Milk production calculation system

Abstract

To provide a milk yield calculation system capable of analyzing a factor even when information pertaining to collected milk yield does not fall within a same range.SOLUTION: A milk yield calculation system 1 comprises a main body 12 (control unit) including: a memory 14 which accumulates factor parameters including a day count after parturition, actual milk yield and climate data as ranch data; an interpolation formula creating unit which creates an interpolation formula to interpolate the day count after parturition and the actual milk yield out of the ranch data for a prescribed period; a regression formula creating unit which creates a regression formula to calculate estimated milk yield using interpolated milk yield of a specific day count after parturition calculated from the interpolation formula as an objective variable and at least one factor parameter selected from the ranch data as an explanatory variable; and a milk yield calculation unit which calculates the estimated milk yield from the regression formula for a value of the inputted factor parameter.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 the mix of feed, the breeding environment, and reduce stress on dairy cows.

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

[0004] Moreover, 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 the greatest effect on milk production in which cases.

[0005] Therefore, a system that can estimate milk production by inputting possible factors would provide a valuable 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 reproductive 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] JP 2020-156359 A [Patent Document 2] JP 2020-020707 A Summary of the Invention [Problem to be solved by the invention]

[0009] Patent Document 1 aims to grasp the current state of dairy cows, and Patent Document 2 aims to estimate future milk production from past milk production. However, they do not enable investigation of factors for increasing milk production in dairy cows.

[0010] Although it is understood that the physiological theory that the 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 and state of the dairy cow itself are comfortable for dairy cows. Therefore, it is thought that a method of collecting many phenomena (information), analyzing the causes, and trying to increase milk production is useful.

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

[0012] However, it is extremely difficult to collect data regularly from multiple farms that are all different in size and circumstances. Depending on the farm, some data are quantified daily, while in other cases data is only collected every few days.

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

[0014] The milk production calculation system of the present invention has been devised in consideration of the above-mentioned problems, and provides a system that enables factor analysis even when the collected information related to milk production is not within the same range.

[0015] More specifically, the milk yield calculation system according to the present invention comprises: days after delivery, For each day after birth a memory for storing factor parameters including actual milk yield and weather data as farm data; A certain period of time within the farm data every The number of days after parturition and the actual milk yield of With continuous functions interpolation The amount of interpolated milk at any postpartum day between the first day of the postpartum period and the last day of the postpartum period can be calculated. an interpolation formula creation unit that creates an interpolation formula; The interpolation formula Any of the above Days after birth Substituting A regression equation creating unit that creates a regression equation for calculating an estimated milk yield using an interpolated milk yield as a response variable and at least one of the factor parameters selected from the farm data as an explanatory variable; The present invention is characterized in that it has a control device including a milk yield calculation unit that calculates an estimated milk yield from the regression equation for the input value of the factor parameter. Effect of the Invention

[0016] The milk production calculation system of the present invention has the advantage that by interpolating collected data on milk production, a certain range of milk production can be obtained even on a daily basis, which is useful for estimating factors that affect milk production.

[0017] Furthermore, if the effect of predictable factors such as temperature on milk production is known, it will be possible to predict future milk production. [Brief description of the drawings]

[0018] [Figure 1] FIG. 1 is a diagram showing a configuration of a milk production calculation system according to the present invention. [Diagram 2]FIG. 1 is a diagram showing the overall (main) flow of the milk production calculation system. [Diagram 3] FIG. 13 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 process flow for calculating an estimated milk yield EY. [Diagram 5] FIG. 13 is a diagram for explaining an interpolation formula. [Figure 6] FIG. 13 is a diagram showing a flow for creating an interpolation formula. [Figure 7] FIG. 13 is a diagram for explaining 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. 4 is an enlarged view of the curves relating to the estimated milk yield EY in FIGS. 3(a) to (e). [Figure 10] FIG. 5 is a diagram showing the process of factor analysis in step S206 in FIG. [Figure 11] FIG. 13 is a diagram showing a comparison between the actual milk yield RY for a certain period m and the estimated milk yield EY obtained from the regression equation in a factor analysis. [Figure 12] FIG. 13 is a diagram showing the processing flow of individual estimated milk yield IY. [Figure 13] FIG. 13 is a diagram showing an example of a display of a step of calculating an individual estimated milk yield IY. [Figure 14] FIG. 13 is a diagram showing the process of calculating the estimated milk yield GY of the herd. [Figure 15] FIG. 13 is a diagram showing an example of a display of a process for calculating an estimated milk yield GY of a herd. [Figure 16] FIG. 13 is a diagram showing the processing flow of the estimated milk production comparison CY. [Figure 17] FIG. 13 shows an example of output from the step of calculating an estimated milk production comparison CY. [Figure 18] FIG. 13 is a diagram showing a process flow for calculating a transitional milk production comparison TY. [Figure 19] FIG. 13 shows an example of output from the step of calculating a transitional milk yield comparison TY. [Figure 20]This figure shows the results when the actual milk yield RY of individual dairy cows is plotted simultaneously on an example output from the process of calculating a comparison of transitional milk yield TY. [Figure 21] This is a graph showing that the milk production calculation system was used on an actual farm, and the estimated milk production can be multiple-regressed using THI and sunshine hours as factor parameters. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0019] The milk production calculation system according to the present invention will be described below with reference to the drawings. Note that the following description is an example of one embodiment of the present invention and an example, and the present invention is not limited to the following description. The following description can be modified without departing from the spirit of the present invention.

[0020] 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 Processing Unit), memory, and a display screen, and can include a mobile communication terminal (a so-called smartphone).

[0021] The main body 12 is composed of a CPU. 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 a memory used by the main body 12.

[0022] 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 16A where dairy cows are actually raised. Here, "installed" may mean that the terminal 10 is physically installed at the ranch 16A, or that a person in charge or responsible person of the ranch 16A possesses a mobile communication terminal. Of course, the terminal 10 may be installed at a location other than the ranch 16A. Moreover, the ranch 16 may be any ranch where dairy cows are raised.

[0023] A plurality of farms 16 may be connected to the main body 12. The milk production calculation system 1 calculates the milk production by collecting and analyzing a large amount of data related to the milk production, so it is desirable to collect a large amount of data. The main body 12 may also be connected to external information 18 via the Internet. This is because not only the data obtained from the farms 16 but also information on the national weather can be obtained via the Internet and used.

[0024] 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 collected data, so a preferred embodiment is one in which the service is provided via a network. In other words, the milk production calculation system 1 may be configured as a cloud.

[0025] Therefore, the cloud form for implementing the milk yield 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).

[0026] The milk yield calculation system 1 according to the present invention calculates an estimated milk yield that can be milked from a dairy cow on the Nth day after calving by inputting a factor parameter P from a terminal 10. N EY (kg) can be obtained. Here, 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, 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 in the milk yield calculation system 1.

[0027] The climate data includes sunrise time, sunset time, temperature, humidity, sunshine hours, solar radiation, wind direction, wind speed, etc. It may also include 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 the following formula (1).

[0028]

number

[0029] 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 area of ​​the farm, the regional climate, or climate data from a national weather forecast service.

[0030] The individual data is data on the dairy cow itself, such as the registration number, pedigree, parity, number of days since calving, milk yield, weight, medical history, owner, and place of rearing.

[0031] Feed data includes the type, composition, and frequency of feed given in the past.

[0032] A terminal 10 installed on a farm 16 transmits individual data related to dairy cows on the farm, data on the actual feed they have consumed, and daily weather data as farm data FD to a main unit 12. The main unit 12 stores these data in a memory 14 and calculates an estimated milk yield according to the number of days since calving N according to factor parameters P input from the terminal 10. N EY is calculated and displayed on terminal 10.

[0033] The main unit 12 may obtain weather data from external information 18. In a closed-type barn, the environment inside the barn is managed, and data can be easily transmitted. However, in an open-type barn, it is difficult to obtain the wind speed inside the barn on that day. In such cases, external weather data can be used as a reference.

[0034] Various types of weather-related data are provided, and these can also be used effectively. The farm data FD may 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.

[0035] Fig. 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 appropriate. The milk production calculation system 1 can return results of the steps of calculating estimated milk production EY, calculating individual estimated milk production IY, calculating group estimated milk production GY, calculating estimated milk production comparison CY, and calculating transitional milk production comparison TY in response to inputs such as calculation conditions and factor parameters P entered from a terminal or the like. These results may be returned to the terminal 10. These calculation steps can be called the menus of the main flow.

[0036] In addition, most of the configuration of the present invention is implemented by software. Therefore, a "step" of processing refers to a group of processes, and the main body 12 may be considered to have "units" that execute the "steps." Specifically, it can be said that the main body 12 (control device) has an estimated milk yield calculation unit, an individual estimated milk yield calculation unit, a group estimated milk yield calculation unit, an estimated milk yield comparison calculation unit, and a transitional milk yield comparison calculation unit.

[0037] Furthermore, if there is a group of lower-level processes within these processing steps, they may also be called a "unit." For example, the estimated milk yield calculation unit has a process of creating a regression equation as one process group. This may be called a "regression equation creation unit." This may further include a "process of creating an interpolation equation" and may be called an interpolation equation creation unit. Furthermore, a process (step S210) of specifically calculating the estimated milk yield when the value of the factor parameter P is substituted into the obtained regression equation may also be called a milk yield calculation unit. Moreover, it is sufficient that the milk yield calculation system 1 has at least a process of calculating the estimated milk yield EY (estimated milk yield calculation unit).

[0038] When the milk production calculation system 1 starts (step S100), a termination decision is made (step S102). The termination decision may be based on a termination instruction from the terminal 10 or on disconnection of communication between the main unit 12 and the terminal 10. If the system is to terminate (Y branch at step S102), the system terminates (step S104). If the system is not to terminate (N branch at step S102), the system proceeds to the next process.

[0039] Next, select whether to calculate estimated milk production EY, individual estimated milk production IY, group estimated milk production GY, estimated milk production comparison CY, or trend milk production comparison TY (steps S106, S108, S110, S112, and S114). For each option, if selected (Y branch), the process moves to the corresponding step, and if not selected (N branch), the process moves to the next step. If the trend milk production comparison TY is not calculated in step S114 (N branch in step S114), the process returns to the end decision (step S102).

[0040] When each selection step (step S106, step S108, step S110, step S112, step S114) is selected (Y branch), the corresponding step is performed and the process returns to the end judgment (step S102). Specifically, when the estimated milk yield EY is to be obtained (Y branch of step S106), a step of calculating the estimated milk yield EY is performed (step S116).

[0041] When an individual estimated milk production IY is obtained (step S108), a step of calculating the individual estimated milk production IY is performed (step S118). When a group estimated milk production GY is obtained (step S110), a step of calculating the group estimated milk production GY is performed (step S120). When an estimated milk production comparison CY is obtained (step S112), a step of calculating the estimated milk production comparison CY is performed (step S122). When a transitional milk production comparison TY is obtained (step S114), a step of calculating the transitional milk production comparison TY is performed (step S124). After each processing step is completed, the process returns to the end judgment (step S102).

[0042] FIG. 3 shows examples of each output. FIG. 3(a) is an example of output of estimated milk production EY. Under a specific condition (e.g., temperature), it is possible to show milk production versus days after parturition. In the figure, estimated milk production curves M when the temperature is T1°C and when it is T2°C are shown. The estimated milk production curve M provided by the milk production calculation system 1 of the present invention is a collection of calculated values ​​(estimated milk production EY) at the minimum on a daily basis. These calculated values ​​may be connected by a straight line or a curve. The estimated milk production curve formed by the estimated milk production EY is represented by the symbol M. In other words, the estimated milk production curve M is obtained by plotting the estimated milk production EY for each appropriate number of days after parturition N and connecting them by a straight line or a curve.

[0043] Figure 3(b) is an example of output of individual estimated milk yield IY. Individual estimated milk yield IY is obtained by plotting the actual milk yield RY of individual dairy cows on the estimated milk yield curve M of a specified population. In Figure 3(b), it is represented by a black circle d. Individuals whose actual milk yield RY is higher or lower than the estimated milk yield EY by a certain amount can be displayed in a particularly noticeable manner (in Figure 3(b), it is shown by the symbols du and dd). Furthermore, if such individuals are found, the owners may be notified.

[0044] Figure 3(c) is an example of the output of group estimated milk yield GY. Group estimated milk yield GY is obtained by plotting the actual milk yield RY of a group of dairy cows managed as a group on the estimated lactation curve M of a specified population. Dairy cows in the same group are plotted with the same mark. In Figure 3(c), for dairy cows belonging to two groups, unit U1 and unit U2, unit U1 is represented by a white circle, and unit U2 is represented by a black circle.

[0045] In addition, if the actual milk yield RY is higher or lower than the estimated milk yield EY by a certain amount, a notification may be given in the same way as for the individual estimated milk yield IY. Note that, even in the same group, a different mark may be used for milk yields that deviate from the estimated milk yield EY by a certain amount or more.

[0046] Figure 3(d) is an example of the output of the estimated milk yield comparison CY. The estimated milk yield comparison CY compares, for example, the estimated milk yield curves M1 and M2 of cows on farms with different environments. Note that the actual milk yield RY of a specific dairy cow may be added to this estimated milk yield curve and displayed.

[0047] Figure 3(e) is an example of output of a comparison of milk yield over time TY. Here, the estimated lactation curves for August of each year at the same farm are compared. Of course, the comparison may also be made with other farms. Also, as described later, the actual milk yield RY of a specific dairy cow may be added to this estimated lactation curve M and plotted. This makes it possible to show how the change in the actual milk yield of that dairy cow over time corresponds to the estimated milk yield EY.

[0048] Please refer to Fig. 2 again. 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.

[0049] <Estimated milk production EY> Figure 4 shows the flow of the calculation process of the estimated milk yield EY. When the calculation process of the estimated milk yield EY starts (step S116), the selection of population data and the creation conditions are decided (step S200). A large amount of farm data is stored in the memory 14 shown in Figure 1. From among them, the necessary location and data period are selected. The conditions for calculating the estimated milk yield EY are also entered here.

[0050] For example, the location is one's own farm, the data period is one year from last year, etc. Also, the conditions include all the conditions for calculating the estimated milk yield EY. For example, it is to decide the length of the period to be compiled when referring to the farm data. More specifically, the actual milk yield may be the milk yield per day, the average milk yield per three days, per week, or per month, or the maximum milk yield per month. Similarly, representative values ​​for other variables may be decided for each fixed period.

[0051] Next, an interpolation formula is created (step S202). The interpolation formula is explained in FIG. 5. In FIG. 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 at a certain farm. Each point plotted here is the actual milk yield RY for each dairy cow, so it is a scatter plot. However, because it is the 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 FIG. 5, this is represented by a no-data area VR.

[0052] In this way, if there are gaps in the farm data, there will be a lack of 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 be used 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 after calving N. Note that, in order to interpolate an actual scatter plot, it is not necessary to be limited to the WOOD curve, and other functions may be used.

[0053]

number

[0054] Here, A, B, and C are constants, Y is milk yield, and N is the number of days since parturition. 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 the number of days since parturition 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).

[0055]

number

[0056] That is, in the interpolation formula (3), milk yield Y is expressed as a function of the number of days since parturition N. As the form of the function, a WOOD curve can be suitably used, but it is not limited to this. The interpolation formula creation process (step S202) in FIG. 4 is a process for determining the interpolation formula (3) from the scatter diagram of the actual milk yield RY as described above.

[0057] [Create Interpolation Formula] Figure 6 shows the process of creating an interpolation formula (step S202). When the process of creating an interpolation formula starts (step S202), data on the number of days since calving N and the actual milk yield RY are 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 the number of days since calving N and the actual milk yield RY for each dairy cow.

[0058] Next, an interpolation formula that best reflects the actual data of the number of days since calving N and the milk yield Y is obtained. 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 FIG. 5). This can be said to be a process of finding a function that fits the actual milk yield RY. Then, the interpolation formula of equation (3) is obtained (step S234). After that, the process returns to the EY calculation routine (FIG. 4) (step S236).

[0059] An interpolation formula can be created for each day, because milking is done almost every day. However, if there are no major changes in the environment or in the individual dairy cows, the average of the actual data every three days or every week can be used to create the interpolation formula (3).

[0060] Also, taking a broader view of the time axis, an interpolation formula may be created by regarding the average value of one month's actual data as the actual data. More specifically, the average monthly milk yield of a certain dairy cow is taken as the actual milk yield RY of that dairy cow. Also, the number of days since calving of 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 FIG. 4.

[0061] 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 within the farm data. For example, it is used when the milk production calculation system 1 is used for the first time, or when the population is changed significantly.

[0062] If the factor analysis is to be performed (Y branch at step S204), the factor analysis process is performed (step S206). Details of the factor analysis process will be described later. If the factor analysis is not to be performed (N branch at step S208), the process proceeds to the next step. The factor analysis will be described in detail with reference to FIG. 10.

[0063] 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 by the interpolation equation (3) (FIG. 7(a)) shown in FIG. 5 and FIG. 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 interpolation formula (3) for dairy cows with N days after calving is Fw 1月 This is represented as (N).

[0064] Here, the number of days after delivery is 50 days. 50 The interpolated milk yield of dairy cows 50 days after parturition in each month from January to December is 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").

[0065] The factor that can best explain the interpolated milk volume for each month is determined as the factor parameter P. The best factor parameter P is known by carrying out the factor analysis process in step S206. If the factor parameter P to be used for calculating the estimated milk volume EY has already been determined, it is used. There does not have to be just one factor parameter P. In other words, this involves finding a regression equation for the interpolated milk volume Y for each month using the least squares method with one or more factors.

[0066] As is well known, the regression equation is expressed as equation (4).

[0067]

number

[0068] where EY is the estimated milk yield and x 1 , x 2 , , x k is the factor (factor parameter P), and a 1 , a 2 , ,a k , where c is a constant.

[0069] Figure 7(c) shows an example of a regression equation when there is one factor. From the interpolation equation (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 a specific day may represent the interpolated milk volume for that month.

[0070] Next, by sorting these monthly interpolated milk yields by factor parameter P (here, for example, the temperature at noon), the graph in Figure 7(c) can be obtained. Note that here, we will show an example of the results assuming that the estimated milk yield EY of dairy cows 50 days after calving in each month can be well explained by the temperature at noon. The regression equation is expressed by equation (5).

[0071]

number

[0072] where x 1 is the temperature at noon, and EY is the estimated milk yield. Such a regression equation is created for each number of days since parturition. In other words, if the final day of postpartum days N is 300 days, then a regression equation for days 1 to 300 can be obtained (see Figure 7(e)).

[0073] 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, it may be possible to closely approximate the relationship between the number of days since calving, N, and the interpolated milk yield, Y. The regression equation obtained in this way is represented as equation (6) (see Figure 7(d)).

[0074]

number

[0075] Where: N EY is called the estimated milk yield EY when a dairy cow is N days postpartum and has factor parameters P.

[0076] [Create regression equation] FIG. 8 shows a 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 FIG. 4 is started, the flow jumps to FIG. 8, where the factor parameter P is 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 p 1 (Temperature).

[0077] Next, the factor parameter p 1 For each specified period m (which is determined in step S200), the estimated milk yield Ym for each postpartum day N is calculated using the interpolation formula (3) (step S252). That is, the interpolated milk yield Ym in FIG. 7(b) is calculated. This makes it possible to plot the scatter diagram in FIG. 7(c).

[0078] Next, the factor parameter p 1 A regression equation is obtained by using the factor parameters p as explanatory variables and the interpolated milk yield Ym as a target variable (step S254). 1 For the same number of days since parturition, N, the relationship between milk yield and the formula (6) (FIG. 7(d)) ​​can be obtained. The regression formula can be obtained by any known method.

[0079] In this way, in the present invention, when calculating the estimated milk yield using the factor parameter P, the actual milk yield RY is not directly used, but the interpolated milk yield calculated from the interpolation formula that complements the actual milk yield RY is used. Therefore, even if there is a gap in the actual milk yield RY, a reasonable value can be calculated as the average milk yield.

[0080] Equation (6) can be used to create a number for postpartum days N from 1 to the final day (the final day here means the longest day after parturition) (see FIG. 7(e)). Regression equation (6) is a summary of FIG. 7(e). Once the regression equation has been found, the process returns to the routine for finding the estimated milk yield EY in FIG. 4 (step S256).

[0081] Referring again to FIG. 4, once the regression equation (6) for calculating the estimated milk yield EY is obtained, the estimated milk yield EY can be calculated by inputting the factor parameter P into the regression equation (6) (step S210).

[0082] When the value of factor parameter P is input into regression equation (6) corresponding to postpartum days N, an estimated milk yield EY is calculated. Therefore, by inputting the same value of factor parameter P sequentially into regression equation (6) representing each postpartum days N, a data set of postpartum days N and estimated milk yield EY for the same value of factor parameter P can be obtained. The curves relating to estimated milk yield in Figures 3(a) to (e) are obtained by plotting the data set of postpartum days N and estimated milk yield EY obtained in this way, or by connecting these points with a straight line.

[0083] Figure 9 shows an enlarged view of the estimated lactation curve M, which plots the estimated lactation yields EY in Figures 3(a) to (e). The estimated lactation yields EY are plotted for each day after parturition. The factor parameter P is p 1 andp 2 This shows the case where a multiple regression equation was calculated using these two factors. The number of days since calving N was calculated using discrete values ​​(10 days, 20 days, 30 days, 50 days, 80 days, 100 days, 150 days, 200 days, 250 days), but it can also be calculated in daily increments. Each estimated milk yield EY was calculated from a different regression equation (different number of days since calving; see Figure 7(e)) expressed by equation (6). In other words, the constants in these equations are different.

[0084] In this step S210 (see FIG. 4) of calculating the estimated milk production amount EY, the client inputs the data from the terminal 10, and the main unit 12 returns the estimated milk production amount EY. The client satisfies the requirements for the milk production amount calculation system 1 according to the present invention if it returns at least one estimated milk production amount EY. The estimated milk production amount EY may be calculated for a plurality of postpartum days N, and the calculated amount may be graphed as shown in FIG. 9.

[0085] Referring again to Fig. 4, if the factor parameter P is to be re-input (Y branch at step S212), step S210 is performed again. If the estimated milk production EY is not to be calculated (N branch at step S212), an end determination is performed (step S214). If the end is to be determined (Y branch at step S214), the process returns to the main routine (step S216). If the end is not to be determined (N branch at step S214), the process returns to step 200, and the EY calculation process is performed again.

[0086] [Factor analysis] Fig. 10 shows the process of factor analysis in step S206 in Fig. 4. The factor analysis process is the same as the creation of the regression equation (step S208) up to a certain point. Specifically, steps S250, S252, and S254 shown in Fig. 8 are the same as those in Fig. 8. Therefore, the same step numbers are also used.

[0087] Therefore, the factor parameter P is input (step S250), the milk yield Ym for each number of days after calving N is calculated using the interpolation formula (3) for each specified period m (which is determined in step S200) for the factor parameter P (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).

[0088] In the factor analysis, the actual milk yield RY for a certain period m is compared with the estimated milk yield EY calculated from the regression equation (step S260). Figure 11 shows this process. The factor parameters are p 1 andp 2 For example, temperature and wind speed. The fixed period is four different days. For example, it may be a representative day in spring, summer, autumn, or winter. Specifically, the days are m1 month n1, m2 month n2, m3 month n3, and m4 month n4. The specific factor parameters p 1 , p 2 The factor parameters can be selected by a separate main factor analysis or by trial and error. 1 , p 2 By inputting the above, the estimated milk yield EY can be calculated from the regression equation (6).

[0089] The number of days since parturition 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 estimated milk yield EY and the actual milk yield RY when the number of days after parturition is 30. This ERR is also calculated when the number of days after parturition is 50, 100, and 150. In the examples described later, an example is shown in which the estimated milk yield EY and the actual milk yield RY are compared when the number of days after parturition is 50.

[0090] The sum of the absolute values ​​ERR of the differences between the estimated milk yield EY and the actual milk yield RY for the four days is represented as the total error TΣ. Thus, in step 260, the estimated milk yield EY and the actual milk yield RY are compared.

[0091] Referring again to FIG. 10, after comparing the estimated milk yield EY and the actual milk yield RY, a termination determination is made (step S262). The termination determination 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 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 estimated milk yield EY and the actual milk yield RY has been described as an absolute value, a squared value may be used.

[0092] If it is to be ended (Y branch of step S262), the process returns to the EY calculation flow (step S264).If it is to be continued, the process returns to step S250 again, and the factor parameter P is re-input.

[0093] Referring again to Figure 4, once the factor analysis (step S206) is completed, the process proceeds to the milk yield calculation step (step S210). Once the factor analysis step (step S206) is completed, it is considered that a suitable factor parameter P has been found, and therefore it is considered that the step of calculating the estimated milk yield EY (step S210) may be performed. The steps thereafter are as described above.

[0094] In the milk production calculation system 1 according to the present invention, the step of obtaining 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 be called a regression equation obtaining step.

[0095] In the calculation of the estimated milk production amount EY (step S116), the estimated milk production amount EY is calculated by inputting the value of the factor parameter P. In this case, the value of the factor parameter P does not need to be past data, and may be a value of the factor parameter P predicted in the future. In other words, it may be the value of the factor parameter P tomorrow or the value of the factor parameter P six months from now.

[0096] As described above, the estimated milk yield calculation returns an estimated average milk yield for a specific population when the factor parameter P has a certain value. In other words, the estimated milk yield value is the expected milk yield that can be expected from dairy cows belonging to that population when the factor parameter P has a certain value.

[0097] <Individual estimated milk production IY> Referring again to FIG. 2, if the estimated milk yield EY is not selected (N branch in step S102), a decision is made as to whether or not to calculate an individual estimated milk yield IY (step S106). This decision can be selected from the client side through a terminal. In the process of calculating the individual estimated milk yield IY, the estimated milk yield EY is compared with the actual milk yield RY, and the difference is calculated. If the individual estimated milk yield is to be calculated (Y branch in step S106), the process proceeds to a calculation process of the individual estimated milk yield IY (step S118).

[0098] Figure 12 shows the flow of the calculation process of the individual estimated milk yield IY. When the process of calculating the individual estimated milk yield IY is started (step S116), the regression equation acquisition process is performed (step S290). This is the same as step S290 in Figure 4. Therefore, when step S290 is executed, the regression equation (6) for calculating the estimated milk yield EY is obtained.

[0099] Next, the actual milk yield IRY of a specific dairy cow I is specified (step S300). This specification can be made by inputting on the client side via a terminal. The actual milk yield IRY may be specified by specifying a specific dairy cow. This is because the actual milk yield of a specific dairy cow for a given period is recorded in memory 14 on the server side. When the actual milk yield IRY is specified, the dairy cow that milked that actual milk yield IRY is identified, and the number of days after parturition N when that actual milk yield IRY was milked and the value of the factor parameter P can also be obtained. The factor parameter P at this time is expressed as p 1 Let us assume that.

[0100] Next, the estimated milk yield EY is calculated and displayed (step S302). Here, a regression equation for multiple postpartum days N is selected from the regression equation (6) obtained from a population for a specified period, and a factor parameter P is input to calculate the estimated milk yield EY. By inputting multiple postpartum days N, the postpartum days and the estimated milk yield EY can be displayed two-dimensionally. The postpartum days N to be input may be determined in advance.

[0101] FIG. 13 shows an example of a display of an individual estimated milk yield IY. Referring to FIG. 13, the horizontal axis is the number of days after parturition (days), and the vertical axis is the milk yield Y. The factor parameter P is p 1 The estimated milk production EY is shown as a smooth line, but in reality it is a line connecting the N points corresponding to the number of days after parturition entered. In other words, it is the estimated milk production curve M.

[0102] Referring again to Figure 12, the actual milk yield of a dairy cow at the next specified postpartum day N is N IRY is compared with the estimated milk yield EY on the number of days after parturition N (step S304). This comparison may be made by displaying the actual milk yield IRY as a plot d on a two-dimensional display of the estimated milk yield EY (see FIG. 13).

[0103] In Figure 13, the number of days after birth, N, is n 1 , n 2 , n 3 , n 4 These data are all plotted with the factor parameter P at p 1In addition, in the step of calculating the individual estimated milk yield IY, the actual milk yields RY of a plurality of individuals may be plotted.

[0104] Referring again to Figure 12, actual milk yield N IRY and estimated milk yield N If the difference in EY is equal to or greater than the excellent threshold value Thu (Y branch in step S304), an excellent notification is made (step S310). The actual milk yield IRY of the specific dairy cow for which the excellent notification was obtained is greater than the estimated milk yield EY, indicating that the cow had an excellent milk yield (excellent dairy cow). The excellent notification may be given by text only, or may be displayed as an excellent score (du) on a graph of the estimated milk yield EY (see FIG. 13). In FIG. 13, the actual milk yield du when the number of days since calving N is n2 is greater than the estimated milk yield. n2 This indicates that EY is greater than or equal to the good threshold Thu.

[0105] Actual milk yield N IR and estimated milk yield N If the difference in EY is not greater than the good threshold Thu (N branch in step S304), the estimated milk yield N EY and actual milk yield N A comparison with IRY is made (step S306). This comparison may be made by displaying the actual milk yield IRY as a plot d on a two-dimensional display of the estimated milk yield EY (see FIG. 13).

[0106] Estimated milk production N EY and actual milk yield N If the difference in IRY is greater than the inferiority threshold Thd (Y distribution in step S306), an inferiority notification is made (step S312). The milk yield IRY of the identified dairy cow is less than the estimated milk yield EY, indicating that the cow had an inferior milk yield (inferior dairy cow). The inferiority notification may be made by text only, or may be displayed as an inferiority point (dd) on a graph of the estimated milk yield EY (see FIG. 3(b)).

[0107] In Figure 13, when the number of days after parturition is n1, the actual milk yield dd is the estimated milk yield. n1 Less than the inferiority threshold Thd for EY (estimated milk yield n1EY is greater than the inferiority threshold Thd than the actual milk yield dd.

[0108] In addition, if the difference with the estimated milk yield EY is not as far as the dominant threshold Thu or the recessive threshold Thd, no notification is required. In FIG. 13, this is the case when the number of days after parturition is n3 and n4. However, even in this case, the actual milk yield N IRY and estimated milk yield N The EY difference is what's needed.

[0109] Estimated milk production N EY and actual milk yield N If the difference IRY is not greater than the inferiority threshold Thd (N branch of step S306), it is determined whether the actual milk volume IRY is to be re-designated (step S308). If the actual milk volume IRY is to be re-designated (Y branch of step S308), the process proceeds to step S302.

[0110] If the actual milk volume IRY is not to be re-designated (N branch of step S308), the process proceeds to the end judgment (step S314). If the process is not to be ended (N branch of step S316), the process returns to the regression equation acquisition step (step S290). If the process is to be ended (Y branch of step S314), the process returns to the main routine (step S316).

[0111] In the process of calculating the individual estimated milk yield IY as described above, the milk yield IRY of the specified dairy cow is compared with the estimated milk yield EY obtained from the population, the difference with the estimated milk yield EY is calculated, and a notification is sent to dairy cows whose milk yield is more than a certain distance away from the estimated milk yield EY.

[0112] <Group estimated milk production GY> Referring again to Figure 2, if the individual estimated milk yield IY is not selected (N branch in step S108), a decision is made as to whether or not to calculate the group estimated milk yield GY (step S110). This decision can be made by selecting from the client side through a terminal. In the process of calculating the group estimated milk yield GY, when the factor parameter P is the same, the milk yields of the groups are compared (the difference is found).

[0113] When the group estimated milk production GY is to be calculated (Y branch of step S110), the process proceeds to a calculation step of the group estimated milk production GY (step S120).

[0114] Figure 14 shows the process of calculating the group estimated milk yield GY. When the process of calculating the group estimated milk yield GY is started (step S120), the regression equation acquisition process is performed (step S290). This is the same as step S290 in Figure 4. Therefore, when step S290 is executed, the regression equation (6) for calculating the estimated milk yield EY is obtained. At this time, the population naturally includes the dairy cows of the group to be compared.

[0115] Next, a group is designated (step S350). A "group" is a group of two or more dairy cows with the same feed, environment, calving number, etc. Hereinafter, there are groups A and B, each of which contains three dairy cows with different postpartum days.

[0116] When the group is specified, the estimated milk yield EY is calculated. Of course, the estimated milk yield curve M may be displayed (step S352). Since the population (here, at least the dairy cows of groups A and B) is determined, the estimated milk yield EY based on the factor parameter P can be calculated.

[0117] Next, for each cow in group A and group B, the actual milk yield was measured. N IRY and estimated milk yield N Then, if the actual milk volume IRY is higher than the superiority threshold value Thu, an excellent notification is made (step S354), and if the actual milk volume IRY is lower than the inferiority threshold value Thd, an inferiority notification is made (step S362).

[0118] Next, it is determined whether evaluation has been performed for all dairy cows in groups A and B (step S358). If not (N branch in step S358), the actual milk yield is calculated again. N IRY and estimated milk yield N EY are compared (steps S354 and S356).

[0119] Once the evaluation of all dairy cows has been completed (Y branch in step S358), the results are displayed (step S364). Figure 15 shows an example of a display of the group estimated milk yield GY. Here, the white circles represent group A and the black circles represent group B. Group A shows a higher milk yield compared to the estimated milk yield EY, and group B shows a lower milk yield compared to the estimated milk yield EY. It is believed that the differences between these groups are due to factors other than the factor parameter P. Those that have received a superiority notice are represented by white stars, and those that have received an inferiority notice are represented by black squares.

[0120] As described above, the herd estimated milk yield GY allows the milk yield of dairy cows in a specified herd to be compared using the estimated milk yield EY as the standard.

[0121] <Estimated milk production comparison CY> Referring again to FIG. 2, if the group estimated milk yield GY is not selected (N branch in step S110), it is determined whether or not to calculate the estimated milk yield comparison CY (step S112). This determination can be selected from the client side through the terminal. The estimated milk yield comparison CY compares the estimated milk yields EY. In other words, it compares the estimated milk yields EY with each other that have different values ​​of the factor parameter P. Alternatively, it may be a comparison between different populations.

[0122] When the estimated milk production amount comparison CY is calculated (Y branch of step S112), the process proceeds to a calculation step of the estimated milk production amount comparison CY (step S122).

[0123] FIG. 16 shows a process flow of the calculation step of the estimated milk yield comparison CY. When the calculation step of the estimated milk yield comparison CY is started (step S122), the regression equation acquisition step is performed (step S290). However, here, a regression equation for a different population may be acquired. For example, a comparison between farms in different regions. This is the same as selecting multiple populations in step S290 in FIG. 4. Of course, the population may be one. Therefore, when step S290 is executed, the regression equation (6) for calculating the estimated milk yield EY is obtained. Here, the population will be denoted by the symbol U in the following description. There may be three or more populations U.

[0124] Once the regression equation is obtained, the population U and the factor parameters P are specified (step S380). Once the population U and the factor parameters P are specified, the estimated milk yield EY is calculated and displayed.

[0125] Thereafter, the user is asked whether or not re-designation is necessary (step S384). If the user wishes to designate the population U and factor parameters P again (Y branch at step S384), the process moves to a process (step S380) in which the population U and factor parameters P are designated; if re-designation is not necessary (N branch at step S384), the process returns to the main flow (step S386).

[0126] Figure 17 shows an example of output from the estimated milk yield comparison CY process. Figure 17(a) shows the case of the same population U, but with different factor parameters P. For example, this is a display of different seasons on the same farm.

[0127] Figure 17(b) shows the results when the factor parameter P is the same and the populations are different, U1 and U2. This can be used between ranches with completely different climates due to differences in location conditions.

[0128] In this way, in the estimated milk yield comparison CY process, the estimated milk yields EY can be compared when the populations U are different, and the estimated milk yields can be calculated and displayed when the factor parameters P are different even for the same population.

[0129] <Comparison of transitional milk production TY> Referring again to Fig. 2, if the estimated milk production comparison CY is not selected (N branch in step S112), it is determined whether or not to calculate the transitional milk production comparison TY (step S114). This determination can be selected from the client side through the terminal.

[0130] In the process of calculating the transitional milk yield comparison TY, the estimated milk yields EY on different time axes are compared. Since the factor parameter P includes climatic data, the comparison of the estimated milk yields EY on different factor parameters P can also be said to be a comparison of the estimated milk yields EY on different time axes. Therefore, the transitional milk yield comparison TY is similar to the estimated milk yield comparison CY in the sense that it compares the estimated milk yields EY on different values ​​of the factor parameter P. However, it differs in that the date and time to be compared is determined first and the factor parameter P on that date and time is used in the process of calculating the estimated milk yield comparison CY. The requirement parameter P is determined first.

[0131] In addition, in the step of calculating the transitional milk yield comparison TY, the change over time in the actual milk yield of a particular dairy cow can also be expressed by comparing it with the factor parameter P at that time.

[0132] When the transitional milk production comparison TY is to be calculated (Y branch of step S114), the process proceeds to a calculation step of the transitional milk production comparison TY (step S124).

[0133] FIG. 18 shows a process flow of the calculation process of the transitional milk yield comparison TY. When the calculation process of the transitional milk yield comparison TY is started (step S124), the regression equation acquisition process is performed (step S290). However, here, regression equations for different populations may be acquired. For example, this is the case when comparing data from different years at the same farm. This is the same as selecting multiple populations in step S290 in FIG. 4. Of course, the population may be one. Therefore, when step S290 is executed, the regression equation (6) for calculating the estimated milk yield EY is obtained. Here, the year is represented by the symbol Q. The year Q may be multiple, or may be a period of several months.

[0134] Once the regression equation is obtained, the display period is specified (step S400). The display period may be, for example, the month, week, or date and time to be displayed and compared for each year. Once the date and time are determined for a specified population, the factor parameter P can be determined. The display period may also include the specification of the dairy cow itself. This is because there may be cases where it is desired to know the change in the milk yield RY of an individual along with the change in the estimated milk yield. Once the factor parameter P is determined, the estimated milk yield EY is calculated and displayed (step S402).

[0135] You are then asked whether or not you want to re-designate the display period (step S404). If you want to re-designate the display period (Y branch at step S404), the process moves to the process where the display period is designated (step S400). If you do not want to re-designate the display period (N branch at step S404), the process returns to the main flow (step S406).

[0136] Fig. 19 shows an example of output from the transitional milk yield comparison TY process. Fig. 19 shows a case where a specific population U has different years, and the estimated milk yield EY for August is displayed for each year Q. The factor parameter P for calculating the estimated milk yield EY for August may be the factor parameter P for a specific day in August, or the factor parameter P for three consecutive days or one week may be averaged.

[0137] Referring to Figure 19, the estimated milk yield (EY) has been increasing overall from 2018 to 2020. In particular, the number of days after parturition (n m Therefore, it can be seen that the milk production improvement implemented each year is having an effect.

[0138] In this way, in the step of comparing the transitional milk yield TY, the estimated milk yield EY for the designated display period can be compared. In Fig. 19, a display example of the same month in different years is shown, but it may be different months in the same year.

[0139] FIG. 20 also shows the results when the actual milk yield RY of individual dairy cows is plotted at the same time. FIG. 20(a) shows, for example, a lactation curve M (May) based on the estimated milk yield EY (May) in May 2019, and FIG. 20(b) shows a lactation curve M (August) based on the estimated milk yield EY (August) in August 2019. In both FIG. 20(a) and (b), the estimated milk yield for the other month is shown with a dotted line. For a specific dairy cow I, n Actual milk yield I n RY is displayed. Note that N 50 indicates the case where the postpartum period is 50 days, and N 150 The figure shows the case where the number of days after calving is 150 days. As time passes from May to August, individual cow I n The amount of milk produced by the cow also changes.

[0140] In Figure 20, this individual cow I n As of May, the estimated milk production was N50 The milk production was higher than in EY (May), but in August the estimated milk production N150 The value is the same as or lower than EY (August). In this way, the process of calculating the transitional milk yield comparison TY makes it possible to know the history of the change in milk yield of a specific dairy cow and the difference between that change and the average level.

[0141] Referring again to Fig. 2, as described above, the milk production calculation system 1 according to the present invention can calculate the estimated milk production EY, the individual estimated milk production IY, the group estimated milk production GY, the estimated milk production comparison CY, and the transitional milk production comparison TY, but it may also be possible to carry out other steps successively with the results of each step.

[0142] Specifically, the combination of menus is possible by saving the results of a certain menu and then displaying or using the saved results while executing another menu. EXAMPLES

[0143] The milk yield 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 was monthly data from 2019. The factor parameter P was THI and sunshine hours. For THI, the average monthly temperature and average monthly humidity values ​​were used. The sunshine hours were the total monthly value. The results are shown in Figure 21.

[0144] Figure 21(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 estimated milk yield EY was calculated using the obtained values ​​by simple regression with only THI as the factor parameter P. Figure 21(b) shows the results. Figure 21(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 parameter P. Note that both are for a case where the number of days since calving is 50 days.

[0145] When only THI was used as the factor parameter P, the standard error was 0.65, which varied considerably. However, when THI and sunshine hours were used as the factor parameter P, the standard error was 0.17, indicating that the actual milk yield could be estimated quite well.

[0146] The fact that milk production can be estimated using THI and sunshine hours was not previously known, and this will have a major impact on how we respond to future increases in milk production. [Industrial Applicability]

[0147] The present invention can be suitably used for computerization of dairy farming. [Explanation of symbols]

[0148] 1. Milk production calculation system 10 Terminal 12 Main unit 14. Memory 16. Ranch 18 External Information

Claims

[Claim 1] a memory for storing, as farm data, factor parameters including the number of days since calving, the actual milk yield for each number of days since calving, and weather data; an interpolation formula creating unit that uses a continuous function to interpolate the number of days since parturition and the actual milk yield for each fixed period in the farm data to create an interpolation formula that can calculate an interpolated milk yield at any number of days since parturition between the first day of the number of days since parturition and the last day of the number of days since parturition; a regression equation creating unit that creates a regression equation for calculating an estimated milk yield by using an interpolated milk yield obtained by substituting the arbitrary number of days since parturition into the interpolation equation as a response variable and at least one of the factor parameters selected from the farm data as an explanatory variable; A milk yield calculation system comprising a control device including a milk yield calculation unit that calculates an estimated milk yield from the regression equation for the inputted values ​​of the factor parameters.

Images