Livestock management support method and livestock management support system

Abstract

Provided is a livestock management support method performed by a livestock management support system. The livestock management support system performs: a step of calculating a candidate rearing parameter for improving an objective function, on the basis of at least a part of a rearing parameter and rearing data which is acquired for livestock reared based on the rearing parameter; and a step of outputting the candidate rearing parameter.

Description

Livestock management support method and livestock management support system

[0001] The present disclosure relates to a livestock management support method and a livestock management support system.

[0002] Livestock production, especially meat and milk production using ruminant livestock, is known to have a high environmental impact. As the global population increases, the global demand for protein is expected to expand, so there is a need to increase meat and milk production. To solve both food and environmental issues, low-environmental-impact ruminant livestock rearing techniques are needed.

[0003] Known prior art technologies aimed at reducing environmental impact include a "grassland management support system" (Non-Patent Document 1) that supports the efficiency of grassland management in public pastures, etc.; reducing N2O emissions from excrement by improving the amino acid balance in feed (Non-Patent Document 2); reducing exhaled methane using physical masks (Non-Patent Document 3); realizing low-environmental-impact agriculture through mountain grazing (Non-Patent Document 4); and an agricultural environment assessment system and agricultural environment assessment device (Patent Document 1).

[0004] Japanese Patent Application Laid-Open No. 2006-061093

[0005] Supporting the Efficiency of Grassland Management on Public Ranches, etc., NARO webpage, https: / / www.naro.go.jp / project / results / 4th_laboratory / nilgs / 2016 / 16_017.html Reducing Greenhouse Gases from Cow Manure with Feed with Improved Amino Acid Balance – Hopes for Environmentally Friendly Livestock Farming –, Agriculture, Forestry and Fisheries Research Council webpage, https: / / www.affrc.maff.go.jp / docs / press / attach / pdf / 211223-3.pdf Could a Mask that “Cleans” Cow Burps Slow the Progress of Climate Change? A British Startup’s Challenge, WIRED, https: / / wired.jp / 2021 / 01 / 19 / cows-climate-change-methane-stop / FaNBa Project, The Uchida Lab webpage, https: / / www.uchidalab.com / projects / fanba

[0006] While prior art documents contain technologies that focus on feed crop cultivation management, animal husbandry management, and manure management, there are no technologies that treat each process as a single system and achieve overall environmental load reduction. Non-Patent Document 4 describes a technology for evaluating the environmental load of an entire dairy farm in terms of nitrogen, and Patent Document 1 describes an agricultural environmental evaluation system, but does not consider proposing optimal management methods based on the evaluation results. The present disclosure has been made in light of the above circumstances, and aims to provide a livestock management support method and livestock management support system that support improving production while reducing environmental load.

[0007] A livestock management support method according to one aspect of the present disclosure is a livestock management support method performed by a livestock management support system, which calculates candidate feeding parameters for improving an objective function based on at least a portion of feeding parameters and feeding data acquired for livestock raised based on the feeding parameters, and outputs the candidate feeding parameters. The livestock management system according to one aspect of the present disclosure includes an input unit that accepts input of at least a portion of the feeding parameters, the feeding data, and the objective function, a calculation unit that calculates candidate feeding parameters for improving the objective function based on at least a portion of the feeding parameters and the feeding data acquired for livestock raised based on the feeding parameters, and an output unit that outputs the candidate feeding parameters.

[0008] According to the present disclosure, it is possible to provide a livestock management support method and a livestock management support system that support improving production volume while reducing environmental load.

[0009] Fig. 1 is a schematic diagram showing material circulation in livestock farming. Fig. 2 is a schematic diagram showing material circulation in livestock farming. Fig. 3 is a functional block diagram showing the configuration of a livestock farming management support system of an embodiment. Fig. 4 is a diagram showing common invariant parameters, individual invariant parameters, variable parameters, and objective function components according to each production form and feeding form. Fig. 5 is a flowchart showing a livestock farming management support method of an embodiment. Fig. 6 is a hardware configuration diagram.

[0010] Non-limiting embodiments of the present disclosure will be described below. The present disclosure is not limited to the examples in the following embodiments.

[0011] Figure 1 is a schematic diagram showing an example of material circulation in livestock farming. The main environmental impacts of raising ruminant livestock include the emission of greenhouse gases, nitrous oxide (NO) and methane (CH), as well as the loss of nitrate nitrogen (NO3 - ) leaching. The greenhouse gases nitrous oxide (N2O) and methane (CH4) are generated in the process of material circulation between plants, livestock, and soil. Fertilizers (chemical fertilizers, compost) added to the soil are decomposed and mineralized by microorganisms in the soil. Feed crops absorb these as a source of nutrients, and also synthesize organic matter using CO2 absorbed from the atmosphere through photosynthesis as a substrate, which they use for growth. Some of the nitrogen not absorbed by plants is released as N2O during the nitrification and denitrification reactions of microorganisms, or is converted into nitrate (NO3 - ) and leach into the groundwater. The grown forage crops are eaten and digested by ruminant livestock. During this digestion process, microorganisms in the rumen of ruminant livestock metabolize carbohydrates to produce CH4, which is excreted as burps. In addition, undigested matter and waste products are excreted as manure. The excreted manure is either put into the soil (farms, pastures) as is or after composting, or treated as industrial waste.

[0012] Figure 2 shows a simplified version of the material cycle in Figure 1. In systems where ruminant livestock production takes place, materials flow in and out in the form of carry-in, carry-out, fixation, and loss. In order to reduce the environmental impact of ruminant livestock production, it is expected that it will be important to (1) reduce the amount of material carried in by reducing the amount of chemical fertilizer and purchased feed used, (2) increase the amount of carbon fixed by plants, and (3) reduce losses by improving material utilization in the material cycle between plants, livestock, and soil.

[0013] <Livestock management support system>

[0014] 3 is a functional block diagram showing a livestock husbandry management support system according to an embodiment of the present invention. The livestock husbandry management support system according to the present invention includes an input unit 1 that receives input of at least some of the feeding parameters, feeding data, and an objective function, a calculation unit 2 that calculates candidate feeding parameters for improving the objective function based on at least some of the feeding parameters and feeding data acquired for livestock raised based on the feeding parameters, and an output unit 3 that outputs the candidate feeding parameters.

[0015] The input unit 1 receives input of at least a portion of the feeding parameters, feeding data, and an objective function.

[0016] In an embodiment, the feeding parameters are parameters used in raising livestock. Examples of feeding parameters are shown in Figure 4. As shown, the feeding parameters may include common invariable parameters, individual invariable parameters, and variable parameters. The common invariable parameters and individual invariable parameters are collectively referred to as invariable parameters. The feeding parameters in the embodiment may be feeding parameters that have been determined in advance by a system other than the livestock farming management support system of the present disclosure and used in raising livestock, or may be parameters that have been calculated as candidate feeding parameters by the livestock farming management support system of the present disclosure and then used in raising livestock.

[0017] The common invariant parameters are fixed parameters that are set in common regardless of the livestock production type and feeding type. In other words, the common invariant parameters are parameters that are not subject to optimization or improvement in the livestock management support method of the embodiment. The common invariant parameters include, for example, climate (variable value), soil classification (fixed value), soil analysis value (fixed value), field area (fixed value), and number of livestock. In the embodiment, the production type includes dairy farming and / or beef cattle fattening, and the feeding type of the livestock includes feeding in barns and / or pastures.

[0018] The individual invariable parameters are parameters that are set according to the livestock production and feeding methods, and are parameters that are not subject to optimization or improvement in the livestock management support method of the embodiment. Examples of the individual invariable parameters include the number of feedings (feed), the type of feed, the nutritional content of the feed, the type of fertilizer, the area of ​​pasture, and the type of grass.

[0019] The variable parameters are parameters that are set according to the livestock production and feeding methods, and are parameters that can be subject to optimization or improvement in the livestock management support method of the embodiment. The variable parameters include, for example, the amount of compost, the amount of chemical fertilizer, the amount of top dressing, the number of top dressings, the dry period, the grazing period, the amount of dry feed (roughage), the amount of dry feed (concentrated feed), the number of top dressings, etc.

[0020] In the example of Figure 4, when the production type is dairy farming and the rearing type is barn-rearing only, the individual invariable parameters include the number of feedings, the type of feed, the nutritional content of the feed, and the type of fertilizer, and the variable parameters include the amount of compost, the amount of chemical fertilizer, the amount of top dressing, the number of top dressings, and the dry period.

[0021] In the example of Figure 4, when the production type is dairy farming and the feeding method involves grazing, the individual invariable parameters include the pasture area, feeding frequency, feed type, feed nutritional content, pasture type, and fertilizer type, and the variable parameters include the grazing period, feed dry matter amount (roughage), feed dry matter amount (concentrated feed), amount of compost, amount of chemical fertilizer, amount of top dressing, number of top dressings, and dry period.

[0022] In the example of Figure 4, when the production method is beef cattle fattening and the feeding method is barn-raising only, the individual invariable parameters include the number of feedings, the type of feed, the nutritional content of the feed, and the type of fertilizer, and the variable parameters include the amount of compost, the amount of chemical fertilizer, the amount of top dressing, the number of top dressings, and the fattening period.

[0023] In the example of Figure 4, when the production type is beef cattle fattening and the feeding type is with grazing, the individual invariable parameters include the pasture area, feeding frequency, feed nutritional content, feed type, pasture type, and fertilizer type, and the variable parameters include the grazing period, feed dry matter amount (roughage), feed dry matter amount (concentrated feed), amount of compost, amount of chemical fertilizer, amount of top dressing, number of top dressings, and fattening period.

[0024] The feeding parameters input in input unit 1 may be part of the feeding parameters used in feeding. Such part of the feeding parameters may be variable parameters. That is, in an embodiment, the feeding parameters include variable parameters and non-variable parameters, and "at least part of the feeding parameters" input in input unit 1 may be variable parameters.

[0025] The feeding data input in the input unit 1 is data obtained from livestock raised using feeding parameters. In this embodiment, the feeding data includes at least one of greenhouse gas emissions, costs, production volume, working hours, and nitrate nitrogen leaching. These data are obtained during livestock raising and are used as components when calculating the objective function.

[0026] The rearing parameters and rearing data input in input unit 1 may include multiple sets of rearing parameters used in multiple rounds of rearing conducted over multiple past periods and / or in multiple rearing facilities, and multiple sets of rearing data obtained from the rearing.

[0027] In an embodiment, the greenhouse gas emissions may be emissions of one or more selected from the group consisting of CO2, CH4, and N2O. The emissions of these gases may be measured by a method including measuring soil-derived gases using a soil respiration measurement device and / or measuring cow-derived gases using a cow respiration measurement device. The emissions of these gases may be weighted according to their contribution to global warming, and then added together, and used as input in input unit 1 or for calculation in calculation unit 2.

[0028] In an embodiment, the costs are those necessary for raising the animals. Such costs may include consumable costs calculated by multiplying the consumption of feed and other consumables used in raising the animals by their unit prices. Costs may also include labor costs calculated by multiplying labor hours by the hourly wage.

[0029] In an embodiment, the production amount may be determined by a method including measuring the amount of various livestock products or measuring livestock gain using a weighing scale. The production amount may be calculated by multiplying the amount of product by the unit price in the market.

[0030] In an embodiment, the amount of nitrate nitrogen leaching can be estimated by measuring the nitrate nitrogen concentration in environmental water such as groundwater. In this case, the nitrate nitrogen concentration itself may be used as the "amount of nitrate nitrogen leaching."

[0031] The objective function input in the input unit 1 is a function for calculating candidate feeding parameters that improve the objective function. Here, improving the objective function may mean minimizing the objective function, or conversely, maximizing the objective function. The objective function may include weighting of its component feeding data. In an embodiment, the objective function may include one or more selected from the group consisting of production volume, greenhouse gas emissions, cost, labor hours, and nitrate nitrogen leaching. Therefore, the objective function may be a linear combination of at least one of production volume, greenhouse gas emissions, cost, labor hours, and nitrate nitrogen leaching. By setting such an objective function, a farmer can improve or optimize feeding parameters using evaluation indexes that give greater weight to components that are particularly prioritized.

[0032] In an embodiment, the objective function may be, for example, a function of the form: f(x)=w1x1+w2x2+...+w n x n Here, the weights w = (w1, w2, ..., w n ) is w1+w2+…+w n = 1, and the components x = (x1, x2, ..., x n) are production volume, greenhouse gas emissions, cost, labor hours, nitrate nitrogen leaching, etc. Furthermore, each weight may be positive or negative depending on whether it is desirable to increase or decrease the component. Alternatively, for some of these components, the reciprocal of the component may be used to calculate the objective function. For example, when production volume and cost are used as components of the objective function, production volume may be weighted with a positive value and cost may be weighted with a negative value, and improving such an objective function may mean making the objective function as large as possible.

[0033] The calculation unit 2 calculates candidate feeding parameters for improving the objective function based on at least some of the feeding parameters and feeding data acquired for the livestock raised based on the feeding parameters. The calculated candidate feeding parameters may be parameters of a type corresponding to some of the feeding parameters (compost amount, chemical fertilizer amount, top dressing amount, number of top dressings, dry period, grazing period, amount of dry matter feed (roughage), amount of dry matter feed (concentrate), number of top dressings, etc.).

[0034] The method by which the calculation unit 2 calculates candidate feeding parameters for improving the objective function is not limited, and for example, a black-box optimization method may be used. Using the black-box optimization method, a parameter set that will produce an optimal result for an objective function whose specific form is unknown can be calculated with low computational cost. The specific technique for the black-box optimization method is not limited, and it may be performed using a Bayesian optimization method and / or a genetic algorithm method.

[0035] When using Bayesian optimization to calculate candidate feeding parameters, Gaussian process regression can be used. In Gaussian process regression, a surrogate model expressed by a Gaussian process is used to approximate a function representing the relationship between at least some of the feeding parameters and the calculated value (objective function f(x)) to be optimized. The initial model of this surrogate model can be calculated by Gaussian process regression based on the initial feeding parameters previously obtained by the calculation unit 2 and the feeding data obtained when these parameters are used. From this surrogate model, an acquisition function is calculated, which serves as an index for determining the next point to be searched in the parameter space (where improvement of the feeding parameters is expected). The next search point is determined based on the acquisition function. The type of acquisition function used is not particularly limited, and examples include upper confidence bound, probability of improvement, and expected improvement. The determined feeding parameters are output from the output device 3 as candidate feeding parameters, and new feeding is performed. The output parameters may be parameters of a type corresponding to some of the feeding parameters input to the input device 1. Feeding data obtained from feeding using candidate feeding parameters is input to the input unit 1, and the calculation unit 2 uses the data to calculate a new surrogate model expressed as a Gaussian process using Gaussian process regression. A new acquisition function is calculated from the new surrogate model, and a new search point is determined based on this. As the number of samples used in the optimization calculation increases, the surrogate model successfully approximates the relationship between the feeding parameters and the objective function f(x), and ultimately, the point that optimizes the surrogate function calculated from the surrogate model also becomes the point that optimizes the objective function f(x). This cycle of determining search points and acquiring feeding data can be repeated until a predetermined objective (e.g., a specific value of the objective function, or an improvement in the objective function value from the previous time being less than a predetermined value) is achieved.

[0036] The output unit 3 outputs the candidate feeding parameters. The candidate feeding parameters output by the output unit 3 are the parameters calculated by the calculation unit 2. Based on the output candidate feeding parameters, the next round of livestock feeding is carried out, and new feeding data can be obtained.

[0037] <Livestock management support method>

[0038] The livestock management support method performed by the livestock management support system of this embodiment calculates candidate feeding parameters for improving the objective function based on at least a portion of the feeding parameters and feeding data obtained for livestock raised based on the feeding parameters, and outputs the candidate feeding parameters.

[0039] An example of a livestock management support method performed by the livestock management support system of the embodiment will be described with reference to the flowchart of FIG.

[0040] In step S1, the farmer begins feeding livestock using initial feeding parameters or previously calculated candidate feeding parameters.

[0041] In step S2, the farmer acquires livestock breeding data for the livestock breeding started in step S1.

[0042] In step S3, the input unit 1 of the livestock management support system accepts input of at least a portion of the feeding parameters, feeding data, and objective function. The input feeding data includes the feeding data acquired in step S2. The input may also include previously acquired feeding data and at least a portion of the feeding parameters used when the feeding data was acquired. Furthermore, if any feeding data, feeding parameters, or objective function other than the feeding data acquired in step S2 has already been input to the livestock management support system, input of this data may be omitted.

[0043] In step S4, the calculation unit 2 calculates candidate feeding parameters based on the input feeding parameters, feeding data, and objective function.

[0044] In step S5, the output unit 3 outputs the candidate feeding parameters calculated in step S4. After the candidate feeding parameters have been output, the process returns to step S1 and the improvement of the feeding parameters is repeated.

[0045] The livestock management support system described above can use, for example, a general-purpose computer system as shown in FIG. 6 . The illustrated computer system includes a CPU (Central Processing Unit, processor) 901, a memory 902, a storage 903 (HDD: Hard Disk Drive, SSD: Solid State Drive), a communication device 904, an input device 905, and an output device 906. The memory 902 and the storage 903 are storage devices. In this computer system, the CPU 901 executes a predetermined program loaded onto the memory 902 to realize the functions of the livestock management support system. The livestock management support system may be implemented on a single computer or multiple computers. The livestock management support system may also be a virtual machine implemented on a computer. The program for the optimization device 3 can be stored on a computer-readable recording medium such as an HDD, SSD, USB (Universal Serial Bus) memory, CD (Compact Disc), or DVD (Digital Versatile Disc), or can be distributed via a network. The computer-readable recording medium is, for example, a non-transitory recording medium.

[0046] The present invention is not limited to the above-described embodiment, and various modifications are possible within the scope of the invention.

[0047] According to the livestock management support method of the embodiment, it is possible to provide a livestock management support method and a livestock management support system that support improving production volume while reducing environmental impact. Furthermore, according to the livestock management support method of the embodiment, some of the feeding parameters used in the calculation and / or parameters output as candidate feeding parameters may be set according to the conditions of each farm, including the production form and feeding form, thereby enabling optimal support to be realized according to the conditions of each farm.

[0048] The present disclosure includes the following embodiments. (Item 1) A livestock husbandry management support method performed by a livestock husbandry management support system, comprising: calculating candidate feeding parameters for improving an objective function based on at least a portion of feeding parameters and feeding data acquired for livestock raised based on the feeding parameters; and outputting the candidate feeding parameters. (Item 2) The livestock husbandry management support method according to Item 1, wherein the step of calculating the candidate feeding parameters is performed by black-box optimization. (Item 3) The livestock husbandry management support method according to Item 1 or 2, wherein the objective function includes at least one of production volume, greenhouse gas emissions, cost, labor hours, and nitrate nitrogen leaching. (Item 4) A livestock husbandry management support system comprising: an input unit that accepts input of at least a portion of the feeding parameters, feeding data, and objective function; a calculation unit that calculates candidate feeding parameters for improving the objective function based on at least a portion of the feeding parameters and feeding data acquired for livestock raised based on the feeding parameters; and an output unit that outputs the candidate feeding parameters. (Item 5) A livestock husbandry management support method according to any one of Items 1 to 3, wherein the feeding parameters include variable parameters and invariable parameters, and at least some of the feeding parameters are the variable parameters. (Item 6) A livestock husbandry management method according to Item 5, wherein the variable parameters and / or the invariable parameters are set according to the livestock production form and / or the feeding form of the livestock. (Item 7) A livestock husbandry management support method according to Item 6, wherein the livestock production form includes dairy farming and / or beef cattle fattening, and the feeding form of the livestock includes raising in barns and / or pastures. (Item 8) A livestock husbandry management support method according to any one of Items 1 to 3 and 5 to 8, wherein the candidate feeding parameters are parameters of a type corresponding to some of the feeding parameters.

[0049] Although the present disclosure has been described with reference to the above several embodiments, the present disclosure is not limited to the examples in the above embodiments. Various modifications can be made to the configuration and details of the present disclosure within the scope of the present disclosure.

[0050] 1 Input section 2 Calculation section 3 Output section

Claims

1. A livestock management support method performed by a livestock management support system, which calculates candidate feeding parameters for improving an objective function based on at least a portion of feeding parameters and feeding data obtained for livestock raised based on the feeding parameters, and outputs the candidate feeding parameters.

2. The livestock management support method according to claim 1, wherein the step of calculating the candidate feeding parameters is performed by black-box optimization.

3. The livestock management support method according to claim 1 or 2, wherein the objective function includes at least one of production volume, greenhouse gas emissions, cost, labor hours, and nitrate nitrogen leaching volume.

4. A livestock management support system comprising: an input unit that accepts input of at least a portion of the feeding parameters, feeding data, and an objective function; a calculation unit that calculates candidate feeding parameters for improving the objective function based on at least a portion of the feeding parameters and feeding data obtained for livestock raised based on the feeding parameters; and an output unit that outputs the candidate feeding parameters.

Images