Milk production calculation system

Abstract

To provide a milk secretion amount calculation system for calculating a milk secretion amount after a predicted start date on the basis of data before the predicted start date.SOLUTION: The milk secretion amount calculation system includes a memory for storing farm data, an interpolation formula creating part for creating an interpolation formula for interpolating a relationship between the number of days after parturition for a certain period in the farm data prior to the predicted starting date and the actual milk amount, a regression formula creating part for creating a regression formula for calculating a herd estimated milk secretion amount for each of the number of days after parturition, with the interpolated milk amount at the specific number of days after postpartum calculated from the interpolation formula as the objective variable, and with at least one factor parameter selected from the farm data as the explanatory variable, and a control device having a herd estimated milk secretion amount calculation part including a herd regression formula calculation part for calculating an estimated milk secretion amount by substituting the value of the explanatory variable assumed on the predicted date into the regression formula corresponding to the postpartum period on the predicted date.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a milk production calculation system that estimates the milk production of a dairy cow according to the number of days since calving based on past data. [Background technology]

[0002] In the dairy industry, increases or decreases in milk production are a major issue that directly affect farm management. In order to increase milk production, efforts are being made to improve feed composition, the rearing environment, and reduce stress on dairy cows.

[0003] On the other hand, these improvements depend largely on the individual farm, making it difficult to compare them with other farms. Dairy cows are generally weak to heat, so it is thought that regions at higher latitudes produce more milk, but it is not clear to what extent differences in climate affect milk production.

[0004] Furthermore, since there are a great many factors that can affect milk production, the current situation is that we rely on experience to determine which factors have a significant impact on milk production in which cases.

[0005] Therefore, a system that can estimate milk production by inputting possible factors would provide a great guideline for dairy farming management.

[0006] Patent Document 1 provides a system that uses an information terminal to keep track of dairy cows in groups according to lactation stage or breeding stage.

[0007] Furthermore, Patent Document 2 discloses a method for calculating a lactation curve for each individual cow using a regression equation based on past milk yields. [Prior art documents] [Patent documents]

[0008] [Patent Document 1] Japanese Patent Application Publication No. 2020-156359 [Patent Document 2] Japanese Patent Publication No. 2020-020707 Summary of the Invention [Problem to be solved by the invention]

[0009] Patent Document 1 attempts to understand the current state of dairy cows, while Patent Document 2 attempts to estimate future milk production from past milk production. However, neither method can investigate factors that increase milk production in dairy cows.

[0010] Although it is understood that the physiological theory that a decrease in oxytocin and the secretion of adrenaline due to high temperatures and stress affect milk production is understood, it is difficult for humans to understand what kind of rearing environment or state of the dairy cow is comfortable for the dairy cow. Therefore, it is thought that a method of collecting many events (information), analyzing the factors, and trying to increase milk production is useful.

[0011] Here, "a lot of information" refers to information on the rearing environment, feed composition, and individual information on the dairy cows themselves from as many farms as possible. This information is updated daily. To ensure diversity of information, it is also desirable to have information from different regions. Therefore, it is desirable to make data from as many farms as possible available.

[0012] However, it is extremely difficult to regularly collect data from multiple farms of different sizes and circumstances. Depending on the farm, various data may be quantified daily, while in other cases data may only be collected every few days.

[0013] In particular, the so-called lactation curve shows the relationship between the number of days since calving and milk production, but unless the farm is very large, it is difficult to predict where dairy cows are all one day apart from each other after calving. Therefore, even with the milk production data that can actually be collected, there are gaps and the range of information (range of number of days since calving) varies, making it impossible to analyze uniformly, which is an issue.

[0014] In addition, predicting future milk production and sales is important for dairy farm management. However, estimating future milk production for multiple dairy cows, whose postpartum days change daily, has not been widely done until now. [Means for solving the problem]

[0015] The milk production calculation system of the present invention was devised in consideration of the above-mentioned problems, and provides a system that enables factor analysis and presents future predicted values ​​of milk production even when the collected information on milk production is not within the same range.

[0016] More specifically, the milk production calculation system according to the present invention comprises: A milk yield calculation system that calculates the milk yield of an individual cow on a predicted date after the prediction start date as an estimated milk yield of a herd based on the number of days since calving and actual milk yield of each individual cow constituting the herd before the prediction start date, and farm data including weather data, a memory for storing the farm data; an interpolation formula creating unit that creates an interpolation formula that interpolates the relationship between the number of days since calving and the actual milk yield for a certain period in the farm data before the prediction start date; a regression equation creation unit that creates a regression equation for calculating the estimated milk yield of the herd for each postpartum day using the interpolated milk yield calculated from the interpolation equation as a response variable and at least one factor parameter selected from the farm data as an explanatory variable; and The regression equation corresponding to the number of days since delivery on the predicted date is: The device is characterized by having a control device having a herd estimated milk yield calculation unit that includes a herd regression equation calculation unit that substitutes the values ​​of the explanatory variables expected on the predicted date and calculates the herd estimated milk yield. [Effects of the Invention]

[0017] The milk production calculation system of the present invention has the advantage that it can interpolate collected data on milk production to obtain a certain range of milk production even on a daily basis, which is useful for estimating factors that affect milk production.

[0018] Furthermore, if the influence of predictable factors such as temperature on milk production is understood, it will be possible to predict future milk production, which will provide important guidelines for dairy farming management. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a diagram showing the configuration of a milk production calculation system according to the present invention. [Figure 2] FIG. 1 is a diagram showing the overall (main) flow of the milk production calculation system. [Figure 3] FIG. 10 is a diagram illustrating an example of an output calculated by the milk production calculation system. [Figure 4] FIG. 1 is a diagram showing a flow of steps for calculating an estimated milk production. [Figure 5] FIG. 10 is a diagram illustrating the creation of an interpolation formula. [Figure 6] FIG. 10 is a diagram showing a flow for creating an interpolation formula. [Figure 7] FIG. 10 is a diagram illustrating the creation of a regression equation. [Figure 8] FIG. 1 is a diagram showing a process flow for creating a regression equation. [Figure 9] FIG. 3(b) is an enlarged view of the curve relating to the estimated milk production in FIG. 3(a). [Figure 10] FIG. 5 is a diagram showing the process of factor analysis in step S206 of FIG. 4. [Figure 11] FIG. 10 is a diagram showing a comparison of actual measured milk yields over a certain period of time with estimated milk yield EY calculated from a regression equation in factor analysis. [Figure 12] FIG. 10 is a diagram showing the flow of herd regression equation calculation. [Figure 13] FIG. 1 is a diagram illustrating the estimated total milk production of a herd. [Figure 14] FIG. 10 is a diagram showing the flow of calculating the total herd regression equation. [Figure 15] This is a diagram showing the flow for calculating predicted sales revenue of a herd using an estimated total milk production of the herd, which has a herd estimated total sales revenue calculation unit. [Figure 16] This is a diagram showing the flow for calculating predicted sales revenue of a herd having a herd estimated total sales revenue calculation unit using the herd estimated milk production volume of each individual cow. [Figure 17] FIG. 1 is a diagram illustrating herd dry days. [Figure 18] This is a diagram showing the flow of determining herd dry cows by comparing the herd's estimated milk production for each individual cow with the dry cow standard value. [Figure 19] This is a diagram showing the flow of determining herd dry cow status by comparing the herd dry date with the number of days since calving for each individual cow. [Figure 20] FIG. 1 is a diagram illustrating a procedure for calculating an individual estimated milk yield. [Figure 21] FIG. 1 is a diagram illustrating an individual estimated total milk production. [Figure 22] FIG. 10 is a diagram showing a flow for calculating an individual estimated milk yield. [Figure 23] FIG. 10 is a diagram showing a flow of individual regression equation calculation. [Figure 24] FIG. 10 is a diagram showing the flow of calculating the individual total regression equation. [Figure 25] FIG. 10 is a diagram showing a flow for calculating predicted sales of an individual using an individual estimated total milk production amount, the individual estimated total sales amount calculation unit. [Figure 26] FIG. 10 is a diagram showing the flow for calculating predicted sales revenue of a herd having an individual estimated total sales revenue calculation unit using the individual estimated milk production volume of each individual cow. [Figure 27] FIG. 1 is a diagram illustrating an individual dry date. [Figure 28] FIG. 10 is a diagram showing the flow of determining whether an individual cow is a dry cow by comparing the individual estimated milk production of each individual cow with the dry cow standard value. [Figure 29] FIG. 10 is a diagram showing the flow of determining individual dry cow status by comparing the individual dry date of each individual cow with the number of days since calving for each individual cow. [Figure 30] This is a diagram showing that the milk production calculation system was used on an actual farm, and estimated milk production can be subjected to multiple regression with THI and sunshine hours as factor parameters. DETAILED DESCRIPTION OF THE INVENTION

[0020] The milk production calculation system according to the present invention will be described below with reference to the drawings. Note that the following description exemplifies one embodiment of the present invention and one example, and the present invention is not limited to the following description. The following description can be modified within the scope of the present invention.

[0021] The configuration of the milk production calculation system according to the present invention is shown in Figure 1. The milk production calculation system 1 according to the present invention is composed of a terminal 10, a main body 12, and a memory 14. The terminal 10 can be composed of a computer having a CPU (Central Processor Unit), memory, and a display screen, and can include a mobile communication terminal (such as a so-called smartphone).

[0022] The main body 12 is composed of a CPU (Central Processor Unit). It may be a single CPU or multiple CPUs connected together. The main body 12 is a device that provides services, so it may be called a server. The memory 14 is memory used by the main body 12. The CPU may be called a control device. Therefore, the main body 12 is a control device. The main body 12 may also have an input unit 12a that receives signals from the terminal 10. In other words, the main body 12 receives commands and numerical values ​​specified by the input unit 12a.

[0023] The main body 12 and the terminal 10 are connected to be able to communicate with each other. The communication may be wired or wireless. The terminal 10 may also be called a client. The terminal 10 is preferably installed at a ranch 16 where dairy cows are actually raised (ranches 16A, 16B, and 16C are shown as examples here). Here, "installed" may mean that the terminal 10 is physically installed at the ranch 16, or it may mean that the person in charge or in charge of the ranch 16 has a mobile communication terminal. Of course, the terminal 10 may be installed at a location other than the ranch 16. Furthermore, the ranch 16 may be any ranch where dairy cows are raised.

[0024] A plurality of farms 16 may be connected to the main body 12. This is because the milk production calculation system 1 calculates milk production by collecting and analyzing a large amount of data related to milk production, and therefore it is desirable to collect a large amount of data. In addition, the main body 12 may be connected to external information 18 via the Internet. This is because not only data obtained from the farms 16 but also information on nationwide weather can be obtained and used via the Internet.

[0025] In this way, the milk production calculation system 1 collects a large amount of data from various places and calculates the milk production amount by analyzing the data, so a preferred embodiment is to provide the service via a network. In other words, the milk production calculation system 1 may be configured as a cloud.

[0026] Therefore, the cloud form for implementing the milk production calculation system is preferably the typical SaaS (Software as a Service), but it may also be in the form of PaaS (Platform as a Service), HaaS (Hardware as a Service), or IaaS (Infrastructure as a Service).

[0027] The milk production calculation system 1 according to the present invention treats a group of dairy cows as a herd and predicts the milk production of the herd. Each dairy cow belonging to the herd is treated as an individual cow (hereinafter simply referred to as an "individual"), and the milk production of each individual cow is predicted. Furthermore, from these predicted milk production amounts, it is possible to estimate sales revenue, dry days, and dry cow status. Dairy cows begin secreting milk after giving birth, but the amount of milk produced decreases after a certain number of days have passed since giving birth. Such cows are rested until their next birth. This is because, after the dry day, no milk production can be expected from the dairy cow until their next birth. These cows are called "dry cows." A "dry day" refers to the day on which a dairy cow is treated as a dry cow.

[0028] More specifically, when the value of the factor parameter P (explanatory variable) expected for a future prediction date from the prediction start date is input from the terminal 10, the milk yield calculation system 1 calculates the estimated milk yield of the herd that can be milked from the dairy cows on the prediction date or the prediction period, which is N days after calving. N EY (kg) or individual estimated milk yield N These are also simply called herd estimated milk yield EY and individual estimated milk yield DEY, omitting the number of days since calving N. When referring to herd estimated milk yield EY or individual estimated milk yield DEY collectively, the term "herd estimated milk yield" is used. N It is called "EY (kg), etc." or "estimated milk yield of the herd, EY, etc."

[0029] The herd estimated milk yield EY is the estimated milk yield of a single dairy cow, even though it is called a herd. To calculate the milk yield of all individual cows belonging to a herd (the "herd estimated total milk yield" described below), it is necessary to calculate the sum of the herd estimated milk yields EY for each individual cow at the number of days after calving.

[0030] Here, the prediction start date is the date on which the prediction is made. The prediction start date can also be a week or a month. Usually, the prediction start date can be the date on which the prediction is made, but it can also be a past day or period. The prediction date is a day in the future from the prediction start date, and is the estimated milk yield of the herd. N The forecast period is the period from the forecast start date to the future, and is the estimated milk yield of the herd. N This is the period for which you want to predict EY (kg), etc. When referring to the forecast date, you can also include the forecast period.

[0031] Furthermore, the factor parameter P is an item that can best explain the milk yield among farm data including weather data, individual data such as the weight of the dairy cow itself and the amount of feed intake, and feed data such as the type of feed consumed. This factor parameter P may be data obtained by processing data of directly measurable items in addition to directly measurable items. This factor parameter P can be found within the milk yield calculation system 1.

[0032] The climate data includes sunrise time, sunset time, temperature, humidity, sunshine hours, solar radiation, wind direction, wind speed, etc. It may also include the temperature and humidity inside the cowshed. It may also include processed data of these, such as THI (Temperature Humidity Index). For example, THI is an index expressed by equation (1).

[0033]

number

[0034] Furthermore, the climate data may not only be the climate inside and outside the barn, but may also be the climate within a 1 km square of the farm, the local climate, or climate data from a national weather forecast service.

[0035] Individual data is data on the dairy cow itself, specifically, registration number, pedigree, parity, number of days since calving, milk yield, weight, medical history, owner, place of rearing, etc.

[0036] The feed data includes the type, composition, and number of times of feeding in the past.

[0037] [Basic data] In this way, the factor parameter P is a concept that includes data before the prediction start date. Data before the prediction start date that is used for estimation is called "basic data." The "basic data" is data for creating a regression equation for prediction. It can also be said to be part of the farm data. The milk yield calculation system 1 according to the present invention finds a regression equation that can explain the milk yield of a certain dairy herd or individual dairy cow as a dependent variable based on the factor parameter P before the prediction start date, and applies the factor parameter P (explanatory variable) assumed for a prediction date after the prediction start date to the regression equation, thereby calculating the herd estimated milk yield that can be milked from a dairy cow on a prediction date after the prediction start date or N days after calving during the prediction period. N This can also be rephrased as the amount by which EY (kg) etc. can be obtained.

[0038] A terminal 10 installed on a ranch 16 transmits individual data related to the dairy cows on the ranch, data on the feed actually consumed, and daily weather data as ranch data FD to the main unit 12. The main unit 12 stores this ranch data in memory 14. When the prediction start date is input from the terminal 10, a formula for calculating the estimated milk yield of the herd, etc. is found based on the factor parameters P prior to the prediction start date. Subsequently, the estimated milk yield of the herd according to the number of days since calving N is calculated according to the factor parameters P assumed for the input prediction date or prediction period. N EY etc. are calculated and displayed on the terminal 10.

[0039] In addition, the estimated milk yield of the herd according to the number of days after calving, N, is N The factor parameter P (explanatory variable) assumed for a forecast date after the forecast start date, which is input when calculating EY etc., is the value of the factor parameter P input into the regression equation, and is the value of the explanatory variable of the regression equation. The value of this factor parameter P (explanatory variable) may not only be a value published in a weather forecast etc., but also a value for a corresponding day or period in a typical year or in the past, or any value calculated based on these.

[0040] In other words, the factor parameter P, which is the explanatory variable, can use past values. Because it is thought that the annual climate does not change significantly throughout the four seasons, the forecast date and forecast period after the forecast start date will not differ significantly from the days and periods corresponding to the forecast date last year or earlier.

[0041] The main unit 12 may also acquire climate data from external information 18. In a closed-type barn, the environment inside the barn is managed, making it easy to transmit data. However, in an open-type barn, it is difficult to acquire the wind speed inside the barn on that day. In such cases, external climate data can be used as a reference.

[0042] Various types of weather data are available, and these can be used effectively. The farm data FD may also be transmitted to the main unit 12 by a method other than the terminal 10. Of course, the terminal 10 that transmits the farm data FD to the main unit 12 and the terminal 10 that inputs the factor parameters P and calculates the milk yield may be separate.

[0043] Figure 2 shows the overall (main) flow of the milk production calculation system 1. It is assumed that each farm transmits farm data FD to the main unit 12 as needed. It is also assumed that the dairy cows belonging to the herd that is the unit of handling are registered in advance.

[0044] The milk production calculation system 1 can return results from the steps of calculating the herd's estimated milk production EY, calculating the herd's predicted sales revenue E$, determining the herd's dry cow DC, calculating the individual's estimated milk production DEY, calculating the individual's predicted sales revenue DE$, and determining the individual dry cow DDC in response to inputs such as calculation conditions and factor parameters P (including explanatory variables) entered from a terminal 10, etc.

[0045] In addition, the estimated total milk production of the herd SUM EY and individual estimated total milk production SUM DEY Furthermore, the step of calculating the herd predicted sales amount E$ and the step of calculating the individual predicted sales amount DE$ may include a step of calculating the herd estimated individual sales amount EI$, a step of calculating the herd estimated total sales amount ΣE$, a step of calculating the individual estimated sales amount DEI$, a step of calculating the individual estimated total sales amount ΣDE$, etc.

[0046] These results may be returned to the terminal 10. Alternatively, these results may be stored in a specific memory on the network. The terminal 10 can display these results. The process of calculating these can be called a menu of the main flow. In other words, the terminal 10 can instruct these calculations.

[0047] Note that the majority of the configuration of the present invention is implemented by software. Therefore, a "step" in a process refers to a group of processes, and the main body 12 can be considered to have "units" that execute the "steps." Specifically, the main body 12 (control device) can be said to have a herd estimated milk yield calculation unit, a herd predicted sales calculation unit, a herd dry cow determination unit, an individual estimated milk yield calculation unit, an individual predicted sales calculation unit, and an individual dry cow determination unit. It can also be said to have a herd estimated total milk yield calculation unit, an individual estimated total milk yield calculation unit, a herd estimated individual sales calculation unit, a herd estimated total sales calculation unit, an individual estimated sales calculation unit, and an individual estimated total sales calculation unit.

[0048] Furthermore, if there is a group of lower-level processes within these processing steps, they may also be referred to as "units." For example, the herd estimated milk yield calculation unit includes a step of creating an interpolation formula and a step of creating a regression formula, which can be referred to as an "interpolation formula creation unit" and a "regression formula creation unit," respectively. Furthermore, there is a step (step S210) of specifically calculating the herd estimated milk yield when the value of the factor parameter P (explanatory variable) is substituted into the obtained regression formula, which can be referred to as a herd regression formula calculation unit. Note that the milk yield calculation system 1 is only required to have at least a step of calculating the herd estimated milk yield EY (herd estimated milk yield calculation unit).

[0049] <Milk production calculation system> 2, when the milk production calculation system 1 starts (step S100), a termination decision is made (step S102). The termination decision may be made by a termination instruction from the terminal 10 or by disconnection of communication between the main body 12 and the terminal 10. If termination is to be made (Y branch of step S102), the process is terminated (step S104). If termination is not to be made (N branch of step S102), the process proceeds to the next step.

[0050] Next, a selection is made as to whether to calculate the herd estimated milk yield EY or the individual estimated milk yield DEY (steps S106, S108). For each option, if a selection is made (Y branch), the process proceeds to the respective processing, and if a selection is not made (N branch), the process proceeds to the next step. If the calculation of the individual estimated milk yield DEY in step S108 is not performed (N branch in step S108), the process returns to the end decision (step S102).

[0051] When the herd estimated milk yield EY is selected (Y branch in step S106), the herd estimated milk yield EY is calculated (step S110). Then, a choice is made between calculating the herd predicted sales amount E$ (step S112) or determining the herd dry cow DC (step S114). When each of these processes is selected (Y branch in steps S112 and S114), the herd predicted sales amount E$ is calculated and the herd dry cow DC is determined (steps S116 and S118), respectively. Then, processing moves to the selection of the individual estimated milk yield DEY (step S108).

[0052] If the individual estimated milk yield DEY is selected (Y branch in step S108), the individual estimated milk yield DEY is calculated (step S120). Then, a choice is made between calculating the individual predicted sales amount DE$ (step S122) or determining the individual dry cow DDC (step S124). If each of these processes is selected (Y branch in steps S122 and S124), the individual predicted sales amount DE$ is calculated and the individual dry cow DDC is determined (steps S126 and S128), respectively. Then, the process proceeds to the end determination (step S102).

[0053] FIG. 3 shows examples of each output. FIG. 3(a) is an output example of the herd estimated milk yield EY. The horizontal axis is the number of days since calving (days), and the vertical axis is the milk yield (here, the unit is "kg"). It is possible to show the herd estimated milk yield EY versus the number of days since calving under a specific factor parameter P (for example, temperature). The figure shows the herd estimated milk yield curves M when the temperature is T1°C and when it is T2°C. The herd estimated milk yield curve M provided by the milk yield calculation system 1 of the present invention is a collection of calculated values ​​(herd estimated milk yield EY) on a minimum daily basis. These calculated values ​​may be connected by a straight line or a curve. The herd estimated milk yield curve formed using the herd estimated milk yield EY is represented by the symbol M. In other words, the herd estimated milk yield curve M is obtained by plotting the herd estimated milk yield EY for each appropriate number of days since calving N and connecting them by a straight line or a curve.

[0054] Figure 3(b) is an output example of predicted herd sales E$. The horizontal axis is the prediction date, and the vertical axis is sales (unit: yen). Once the herd's estimated milk yield EY is calculated, the herd's predicted sales E$ on the prediction date can be calculated by multiplying it by the unit price of milk (yen / kg). Note that the process of calculating the herd's predicted sales E$ may also involve calculating the herd's estimated individual sales EI$, which is the sales for each individual cow, and the herd's estimated total sales ΣE$, which is the sales from all the dairy cows that make up the herd.

[0055] Figure 3(c) shows the results of the herd dry cow DC determination. The horizontal axis is the number of days since calving (days), and the vertical axis is the milk yield (unit: kg). The herd estimated milk yield EY (herd estimated milk yield curve M) on a certain predicted date η is calculated, and the dry cow standard value DY for determining a cow as dry is calculated. TH The following is the number of days since calving. This day is called the herd dry day, DYday. The herd dry day, DYday, can be said to be the day when the herd's estimated milk production decreases and reaches the dry standard value.

[0056] Therefore, an individual cow whose number of days since calving is greater than the herd dry day DYday is determined to be a herd dry cow DC. In Figure 3(c), the symbol DC (black triangle) is a herd dry cow DC because the number of days since calving is greater than DYday. On the other hand, NOT DC (black circle) is not a herd dry cow DC because the number of days since calving is less than DYday.

[0057] Figure 3(d) is an example of an output of an individual estimated milk yield DEY. The horizontal axis is the number of days since calving (days), and the vertical axis is the milk yield (units are "kg" here). The individual estimated milk yield DEY is calculated as a percentage of the herd estimated milk yield EY. The curve representing the individual estimated milk yield DEY is the individual estimated milk yield curve DM. In Figure 3(d), the curve DM representing the individual estimated milk yield DEY is below the herd estimated milk yield curve M, which represents the herd estimated milk yield EY. This indicates that this particular individual cow is a dairy cow that secretes less milk than the milk yield estimated for the herd. Therefore, for other individual cows, the curve DM may be drawn above the curve M.

[0058] Figure 3(e) is a graph showing the individual predicted sales amount DE$. The horizontal axis is the predicted date, and the vertical axis is the sales amount (for example, yen). As shown in Figure 3(d), the individual cow shown as an example here has a production amount less than the herd's estimated milk yield EY, so the individual predicted sales amount DE$ is also less than the herd's predicted sales amount E$. The individual predicted sales amount DE$ may also be calculated by calculating the individual estimated sales amount DEI$ and the individual estimated total sales amount ΣDE$, which are sales for each individual cow.

[0059] Figure 3(f) is a graph showing the individual dry cow DDC. The horizontal axis is the number of days since calving (days), and the vertical axis is the milk yield (here, the unit is "kg"). The individual cow shown in the example has a milk yield less than the estimated milk yield EY calculated for the herd, so the individual dry day DDYday will also be earlier than the herd dry day DYday. In Figure 3(f), if the number of days since calving for this individual cow is equal to or greater than the individual dry day DDYday on the predicted day η, it will be determined to be an individual dry cow DDC, and if it is less than the individual dry day DDYday, it will not be determined to be an individual dry cow DDC ( NOT When the number of parities is low, individual cows with low milk production can be treated as part of the herd, but when the number of parities increases and removal from the herd is considered, determining the individual dry cow DDC is useful.

[0060] Referring again to Figure 2, the above overall flow may be shared between the main body 12 and the terminal 10. For example, the selection of each process and the display of the final result may be performed by the terminal 10, and each process may be performed by the main body 12. Each step will be described in detail below.

[0061] <Herd estimated milk production EY> Figure 4 shows the flow of the process for calculating the herd's estimated milk yield EY. When the process for calculating the herd's estimated milk yield EY begins (step S110), the selection of population data and the conditions for creating it are determined (step S200). A large amount of farm data is stored in memory 14 shown in Figure 1. From this, the necessary location (including the identification of the individual cows that make up the herd) and the period of basic data are selected. The conditions for calculating the herd's estimated milk yield EY are also entered here.

[0062] For example, the location may be your own farm, the herd may be all dairy cows, and the basic data period may be the previous year. The conditions may include all conditions for calculating the estimated milk yield EY of the herd. For example, it may be determining the length of the period to be compiled when referencing farm data. More specifically, the actual milk yield may be the milk yield per day, or the average milk yield every three days, every week, or every month, or the maximum milk yield for one month. Similarly, representative values ​​may be determined for other variables at regular intervals.

[0063] Also, the forecast start date and forecast date are input. The forecast start date is usually "today, the day the forecast operation is to be performed," but it may be a day in the past. Furthermore, the forecast date is essentially the factor parameter P assumed for the forecast date. As already mentioned, the actual value for the day (or period) corresponding to the past forecast date may be used as the factor parameter P assumed for the forecast date. Therefore, if the factor parameter P is derived from climate data, it is possible to make long-term forecasts.

[0064] Next, an interpolation formula is created (step S202). The interpolation formula is explained in Figure 5. In Figure 5, the horizontal axis is the number of days since calving N (days), and the vertical axis is the actual milk yield RY (kg). For example, this is data for a certain period (e.g., August 1, 2020) for multiple dairy cows on a certain farm. Each point plotted here is the actual milk yield RY for each individual cow, so it becomes a scatter plot. However, because it is actual milk yield, there are also ranges that cannot be plotted. This is the case when there are no dairy cows with the corresponding number of days since calving. In Figure 5, this is represented by a no-data area VR.

[0065] In this way, if there are gaps in the farm data, there will be insufficient data when creating the regression equation later, and the accuracy of the regression will decrease. Therefore, this scatter plot is approximated with an appropriate function. A suitable function to use is the Wood curve. The Wood curve is a curve expressed by equation (2) and is well known as a curve that represents milk production Y against the number of days since calving N. However, to interpolate an actual scatter plot, it is not necessary to be limited to the Wood curve, and other functions can also be used.

[0066]

number

[0067] Here, A, B, and C are constants, Y is milk yield, N is number of days since calving, and e is Napier's number. The constants A, B, and C can be determined so that this Wood curve fits the scatter plot using the least squares method. In this way, fitting the relationship between number of days since calving N and milk yield Y with a continuous function is called "creating an interpolation equation." The equation used to interpolate the scatter plot can be more generalized and expressed as equation (3).

[0068]

number

[0069] In other words, in the interpolation formula (3), milk yield Y is expressed as a function of the number of days since calving N. The Wood curve is preferably used as the form of the function, but is not limited to this. The interpolation formula creation step (step S202) in Figure 4 is a step of determining the interpolation formula (3) from the scatter diagram of actual milk yield RY as described above.

[0070] [Create Interpolation Formula] Figure 6 shows the process of creating the interpolation formula in Figure 4 (step S202). When the process of creating the interpolation formula begins (step S202), data on postpartum days N and actual milk yield RY is extracted (step S230) from the population and conditions determined in step S200 of Figure 4 (EY calculation flow). These can be called actual data. Specifically, a scatter plot may be drawn. The actual data is a set of postpartum days N and actual milk yield RY for each dairy cow.

[0071] Next, an interpolation formula that best reflects the actual data of postpartum days N and milk yield Y is found. For example, the least squares method using the Wood curve shown above is applied to determine the constants A, B, and C (step S232) (see also Figure 5). This can be said to be a process of finding a function to fit to the actual milk yield RY. Then, the interpolation formula (3) is obtained (step S234). After that, the process returns to the EY calculation routine (Figure 4) (step S236).

[0072] An interpolation formula can be created for each day, since milking occurs almost every day. However, if there are no major changes in the environment or individual cows, you can create interpolation formula (3) by treating the average of the actual data every three days or every week as the actual data.

[0073] Also, taking a broader view of the time axis, an interpolation formula may be created by regarding the average value of actual data for one month as actual data. More specifically, the average monthly milk yield of a certain individual cow is taken as the actual milk yield RY of that dairy cow. Also, the number of days since calving for that dairy cow is taken as the average for that month (i.e., if N=10 days at the beginning of the month, the number of days since calving for that month is taken as 25 days). These may be determined in the "Population Data, Creation Conditions" (step S200) in Figure 4.

[0074] Continuing to refer to Figure 4, once the interpolation formula has been created, it is determined whether or not to perform factor analysis (step S204). This determination can be made by the user of the milk production calculation system 1 via the terminal 10. Factor analysis is a determination as to whether or not to investigate the factor parameter P that can best explain the milk production Y obtained from the interpolation formula (2) from the farm data. For example, this is used when using the milk production calculation system 1 for the first time or when the population is significantly changed.

[0075] If factor analysis is to be performed (Y branch in step S204), the factor analysis process is performed (step S206). Details of the factor analysis process will be described later. If factor analysis is not to be performed (N branch in step S204), the process proceeds to the next step. The factor analysis will be described in detail with reference to FIG. 10.

[0076] Next, a regression equation is created (step S208). The creation of the regression equation is explained in FIG. 7. For example, the interpolated milk yield of a dairy cow for each month with N days since calving can be calculated using the interpolation equation (3) (FIG. 7(a)) shown in FIGS. 5 and 6. Here, the interpolation equation is expressed in the general form Fw(), and the milk yield Y is the average milk yield for each month. Therefore, Y 1月 represents the interpolated milk yield in January, and the formula for calculating the interpolated formula (3) for dairy cows with N days postpartum is Fw 1月 It is represented as (N).

[0077] Here, 50 days after delivery is N 50 The interpolated milk yield of dairy cows 50 days after calving in each month from January to December is expressed as Fw 1月 (N 50 ), Fw 2月 (N 50 ),...,Fw 12月 (N 50 ) is calculated. Of course, Fw m (N) is an interpolation formula created for each month ("m" stands for "month").

[0078] The factor that can best explain the interpolated milk yield for each month is determined as the factor parameter P. The type of factor parameter P that is best is determined by carrying out the factor analysis process in step S206. For example, temperature, humidity, hours of sunlight, etc. If the factor parameter P to be used to calculate the herd's estimated milk yield EY has already been determined, use that parameter. There does not have to be just one factor parameter P. In other words, this involves finding a regression equation for the interpolated milk yield Y for each month using one or more factors, using the least squares method or the like.

[0079] As is well known, the regression equation is expressed as equation (4).

[0080]

number

[0081] where EY is the herd estimated milk yield, and x1, x2, . . ., x k is a factor (factor parameter P), and a1, a2, . . ., a k , where c is a constant.

[0082] Figure 7(c) shows an example of a regression equation when there is one factor. From the interpolation formula (3) shown in Figure 7(a), the interpolated milk yield Y m(See FIG. 7(b)). Here, m represents the month. The interpolated milk volume for each month may be the average of the first three days of the month. The data for only one specific day may represent the interpolated milk volume for that month.

[0083] Next, by sorting these monthly interpolated milk yields by factor parameter P (here, for example, noon temperature), the graph shown in Figure 7(c) can be obtained. Note that here, we will show an example of the results assuming that the herd estimated milk yield EY of dairy cows 50 days after calving in each month can be well explained by the noon temperature. The regression equation is expressed as equation (5).

[0084]

number

[0085] Here, x1 is the temperature at noon, and EY is the herd's estimated milk production. Note that such a regression equation is created for each number of days since calving. In other words, if the final day of post-calving day N is 300 days, then a regression equation from 1 to 300 days later can be obtained (see Figure 7(e)).

[0086] When there are multiple factors, it is not possible to describe it two-dimensionally as in Figure 7(c), but if a multiple regression equation is obtained using multiple factor parameters P, it may be possible to closely approximate the relationship between postpartum days N and interpolated milk yield Y. The regression equation obtained in this way is expressed as equation (6) (see Figure 7(d)).

[0087]

number

[0088] where: N EY is called the herd estimated milk yield EY when the dairy cow on postpartum days N has factor parameter P. Also, substituting factor parameter P into equation (6) to obtain the result can be said to "calculate the herd estimated milk yield." The calculated estimated milk yield is the herd estimated milk yield EY. In addition, the herd estimated milk yield on postpartum days N when factor parameter P is NIt is expressed as EY(P) (N may be omitted). Also, the i-th individual cow CW belonging to the herd i When expressing the estimated milk production of a herd, N EY [CW i ] etc. and [] (square brackets).

[0089] [Create regression equation] Figure 8 shows the flow of the process of creating a regression equation. When the process of creating a regression equation (step S208) in the EY calculation flow shown in Figure 4 is started, the process jumps to Figure 8, where the factor parameter P is first input (step S250). The factor parameter P can be input from the terminal 10. In other words, it is input by the user of the milk production calculation system 1. Here, it is assumed that the factor parameter P is p1 (temperature).

[0090] Next, for each fixed period m (defined in step S200) specified for the factor parameter p1, the estimated milk yield Ym for each postpartum day N is calculated using the interpolation formula (3) (step S252). In other words, the interpolated milk yield Ym in Figure 7(b) is calculated. This allows the scatter diagram in Figure 7(c) to be plotted.

[0091] Next, a regression equation is obtained using the factor parameter p1 as the explanatory variable and the interpolated milk yield Ym as the target variable (step S254). That is, at this time, for the factor parameter p1, the relationship between milk yield and the number of days since calving N can be obtained as shown in equation (6) (FIG. 7(d)). The regression equation can be obtained by any known method.

[0092] In this way, in the present invention, when calculating the herd's estimated milk yield using the factor parameter P, the actual milk yield RY is not used directly, but the interpolated milk yield calculated from the interpolation formula that interpolates the actual milk yield RY is used. Therefore, even if there is missing data in the actual milk yield RY, a reasonable value can be calculated as the average milk yield.

[0093] Equation (6) can be used to generate numbers where the number of days since calving N ranges from 1 to the final day (the final day here means the longest day since calving) (see Figure 7(e)). Regression equation (6) summarizes Figure 7(e). Once the regression equation has been determined, the process returns to the routine for determining the herd's estimated milk yield EY in Figure 4 (step S256).

[0094] Referring again to FIG. 4, once the regression equation (6) for calculating the herd estimated milk yield EY is obtained, the herd estimated milk yield EY can be calculated by inputting the factor parameter P into the regression equation (6) (step S210).

[0095] [Cow herd regression equation calculation EYI] Figure 12 shows the flow for actually calculating regression equation (6) and finding the herd's estimated milk yield EY. This is the herd regression equation calculation (EYI) process. When the herd regression equation calculation is started (step S210), the number of days since calving N and factor parameter P are input (step S280). Next, the input number of days since calving N and factor parameter P are used with equation (6) to find the herd's estimated milk yield at the number of days since calving N when factor parameter P is used. N EY is calculated (step S282), and may be displayed after calculation.

[0096] Next, the user confirms the re-entry (step S284). After re-entry, the estimated milk yield of the herd for other postpartum days N and factor parameters P can be calculated. N If EY is desired (Y branch at step S284), return to input (step S280), otherwise return to the EY routine (step S286).

[0097] When the value of factor parameter P (explanatory variable) is input into regression equation (6) corresponding to postpartum days N, the herd estimated milk yield EY is calculated. Therefore, by sequentially inputting the same value of factor parameter P into regression equation (6) representing each postpartum day N, a data set of postpartum days N and herd estimated milk yield EY for the same value of factor parameter P can be obtained. The curve relating to herd estimated milk yield in Figure 3(a) is obtained by plotting the data set (pair of data) of postpartum days N and herd estimated milk yield EY obtained in this way, or by connecting these points with a straight line.

[0098] Figure 9 shows an enlarged view of the herd estimated milk production curve M, which plots the herd estimated milk production EY in Figure 3(a). The herd estimated milk production EY is plotted for each number of days postpartum. The factor parameter P represents a multiple regression equation calculated using two factors, p1 and p2. Therefore, the explanatory variables are also the two factor parameters P, p1 and p2. The number of days postpartum N was calculated using discrete values ​​(10, 20, 30, 50, 80, 100, 150, 200, and 250 days), but it can also be calculated in daily increments. The herd estimated milk production EY was calculated using a different regression equation (6) (different number of days postpartum; see Figure 7(e)). In other words, the constants in these equations (e.g., a1 and c in equation (5)) are different.

[0099] In step S210 (see Figure 4) of calculating the herd regression equation, the client inputs the value of the factor parameter P (explanatory variable) expected on the predicted date and the desired number of days since calving N from the terminal 10, and the main unit 12 returns the herd estimated milk yield EY. The client satisfies the requirements for the milk yield calculation system 1 according to the present invention if it returns at least one herd estimated milk yield EY. The herd estimated milk yield EY may be calculated for multiple numbers of days since calving N and graphed as shown in Figure 9.

[0100] Next, the total herd regression equation calculation EYS is performed to calculate the estimated total herd milk yield SUM EY (Step S212). EY A conceptual diagram of the calculation of is shown below. The horizontal axis is the number of days since calving (days), and the vertical axis is the herd's estimated milk yield (kg). Once the herd's estimated milk yield EY is determined, the herd's estimated milk yield curve M can be drawn. Specifically, by substituting the factor parameter P assumed on the predicted day into all of the regression equations in equation (6), it is possible to plot the herd's estimated milk yield EY for all of the number of days since calving N. Here, it is assumed that the herd consists of nine dairy cows.

[0101] The factor parameter P (explanatory variable) for the predicted date is p1, and the number of days since calving for each dairy cow on the predicted date is N1, N2, N3, N4, and N5. The number of dairy cows at each number of days since calving is 1, 2, 4, 1, and 1, respectively. In Figure 13, the number of days since calving on the horizontal axis is shown in parentheses.

[0102] Then, the estimated total milk production of the herd SUM EY is calculated as the sum of the herd estimated milk production EY obtained from each postpartum day N of each dairy cow.

[0103] In other words, the total milk production of the herd is generally estimated as SUM EY If there are Ck dairy cows each k days after calving, then this can be expressed as in equation (7).

[0104]

number

[0105] In addition, the estimated milk yield of the herd is calculated from the number of days since birth N of each dairy cow. N Calculate EY and estimate the herd's milk production N EY is the milk yield of the dairy cow, and the total milk yield of the herd is calculated by adding the milk yield of all dairy cows. EY This gives the same result as equation (7).

[0106] If the herd estimated milk production EY is calculated for a certain period (for example, one week, one month, etc.), formula (7) may be further multiplied by the period (number of days).

[0107] [Cattle herd total regression equation calculation] Figure 14 shows the flow for calculating the estimated total milk yield of a herd. This is the herd total regression equation calculation process. When the herd total regression equation calculation is started (step S212), initial settings are made (step S300). As the initial settings, the estimated total milk yield of a herd SUM EY Then, the discrimination index i of the individual cow is initialized, and the final value of the discrimination index (END) is set to n+1, where n is the number of individual cows belonging to the herd.

[0108] Next, the i-th individual cow CW i The number of days since parturition Ni is obtained (step S302). Then, the estimated milk yield of the herd is calculated using the regression equation (6). Ni EY is calculated and the estimated total milk production of the herd is SUM EY (Step S304). The factor parameter P (explanatory variable) may be the value used in step S210 (calculation step of the herd regression equation: see FIG. 12), or the user may be queried before or after the initial setting (Step S300).

[0109] It is determined whether the calculation for all individual cows has been completed (step S306), and if any remain (N branch in step S306), the discrimination index i is incremented (step S310), and the individual cow CW i The process returns to the step of acquiring the postpartum days Ni (step S302). When the sum of the estimated herd milk yields for all individual cows has been completed (Y branch in step S306), the results are displayed (step S308), and the process returns to the EY routine (step S312).

[0110] Referring again to Figure 4, if the cow herd regression equation calculation or the cow herd total regression equation calculation is to be repeated (Y branch of step S214), step S210 is performed again. If repetition is not to be performed (N branch of step S214), an end determination is made (step S216). If the end is to be performed (Y branch of step S216), the process returns to the main routine (step S218). If the end is not to be performed (N branch of step S216), the process returns to step S200 and the EY calculation process is performed again.

[0111] [Factor analysis] Figure 10 shows the process of factor analysis in step S206 in Figure 4. The factor analysis process is the same as the creation of the regression equation (step S208) up to some point. Specifically, steps S250, S252, and S254 shown in Figure 8 are the same as those in Figure 8. Therefore, the same step numbers are used.

[0112] Therefore, the factor parameter P is input (step S250), and for each specified period m (which is determined in step S200) for the factor parameter P, the milk yield Ym for each postpartum day N is calculated using the interpolation formula (3) (step S252), and a regression formula is calculated using the factor parameter P as the explanatory variable and the milk yield Ym as the target variable (step S254).

[0113] In the factor analysis, the actual milk yield RY for a certain period m is compared with the herd's estimated milk yield EY calculated from the regression equation (step S260). Figure 11 illustrates this process. Two factor parameters, p1 and p2, are selected. These parameters can be temperature and wind speed, for example. The example shows a case where the certain period is four different days. For example, it can be one representative day in each of spring, summer, fall, and winter. Specifically, the days are m1 month, n1, m2 month, n2, m3 month, n3, and m4 month, n4. The specific factor parameters p1 and p2 for these days are kept as records. The factor parameters can be selected by a separate main factor analysis or by trial and error. By inputting the factor parameters p1 and p2, the herd's estimated milk yield EY can be calculated from the regression equation (6).

[0114] The number of days since calving N can be calculated from 1 to the maximum day, but here, the results are shown for four types: 30 days, 50 days, 100 days, and 150 days. These values ​​are taken as the absolute value ERR of the difference with the actual milk yield RY at each date and time. 30 RY is the actual milk yield of a dairy cow 30 days after calving. 30 ERR indicates the absolute value of the difference between the herd's estimated milk yield EY and the actual milk yield RY when the number of days since calving is 30. This ERR is also calculated when the number of days since calving is 50, 100, and 150. In the example described below, an example is shown in which the herd's estimated milk yield EY and the actual milk yield RY are compared when the number of days since calving is 50.

[0115] The sum of the absolute values ​​ERR of the differences between the herd's estimated milk yield EY and the actual milk yield RY for four days is represented as the total error TΣ. In this way, in step S260, the herd's estimated milk yield EY and the actual milk yield RY are compared.

[0116] Referring again to Figure 10, after comparing the herd estimated milk yield EY and the actual milk yield RY, an end decision is made (step S262). The end decision is made based on whether the factor parameter P input in step S250 is appropriate for the absolute value ERR of the difference between the herd estimated milk yield EY and the actual milk yield RY. Whether ERR is appropriate or not may be determined in advance or may be determined by the user. Also, a program (which may include so-called "AI") may be set to select an optimal value. Also, although the difference between the herd estimated milk yield EY and the actual milk yield RY has been described as an absolute value, a squared value may also be used.

[0117] If it is to be ended (Y branch at step S262), the process returns to the EY calculation flow (step S264). If it is to be continued (N branch at step S262), the process returns to step S250 again, and the factor parameter P is input again.

[0118] Referring again to Figure 4, once the factor analysis (step S206) is complete, processing moves to the herd regression equation calculation step (step S210). Once the factor analysis step (step S206) is complete, it is considered that the regression equation and suitable factor parameter P have been found, so it is considered that the herd regression equation calculation step (step S210) for calculating the herd's estimated milk yield EY can be carried out. As already mentioned, in the herd regression equation calculation step (step S210), the herd's estimated milk yield EY is calculated and found using the regression equation of equation (6) from the value of factor parameter P (explanatory variable) assumed on the predicted day and the desired number of days postpartum N. The steps thereafter are as described above.

[0119] [Regression equation construction process] In the milk production calculation system 1 according to the present invention, the step of finding the regression equation (6) is used in any menu. Therefore, steps S200, S202, S204, S206, and S208 are collectively referred to as step S290. Step S290 may also be called the regression equation construction step.

[0120] Calculation of the herd estimated milk yield EY (step S110) involves inputting the value of the factor parameter P (explanatory variable) to calculate the herd estimated milk yield EY. Unless otherwise specified, the herd estimated milk yield EY is calculated for all postpartum days. Of course, it may also be calculated only for a specific postpartum day. In this case, the value of the factor parameter P is the factor parameter P assumed on the predicted date. The factor parameter P assumed on the predicted date may have been input when setting the population data creation conditions (step S200). It may also be input in step S210, which is the process of calculating the herd estimated milk yield EY. It may also be input again when re-inputting the factor parameter P (when returning to step S210 via the Y branch in step S214).

[0121] As described above, the herd estimated milk yield calculation returns an estimated value for the average milk yield of dairy cows belonging to a specific population when the factor parameter P has a certain value. In other words, the herd estimated milk yield value is the predicted milk yield that can be expected from dairy cows belonging to that population when the factor parameter P has a certain value.

[0122] <Estimated herd sales amount E$> Referring again to Figure 2, when the herd estimated milk yield EY is calculated (step S110), the herd predicted sales revenue E$ can be obtained from the calculated herd predicted sales revenue E$. If it is selected to calculate the herd predicted sales revenue E$ (Y branch in step S112), a step of calculating the herd predicted sales revenue E$ (step S116) is carried out. In the step of calculating the herd predicted sales revenue E$, the herd estimated individual sales revenue EI$ and the herd estimated total sales revenue ΣE$ can be calculated. The herd estimated individual sales revenue EI$ is the predicted sales revenue of an individual cow calculated based on the herd estimated milk yield EY. Furthermore, the herd estimated total sales revenue ΣE$ is the total sales revenue of all individual cows belonging to the herd calculated based on the herd estimated milk yield EY.

[0123] The estimated herd sales amount EI$ is calculated based on the CW of a specific individual cow. i Estimated milk production of the cow herd EY[CW i ] can be calculated from equation (6) and multiplied by the milk price q (yen / kg). In addition, the estimated total sales of the herd ΣE$ can be calculated by multiplying the individual cow CWi Estimated milk production of the cow herd EY[CW i ] for all individual cows, or add the estimated total milk production of the herd SUM EY This can be obtained by multiplying the price by the milk price q (yen / kg). Note that the milk price q is the amount expected on the forecast date, but no specific basis is required and it can be past performance data for each season.

[0124] [Calculation of predicted sales revenue for herd 1] Figure 15 shows the herd estimated total sales revenue ΣE$ and the herd estimated total milk production SUM EY The flow is shown below when the amount is obtained by multiplying by the milk unit price q (yen / kg).

[0125] When the process of calculating the estimated herd sales revenue E$ starts (step S116A), you select whether to calculate the estimated herd individual sales revenue EI$ (step S330) or the estimated herd total sales revenue ΣE$ (step S332).

[0126] When calculating the estimated individual sales amount EI$ of a herd (Y branch in step S330), a specific individual cow CW i Enter the number of days since calving Ni and the factor parameter P (explanatory variable) to calculate the estimated milk yield of the herd. Ni EY is calculated and multiplied by the milk unit price q (yen / kg) to calculate the estimated individual sales revenue of the herd EI$ (step S340). The results may be displayed.

[0127] If it is to be repeated (Y branch at step S342), step S340 is repeated again, and if it is not (N branch at step S342), the process proceeds to the step of calculating the estimated total sales amount of the herd ΣE$ (step S332).

[0128] When calculating the estimated total sales amount of the herd ΣE$ (Y branch of step S332), first, the estimated total milk production amount of the herd SUM EY (Step S212). In the step of calculating the herd estimated milk yield EY (Step S110: see FIG. 2), the herd estimated total milk yield SUM EY If you have calculated it and want to use that value, you can skip this step.

[0129] Next, the estimated total milk production of the herd, SUM EY is multiplied by the milk price q (yen / kg) to calculate the estimated total sales amount of the herd ΣE$ (step S350). The result may be displayed.

[0130] If repetition is to be performed (Y branch at step S352), the process returns to step S212. If not (N branch at step S352), the process returns to the herd predicted sales amount E$ step (step S354).

[0131] By the above process, the herd estimated total milk yield SUM, which is the sum of the herd estimated milk yields EY of the individual cows belonging to the herd on the prediction day, is calculated. EY The estimated total sales revenue of the herd, ΣE$, can be calculated by multiplying this by the unit price of milk, q.

[0132] [Herd forecast sales calculation 2] Figure 16 shows the estimated total sales of the herd ΣE$ and the individual cow CW i Estimated milk production of the cow herd EY[CW i The flow chart shows the case where the total amount is obtained by multiplying [ ] by the milk price q and adding the result for all individual cows.

[0133] When the process of calculating the estimated herd sales revenue E$ starts (step S116B), you select whether to calculate the estimated herd individual sales revenue EI$ (step S330) or the estimated herd total sales revenue ΣE$ (step S334).

[0134] The calculation of the herd's estimated individual sales revenue EI$ (Y branch of step S330) is the same as in Figure 15, so it will be omitted. When calculating the herd's estimated total sales revenue ΣE$ (Y branch of step S334), the herd's estimated total sales revenue ΣE$ and the individual cow's identification index i are initialized, and the final value of the identification index (END) is set to n+1 (step S360). Here, n is the number of individual cows belonging to the herd. In addition, the value of the factor parameter P (explanatory variable) for the prediction date is obtained. This may be entered separately, or the value used when the herd's estimated milk yield was calculated (step S110) may be used.

[0135] Next, the i-th individual cow CW i The number of days since delivery Ni is obtained (step S362). This can be obtained by adding the number of days since delivery on the prediction start date to the number of days until the prediction date.

[0136] And the estimated milk yield of individual cows in the herd Ni EY is calculated, multiplied by the milk price q (yen / kg), and added to the estimated total sales amount of the herd ΣE$ (step S364). i The estimated sales amount of each individual cattle herd, EI$, is calculated and added to the estimated total sales amount of the cattle herd, ΣE$.

[0137] If the addition has not been completed for all individual cows (N branch in step S366), the discrimination index i is incremented (step S370), and the process returns to the step of acquiring the postpartum days Ni of the individual cow CWi on the predicted day (step S362).

[0138] When the addition is complete for all individual cows (Y branch in step S366), the results are displayed and a request for re-input is made (step S368). If the value of the factor parameter P needs to be changed (to change the predicted date), re-input is made (Y branch in step S368) and the process returns to the initial setting (step S360). If the process does not need to be repeated (N branch in step S368), the process returns to the predicted herd sales amount E$ (step S372).

[0139] By the above steps, the herd's estimated total sales revenue ΣE$ can be calculated as the sum of the herd's estimated individual sales revenue EI$, which is obtained by multiplying the herd's estimated milk production EY of individual cows belonging to the herd on the prediction date by the milk price q.

[0140] Both A and B of the above step S116 can be the step of calculating the predicted sales amount of the herd of cattle (step S116) in FIG.

[0141] <Determining herd dry cow DC> Please refer to Figure 2. Once the herd estimated milk production EY has been calculated (step S110), the herd dry cow DC can be determined. The determination of the herd dry cow DC will be explained using Figure 17. In Figure 17, the horizontal axis is the number of days since calving (days) and the vertical axis is the herd estimated milk production EY (kg). Since the herd estimated milk production EY has been calculated, the herd estimated milk production curve M can be drawn.

[0142] The dry cow DC of a herd is the dry cow standard value DY that can determine whether a cow is a candidate for dry cows, based on the herd estimated lactation curve M. TH The dry milk reference value DY can be determined by applying TH is the milk yield at which a dairy cow is judged to be a dry cow candidate after a certain period of time has passed since calving. In Figure 17, for an individual cow with a postpartum period of N5, the herd estimated milk yield EY is less than the dry cow standard value DY. TH On the other hand, if the cow is less than N4 days postpartum, the herd estimated milk yield EY is lower than the dry cow standard value DY. TH higher, so the cows are not candidates for dry cow DC ( NOT DC).

[0143] In addition, the herd estimated lactation curve M and dry milk standard value DY TH The day of intersection with is the herd dry day DYday. TH is determined empirically and can be input in step S200 of Fig. 4. Also, individual cows that have just given birth and have low milk production are excluded from candidates for dry cows.

[0144] [Group Dry Cow Judgment 1] Figure 18 shows the herd's estimated milk production EY and dry milk standard value DY. TH The flow for displaying candidates for herd dry cow DC by directly comparing the above is shown below. When the process for determining herd dry cow DC starts (step S118A), initial settings are made (step S400). As initial settings, the individual identification index i is initialized (i=1) and the final value (END) of the identification index is set to n+1, where n is the number of dairy cows in the herd.

[0145] Next, the i-th individual cow CW iThe number of days since calving Ni on the predicted date is obtained (step S402). i The estimated milk yield of the herd at the time of postpartum days Ni is calculated by adding the number of days after separation on the predicted start date to the number of days until the predicted date. Ni EY is calculated and the dry milk standard value DY is calculated. TH (step S404).

[0146] herd milk production Ni EY is the dry milk reference value DY TH If it is less than (Y branch in step S404), the individual cow CW i are displayed as candidates for the herd dry cow DC (step S406), and CW i It is determined whether the individual cow is the last one (step S408). Ni EY is the dry milk reference value DY TH If it is not less than the number (N branch in step S404), step S406 is skipped.

[0147] CW i If it is not the last individual cow (N branch in step S408), the discrimination index i is incremented (step S410), and the i-th individual cow CW i The process returns to the step of acquiring the number of days since calving Ni on the predicted date (step S402). When the determination is completed up to the last individual cow (Y branch in step S408), the process returns to the DC determination (step S412).

[0148] Through the above process, the estimated milk yield of the herd on the predicted day EY is calculated based on the dry milk standard value DY. TH Lower individual cows can be designated as herd dry cow candidates.

[0149] [Cow herd dry cow judgment 2] In addition, the determination of herd dry cows can also be made by calculating the herd dry day DYday on the herd estimated lactation curve M on the predicted day, and displaying the individual cows whose postpartum days are equal to or longer than the herd dry day DYday. In addition, the herd estimated lactation amount on the predicted day for all individual cows belonging to the herd is calculated based on the dry standard value DYday. TH It can also be gained by showing fewer individual cattle.

[0150] 19 shows a flow for determining the herd dry cow DC from the herd dry date. When the herd dry cow DC determination is executed (step S118B), day a is entered into the number of days since calving N (step S430). Day a is the day when the herd estimated milk production EY first exceeds the dry cow reference value DY TH Enter the day after the day exceeding (DC0: see Figure 17). Day a can usually be set to 30 to 50 days. "Day a" is called the early number of days.

[0151] Next, it is determined whether the number of days since delivery N is the last day (step S432). If it is the last day (Y branch in step S432), it is determined that "DC is not applicable" (step S442), and the process returns to the DC determination step (S118B) (step S444).

[0152] If the number of days since calving N is not the last day (branch N in step S432), the estimated milk yield of the herd when the number of days since calving is N is N EY is the dry milk reference value DY TH It is determined whether or not it is equal to or less (step S434).

[0153] Estimated milk production of the herd when the number of days since calving is N N EY is the dry milk reference value DY TH If it is equal to or less than this (Y branch in step S434), the herd dry day DYday is set to N (step S436), and the individual cow i whose postpartum days Ni are greater than the herd dry day DYday is displayed as a herd dry cow DC candidate (step S438).Then, the process returns to the DC calculation step (S118) (step S444).

[0154] Estimated milk production of the herd when the number of days since calving is N N EY is the dry milk reference value DY TH If it is not equal to or less than this (N branch in step S434), the number of days since delivery N is incremented (step S440), and the process returns to step S432.

[0155] By the above process, the estimated milk yield of the herd on the predicted date EY is equal to the dry milk standard value DY. THIndividual cows whose postpartum days are longer than the herd dry day DYday can be displayed as candidates for herd dry cow DC.

[0156] Both of the above steps S118 A and B can be the step of determining the dry cows in the herd in FIG. 2 (step S118).

[0157] <Individual estimated milk production DEY> Referring again to Figure 2, we will now explain individual estimated milk yield DEY. In herd estimated milk yield EY, all dairy cows follow the same herd estimated milk yield EY. However, individual estimated milk yield DEY estimates the milk yield for each individual cow. Individual estimated milk yield DEY is obtained by multiplying the herd estimated milk yield EY by the milk yield coefficient calculated for each individual cow.

[0158] Figure 20 shows how to calculate the individual estimated milk yield DEY. i Individual cow CW i The date on which the milk yield coefficient of the individual cow is calculated for the herd's estimated milk yield EY is called the "individual identification date." The individual identification date can be any date before the prediction start date, and the individual cow CW i Actual milk yield RY[CW i ] is the latest date that has been determined. i ] is an individual cow CW i This is the actual milk yield.

[0159] Figure 20(a) shows the herd estimated milk yield EY on the individual identification day (the factor parameter for this day is p1). The herd estimated milk yield EY is expressed by the herd estimated milk yield curve M. The horizontal axis is the number of days after calving, and the vertical axis is the herd estimated milk yield EY. Individual cow CW on the individual identification day i is the number of days since calving, Nr, and the actual milk yield is RY[CW i ]. In addition, the individual cow CW whose postpartum days are Nr i The actual milk yield is Nr RY[CW i ]. The estimated milk yield of the herd at the time of postpartum days Nr is Nr It is represented by EY.

[0160] In addition, here, individual cow CW i Actual milk yield Nr RY[CW i ] is the herd's estimated milk production. Nr It is assumed that the CW is larger than EY. Of course, it can be the other way around. i Milk yield coefficient α i is calculated as in equation (8). This step can be said to be the milk yield coefficient calculation process.

[0161]

number

[0162] α i is a value determined for each individual cow. i Individual cattle CW for which calculation is made i is also called a specific individual cow. i is the actual milk yield Nr RY[CW i ] and estimated herd milk yield Nr Both EY and EY may be average values ​​for a certain period. Also, the milk production coefficients for a certain consecutive period may be averaged. For example, the actual milk production for each month in the last three months up to the forecast draft date Nr RY[CW i ] and estimated herd milk yield Nr Milk yield coefficient α for each month from EY i The milk yield coefficient α is calculated by averaging the milk yield coefficients for three months. i In addition, since the prediction start date can include not only the current day but also a certain period before the prediction start date, even if calculated in this way, the milk yield coefficient αi can be said to be the value on the prediction start date.

[0163] Referring to Figure 20(b), next, the forecast date (factor parameters for this date are p k ) is calculated. In Figure 20(b), it is shown as the herd estimated milk production curve M. If the value of the factor parameter P is different, the herd estimated milk production curve M in Figure 20(a) and the herd estimated milk production curve M in Figure 20(b) will be different curves. Individual cow CW i Individual estimated milk yield DEY[CWi ] is the α of the estimated milk yield of the herd EY i Calculate as a double.

[0164] The herd estimated milk yield EY is expressed by the regression equation (6). Therefore, the process of calculating the individual estimated milk yield DEY is to use the regression equation (6) as i It can be said that this is the process of calculating the individual regression equation obtained by multiplying the individual estimated milk yield DEY[CW i ] can also be drawn as an individual estimated lactation curve DM. Therefore, the individual estimated lactation curve DM is the α of the herd estimated lactation curve M. i times (see Figure 20(b)).

[0165] Next predicted date individual cow CW i The number of days since calving Nr+X is calculated. i The number of days since parturition can be calculated by the number of days since parturition. Here, X is the number of days from the individual identification date to the prediction date. Nr+X DEY [CW i ] is calculated. This is i ( Nr+X EY). Nr+X DEY [CW i ] is the CW of an individual cow whose postpartum days are Nr+X (days) i The estimated milk production of each individual cow is also Nr+X EY is the herd estimated milk yield at Nr+X days postpartum.

[0166] In other words, the estimated milk yield of an individual on the prediction day, DEY, is calculated by multiplying the milk yield of a herd, EY, by α, where the number of days since calving on the prediction day is Nr+X (days). i This step is a step for calculating the estimated milk yield of an individual.

[0167] Referring to Figure 21, the estimated total milk production of an individual, SUM DEY The individual estimated milk yield DEY is calculated as shown in Figure 20. Therefore, the individual estimated milk yield DEY is calculated for each of all dairy cows belonging to the herd. The sum of the individual estimated milk yields DEY on the prediction day is the individual estimated total milk yield SUM DEYis.

[0168] In Figure 21, the horizontal axis is the number of days since calving (days) and the vertical axis is milk yield (kg). For the sake of simplicity, let's assume that the current herd is made up of individual cow A and individual cow B. The graph shows the herd's estimated milk yield EY (herd estimated milk yield curve M) on the predicted day and the individual estimated milk yields of individual cow A and individual cow B. NA DEY[A], NB DEY[B] is shown. NA and NB are the number of days after parturition on the predicted date for individual cow A and individual cow B. The individual estimated milk yields DEY[A] and DEY[B] are calculated using the individual estimated milk yield curve DM A , D.M. B The milk yield coefficients of individual cow A and individual cow B were expressed as α A , α B It was decided.

[0169] Additionally, the individual estimated milk yield DEY[A] of individual cow A was greater than the herd estimated milk yield EY, and the individual estimated milk yield DEY[B] of individual cow B was less than the herd estimated milk yield EY. Additionally, the number of days since calving on the predicted day for individual cow A and individual cow B was set as NA (days) and NB (days), respectively.

[0170] Individual estimated total milk production SUM DEY is the individual estimated milk yield of individual cow A at postpartum days NA NA DEY[A] and individual estimated milk production at postpartum days NB for individual cow B NB This step is used to calculate the estimated milk yield of an individual.

[0171] Figure 22 shows the flow of calculating the individual estimated milk yield DEY. When the process of calculating the individual estimated milk yield DEY in Figure 2 starts (step S120), the process of constructing a regression equation for the individual identification day is carried out (step S500). This is equivalent to executing step S290 in Figure 4 to obtain the regression equation of equation (6) for the individual identification day (the day on which the milk yield coefficient α is determined). This can also be said to determine the herd milk yield EY for the individual identification day.

[0172] Next, the process of calculating the individual regression equation DEYI is performed (step S502). In step S500, the regression equation (6) is obtained, so the milk yield coefficient α can be determined by using the actual milk yield RY of the individual specific day. As a result, the individual cow CW on the predicted day i The process of calculating the individual regression equation DEYI results in the CW i Calculate the estimated milk yield (DEY) of each cow.

[0173] Next, the step of calculating the individual total regression equation DEYS (step S504) is performed. Since it is now possible to calculate the individual estimated milk yield DEY on the prediction day, the individual estimated milk yield DEY for all individual cows belonging to the herd is added up to obtain the individual estimated total milk yield SUM DEY The process of calculating the individual total regression equation DEYS results in the individual estimated total milk yield SUM DEY Calculate.

[0174] Then, a question is asked as to whether or not re-input is required (step S506). If re-input is required (Y branch in step S506), the process returns to the step of constructing a regression equation for the individual identification date (step S500), and the process can be repeated by changing the individual identification date, etc. If re-input is not required (N branch in step S506), the process returns to the DEY calculation step (step S120) (step S508).

[0175] [Individual regression equation calculation] The flow of the individual regression equation calculation process is shown in Figure 23. When the flow of individual regression equation calculation starts (step S502), the individual cow CW i Number of days after parturition (Nr) and actual milk yield on a specific individual day Nr RY[CW i ] is obtained (step S520). This is recorded in the basic data.

[0176] Next, the estimated milk production of the herd on the individual specific date Nr EY is calculated (step S522). Specifically, equation (6) is obtained. The result obtained in step S500 of FIG. 22 may be used as is.

[0177] Next, actual milk volume Nr RY[CW i ] and the estimated milk yield of the herd EY based on equation (8), the milk yield coefficient α i Then, the estimated milk production of the herd on the predicted day is calculated (step S524). Nr+X EY is calculated (step S526), ​​and then α i By multiplying this, the individual estimated milk yield DEY is obtained (step S528). X is the number of days from the individual identification date to the prediction date. Then, the process returns to the individual regression equation calculation step (step S530). In step S528, the calculation of equation (9) is performed.

[0178]

number

[0179] The estimated milk yield DEY obtained here is written as follows: Nr+X DEY [CW i ]. This is the i-th individual cow CW i The number of days since parturition is Nr + X (days) and the estimated milk yield of the individual is Nr + X (days). The calculated estimated milk yield of the individual can be displayed on a terminal or the like. Nr (days) is the CW of the individual cow on the specific day. i is the number of days since parturition, and X (days) is the number of days from the individual identification date to the prediction date.

[0180] [Individual Total Regression Equation Calculation] Figure 24 shows the process flow for calculating the individual total regression equation. When the process for calculating the individual total regression equation starts (step S504), initial settings are first made (step S540). The initial settings are the discrimination index i and the individual estimated total milk yield SUM DEY Initialization of (i=1, SUM DEY = 0) and setting the final value (END) of the population. The final value (END) is set to n+1, where "n" is the number of cows in the herd.

[0181] Next, the estimated milk yield (DEY) of the individual on the predicted date (more precisely, Nr+X DEY [CW i]") is calculated (step S542). Nr (days) is the number of days since parturition on the individual specific day, and Nr + X (days) is the number of days since parturition on the predicted day. Then, the individual estimated milk yield DEY and the individual estimated total milk yield SUM DEY Add the new estimated total milk production SUM DEY (step S544).

[0182] It is determined whether addition has been completed for all individual cows (step S546), and if not (N branch of step S546), the discrimination index i is incremented (step S548) and the process returns to step S542. If completion has been completed (Y branch of step S546), the process returns to the individual total regression equation calculation step (step S550).

[0183] The estimated total milk production of the individual obtained here is SUM DEY is the sum of the individual estimated milk production on the prediction day. The calculated individual estimated total milk production SUM DEY can be displayed on a terminal or the like.

[0184] <Individual Estimated Sales Amount DE$> 2, once the individual estimated milk yield DEY has been determined, the individual estimated sales revenue DE$ can be calculated. If it is selected to calculate the individual estimated sales revenue DE$ (Y branch in step S122), the individual estimated sales revenue DE$ is calculated (step S126). The individual estimated sales revenue DE$ can be calculated by calculating the individual estimated sales revenue DEI$, which is the estimated sales revenue of each individual cow, and the individual estimated total sales revenue ΣDE$, which is the estimated sales revenue of all individual cows belonging to the herd.

[0185] The individual estimated sales amount DEI$ can be obtained by multiplying the individual estimated milk yield DEY calculated in step S120 (see Figure 2) by the unit price of milk q (yen / kg). On the other hand, the individual estimated total sales amount ΣDE$ can be calculated by multiplying the individual estimated milk yield DEY by the unit price of milk q to calculate the individual estimated sales amount for each individual cow and then calculating the sum of these amounts, or by multiplying the individual estimated total milk yield SUM DEY One way to calculate this is to multiply by the unit price of milk q. Either process can be used to calculate the individual estimated total sales amount ΣDE$.

[0186] 25 and 26 show examples of the respective flows, which are referred to as step S126A and step S126B, respectively. Either process can be used as step S126 in FIG.

[0187] [Estimated total sales amount calculation 1] Figure 25 shows the flow of the process for calculating the individual estimated sales amount DE$. Note that this flow uses the individual estimated total milk yield SUM DEY When the process of calculating the individual predicted sales amount DE$ starts (step S126A), a selection is made between calculating the individual predicted sales amount DEI$ or calculating the individual predicted total sales amount ΣDE$ (steps S560, S562). If no selection is made, the process proceeds to the next step. That is, in the N branch of step S560, the process proceeds to step S562, and in the N branch of step S562, the process returns to the individual predicted sales amount DE$ calculation process.

[0188] When the individual estimated sales amount DEI$ is selected (Y branch in step S560), the individual estimated milk yield of the individual cow CWi on the prediction date is calculated, and multiplied by the milk unit price q to obtain the individual estimated milk yield DEI$ (step S570). It is assumed that the individual cow to be selected was input at the beginning of this step. Next, re-input is confirmed (step S572), and if re-input is to be performed (Y branch in step S572), the individual cow is changed, and step S570 is executed again. If re-input is not to be performed (N branch in step S572), the process proceeds to the next step (step S562).

[0189] When the individual estimated total sales amount ΣDE$ is selected (Y branch in step S562), first, the individual estimated total milk production amount SUM DEY (Step S504). This can be obtained by executing step S504 in FIG. 24. However, the individual estimated total milk production SUM DEY If it has already been calculated, that value may be used.

[0190] And the individual estimated total milk production SUM DEYThe value obtained by multiplying this by the unit price q of milk is set as the individual estimated total sales amount ΣDE$ (step S580). The calculated individual estimated total sales amount ΣDE$ may be displayed on a terminal or the like.

[0191] Next, the user confirms whether or not to re-enter the data (step S582). If the user decides to re-enter the data (Y branch at step S582), the user makes changes to the predicted date, etc., and executes step S504 again. If the user does not decide to re-enter the data (N branch at step S582), the process returns to the step of the predicted individual sales amount DE$ (step S584).

[0192] The above process yields the total estimated milk yield SUM, which is the sum of the estimated milk yields DEY on the expected date for all individual cows in the herd. DEY The estimated total sales amount for an individual can be calculated by multiplying this by the unit price of milk q.

[0193] [Estimated total sales amount calculation 2] 26 shows a flow chart for a method of calculating the individual estimated total sales amount ΣDE$ as the sum of the estimated sales amount of each individual cow. When the process of calculating the individual estimated sales amount DE$ starts (step S126B), a selection is made between calculating the individual estimated sales amount DEI$ or calculating the individual estimated total sales amount ΣDE$ (steps S560, S564). If no selection is made, the process proceeds to the next step. That is, in the N branch of step S560, the process proceeds to step S564, and in the N branch of step S564, the process returns to the individual estimated sales amount DE$ calculation process (step S600).

[0194] The calculation flow for the individual estimated sales amount DEI$ is the same as that in Figure 25, so it will be omitted here. When calculation of the individual estimated total sales amount ΣDE$ is selected (Y branch in step S564), initial settings are made (step S590). The initial settings include initialization of the discrimination index i and the individual estimated total sales amount ΣDE$ (i=1, ΣDE$=0), and setting the final value of the population (END). The final value END is set to n+1, where "n" is the number of dairy cows in the herd.

[0195] Next, the individual's estimated milk yield DEY on the prediction date is calculated, multiplied by the milk price q, and added to the individual's estimated total sales amount ΣDE$ to obtain the new individual's estimated total sales amount ΣDE$ (step S592). It is confirmed whether addition up to the last individual cow has been completed (step S594), and if not (N branch in step S594), the discrimination index i is incremented (step S598), and the addition of the individual's estimated total sales amount ΣDE$ (step S592) is repeated.

[0196] When calculations have been completed up to the last individual cow (Y branch at step S594), it is checked whether or not re-input is required (step S596). At this time, the obtained individual estimated total sales amount ΣDE$ may be displayed on a terminal or the like. If re-input is required (Y branch at step S596), the predicted date, etc. is changed, and the process is repeated from the initialization step (step S590). If re-input is not required (N branch at step S596), the process returns to the individual estimated sales amount DE$ (step S600).

[0197] By following the above steps, the sales revenue for each individual cow in the herd can be calculated by multiplying the individual estimated milk yield by the milk price q, and the sum of these can be calculated as the individual estimated total sales revenue DE$.

[0198] Both A and B of the above step S126 can be the step of calculating the individual predicted sales amount DE$ (step S126) in FIG.

[0199] <Determining individual dry cow DDC> Please refer to Figure 2. When the individual estimated milk yield DEY is calculated (step S120), the individual dry cow DDC can also be determined. The determination of the individual dry cow DDC will be explained using Figure 27. In Figure 27, the horizontal axis is the number of days since calving (days) and the vertical axis is the individual estimated milk yield DEY (kg). The herd estimated milk yield EY is also depicted as the herd estimated milk yield curve M.

[0200] The individual dry cow DDC is the dry cow reference value DY that can be used to determine whether a cow is dry, based on the individual estimated lactation curve DM. THFigure 27 shows that if the number of days since calving on the predicted date for this individual cow is more than N6, this individual cow is a candidate for individual dry cow DDC. If the number of days since calving is less than N6, this individual cow is not an individual dry cow ( NOT DDC). The individual estimated lactation curve DM and dry milk reference value DY TH The day of intersection (N6) is the individual dry day (DDYday).

[0201] In addition, the dry milk standard value DY TH is empirically determined and is input when step S200 in FIG. 4 or step S500 in FIG. 22 is executed (both in the regression equation construction step (step S290)).

[0202] There are two ways to distinguish between dry and dry cows. Calculate the estimated milk yield (DEY) for each individual cow and compare it with the dry cow standard value (DY). TH and the method of comparing the estimated milk yield DEY and dry standard value DY for each individual cow. TH The individual dry day DDYday is calculated from the calculated number of days since parturition on the predicted date and the individual dry day DDYday are compared. Either method can be used as the individual dry cow discrimination step (step S128) in FIG.

[0203] [Individual dry cow determination 1] Figure 28 shows the calculation of the individual estimated milk yield DEY for each individual cow and the dry milk standard value DY TH The flow for the method of comparing the above is shown below. When the process of determining individual dry cows starts (step S128A), initial settings are made (step S610). The initial settings involve initializing the discrimination index i and setting the final value (END) of the population. The discrimination index i is set to i=1, and the final value (END) is set to n+1, where "n" is the number of dairy cows in the herd.

[0204] Next, the i-th individual cow CW on the prediction date i The number of days since parturition Nr+X is obtained (step S612), where Nr is the number of days since parturition on the individual identification date, and X is the number of days from the individual identification date to the predicted date.

[0205] Then, it is determined whether the number of postpartum days Nr+X on the predicted day is equal to or less than the individual early days Da (step S614). The individual early days Da is determined by the time when the herd estimated milk yield EY first falls below the dry milk reference value DY TH This is the day corresponding to the early days a, which is the day when the individual estimated milk yield DEY of the individual cow CWi first exceeds the dry standard value DY TH This is the number of days that will exceed the specified number. Usually, you can set it to 30 to 50 days.

[0206] If the number of days since calving (Nr+X) is equal to or less than the number of early days (Da), the discrimination index i is incremented and the process of obtaining the number of days since calving (Nr+X) for another individual cow is repeated (step S612). This is because individual cows that have just calved are excluded from the determination of dry cows.

[0207] And individual cow CW i The estimated milk yield DEY of the individual is the dry milk standard value DY TH It is determined whether or not the individual cow CW is equal to or less than the above (step S616). i The estimated milk yield DEY of the individual is the dry milk standard value DY TH If it is below (Y branch in step S616), the individual cow CW i are displayed as candidates for individual dry cows (step S618), and it is determined whether or not the determination for all individual cows has been completed (step S620). i The estimated milk yield DEY of the individual is the dry milk standard value DY TH If it is not equal to or less than this (N branch in step S616), step S618 is skipped and the process proceeds to step S620.

[0208] If the judgment has not been completed up to the last individual cow (N branch in step S620), the discrimination index i is incremented, and the i-th individual cow CW on the prediction date is calculated. i The process returns to the step of acquiring the number of days since calving Nr+X (step S612). If the determination has been completed up to the last individual cow (Y branch in step S620), the process returns to the step of determining whether the individual is a dry cow (step S624).

[0209] Through the above process, the estimated milk yield DEY of the individual on the predicted day is calculated based on the dry milk standard value DY TH Individual cows that are lower can be displayed as dry cow candidates.

[0210] [Individual dry cow determination 2] Figure 29 shows the flow for determining whether an individual dry cow is a cow using the individual dry day DDYday. When the process for determining whether an individual dry cow is a cow starts (step S128B), initial settings are made (step S640). The initial settings involve initializing the discrimination index i and setting the final value (END) of the number of cows. The discrimination index i is set to i=1, and the final value (END) is set to n+1. "n" is the number of cows in the herd.

[0211] Next, the number of postpartum days N is set to the individual early days Da (step S642). Then, it is determined whether the determination has been completed for all individual cows belonging to the herd (step S644). If it has been completed (Y branch in step S644), the process returns to the individual dry cow determination step (step S660). If there are still individual cows to be determined (N branch in step S644), the process proceeds to the next step.

[0212] In the next process, it is determined whether the number of days since delivery N is final (step S646). The final value of the number of days since delivery N is set in advance. It is usually set to 260 days or more. If N is not the final value (N branch of step S646), the process proceeds to the next process (step S648); if N is the final value (Y branch of step S646), the identification index i is incremented (step S658), and the process returns to the step of setting the number of days since delivery N (step S642).

[0213] If N is not the final value (N branch in step S646), the individual estimated milk yield DEY on the prediction date of the i-th individual is calculated by subtracting the individual estimated milk yield DEY when the number of days since parturition is N from the dry milk standard value DY TH It is determined whether the individual cow CW is equal to or less than the above (step S648). i It can be calculated from the individual regression equation on the predicted date. iThe individual regression equation for the predicted date is the individual cow CW in equation (6). i Milk yield coefficient α i This is because it can be obtained by multiplying by and then finding the individual regression equation when the number of days since delivery is N.

[0214] Individual estimated milk yield DEY is the dry milk standard value DY TH If it is not equal to or less than the dry milk standard value DY (N branch in step S648), N is incremented (step S656), and the process returns to the determination of whether N is the last day or not (step S646). TH If the number of days since sorting N is equal to or less than the above, it is determined that the individual dry day DDYday for this individual cow (step S650).

[0215] And the individual cow CW on the predicted date i It is determined whether the number of days since parturition is greater than the individual dry day DDYday (step S652). i ] is the individual cow CW on the individual identification date i is the number of days since calving, and X is the number of days from the individual identification date to the prediction date. i If the number of days since calving is greater than the individual dry day DDYday (Y branch in step S652), the individual cow CWi is displayed as an individual dry cow DDC (step S654).

[0216] Cow CW on the predicted date i If the number of days since calving is not greater than the individual dry day DDYday (N branch in step S652), the discrimination index i is incremented and the process returns to the step of setting the number of days since calving N (step S642). In this way, the individual dry day DDYday of each individual cow belonging to the herd is calculated and compared with the number of days since calving of that individual cow on the predicted date to determine whether or not the individual is a dry cow DDC.

[0217] Through the above process, the estimated milk yield DEY of the individual on the predicted day is calculated based on the dry milk standard value DY TH Individual cows whose number of days since calving Nr+X (days) is longer than the individual dry day DDYday can be displayed as individual dry cow candidates.

[0218] Both A and B of the above step S128 can be the step of determining the individual dry cow in FIG. 2 (step S128).

[0219] As described above, the milk production calculation system according to the present invention predicts future milk production based on past basic data. If future milk production can be predicted, sales can be predicted based on that amount.

[0220] Furthermore, once sales figures can be predicted, it becomes possible to compare them with feed costs (which are expenses), allowing for productivity estimates. For example, by inputting feed costs, the system can compare them with predicted sales figures and display results such as feed costs being 30% of predicted sales figures. This makes it possible to consider feed loss and efficient feed formulations. This also makes it possible to consider the formulation of concentrated feed in particular.

[0221] In addition, milk components that represent the content of fat, protein, etc. may also be predicted and displayed in a similar manner. In the case of a transaction in which the milk price increases when the concentration is above a predetermined level and decreases when the concentration is below the predetermined level, the sales amount may be displayed taking into account the increase or decrease in the milk price according to the predetermined level. [Example]

[0222] The milk production calculation system according to the present invention was implemented using data obtained from a farm in Japan. The farms were multiple farms in a certain region, with approximately 5,000 dairy cows. The population data was for each month in 2019. The factor parameter P was THI and sunshine hours. THI used the average monthly temperature and average humidity values. Sunshine hours was the total monthly value. The results are shown in Figure 30.

[0223] Figure 30(a) shows the values ​​of THI and sunshine hours for the five months of February, May, August, and November in 2019. The average actual milk yield (kg / day) and number of days since calving N of the target farm were approximated with a Wood curve, and the values ​​obtained from this were used to calculate the estimated milk yield EY by simple regression using only THI as the factor parameter P. Figure 30(b) shows the results. Figure 30(c) also shows the results of calculating the estimated milk yield EY by multiple regression using THI and sunshine hours (monthly total) as the factor parameters P. Note that both are for a case where the number of days since calving is 50.

[0224] When only THI was used as the factor parameter P, the standard error was 0.65, which was quite variable. However, when THI and sunshine duration were used as the factor parameter P, the standard error was 0.17. This shows that the actual milk yield can be estimated to a great extent.

[0225] The fact that milk production can be estimated using THI and sunshine hours has not been known until now, and this will have a major impact on how we respond to future increases in milk production. [Industrial Applicability]

[0226] The present invention can be suitably used for computerizing dairy farming. [Explanation of symbols]

[0227] 1. Milk production calculation system 10 devices 12 Main Unit 14 Memory 16 Ranch 18 External Information

Claims

1. A milk yield calculation system that calculates the milk yield of an individual cow on a predicted date after the prediction start date as an estimated milk yield of a herd based on the number of days since calving and actual milk yield of each individual cow constituting the herd before the prediction start date, and farm data including weather data, a memory for storing the farm data; an interpolation formula creating unit that creates an interpolation formula that interpolates the relationship between the number of days since calving and the actual milk yield for a certain period in the farm data before the prediction start date; a regression equation creation unit that creates a regression equation for calculating the estimated milk yield of the herd for each postpartum day using the interpolated milk yield calculated from the interpolation equation as a response variable and at least one factor parameter selected from the farm data as an explanatory variable; and The regression equation corresponding to the number of days since delivery on the predicted date is: A milk production calculation system having a control device with a herd estimated milk production calculation unit including a herd regression equation calculation unit that substitutes the values ​​of the explanatory variables expected on the predicted date and calculates the herd estimated milk production.

2. The milk production calculation system according to claim 1, further comprising a herd predicted sales amount calculation unit that calculates sales amount based on the herd estimated milk production amount and the expected milk unit price on the predicted date.

3. The milk production calculation system described in claim 2, characterized in that the herd predicted sales calculation unit has a herd estimated individual sales calculation unit that calculates the herd estimated individual sales by multiplying the herd's estimated milk production on the predicted date by the milk unit price.

4. The milk production calculation system described in claim 2 or 3, characterized in that the herd predicted sales calculation unit has a herd estimated total sales calculation unit that calculates the herd estimated total sales by multiplying the herd estimated total milk production, which is the sum of the herd estimated milk production of individual cows belonging to the herd on the predicted date, by the milk unit price.

5. The milk production calculation system described in claim 2 or 3, characterized in that the herd predicted sales calculation unit has a herd estimated total sales calculation unit that calculates the sum of the herd estimated individual sales obtained by multiplying the herd estimated milk production of individual cows belonging to the herd on the predicted date by the milk unit price as the herd estimated total sales.

6. A milk production calculation system as described in claim 1, further comprising a herd dry cow determination unit that indicates candidate dry cows among the dairy cows belonging to the herd on the predicted date based on the herd's estimated milk production and the dry milk standard value.

7. The milk production calculation system described in claim 6, characterized in that the herd dry cow determination unit displays individual cows whose herd estimated milk production on the prediction date is lower than the dry milk standard value as candidate dry cows.

8. The milk production calculation system described in claim 6, characterized in that the herd dry cow determination unit displays as a candidate herd dry cow an individual cow whose post-calving number of days is greater than the herd dry day on which the herd's estimated milk production on the predicted date becomes the dry milk standard value.

9. A milk yield calculation system that calculates an estimated milk yield of an individual cow on a prediction date after the prediction start date as an individual estimated milk yield based on the number of days since calving and actual milk yield of each individual cow constituting a herd before the prediction start date, and farm data including weather data, a memory for storing the farm data; an interpolation formula creating unit that creates an interpolation formula that interpolates the relationship between the number of days since calving and the actual milk yield for a certain period in the farm data before the prediction start date; a regression equation creation unit that creates a regression equation for calculating an estimated milk yield of a herd for each postpartum day using the interpolated milk yield calculated from the interpolation equation as a response variable and at least one factor parameter selected from the farm data as an explanatory variable; and The regression equation corresponding to the number of days since calving of the individual cow on the prediction start date is: A herd regression equation calculation unit that substitutes the values ​​of the explanatory variables assumed on the prediction start date and calculates the herd estimated milk yield; A milk yield coefficient calculation unit that calculates the ratio of the specific individual cow to the regression equation as a milk yield coefficient based on the herd estimated milk yield on the prediction start date and the actual milk yield of the specific individual cow on the prediction start date; The regression equation corresponding to the number of days since calving of the specific individual cow on the predicted date is: A milk production calculation system having a control device with an individual estimated milk production calculation unit that includes an individual regression equation calculation unit that substitutes the values ​​of the explanatory variables expected on the predicted date and further multiplies the obtained result by the milk production coefficient to calculate the individual estimated milk production.

10. The milk production calculation system according to claim 9, further comprising an individual predicted sales amount calculation unit that calculates sales amount based on the individual estimated milk production amount and the expected unit price of milk on the predicted date.

11. The milk production calculation system described in claim 10, wherein the individual predicted sales calculation unit further includes an individual estimated sales calculation unit that multiplies the individual estimated milk production by the milk unit price to calculate the individual estimated sales of the specific individual cow on the predicted date.

12. A milk production calculation system as described in claim 10 or 11, wherein the individual predicted sales calculation unit further includes an individual estimated total sales calculation unit that calculates the individual estimated sales by multiplying the individual estimated total milk production, which is the sum of the individual estimated milk production of all the individual cows belonging to the herd on the predicted date, by the milk unit price.

13. A milk production calculation system as described in claim 10 or 11, wherein the individual predicted sales calculation unit further includes an individual estimated total sales calculation unit that calculates the sales for each individual cow by multiplying the individual estimated milk production by the milk unit price for all individual cows in the herd, and calculates the sum of these as the individual estimated total sales.

14. A milk production calculation system as described in claim 9, further comprising an individual dry cow determination unit that indicates candidate dry cows among the dairy cows belonging to the herd on the predicted date based on the individual estimated milk production and the dry milk standard value.

15. The milk production calculation system described in claim 14, characterized in that the individual dry cow determination unit displays the individual cow whose individual estimated milk production on the prediction date is lower than the dry milk standard value as the candidate dry cow.

16. The milk production calculation system described in claim 14, characterized in that the individual dry cow determination unit displays as an individual dry cow candidate an individual cow whose number of days since calving is greater than the individual dry day on which the individual estimated milk production on the prediction date becomes the dry milk standard value.

Images