ​​related to manufacturing conditions in the beer-flavored beverage manufacturing process is used to predict the bitterness transfer rate corresponding to the input value set. By introducing this predictive model, the bitterness transfer rate in the beer-flavored beverage manufacturing process can be predicted with higher accuracy. [Effects of the Invention]

[0007] According to one aspect of this disclosure, the rate of bitterness transfer in the manufacturing process of beer-flavored beverages can be predicted with greater accuracy. [Brief explanation of the drawing]

[0008] [Figure 1] This figure shows an example of the functional configuration of a prediction system. [Figure 2] This figure shows an example of the hardware configuration of the computer that makes up the prediction system. [Figure 3] A flowchart illustrating an example of how the prediction system works. [Modes for carrying out the invention]

[0009] The following describes various examples in this disclosure in detail with reference to the attached drawings. In the description of the drawings, identical or equivalent elements are denoted by the same reference numeral, and redundant descriptions are omitted.

[0010] [System Overview] The prediction system described herein is a computer system that predicts the bitterness transfer rate in the manufacturing process of beer-flavored beverages. In this disclosure, the bitterness transfer rate is also referred to as the bitterness transfer rate.

[0011] The prediction system predicts the bitterness transfer rate using a pre-generated prediction model. The prediction model is a trained model that accepts data values ​​related to manufacturing conditions in a process that constitutes at least a part of the manufacturing process of a beer-flavored beverage and is trained to calculate the bitterness transfer rate in that process. In this disclosure, the process for which the bitterness transfer rate is predicted is also referred to as the "target process." The bitterness transfer rate in the target process is calculated as the ratio of the bitterness of the product in the target process to the bitterness of the raw materials in the target process.

[0012] In the following, the transfer rate of bitterness value (BU) is shown as an example of the bitterness transfer rate. In this disclosure, the transfer rate of bitterness value is also referred to as the BU transfer rate. Bitterness value is an important factor for the flavor and quality of beer-flavored beverages. Bitterness value is brought about by the hop raw material (hops) and fluctuates during the manufacturing (brewing) process of beer-flavored beverages. In order to achieve the desired bitterness value in the final product of a beer-flavored beverage, it is necessary to set an appropriate BU transfer rate at each stage of the manufacturing process. However, there are many factors involved in the BU transfer rate, and the relationships between these factors are complex, so setting an appropriate BU transfer rate at each stage largely depends on the experience or intuition of the manufacturing personnel. Therefore, setting the BU transfer rate is not easy. By using a prediction system that can automatically predict the BU transfer rate with high accuracy, brewers can set the BU transfer rate in the manufacturing process of beer-flavored beverages more easily than before. For example, even inexperienced manufacturing personnel can easily set the BU transfer rate. In addition, since the prediction system can accurately predict the BU transfer rate, it becomes possible to easily obtain a beer-flavored beverage with the desired bitterness value. Furthermore, by making it easier to set the BU transition rate, the number of trials in the brewing process can be reduced, for example, thereby lowering the time and cost required to develop beer-flavored beverages. In addition, by obtaining the desired beer-flavored beverage with greater accuracy, the possibility of producing substandard products decreases, thus reducing production losses.

[0013] The bitterness value of beer-flavored beverages can be measured by the method described in "7.12 Bitterness Value" or "8.15 Bitterness Value" of the Revised BCOJ Beer Analysis Method (published by the Japan Brewing Association, edited by the International Technical Committee [Analysis Committee] of the Beer Brewers Association, revised and augmented in 2013). The bitterness value can be appropriately set by adjusting, for example, the type and amount of raw materials used. The BU transfer rate in the target process is calculated as the ratio of the BU of the product in the target process to the BU of the raw materials in the target process. For the BU of hop raw materials (hops), a solution containing a unit weight (e.g., 1 g) of hop raw materials (e.g., a solution obtained by mixing hop raw materials with water) is boiled, and the BU of the solution is measured in the same way as for beer-flavored beverages described above, thereby determining the estimated BU that will be imparted to the beverage by the hop raw materials per unit weight (hereinafter referred to as "estimated BU per hop raw material"). The BU transfer rate from hop raw materials to beer-flavored beverages is then calculated as the ratio (%) of the BU of the beer-flavored beverage to the estimated BU per hop raw material.

[0014] [Beer-flavored beverage] In this disclosure, "beer-flavored beverage" refers to a beverage having a beer-like flavor. Examples of beer-flavored beverages, though not limited to those mentioned above, include beer, sparkling alcoholic beverages, and other carbonated alcoholic beverages as defined in Article 3 of the Liquor Tax Act (Act No. 6 of 1953). Furthermore, beverages not classified as carbonated alcoholic beverages under the Liquor Tax Act, as well as soft drinks (e.g., non-alcoholic beer-flavored beverages), can also be considered beer-flavored beverages. The beer-flavored beverages related to this disclosure are not limited to those exemplified above.

[0015] Beer-flavored beverages may or may not contain malt as an ingredient. Malt ingredients refer to malt or processed malt products. Examples of malt include barley, wheat, rye, oats, oats, pearl oats, and oats. Examples of processed malt products include malt extract, malt, and malt extract. Malt extract is obtained by extracting malt extract containing sugars and nitrogen from malt. Malt is obtained by germinating malt. Malt extract is obtained by extracting extract containing sugars and nitrogen from malt. Beer-flavored beverages may have a malt usage ratio of 0% by mass or more and 100% by mass or less. The malt usage ratio refers to the proportion of malt to ingredients other than water and hops.

[0016] Beer-flavored beverages may or may not contain ingredients other than malt. These other ingredients may include, for example, grains such as corn, rice, and sorghum; potatoes and sweet potatoes; legumes such as soybeans and peas; herbs and spices; or carbohydrates such as starch, grits, and liquid sugar.

[0017] Beer-flavored beverages may or may not contain hops as an ingredient. In this disclosure, hops include, for example, fresh hops, dried hops, hop pellets, and hop extracts, and also include hop processed products such as raw hops, hexahops, tetrahops, and isopropyl hop extracts.

[0018] [System Configuration] Figure 1 shows the functional configuration of a prediction system 10 in one example. In this example, the prediction system 10 comprises a learning unit 11, an acquisition unit 12, and a prediction unit 13 as functional modules. The learning unit 11 is a functional module that generates a prediction model 20. The acquisition unit 12 is a functional module that acquires one or more target data values ​​related to the manufacturing conditions in a target process that constitutes at least a part of the manufacturing process of a beer-flavored beverage as an input value set. The prediction unit 13 is a functional module that inputs the input value set into the prediction model 20 and predicts the BU transition rate in the target process.

[0019] In one example, the prediction system 10 is connected to the database 30 via a communication network. The communication network may be constructed by the Internet, an intranet, or a combination thereof. The communication network may be constructed by a wired network, a wireless network, or a combination thereof. The database 30 is a storage device that stores training data used to generate the prediction model 20. The database 30 may be a component of the prediction system 10 or may be provided outside the prediction system 10.

[0020] FIG. 2 is a diagram showing an example of the hardware configuration of the computer 100 that constitutes the prediction system 10. For example, the computer 100 includes a processor 101, a main memory unit 102, an auxiliary storage unit 103, a communication control unit 104, an input device 105, and an output device 106. The processor 101 executes an operating system and an application program. The main memory unit 102 is composed of, for example, ROM and RAM. The auxiliary storage unit 103 is composed of, for example, a hard disk or a flash memory and generally stores a larger amount of data than the main memory unit 102. The communication control unit 104 is composed of, for example, a network card or a wireless communication module. The input device 105 is composed of, for example, a keyboard, a mouse, a touch panel, etc. The output device 106 is composed of, for example, a monitor and a speaker.

[0021] Each functional module of the prediction system 10 is realized by a prediction program 110 stored in advance in the auxiliary storage unit 103. Each functional module is realized by causing the processor 101 or the main memory unit 102 to load the prediction program 110 and causing the processor 101 to execute the prediction program 110. The processor 101 operates the communication control unit 104, the input device 105, or the output device 106 according to the prediction program 110 and reads and writes data in the main memory unit 102 or the auxiliary storage unit 103.

[0022] The prediction program 110 may be provided after being recorded on a non-temporary recording medium such as a CD-ROM, DVD-ROM, semiconductor memory, etc. Alternatively, the prediction program 110 may be provided via a communication network as a data signal superimposed on a carrier wave.

[0023] The prediction system 10 may be composed of one computer 100, or may be composed of a plurality of computers 100. When a plurality of computers 100 are used, these computers 100 are connected via a communication network such as the Internet or an intranet, thereby logically constructing one prediction system 10. The prediction system 10 may be constructed by combining multiple types of computers.

[0024] [Prediction Model] For each of one or more target processes, the prediction system 10 can predict the BU transfer rate using the prediction model 20 corresponding to the target process. The prediction system 10 may treat a part of the manufacturing process as the target process, or may treat the entire manufacturing process as the target process.

[0025] In one example, the learning unit 11 executes machine learning using the training data in the database 30 to generate individual prediction models 20. Machine learning refers to a method of autonomously finding laws or rules by repeatedly learning based on given information. It should be noted that the prediction model 20 obtained by training a certain machine learning model is a computational model estimated to be optimal, but is not necessarily the "truly optimal computational model" in reality. The machine learning model (prediction model 20) may be realized by, for example, XGBoost or random forest. <0000​​Each data record in the training data used in machine learning to generate a predictive model 20 for a given target process represents a combination of at least one set of data values ​​related to the manufacturing conditions in the target process and the BU transition rate in the target process, which is used as the ground truth. Such training data is prepared, for example, based on data collected in past trials, analyses, or production of beer-flavored beverages.

[0027] The learning unit 11 generates a predictive model 20 for a given target process as follows: The learning unit 11 inputs a set of data values ​​contained in one data record of the training data corresponding to the target process into a given machine learning model and obtains an estimated value of the BU transition rate output from the machine learning model. The learning unit 11 updates the machine learning model based on the error between the estimated value and the correct answer. The learning unit 11 repeats the learning process using each data record of the training data until a given termination condition is met, thereby obtaining the predictive model 20. The generation of the predictive model 20 by the learning unit 11 corresponds to the learning phase.

[0028] In one example, the prediction system 10 treats at least one of the following as target processes: a mashing process, which is the process of producing wort (e.g., cold wort) using hops (hop raw material); a fermentation process, which is the process of fermenting the wort with yeast to produce a semi-finished product (post-fermentation liquid) of a beer-flavored beverage from the wort; and a post-fermentation process, which is the process of producing a final product of a beer-flavored beverage from the semi-finished product. The mashing process may include a boiling process. The fermentation process may include a storage process. The post-fermentation process may include a filtration process. A semi-finished product is a product that is not yet completed as a final product but can be stored and transferred in that state. The prediction system 10 generates at least one of the following: a prediction model 21 that predicts the BU conversion rate from hops (hop raw material) to wort in the mashing process; a prediction model 22 that predicts the BU conversion rate from wort to semi-finished product or final product in the fermentation process; and a prediction model 23 that predicts the BU conversion rate from semi-finished product to final product in the post-fermentation process. In this example, the database 30 stores at least one of the following: training data for generating prediction model 21, training data for generating prediction model 22, and training data for generating prediction model 23.

[0029] The predictive model 21 accepts at least one (or at least two) of the following factors F1 to F15 as data values ​​related to the manufacturing conditions in the mashing process, and calculates the BU transfer rate from hops (hop raw material) to wort based on these data values. (F1) Amount of hop bitterness added in the brewing process after the boiling process. (F2) Bitterness per unit volume of hops added in the brewing process after the boiling process. (F3) The ratio of the amount of hop bitterness added in the brewing process after the boiling process to the total amount of hop bitterness added in the brewing process. (F4) Amount of hop bitterness added 60 minutes after the start of the boiling process (F5) Bitterness per unit volume of hops added 60 minutes or more after the start of the boiling process. (F6) The ratio of the amount of bitterness from hops added 60 minutes or more after the start of the boiling process to the total amount of bitterness from hops added throughout the entire brewing process. (F7) Amount of bitterness from hops added in the latter half of the boiling process (F8) Bitterness per unit volume of hops added in the latter half of the boiling process. (F9) The ratio of the amount of bitterness from hops added in the latter half of the boiling process to the total amount of bitterness from hops added throughout the entire brewing process. (F10) The total weight of hops added throughout the brewing process. (F11) Weight of hops per unit volume added throughout the entire brewing process (F12) Weight of the acid used in the preparation process (F13) Weight per unit volume of acid used in the preparation process (F14) Amount of bitterness of hop extract added during the boiling process (F15) Amount of bitterness per unit volume of hop extract added during the boiling process. Here, "the latter half of the boiling process" refers to the time interval from the midpoint between the start and end times of the boiling process to the end time.

[0030] The predictive model 22 accepts at least one (or at least two) of the factors F1-F3 and F16-F25 listed above as data values ​​related to the manufacturing conditions in the fermentation process, and calculates the BU transfer rate from wort to semi-finished or final product based on these data values. The yeast type, raw wort extract value, and pseudo-non-fermented extract value are all selected from a predetermined set of corresponding types or values. (F16) Amount of hop bitterness added during the fermentation or storage process (F17) Bitterness per unit volume of hops added during the fermentation or storage process. (F18) Fermentation temperature (F19) Yeast type (F20) Amount of yeast added per unit volume (F21) Amount of acid used in the fermentation or storage process (F22) Amount of acid used per unit volume in the fermentation or storage process (F23) Malt usage ratio (F24) Value of raw wort extract or pseudo-unfermented extract of the wort to be fermented. (F25) Amount of air per unit time that is passed through the wort

[0031] The predictive model 23 accepts at least one of the flavoring hop ratio and unit bitterness amount, along with extract concentration, as data values ​​related to the manufacturing conditions in the post-fermentation process, and calculates the BU conversion rate from semi-finished product to final product based on these data values. In this disclosure, flavoring hop ratio means the proportion of hops used for flavoring to the total amount of hops. Unit bitterness amount means the amount of bitterness per unit volume of wort. Extract concentration means the concentration of extract in the wort.

[0032] [System operation] Referring to Figure 3, the operation of the prediction system 10 will be described as an example of the prediction method relating to this disclosure. Figure 3 is a flowchart showing the operation of the prediction system 10 as a processing flow S1. That is, the prediction system 10 executes processing flow S1. Processing flow S1 is executed after the prediction unit 13 generates one or more prediction models 20, and therefore corresponds to the operation phase or prediction phase. In the following, the operation of the prediction system 10 will be described assuming that multiple prediction models 21 to 23 are prepared as multiple prediction models 20.

[0033] In step S11, the acquisition unit 12 acquires the input value set and the specified data. The specified data refers to the data used to specify the prediction model 20 used to predict the BU transition rate. For example, the user of the prediction system 10 selects the target process for which they want to predict the BU transition rate and specifies the input value set required for that prediction. In response to the user's operation, the acquisition unit 12 acquires the specified data indicating the selected target process and the specified input value set. The acquisition unit 12 may receive this data via the input device 105, read it from a predetermined storage device, or receive it from another computer.

[0034] In step S12, the prediction unit 13 selects a prediction model 20 from among multiple prediction models 20 that corresponds to the specified data.

[0035] In step S13, the prediction unit 13 inputs the input value set to the selected prediction model 20 to predict the BU transition rate in the target process. The combination of data items of one or more target data values ​​indicated by the input value set is the same as the combination of data items of one or more data values ​​indicated by the training data used to generate the selected prediction model 20. The selected prediction model 20 calculates the BU transition rate based on its input value set. The prediction unit 13 obtains this BU transition rate as the prediction result.

[0036] In step S14, the prediction unit 13 outputs the predicted BU transition rate. The prediction unit 13 may display the BU transition rate on a display device, store it in a predetermined storage device, or transmit it to another computer system.

[0037] The acquisition unit 12 can acquire specified data indicating the brewing process and a set of input values ​​corresponding to the brewing process (step S11). In this case, the set of input values ​​may include at least one of the factors F1 to F15 described above. Alternatively, the set of input values ​​may include at least two or more of the factors F1 to F15. The prediction unit 13 selects a prediction model 21 based on the specified data (step S12), inputs the set of input values ​​into the prediction model 21 to predict the BU transfer rate from hops (hop raw material) to wort (step S13), and outputs the predicted BU transfer rate (step S14).

[0038] The acquisition unit 12 may acquire specified data indicating the fermentation process and a set of input values ​​corresponding to the fermentation process (step S11). In this case, the set of input values ​​may include at least one of the factors F1 to F3 and F16 to F25 described above. Alternatively, the set of input values ​​may include at least two or more of the factors F1 to F3 and F16 to F25. The prediction unit 13 selects a prediction model 22 based on the specified data (step S12), inputs the set of input values ​​into the prediction model 22 to predict the BU conversion rate from wort to semi-finished or final product (step S13), and outputs the predicted BU conversion rate (step S14).

[0039] The acquisition unit 12 can acquire specified data indicating the post-fermentation process and an input value set corresponding to the post-fermentation process. In this case, the input value set may include at least one of the flavoring hop ratio and bitterness unit amount, and extract concentration. The prediction unit 13 selects a prediction model 23 based on the specified data (step S12), inputs the input value set to the prediction model 23 to predict the BU transition rate from semi-finished product to final product (step S13), and outputs the predicted BU transition rate (step S14).

[0040] [Differentiation] The above provides a detailed explanation of various examples provided in this disclosure. However, this disclosure is not limited to the examples given above. Various modifications are possible with respect to this disclosure, as long as they do not deviate from its essence.

[0041] As mentioned above, the BU transfer rate is just one example of the bitterness transfer rate. The prediction system may also predict bitterness transfer rates other than the BU transfer rate. For example, the prediction unit (prediction model) may predict (calculate) the transfer rate of α-acid content per unit amount of hops instead of the estimated BU transfer rate per hop raw material. Alternatively, the prediction unit (prediction model) may predict (calculate) the transfer rate of iso-α-acid content (concentration) instead of the BU transfer rate for wort, semi-finished products, or final products.

[0042] Predictive models are portable between computer systems. Therefore, a predictive system may use a predictive model generated on another computer system without having a learning unit. When a predictive system uses a single predictive model, the acquisition unit does not need to acquire specified data, and the prediction unit does not need to select a predictive model.

[0043] In this disclosure, the expression "at least one processor executes the first process, the second process, ... the nth process," or a corresponding expression, is a concept that includes cases where the entity executing the n processes from the first process to the nth process changes midway through. In other words, this expression is a concept that includes both cases where all n processes are executed by the same processor and cases where the processor changes at an arbitrary rate for the n processes.

[0044] The processing steps for a method executed by at least one processor are not limited to the examples above. For example, some of the steps or processes described above may be omitted, or each step may be performed in a different order. Also, any two or more of the steps described above may be combined, or some of the steps may be modified or deleted. Alternatively, other steps may be performed in addition to each of the steps described above.

[0045] When comparing the relative magnitudes of two numbers in a computer system or within a computer, either of the two criteria, "greater than or equal to" and "greater than," may be used, or either of the two criteria, "less than or equal to" and "less than."

[0046] [Note] As can be seen from the various examples above, this disclosure includes the following aspects: (Note 1) An acquisition unit that acquires at least one target data value related to the manufacturing conditions in a process that constitutes at least a part of the manufacturing process of a beer-flavored beverage as an input value set, A prediction unit that receives the input set of data values ​​and predicts the bitterness transfer rate in the process, into a prediction model that has been trained to receive data values ​​related to the manufacturing conditions and calculate the bitterness transfer rate, which is the rate of bitterness transfer in the process, A prediction system equipped with the following features. (Note 2) The acquisition unit further acquires the specified data for specifying the prediction model, The prediction unit, From among the multiple prediction models corresponding to the multiple processes, select the prediction model corresponding to the specified data. The selected prediction model is input to the set of input values ​​to predict the bitterness transfer rate. The prediction system described in Appendix 1. (Note 3) The aforementioned set of input values The amount of bitterness from the hops added in the brewing process after the boiling process, The amount of bitterness per unit volume of hops added in the brewing process after the boiling process, The ratio of the amount of hop bitterness added in the brewing process after the boiling process to the total amount of hop bitterness added in the brewing process, The amount of bitterness from hops added 60 minutes or more after the start of the boiling process, The amount of bitterness per unit volume of hops added 60 minutes or more after the start of the boiling process, The ratio of the amount of bitterness from hops added 60 minutes or more after the start of the boiling process to the total amount of bitterness from hops added throughout the entire brewing process, The amount of bitterness from the hops added in the latter half of the boiling process, The amount of bitterness per unit volume of hops added in the latter half of the boiling process, The ratio of the amount of bitterness from hops added in the latter half of the boiling process to the total amount of bitterness from hops added throughout the entire brewing process, The total weight of hops added throughout the brewing process, The weight of hops added per unit volume throughout the entire brewing process, The weight of the acid used in the preparation process, The weight per unit volume of acid used in the preparation process, The amount of bitterness from the hop extract added during the boiling process, The amount of bitterness per unit volume of hop extract added during the boiling process, The amount of bitterness from the hops added during the fermentation or storage process, The amount of bitterness per unit volume of hops added during the fermentation or storage process, Fermentation temperature and Types of yeast, The amount of yeast added per unit volume of liquid, The amount of acid used in the fermentation or storage process, The amount of acid used per unit volume in the fermentation or storage process, The ratio of malt used, The value of the raw wort extract or pseudo-unfermented extract of the wort to be fermented, The amount of air per unit time that is aerated through the wort, including at least one of the following: The prediction system described in Appendix 1 or 2. (Note 4) The aforementioned process is a process of producing wort using hops, The aforementioned bitterness transfer rate is the bitterness transfer rate from the hops to the wort. The aforementioned set of input values The amount of bitterness from the hops added in the brewing process after the boiling process, The amount of bitterness per unit liquid volume of hops added in the brewing process after the boiling process, The ratio of the amount of bitterness from hops added in the brewing process after the boiling process to the total amount of bitterness from hops added in the brewing process, The amount of bitterness from hops added 60 minutes or more after the start of the boiling process, The amount of bitterness per unit volume of hops added 60 minutes or more after the start of the boiling process, The ratio of the amount of bitterness from hops added 60 minutes or more after the start of the boiling process to the total amount of bitterness from hops added throughout the entire brewing process, The amount of bitterness from the hops added in the latter half of the boiling process, The amount of bitterness per unit volume of hops added in the latter half of the boiling process, The ratio of the amount of bitterness from hops added in the latter half of the boiling process to the total amount of bitterness from hops added throughout the entire brewing process, The total weight of hops added in the aforementioned brewing process, The weight of hops added per unit liquid volume throughout the aforementioned brewing process, The weight of the acid used in the aforementioned preparation process, The weight per unit volume of acid used in the aforementioned preparation process, The amount of bitterness of the hop extract added in the boiling process, The amount of bitterness per unit volume of hop extract added in the boiling process, including at least one of the following: The prediction system described in Appendix 3. (Note 5) The aforementioned process is a process for producing a semi-finished or final product of the beer-flavored beverage from the wort, The bitterness transfer rate is the bitterness transfer rate from the wort to the semi-finished or final product. The aforementioned set of input values The amount of bitterness from the hops added in the brewing process after the boiling process, The amount of bitterness per unit liquid volume of hops added in the brewing process after the boiling process, The ratio of the amount of bitterness from hops added in the brewing process after the boiling process to the total amount of bitterness from hops added in the brewing process, The amount of bitterness from the hops added in the fermentation or storage process, The amount of bitterness per unit volume of hops added in the fermentation process or storage process, The fermentation temperature and, The aforementioned types of yeast, The amount of yeast added per unit volume of liquid, The amount of acid used in the fermentation process or storage process, The amount of acid used per unit liquid volume in the fermentation process or storage process, The malt usage ratio mentioned above, The value of the raw wort extract or pseudo-unfermented extract of the wort to be fermented, The amount of air per unit time that is passed through the wort, including at least one of the following: The prediction system described in Appendix 3. (Note 6) A prediction method performed by a prediction system comprising at least one processor, A step of obtaining at least one target data value as an input value set that relates to the manufacturing conditions in a process that constitutes at least a part of the manufacturing process of a beer-flavored beverage, A step of inputting the set of input values ​​into a predictive model that has been trained to accept data values ​​related to the manufacturing conditions and calculate the bitterness transfer rate, which is the rate of bitterness transfer in the process, in order to predict the bitterness transfer rate in the process. A prediction method that includes this.

[0047] According to Appendix 1,6, a predictive model pre-trained to calculate the bitterness transfer rate from data values ​​related to manufacturing conditions in the beer-flavored beverage manufacturing process is used to predict the bitterness transfer rate corresponding to the input value set. By introducing this predictive model, the prediction results for the bitterness transfer rate can be easily obtained, and by using these prediction results, the bitterness transfer rate in the beer-flavored beverage manufacturing process can be predicted with higher accuracy.

[0048] According to Appendix 2, multiple prediction models corresponding to multiple processes are prepared, and the bitterness transfer rate is predicted using the specified prediction model. Since a prediction model is prepared for each process, the bitterness transfer rate can be predicted more accurately for each of the multiple processes. In addition, since multiple processes are covered by the prediction system, the practicality and convenience of the prediction system can be improved.

[0049] According to Appendix 3, at least one of the various factors that may contribute to the bitterness transfer rate throughout the entire manufacturing process of beer-flavored beverages is used as input to the predictive model, so the bitterness transfer rate can be predicted more accurately.

[0050] According to Appendix 4, at least one of the various factors that may contribute to the rate of bitterness transfer from hops to wort is used as input to a predictive model corresponding to the process of producing wort using hops, so that the rate of bitterness transfer can be predicted more accurately.

[0051] According to Appendix 5, at least one of the various factors that may contribute to the rate of bitterness transfer from wort to the semi-finished or final product of a beer-flavored beverage is used as input to a predictive model corresponding to the process of manufacturing the semi-finished or final product from wort, so that bitterness transfer rate can be predicted more accurately. [Explanation of symbols]

[0052] 10...Prediction system, 11...Learning unit, 12...Acquisition unit, 13...Prediction unit, 20-23...Prediction model, 30...Database, 110...Prediction program.

Claims

1. An acquisition unit that acquires at least one target data value related to the manufacturing conditions in a process that constitutes at least a part of the manufacturing process of a beer-flavored beverage as an input value set, A prediction unit that receives the input set of data values ​​and predicts the bitterness transfer rate in the process, into a prediction model that has been trained to receive data values ​​related to the manufacturing conditions and calculate the bitterness transfer rate, which is the rate of bitterness transfer in the process, A prediction system equipped with the following features.

2. The acquisition unit further acquires the specified data for specifying the prediction model, The prediction unit, From among the multiple prediction models corresponding to the multiple processes, select the prediction model corresponding to the specified data. The selected prediction model is input to the set of input values ​​to predict the bitterness transfer rate. The prediction system according to claim 1.

3. The aforementioned set of input values The amount of bitterness from the hops added in the brewing process after the boiling process, The amount of bitterness per unit volume of hops added in the brewing process after the boiling process, The ratio of the amount of hop bitterness added in the brewing process after the boiling process to the total amount of hop bitterness added in the brewing process, The amount of bitterness from hops added 60 minutes or more after the start of the boiling process, The amount of bitterness per unit volume of hops added 60 minutes or more after the start of the boiling process, The ratio of the amount of bitterness from hops added 60 minutes or more after the start of the boiling process to the total amount of bitterness from hops added throughout the entire brewing process, The amount of bitterness from the hops added in the latter half of the boiling process, The amount of bitterness per unit volume of hops added in the latter half of the boiling process, The ratio of the amount of bitterness from hops added in the latter half of the boiling process to the total amount of bitterness from hops added throughout the entire brewing process, The total weight of hops added throughout the brewing process, The weight of hops added per unit volume throughout the entire brewing process, The weight of the acid used in the preparation process, The weight per unit volume of acid used in the preparation process, The amount of bitterness from the hop extract added during the boiling process, The amount of bitterness per unit volume of hop extract added during the boiling process, The amount of bitterness from the hops added during the fermentation or storage process, The amount of bitterness per unit volume of hops added during the fermentation or storage process, Fermentation temperature and Types of yeast, The amount of yeast added per unit volume of liquid, The amount of acid used in the fermentation or storage process, The amount of acid used per unit volume in the fermentation or storage process, The ratio of malt used, The value of the raw wort extract or pseudo-unfermented extract of the wort to be fermented, The amount of air per unit time that is aerated through the wort, including at least one of the following: The prediction system according to claim 1 or 2.

4. The aforementioned process is a process of producing wort using hops, The aforementioned bitterness transfer rate is the bitterness transfer rate from the hops to the wort. The aforementioned set of input values The amount of bitterness from the hops added in the brewing process after the boiling process, The amount of bitterness per unit liquid volume of hops added in the brewing process after the boiling process, The ratio of the amount of bitterness from hops added in the brewing process after the boiling process to the total amount of bitterness from hops added in the brewing process, The amount of bitterness from hops added 60 minutes or more after the start of the boiling process, The amount of bitterness per unit volume of hops added 60 minutes or more after the start of the boiling process, The ratio of the amount of bitterness from hops added 60 minutes or more after the start of the boiling process to the total amount of bitterness from hops added throughout the entire brewing process, The amount of bitterness from the hops added in the latter half of the boiling process, The amount of bitterness per unit volume of hops added in the latter half of the boiling process, The ratio of the amount of bitterness from hops added in the latter half of the boiling process to the total amount of bitterness from hops added throughout the entire brewing process, The total weight of hops added in the aforementioned brewing process, The weight of hops added per unit liquid volume throughout the aforementioned brewing process, The weight of the acid used in the aforementioned preparation process, The weight per unit volume of acid used in the aforementioned preparation process, The amount of bitterness of the hop extract added in the boiling process, The amount of bitterness per unit volume of hop extract added in the boiling process, including at least one of the following: The prediction system according to claim 3.

5. The aforementioned process is a process for producing a semi-finished or final product of the beer-flavored beverage from the wort, The bitterness transfer rate is the bitterness transfer rate from the wort to the semi-finished or final product. The aforementioned set of input values The amount of bitterness from the hops added in the brewing process after the boiling process, The amount of bitterness per unit liquid volume of hops added in the brewing process after the boiling process, The ratio of the amount of bitterness from hops added in the brewing process after the boiling process to the total amount of bitterness from hops added in the brewing process, The amount of bitterness from the hops added in the fermentation or storage process, The amount of bitterness per unit volume of hops added in the fermentation process or storage process, The fermentation temperature and, The aforementioned types of yeast, The amount of yeast added per unit volume of liquid, The amount of acid used in the fermentation process or storage process, The amount of acid used per unit liquid volume in the fermentation process or storage process, The malt usage ratio mentioned above, The value of the raw wort extract or pseudo-unfermented extract of the wort to be fermented, The amount of air per unit time that is passed through the wort, including at least one of the following: The prediction system according to claim 3.

6. A prediction method performed by a prediction system comprising at least one processor, A step of obtaining at least one target data value as an input value set that relates to the manufacturing conditions in a process that constitutes at least a part of the manufacturing process of a beer-flavored beverage, A step of inputting the set of input values ​​into a predictive model that has been trained to accept data values ​​related to the manufacturing conditions and calculate the bitterness transfer rate, which is the rate of bitterness transfer in the process, in order to predict the bitterness transfer rate in the process. A prediction method that includes this.

Images