Agricultural tool shearing support system, agricultural tool shearing support program and agricultural tool shearing support method

Abstract

To provide a scheduling system, a scheduling program and a scheduling method which can promote the increase of borrowing users.SOLUTION: An agricultural tool shearing support system comprises: a borrowing request specification unit 13a for allowing a borrowing user SU to specify a temporary borrowing agricultural tool KN and a temporary borrowing date / time KT; a first learnt model 13b for outputting compliance prediction data YD using weather prediction data KY of a specified date / time; a compliance prediction display unit 13e for displaying compliance prediction TY to the borrowing user SU based on the compliance prediction data YD outputted by the first learnt model 13b; a borrowing determination unit 13f for allowing the borrowing user SU to determine a borrowing agricultural tool SN and a borrowing date / time ST; and an evaluation material data acquisition unit 13g for acquiring evaluation material data ZD. The first learnt model 13b machine-learns the compliance prediction data YD, the data used for the output of the compliance prediction data YD and the evaluation material data ZD.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to an agricultural machinery sharing support system, an agricultural machinery sharing support program, and an agricultural machinery sharing support method that support the sharing of agricultural machinery between users. [Background technology]

[0002] In recent years, so-called car sharing, where multiple people share a vehicle, has become popular. Car sharing has the advantage of eliminating the need for vehicle purchase and maintenance costs (maintenance costs, parking fees, etc.), and car sharing is becoming more widespread and established.

[0003] As car sharing becomes more popular, the number of services that support car sharing is also increasing. A "service that supports car sharing" is, for example, a service that matches users who want to rent a vehicle with users who want to borrow a vehicle and acts as an intermediary between the two.

[0004] As an example of technology related to such a service, Patent Document 1 discloses technology for a car sharing system equipped with a server that manages the rental of shared vehicles. The car sharing system disclosed in Patent Document 1 has a function in which the server obtains the operating rate of unused vehicles and discounts the usage fee based on the operating rate, which is said to increase the number of users.

[0005] The applicant was inspired by the car sharing business described above and thought that he could support agricultural workers by supporting the sharing of agricultural machinery. In other words, he thought that by sharing agricultural machinery and reducing the purchase and maintenance costs of the machinery, the burden on agricultural workers could be reduced. In addition, agricultural machinery is often a specialized item that can only be used for specific farm work, and there are also tools that are necessary but not used frequently, so the advantage of sharing agricultural machinery is greater than that of car sharing. The technology disclosed in Patent Document 2 is mentioned in paragraphs

[0062] and

[0063] . [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Patent Publication No. 2022-082155 [Patent Document 2] JP 2019-128741 A Summary of the Invention [Problem to be solved by the invention]

[0007] However, if the car sharing method is simply applied to the sharing of agricultural machinery, it is predicted that problems will arise in terms of the balance between users who lend out agricultural machinery (hereinafter referred to as "lending users") and users who borrow agricultural machinery (hereinafter referred to as "borrowing users"). Specifically, there are fewer people who use agricultural machinery compared to those who use cars, and since there are many types of agricultural machinery, the demand for one piece of agricultural machinery is much lower than the demand for one vehicle. In other words, it is thought that there will be fewer users who want to borrow agricultural machinery compared to users who lend it out.

[0008] In such a situation, if a service supporting sharing of agricultural machinery is provided, even if a lending user wishes to lend agricultural machinery, the agricultural machinery may not be borrowed or the frequency of borrowing may decrease, so the lending user may stop using the service. As a result, if the number of lending users decreases and the amount of lending agricultural machinery decreases, the borrowing user may be less likely to be able to borrow the agricultural machinery he or she wants to borrow, and the borrowing user may also stop using the service, and ultimately, it may become difficult to continue the service. In other words, in order to continue a service supporting sharing of agricultural machinery, it is important to increase the number of borrowing users.

[0009] As a method for increasing the number of users, a method of discounting the usage fee based on the operation rate, as disclosed in Patent Document 1, is conceivable, but in the case of agricultural machinery sharing, the effect of this method cannot be expected compared to car sharing. This is because, while many agricultural tasks need to be performed at specific times (seasons and weather), agricultural machinery with a low operation rate is likely to be used for agricultural tasks that are not performed at the aforementioned times, so even if there is a discount, it is difficult to say that the number of users who rent the agricultural machinery will increase.

[0010] The present invention has been made in consideration of the above-mentioned conventional circumstances. It is an object of the present invention to provide an agricultural machinery sharing support system, an agricultural machinery sharing support program, and an agricultural machinery sharing support method that can promote an increase in the number of users of the agricultural machinery sharing system.

[0011] An agricultural machinery sharing support system according to a first aspect of the present invention is an agricultural machinery sharing support system for supporting the sharing of agricultural machinery between users, comprising: a borrowing desire designation unit that causes a borrowing user, who is a user who borrows agricultural machinery, among the users to designate a temporary borrowing agricultural machinery which is the agricultural machinery that the user wishes to borrow, and a temporary borrowing date and time which is the date and time at which the user wishes to borrow the temporary borrowing agricultural machinery; a first trained model that uses weather forecast data for a specific date and time to output suitability prediction data which is data relating to a prediction as to whether the specific date and time is a suitable date and time for borrowing a specific agricultural machinery; and a borrowing desire designation unit that outputs suitability prediction data which is data relating to a prediction as to whether the specific date and time is a suitable date and time for borrowing a specific agricultural machinery, the temporary borrowing date and time designated by the borrowing desire designation unit ... The system includes a suitability prediction display unit that displays to the borrowing user a suitability prediction, which is a prediction of whether the provisional borrowing date and time is a suitable date and time for borrowing the provisional borrowed agricultural machinery based on suitability prediction data related to the agricultural machinery and the provisional borrowing date and time, a borrowing determination unit that allows the borrowing user to decide the borrowed agricultural machinery, which is the agricultural machinery to actually borrow, and the borrowing date and time, which is the date and time when the borrowed agricultural machinery will actually be borrowed, and an evaluation material data acquisition unit that acquires evaluation material data, which is data used to evaluate whether the suitability prediction data output by the first trained model was appropriate, and the first trained model can be configured to be generated by machine learning using the suitability prediction data, the data used to output the suitability prediction data, and the evaluation material data as learning data.

[0012] As mentioned above, the suitability prediction is a prediction of whether the provisional borrowing date and time is a suitable date and time for "borrowing the provisionally borrowed agricultural machinery," or in other words, it can be said to be an indirect prediction of whether the provisional borrowing date and time is a suitable date and time for "agricultural work to be carried out using the provisionally borrowed agricultural machinery."

[0013] In other words, according to the above configuration, the agricultural machinery sharing support system can display a suitability prediction and suggest to the borrowing user an "appropriate farm work schedule," which is useful information, thereby promoting an increase in the number of borrowing users.

[0014] Furthermore, it is expected to have a particularly excellent effect of encouraging use by people who do not normally do farm work (people other than agricultural workers), thereby expanding the user base. For people who do not normally do farm work and have little knowledge about farm work, it is very difficult to decide when is the best time to do farm work, but by using the appropriate farm work schedule proposed by the agricultural machinery sharing support system configured as described above, they can do farm work without having to make that decision themselves.

[0015] The agricultural machinery sharing support system relating to the second aspect of the present invention can be configured so that the evaluation material data is at least one of past weather data, data relating to an evaluation by the borrowing user of whether the borrowing date and time was a suitable date and time for borrowing the borrowed agricultural machinery, or data relating to an evaluation of the borrowed agricultural machinery by the borrowing user.

[0016] In the agricultural machinery sharing support system according to the third aspect of the present invention, the first trained model can be configured to output suitability prediction data using data related to agricultural work performed in the past by the borrowing user. In addition, since the first trained model is generated by machine learning using the data used to output the suitability prediction data as training data, the configuration can also be said to be a configuration in which data related to agricultural work performed in the past by the borrowing user can be used as training data for generating the first trained model.

[0017] According to the above configuration, the first trained model learns data related to agricultural work performed in the past, and outputs suitability prediction data using the data related to agricultural work performed in the past, so that the suitability prediction data can also take into account the influence of agricultural work performed in the past by the borrowing user.

[0018] The agricultural machinery sharing support system according to the fourth aspect of the present invention can be configured so that the first trained model further outputs suitability prediction data using user data including at least one of the borrowing user's height, farming experience, age, sex, shipping form of agricultural products, or sales scale. In addition, since the first trained model is generated by machine learning using the data used to output the suitability prediction data as training data, the configuration can also be said to be a configuration in which user data can be further used as training data for generating the first trained model.

[0019] According to the above configuration, the first trained model learns user data and also uses the user data to output suitability prediction data, so that the suitability prediction data can take into account the influence of differences between borrowing users (individual differences) and the accuracy and suitability of the evaluation material data ZD provided by the user US.

[0020] The agricultural machinery sharing support system according to a fifth aspect of the present invention can be configured so that the evaluation material data acquisition unit acquires a part or all of the evaluation material data from the borrowing user.

[0021] A large amount of training data is required to generate a highly accurate trained model. In particular, predictions of the suitability of agricultural work are easily affected by differences in the environment (for example, the climate and geology differ depending on the region where agricultural work is performed), so in order to generate a highly accurate trained model, it is necessary to collect a large amount of training data for each environment.

[0022] In spite of these problems, according to the above configuration, evaluation material data (one type of learning data) can be obtained from the borrowing user, so that learning data in the borrowing user's environment can be naturally collected in the course of operating the service that supports agricultural machinery sharing, thereby reducing the effort and cost required for the administrator of the agricultural machinery sharing support system to collect learning data.

[0023] The agricultural machinery sharing support system according to a sixth aspect of the present invention includes a second trained model that is generated by machine learning using the suitability prediction data, the data used to output the suitability prediction data, and the evaluation material data obtained from the borrowing user as learning data, and outputs second correction data, which is data for correcting the suitability prediction data based on the preferences of each borrowing user, and the suitability prediction display unit can be configured to display a suitability prediction that reflects the preferences of each borrowing user based on the second correction data.

[0024] According to the above configuration, since the suitability prediction can reflect the preferences of the borrowing user, the suitability prediction becomes more useful information for the borrowing user, which can further contribute to promoting an increase in the number of borrowing users.

[0025] The agricultural machinery sharing support system according to the seventh aspect of the present invention can be configured to include an incentive granting unit that grants an incentive to the borrowing user who has provided the evaluation material data.

[0026] According to the above configuration, an incentive can be given to borrowing users who provide evaluation material data, which is one type of learning data. This encourages borrowing users to proactively provide evaluation material data in order to receive the incentive, thereby enabling more learning data to be collected.

[0027] In the agricultural machinery sharing support system relating to an eighth aspect of the present invention, the borrowing desire designation unit allows the borrowing user to designate multiple provisional borrowing dates and times, the first trained model outputs multiple suitability prediction data relating to the provisional borrowing agricultural machinery and the multiple provisional borrowing dates and times designated by the borrowing desire designation unit, the suitability prediction display unit displays a suitability prediction list that is a list of the multiple provisional borrowing dates and times and suitability predictions for each provisional borrowing date and time based on the multiple relevant suitability prediction data output by the first trained model, and the borrowing determination unit can be configured to allow the borrowing user to determine the borrowing date and time from the suitability prediction list displayed by the suitability prediction display unit.

[0028] According to the above configuration, the borrowing user can specify multiple temporary borrowing dates and times as candidates for the borrowing date and time, and check their suitability predictions in a list, so that the borrowing date and time can be determined while comparing the suitability predictions for the multiple temporary borrowing dates and times.

[0029] The agricultural machinery sharing support system according to a ninth aspect of the present invention includes a third learned model that is generated by machine learning using the suitability prediction data, the data used to output the suitability prediction data, and the decision of the borrowing user in the borrowing decision unit as learning data, and outputs third correction data, which is data for correcting the suitability prediction data based on the preferences of each borrowing user, and the suitability prediction display unit can be configured to display a suitability prediction list reflecting the preferences of each borrowing user based on the third correction data.

[0030] According to the above configuration, since the suitability prediction list can reflect the preferences of the borrowing user, the suitability prediction list becomes more useful information for the borrowing user, and can further contribute to promoting an increase in the number of borrowing users.

[0031] A tenth aspect of the present invention provides an agricultural machinery sharing support program for supporting the sharing of agricultural machinery between users, the agricultural machinery sharing support program including a borrowing request designation function that causes a borrowing user, who is a user who borrows agricultural machinery, among the users to designate a temporary borrowing agricultural machinery that is the agricultural machinery that the user wishes to borrow and a temporary borrowing date and time that is the date and time at which the user wishes to borrow the temporary borrowing agricultural machinery; a suitability prediction data output function that causes a first trained model to output suitability prediction data that is data relating to a prediction as to whether or not a specific date and time is a suitable date and time for borrowing a specific agricultural machinery, using weather forecast data for the specific date and time; and a suitability prediction data output function that causes the first trained model to output suitability prediction data specified by the borrowing request designation function using the weather forecast data for the temporary borrowing date and time specified by the borrowing request designation function. The computer can be configured to realize a suitability prediction display function that displays to the borrowing user a suitability prediction, which is a prediction of whether the provisional borrowing date and time is a suitable date and time for borrowing the provisional borrowed agricultural machinery based on suitability prediction data related to the provisional borrowed agricultural machinery and the provisional borrowing date and time; a borrowing decision function that allows the borrowing user to decide the borrowed agricultural machinery, which is the agricultural machinery to actually borrow, and the borrowing date and time, which is the date and time when the borrowed agricultural machinery will actually be borrowed; and an evaluation material data acquisition function that acquires evaluation material data, which is data used to evaluate whether the suitability prediction data output by the first trained model was appropriate, and the first trained model can be configured to be generated by machine learning using the suitability prediction data, the data used to output the suitability prediction data, and the evaluation material data as learning data.

[0032] As mentioned above, the suitability prediction is a prediction of whether the provisional borrowing date and time is a suitable date and time for "borrowing the provisionally borrowed agricultural machinery," or in other words, it can be said to be an indirect prediction of whether the provisional borrowing date and time is a suitable date and time for "agricultural work to be carried out using the provisionally borrowed agricultural machinery."

[0033] In other words, according to the above configuration, the agricultural machinery sharing support program can display a suitability prediction and suggest to the borrowing user an "appropriate farm work schedule," which is useful information, thereby promoting an increase in the number of borrowing users.

[0034] Furthermore, it is expected to have a particularly excellent effect of encouraging use by people who do not normally do farm work (people other than agricultural workers), thereby expanding the user base. For people who do not normally do farm work and have little knowledge about farm work, it is very difficult to decide when is the best time to do farm work, but by using the appropriate farm work schedule proposed by the agricultural machinery sharing support program configured as above, they can do farm work without having to make that decision themselves.

[0035] An agricultural machinery sharing support method according to an eleventh aspect of the present invention is an agricultural machinery sharing support method for supporting sharing of agricultural machinery between users using an agricultural machinery sharing support system, the agricultural machinery sharing support method including a borrowing user, who is a user who borrows agricultural machinery, among the users, specifying, via the agricultural machinery sharing support system, a temporary borrowing agricultural machinery which is the agricultural machinery that the user wishes to borrow, and a temporary borrowing date and time which is a date and time at which the user wishes to borrow the temporary borrowing agricultural machinery; a suitability prediction data output step of causing the agricultural machinery sharing support system to output, to a first trained model, suitability prediction data which is data relating to a prediction as to whether or not the specific date and time is a suitable date and time for borrowing a specific agricultural machinery, and a suitability prediction data output step of the first trained model using the weather forecast data for the temporary borrowing date and time specified by the borrowing request designation function. The agricultural machinery sharing support system includes a suitability prediction display process for displaying to the borrowing user a suitability prediction, which is a prediction of whether the provisional borrowing date and time is a suitable date and time for borrowing the provisional borrowing agricultural machinery based on suitability prediction data relating to the provisional borrowing agricultural machinery and the provisional borrowing date and time specified by the desired designation function; a borrowing determination process for having the borrowing user determine, via the agricultural machinery sharing support system, the borrowing agricultural machinery that is actually to be borrowed, and the borrowing date and time that is the date and time at which the borrowing agricultural machinery will actually be borrowed; and an evaluation material data acquisition process for the agricultural machinery sharing support system to acquire evaluation material data that is data used to evaluate whether the suitability prediction data output by the first trained model was appropriate, and the first trained model can be configured to be generated by machine learning using the suitability prediction data, the data used to output the suitability prediction data, and the evaluation material data as training data.

[0036] As mentioned above, the suitability prediction is a prediction of whether the provisional borrowing date and time is a suitable date and time for "borrowing the provisionally borrowed agricultural machinery," or in other words, it can be said to be an indirect prediction of whether the provisional borrowing date and time is a suitable date and time for "agricultural work to be carried out using the provisionally borrowed agricultural machinery."

[0037] In other words, according to the above configuration, the agricultural machinery sharing support method can display a suitability prediction and suggest to the borrowing user an "appropriate farm work schedule," which is useful information, thereby promoting an increase in the number of borrowing users.

[0038] Furthermore, it is expected to have a particularly excellent effect of encouraging use by people who do not normally do farm work (people other than agricultural workers), thereby expanding the user base. For people who do not normally do farm work and have little knowledge about farm work, it is very difficult to decide when is the best time to do farm work, but by using the appropriate farm work schedule proposed by the agricultural machinery sharing support method configured as described above, they can do farm work without having to make that decision themselves. [Brief description of the drawings]

[0039] [Figure 1] 1 is a system configuration diagram showing a configuration of a scheduling system according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a functional block diagram showing functions of a scheduling system according to an embodiment of the present invention. [Diagram 3] 13 is a schematic diagram of an operation screen displayed by a borrowing request designation unit. FIG. [Figure 4] 13 is a schematic diagram of an operation screen displayed by a borrowing request designation unit. FIG. [Diagram 5] 13 is a schematic diagram of an operation screen displayed by a suitability prediction display unit. FIG. [Figure 6] 13 is a schematic diagram of an operation screen displayed by a suitability prediction display unit. FIG. [Figure 7] 1 is a flowchart showing an agricultural machinery sharing support method using the agricultural machinery sharing support system. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0040] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. However, the embodiment shown below is an example for embodying the technical idea of ​​the present invention, and the present invention is not limited to the following. In addition, this specification never specifies the members shown in the claims to the members of the embodiment. In particular, the dimensions, materials, shapes, and relative positions of the components described in the embodiment are not intended to limit the scope of the present invention, and are merely explanatory examples, unless otherwise specified. Note that the size and positional relationship of the members shown in each drawing may be exaggerated to clarify the explanation. Furthermore, in the following explanation, the same name and symbol indicate the same or similar members, and detailed explanation will be omitted as appropriate. Furthermore, each element constituting the present invention may be configured with the same member so that multiple elements are shared by one member, or conversely, the function of one member can be shared and realized by multiple members. Furthermore, the function or processing of each element constituting the present invention can be realized by artificial intelligence (AI) to the extent possible.

[0041] Furthermore, in explaining the present invention, the term "user" is used to mean "a person (including an organization) who uses the agricultural machinery sharing support system, agricultural machinery sharing support program, or agricultural machinery sharing support method to share, or a person (including an organization) who plans to do so" (it is not intended to be interpreted as meaning "a person who has user authority (user account) in the agricultural machinery sharing support system"). For example, if an administrator of an agricultural machinery sharing support system is using the system to rent out agricultural machinery, the administrator of the system is understood to be included in the "user".

[0042] In addition, in explaining the present invention, the term "sharing" is interpreted broadly and, when used with respect to agricultural machinery, generally means sharing agricultural machinery among users. In other words, the term "sharing" is understood to include cases where an administrator unilaterally lends agricultural machinery to a user (such as a rental service).

[0043] In addition, in explaining the present invention, a user who lends out agricultural machinery is referred to as a "lending user" and a user who borrows agricultural machinery is referred to as a "borrowing user", but these expressions do not mean that users are unequivocally classified into "lending users" and "borrowing users". These expressions are merely used for convenience to represent a user who lends out agricultural machinery as a "lending user" and a user who borrows agricultural machinery as a "borrowing user" in one scene when the present invention is used. In other words, a user who is a "lending user" in one scene is not necessarily a "lending user" in another scene, and a user who is a "borrowing user" in one scene is not necessarily a "borrowing user" in another scene.

[0044] In addition, in explaining the present invention, the term "agricultural machinery" is interpreted in a broad sense and includes all machines and implements used in agriculture. Specifically, agricultural machinery includes tractors, combine harvesters, rice planters, grass cutters, fertilizer spreaders, cultivators, herbicide sprayers, agricultural drones, etc. [Agricultural machinery sharing support system 1]

[0045] An agricultural machine sharing support system 1 according to one embodiment of the present invention will be described with reference to Figs. 1 to 6.

[0046] The agricultural machinery sharing support system 1 is a system that supports the sharing of agricultural machinery between users US, and is composed of a network 11, a plurality of user terminals 12, and a server 13, as shown in FIG. [Network 11]

[0047] As shown in FIG. 1, the network 11 is a communication system for connecting a plurality of user terminals 12 and a server 13 so that they can communicate with each other, and the form of the network is not particularly limited. [User terminal 12]

[0048] 1, the user terminal 12 is an information terminal used by the user US, and corresponds to, for example, a personal computer, a tablet, a smartphone, etc. The user terminal 12 does not mean a specific information terminal, and when the user US uses multiple information terminals, all of them correspond to the user terminal 12.

[0049] The user terminal 12 is also communicatively connected to a server 13 via the network 11, and transmits instructions from the user US to the server 13, receives information from the server 13, and so on. [Server 13]

[0050] The server 13 is a server managed by an administrator AD, as shown in FIG. 1. The server 13 is communicably connected to a user terminal 12 via a network 11, and performs processing to support sharing of agricultural machinery between users US based on instructions from the users US transmitted via the user terminal 12. The support may be provided via a web browser or a dedicated application, and the method is not particularly limited. The program for realizing the "processing" referred to here corresponds to an example of an "agricultural machinery sharing support program" in the claims.

[0051] As shown in FIG. 2, the server 13 further includes a borrowing request designation unit 13a, a first trained model 13b, a second trained model 13c, a third trained model 13d, a suitability prediction display unit 13e, a borrowing decision unit 13f, an evaluation material data acquisition unit 13g, an incentive granting unit 13h, a model generation unit 13i, and a memory unit 13j. <Borrowing request specification section 13a>

[0052] As shown in Figure 2, the borrowing request designation unit 13a is a component that allows a user US who borrows agricultural machinery (hereinafter referred to as the "borrowing user SU") to designate the agricultural machinery they wish to borrow (hereinafter referred to as the "temporary borrowed agricultural machinery KN") and the date and time they wish to borrow the temporary borrowed agricultural machinery KN (hereinafter referred to as the "temporary borrowing date and time KT"), and transmits data related to the specified temporary borrowed agricultural machinery KN and the temporary borrowing date and time KT to the first trained model 13b.

[0053] Specifically, the borrowing request designation unit 13a displays a calendar-like operation screen as shown in Fig. 3 on the user terminal 12 used by the borrowing user SU, and allows the borrowing user SU to designate a date. For example, if the user terminal 12 is a personal computer, the date can be designated by clicking on the date, and if the user terminal 12 is a tablet or smartphone, the date can be designated by tapping on the date.

[0054] Thereafter, the borrowing request designation unit 13a displays an operation screen for designating a time, and allows the borrowing user SU to designate the time. The form of the operation screen and the method of designating the time are not particularly limited, but can be realized, for example, by using a general graphical user interface (GUI) used for designating a time.

[0055] The borrowing request designation unit 13a allows the borrowing user SU to designate one provisional borrowing date and time KT by the above operations, and allows the borrowing user SU to designate multiple provisional borrowing dates and times KT by repeating the above operations. If the borrowing of the agricultural machinery is limited to borrowing on a daily basis, the designation of the time can be omitted (designating only the date will designate the provisional borrowing date and time KT).

[0056] Furthermore, when the borrowing user SU specifies the provisional borrowing date and time KT, the borrowing request specification unit 13a displays an operation screen having a list of agricultural implements as shown in Fig. 4 (a schematic diagram showing the case where "Monday, MM, 24th, 20YY, 10:00 to 18:00 is specified as the provisional borrowing date and time KT") on the user terminal 12 used by the borrowing user SU, and allows the borrowing user SU to specify the provisionally borrowed agricultural implement KN from the list. For example, if the user terminal 12 is a personal computer, the date can be specified by clicking on the name of the agricultural implement, and if it is a tablet or smartphone, the date can be specified by tapping on the name of the agricultural implement.

[0057] In addition, the borrowing request designation unit 13a only needs to have a function to allow the borrowing user SU to designate the temporarily borrowed agricultural machinery KN and the temporary borrowing date and time KT, and the manner of realizing the function, the manner of the operation screen, etc. are not limited to those described above. For example, the temporarily borrowed agricultural machinery KN may be designated first, and then the temporary borrowing date and time KT may be designated. In addition, the list of the temporarily borrowed agricultural machinery KN displayed when the borrowing user SU designates the temporarily borrowed agricultural machinery KN may display all of the agricultural machinery that can be borrowed in the agricultural machinery sharing support system 1, or may display only the agricultural machinery that can be borrowed at the temporary borrowing date and time KT designated by the borrowing user SU. In addition, in the list of agricultural machinery in FIG. 4, the names of the agricultural machinery are displayed in the form of characters, but this is not limited. For example, even if the agricultural machinery is of the same type, the manufacturer and model number may be written next to the name of the agricultural machinery, and the agricultural machinery may be displayed separately for each manufacturer and model number, or the name of the lending user may be written next to the name of the agricultural machinery, and the agricultural machinery may be displayed separately for each lending user. In these cases, the model generation unit 13i described later can include information data on the manufacturer, model number, and lending user in the temporarily borrowed agricultural machinery KN and the borrowed agricultural machinery SN as the learning data GD, and it is also possible to generate the first trained model 13b, the second trained model 13c, and the third trained model 13d based on these data. In addition, instead of or in addition to the display in the form of characters, an image (photograph) of the agricultural machinery may be displayed. In addition, a designation other than the temporarily borrowed agricultural machinery KN and the temporary borrowing date and time KT, for example, a designation of the place where the temporarily borrowed agricultural machinery KN is used may be configured. In addition, the function may be realized by applying a means for designating a car and its borrowing date and time, or an auxiliary means thereof, which is used in an existing system for supporting car sharing. First trained model 13b

[0058] The first trained model 13b is a trained model that uses weather forecast data KY (which may include all data predicting weather-related matters such as weather, temperature, humidity, precipitation, wind direction, and wind volume) for a specific date and time to output data related to a prediction as to whether the specific date and time is suitable for borrowing a specific agricultural tool (hereinafter referred to as "suitability prediction data YD"). In other words, it can be said to be a trained model that indirectly outputs prediction data as to whether a specific date and time is suitable for agricultural work performed using a specific agricultural tool.

[0059] Specifically, as shown in FIG. 2, the first trained model 13b receives data relating to the temporarily borrowed agricultural machinery KN and the temporary borrowing date and time KT designated by the borrowing user SU from the borrowing request designation unit 13a, and receives weather forecast data KY for the temporary borrowing date and time KT from the database of the Japan Meteorological Agency. The first trained model 13b then uses the received data to output suitability prediction data YD relating to the temporarily borrowed agricultural machinery KN and the temporary borrowing date and time KT. The output suitability prediction data YD is transmitted to the suitability prediction display unit 13e and is used for displaying to the borrowing user SU a prediction of whether or not the temporary borrowing date and time KT is a suitable date and time for borrowing the temporarily borrowed agricultural machinery KN (hereinafter referred to as the "suitability prediction TY").

[0060] In addition, as shown in Figure 2, the first trained model 13b transmits the output suitability prediction data YD, data related to the temporarily borrowed agricultural machinery KN and the temporarily borrowed date and time KT used for the output, and the weather forecast data KY to the memory unit 13j, and stores and accumulates them as data necessary to regenerate the first trained model 13b, the second trained model 13c, and the third trained model 13d.

[0061] The first trained model 13b is generated by the model generation unit 13i through machine learning using previously output suitability prediction data YD, data used to output the suitability prediction data YD, and evaluation material data ZD (details will be described in <Evaluation material data acquisition unit 13g>) as training data GD. That is, the first trained model 13b is regenerated (re-trained) using the above-mentioned data accumulated by sharing agricultural machinery via the agricultural machinery sharing support system 1, and can continue to output highly accurate suitability prediction data YD.

[0062] Many technologies have been developed for outputting optimal agricultural work schedules using artificial intelligence, such as that disclosed in JP 2019-128741 A (hereinafter referred to as "Patent Document 2"). Such technologies improve accuracy by subdividing the prediction range in terms of area ("micro-plot units" in Patent Document 2) and using a large amount of learning data for each prediction range, but the problem is that it is necessary to collect a large amount of learning data (in particular, since predictions of the suitability of agricultural work are easily affected by differences in the environment (for example, the climate and geology differ depending on the area where agricultural work is performed), it can be said that in order to generate a highly accurate trained model, it is necessary to collect a large amount of learning data for each environment).

[0063] The suitability prediction by the first trained model 13b in the present invention can also be considered as a type of technology that uses artificial intelligence to output an optimal agricultural work schedule. However, in the present invention, since a part or all of the evaluation material data ZD can be acquired from the borrowing user SU, a large amount of learning data GD can be easily collected. In addition, since the evaluation material data ZD can be acquired from the borrowing user SU, the learning data GD in the environment of the borrowing user SU can be naturally collected in the process of operating a service that supports sharing of agricultural machinery. Note that the incentive granting unit 13h described later can grant an incentive to the borrowing user SU who provided the evaluation material data ZD, so that the evaluation material data ZD can be collected more efficiently.

[0064] With the above-described configuration, the administrator AD can use the borrowing user SU to accomplish some or all of the collection of evaluation material data ZD, which was previously difficult, and the borrowing user SU can obtain a more accurate suitability prediction TY. 《Aspect in which the first trained model 13b uses data related to agricultural work performed in the past by the borrowing user SU》

[0065] The first trained model 13b can also be configured to output the suitability prediction data YD using data related to farm work performed in the past by the borrowing user SU. The method of acquiring data related to farm work performed in the past by the borrowing user SU is not particularly limited, and may be provided by the borrowing user SU via the operation of the user terminal 12, or may be acquired from a system, service, etc. used by the borrowing user SU that records farm work performed in the past. In addition, data related to the borrowed farm equipment SN that the borrowing user SU borrowed in the past via the farm equipment sharing support system 1 can be stored in the storage unit 13j, and the data related to the borrowed farm equipment SN can be used as data related to farm work performed in the past.

[0066] Since the first trained model 13b is generated by machine learning using the data used to output the suitability prediction data YD as the training data GD, the configuration can be said to be capable of using data related to farm work performed in the past by the borrowing user SU as the training data GD related to the generation of the first trained model 13b. In other words, since the first trained model 13b in the above configuration is configured by learning data related to farm work performed in the past and outputs the suitability prediction data YD using data related to farm work performed in the past, the suitability prediction data YD can also take into account the influence of farm work performed in the past by the borrowing user SU.

[0067] For example, assume that the first trained model 13b has been generated by learning that "crops will grow well if fertilizer is spread two weeks after sowing." This first trained model 13b is considered to output suitability prediction data YD such that the suitability prediction TY of the agricultural equipment used to spread fertilizer is good (for example, a large value when the suitability prediction TY is displayed as a percentage) at a date and time before or after two weeks counting from the date and time of sowing seeds performed in the past by the borrowing user SU.

[0068] Also, for example, it is assumed that the first trained model 13b has been generated by learning that "if the cumulative precipitation after spraying the pesticide exceeds 240 mm, it is efficient to spray the pesticide again." This first trained model 13b is considered to output suitability prediction data YD such that the suitability prediction TY of the agricultural machinery used to spray the pesticide is good (prediction that it is a "suitable date and time") on or after the date on which the cumulative precipitation is predicted to exceed 240 mm, calculated from the first pesticide spraying performed in the past by the borrowing user SU.

[0069] The above example merely describes the manner in which the first trained model 13b uses data related to farm work performed by the borrowing user SU in the past, and describes only the portion in which the suitability prediction data YD is affected by the data related to the farm work. In reality, the suitability prediction data YD is also affected by data other than the data related to the farm work, so the condition in the above example in which the suitability prediction TY is good is not necessarily the best condition. 《Aspect of the first trained model 13b using user data UD》

[0070] The first trained model 13b can also be configured to output suitability prediction data using the user data UD of the borrowing user SU.

[0071] "User data UD" is data or a data set provided for each user US, and includes at least data for identifying the user US (e.g., an identification number). However, in the embodiment described here, the user data UD further includes data related to information of other users US, and the suitability prediction data YD is output using the data. Specifically, examples of data related to information of other users US include the height, farming history, age, sex, shipping form of agricultural products, and sales scale of the user US.

[0072] The method of acquiring data relating to other information of user US is not particularly limited, and for example, the data may be provided by the borrowing user SU through operation of the user terminal 12 when registering information relating to user US (user registration) which is performed in the preparation process ST01 described below.

[0073] Since the first trained model 13b is generated by machine learning using the data used to output the suitability prediction data YD as the training data GD, the above configuration can be said to be capable of using the user data UD as the training data GD related to the generation of the first trained model 13b. In other words, since the first trained model 13b in the above configuration is formed by training the user data UD and outputs the suitability prediction data YD using the user data UD as well, the suitability prediction data YD can be made to take into account the accuracy and suitability of the evaluation material data ZD provided by the user US, taking into account the influence of differences (individual differences) between borrowing users SU.

[0074] For example, when using height data of the user US, the first trained model 13b is thought to output suitability prediction data YD that takes into account whether the temporarily borrowed agricultural machinery KN is suitable for the user US from the perspective of the size of the temporarily borrowed agricultural machinery KN.

[0075] Furthermore, for example, when data on the gender of the user US is used, the first learned model 13b is thought to output suitability prediction data YD that takes into account whether the temporarily borrowed agricultural machinery KN is suitable for the physique of the user US from the standpoint of operability of the temporarily borrowed agricultural machinery KN (whether force is required, whether delicate operation is required, etc.).

[0076] In addition, when the user data UD is used, the first trained model 13b judges the accuracy and suitability of the evaluation material data ZD provided by the user US in the past, adjusts the weighting of the evaluation material data ZD, and then outputs the suitability prediction data YD. Specifically, for example, if the user US who provided the evaluation material data ZD has a long agricultural career, is older, or has a large sales volume, it is considered that the accuracy of the evaluation material data ZD is high, and the weighting of the evaluation material data ZD is increased before outputting the suitability prediction data YD. In addition, it is considered that the evaluation material data ZD provided by the user US who has similar user data UD (especially height, age, sex, and shipping form) to the borrowing user SU is high in suitability, and the weighting of the evaluation material data ZD is increased before outputting the suitability prediction data YD. 《An embodiment in which the first trained model 13b uses data related to the location where the agricultural machinery is used》

[0077] The first trained model 13b can also be configured to output the suitability prediction data using data related to the location where the borrowing user SU plans to use the temporarily borrowed agricultural machinery KN. The method of acquiring the data related to the location where the agricultural machinery will be used is not particularly limited, and for example, the data may be provided by the borrowing user SU via the operation of the user terminal 12, or the location information of the user terminal 12 may be acquired.

[0078] Since the first trained model 13b is generated by machine learning using the data used to output the suitability prediction data YD as the training data GD, the configuration can be said to be capable of using data relating to the location where the borrowing user SU plans to use the temporarily borrowed agricultural machinery KN when the temporarily borrowed agricultural machinery KN is actually borrowed (in other words, data relating to the location where the borrowed agricultural machinery SN is used) as the training data GD for generating the first trained model 13b. In other words, the first trained model 13b in the above configuration learns data relating to the location where the agricultural machinery is used, and outputs the suitability prediction data YD using data relating to the location where the agricultural machinery is planned to be used, so that the suitability prediction data YD can also take into account the influence of the location where the agricultural machinery is used. Since the suitability prediction of agricultural work is easily influenced by differences in the environment (for example, the climate and geology differ depending on the area where the agricultural work is performed), if the influence of the location where the agricultural machinery is used can be taken into account, a trained model with high accuracy can be generated.

[0079] The first trained model 13b may be modified as appropriate without being limited to the above-mentioned embodiment. Furthermore, the learning data for generating the first trained model 13b and the data used by the first trained model 13b to output the suitability prediction data YD are not limited to the above-mentioned data and may be modified as appropriate. <Second trained model 13c>

[0080] The second trained model 13c is a trained model that outputs data (hereinafter referred to as "second correction data HD2") for correcting the suitability prediction data YD based on the preferences of each borrowing user SU.

[0081] 2, the second trained model 13c outputs second correction data HD2 related to the borrowing user SU (user data UD) by using the user data UD of the borrowing user SU received from the user terminal 12. The output second correction data HD2 is transmitted to the suitability prediction display unit 13e and is used for the purpose of reflecting the preferences of the borrowing user SU in the suitability prediction TY.

[0082] In addition, as shown in Figure 2, the second trained model 13c transmits the output second correction data HD2 and the user data UD related to the second correction data HD2 to the memory unit 13j, and stores and accumulates them as data necessary to regenerate the first trained model 13b, the second trained model 13c, and the third trained model 13d.

[0083] In addition, the second trained model 13c is generated or regenerated by the model generation unit 13i by machine learning the suitability prediction data YD output in the past, the data used to output the suitability prediction data YD, and the evaluation material data ZD obtained from the borrowing user SU as training data GD.

[0084] In other words, when a specific borrowing user SU provides evaluation material data ZD, the evaluation material data ZD can be used to predict the preferences of the borrowing user SU, and the next time the borrowing user SU borrows agricultural machinery, the evaluation material data ZD can be used to display a suitability prediction TY that reflects the preferences of the borrowing user SU. Note that the "preference" here is, for example, "I prefer cool days even if the weather is somewhat bad," and features can be extracted from the evaluation material data ZD by using, for example, deep learning as a machine learning algorithm.

[0085] With the above-described configuration, the suitability prediction TY can reflect the preferences of the borrowing user SU, so that the suitability prediction TY becomes more useful information for the borrowing user SU, and can further contribute to promoting an increase in the number of borrowing users SU.

[0086] The second trained model 13c is not limited to the above-mentioned embodiment and may be modified as appropriate. Also, the second trained model 13c is not necessarily required to be provided. <Third trained model 13d>

[0087] The third learned model 13d is a learned model that outputs data (hereinafter referred to as "third correction data HD3") for correcting the suitability prediction data YD based on the preferences of each borrowing user SU when the borrowing user SU determines the provisional borrowing date and time KT from the suitability prediction list displayed by the suitability prediction display unit 13e (details are explained in "Suitability prediction display unit 13e").

[0088] 2, the third trained model 13d outputs third correction data HD3 related to the borrowing user SU (user data UD) by using the user data UD of the borrowing user SU received from the user terminal 12. The output third correction data HD3 is transmitted to the suitability prediction display unit 13e and is used for the purpose of reflecting the preferences of the borrowing user SU in the suitability prediction TY.

[0089] In addition, as shown in Figure 2, the third trained model 13d transmits the output third correction data HD3 and the user data UD related to the third correction data HD3 to the memory unit 13j, and stores and accumulates them as data necessary to regenerate the first trained model 13b, the second trained model 13c, and the third trained model 13d.

[0090] The third correction data HD3 is generated or regenerated by the model generation unit 13i by machine learning the suitability prediction data YD output in the past as the learning data GD, the data used to output the suitability prediction data YD, and the determination of the borrowing user SU by the borrowing determination unit 13f. Note that the "determination of the borrowing user SU" here refers to data that indicates which temporary borrowing date and time KT has been determined from among multiple temporary borrowing dates and times KT (suitability prediction list), and corresponds to, for example, data that compiles data related to multiple temporary borrowing dates and times KT as options and data related to the borrowing date and time ST (the selected temporary borrowing date and time KT).

[0091] In other words, the third learned model 13d is a component that, when a specific borrowing user SU provides evaluation material data ZD, makes it possible to predict the preferences of the borrowing user SU from the evaluation material data ZD, and displays a suitability prediction TY that reflects the preferences of the borrowing user SU the next time the borrowing user SU borrows agricultural machinery.

[0092] The second trained model 13c outputs the second correction data HD2 based on the absolute evaluation (preference) of the borrowing user SU, whereas the third trained model 13d outputs the third correction data HD3 based on the relative evaluation (preference) of the borrowing user SU. In other words, the third trained model 13d is a component that can learn and reflect points such as which tentative borrowing date and time KT the borrowing user SU will specify when there are multiple tentative borrowing dates and times as candidates for the borrowing date and time.

[0093] With the above-described configuration, the suitability prediction TY can reflect the preferences of the borrowing user SU, so that the suitability prediction TY becomes more useful information for the borrowing user SU, and can further contribute to promoting an increase in the number of borrowing users SU.

[0094] The form of the third trained model 13d is not limited to the above and can be changed as appropriate. Also, the third trained model 13d does not necessarily need to be provided. <Suitability prediction display section 13e>

[0095] As shown in FIG. 2, the suitability prediction display unit 13e is a component that receives suitability prediction data YD from the first trained model 13b and displays the suitability prediction TY on the user terminal 12 based on the suitability prediction data YD.

[0096] In addition, when the agricultural machinery sharing support system 1 has the second learned model 13c, the suitability prediction display unit 13e receives the second correction data HD2 from the second learned model 13c, and displays the suitability prediction TY on the user terminal 12 based on the suitability prediction data YD corrected by the second correction data HD2. Similarly, when the agricultural machinery sharing support system 1 has the third learned model 13d, the suitability prediction display unit 13e receives the third correction data HD3 from the third learned model 13d, and displays the suitability prediction TY on the user terminal 12 based on the suitability prediction data YD corrected by the third correction data HD3.

[0097] Specifically, when a borrowing user SU specifies a provisional borrowing date and time KT, the suitability prediction display unit 13e displays an operation screen such as that shown in Figure 5 on the user terminal 12, and displays the specified provisional borrowing agricultural machinery KN and provisional borrowing date and time KT, as well as a suitability prediction TY based on the suitability prediction data YD relating to the provisional borrowing agricultural machinery KN and the provisional borrowing date and time KT.

[0098] In addition, when the borrowing user SU specifies multiple provisional borrowing dates and times KT, the suitability prediction display unit 13e displays an operation screen such as that shown in Figure 6 on the user terminal 12, and displays a list of the specified provisional borrowing agricultural machinery KN, the specified multiple provisional borrowing dates and times KT, and the suitability prediction TY for each provisional borrowing date and time KT (corresponding to an example of a ``suitability prediction list'' in the claims).

[0099] In addition, as shown in Figures 5 and 6, the suitability prediction TY is displayed in a percentage form consisting of three evaluation items (the closer to 100%, the more suitable the provisional borrowing date and time KT is for borrowing the provisional borrowed agricultural machinery KN), of which evaluation items A and B are output by a mathematical model, and evaluation item C is output based on the suitability prediction data YD.

[0100] In addition, the suitability prediction is a prediction of whether the provisional borrowing date and time is suitable for "borrowing the provisionally borrowed agricultural machinery", or in other words, it can be said to be an indirect prediction of whether the provisional borrowing date and time is suitable for "farm work performed using the provisionally borrowed agricultural machinery". In other words, according to the above configuration, the agricultural machinery sharing support system can suggest to the borrowing user "an appropriate farm work schedule", which is useful information, by displaying the suitability prediction, thereby promoting an increase in the number of borrowing users.

[0101] Furthermore, it is expected to have a particularly excellent effect of encouraging use by people who do not normally do farm work (people other than agricultural workers), thereby expanding the user base. For people who do not normally do farm work and have little knowledge about farm work, it is very difficult to decide when is the best time to do farm work, but by using the appropriate farm work schedule proposed by the agricultural machinery sharing support system configured as described above, they can do farm work without having to make that decision themselves.

[0102] In this way, the agricultural machinery sharing support system 1 can provide the borrowing users SU with the suitability predictions TY, which are useful information for the borrowing users SU, and can therefore promote an increase in the number of borrowing users SU.

[0103] In addition, the suitability prediction TY is displayed in the form of a pie chart on the operation screen of Figure 5 and in the form of a bar graph on the operation screen of Figure 6, so that even if the borrowing user SU has little knowledge about agricultural work, he or she can intuitively grasp whether the provisional borrowing date and time KT is a suitable date and time for borrowing the provisional borrowing agricultural machinery KN.

[0104] The form of the suitability prediction display section 13e is not limited to that described above, and the form of the operation screen can be changed as appropriate.

[0105] Specifically, the suitability prediction display unit 13e may display the suitability prediction TY in a manner other than a percentage, for example, a sentence explaining the suitability of the provisional borrowing date KT may be displayed, and the suitability prediction TY may be displayed in the form of a sentence. Furthermore, the display manner of the suitability prediction TY may be any manner that allows the borrowing user SU to understand whether the provisional borrowing date KT is a suitable date and time for borrowing the provisional borrowing agricultural implement KN, and may be a manner using figures or colors other than a percentage or sentences.

[0106] In the above description, the suitability prediction TY consists of three evaluation items, one of which is output based on the suitability prediction data YD, but this is not limited to this. For example, the evaluation items may be three or more or three or less, and the method of outputting the evaluation items is not particularly limited (all may be output by a trained model (AI)). <Borrowing determination section 13f>

[0107] The borrowing determination unit 13f is a component that allows the borrowing user SU to determine the agricultural machinery to actually borrow (hereinafter referred to as the "borrowed agricultural machinery SN") and the date and time when the borrowed agricultural machinery SN will actually be borrowed (hereinafter referred to as the "borrowing date and time ST").

[0108] Specifically, when the borrowing user SU specifies a temporary borrowing date and time KT, the borrowing user SU can press the decision button KB shown on the operation screen (Figure 5) displayed on the user terminal 12 to determine the specified temporary borrowing agricultural machinery KN as the borrowed agricultural machinery SN, and the specified temporary borrowing date and time KT as the borrowing date and time ST.

[0109] In addition, if the borrowing user SU specifies multiple provisional borrowing dates and times KT, the user can select one provisional borrowing date and time KT from the suitability prediction list on the operation screen (Figure 6) displayed on the user terminal 12 and then press the decision button KB to determine the specified provisional borrowing agricultural machinery KN as the borrowed agricultural machinery SN and to determine the selected provisional borrowing date and time KT as the borrowing date and time ST.

[0110] With the above-mentioned configuration, the agricultural machinery sharing support system 1 allows the borrowing user SU to determine the borrowed agricultural machinery SN and borrowing date and time ST while being able to visually confirm the suitability prediction TY, so the borrowing user SU can determine the borrowed agricultural machinery SN and borrowing date and time ST after confirming the date and time suitable for borrowing the agricultural machinery, making it easier to determine the borrowed agricultural machinery SN and borrowing date and time ST. In addition, when the borrowing user SU specifies multiple provisional borrowing dates and times KT, the borrowed agricultural machinery SN and borrowing date and time ST can be determined while comparing the suitability prediction TY for each of the multiple provisional borrowing dates and times KT using the suitability prediction list, making it easier to determine the borrowed agricultural machinery SN and borrowing date and time ST.

[0111] In addition, when the borrowing user SU specifies multiple provisional borrowing dates and times KT, the borrowing determination unit 13f transmits data relating to the determined borrowed agricultural machinery SN and borrowing date and time ST to the memory unit 13j, as shown in Figure 2, and stores and accumulates it as data necessary to regenerate the third learned model 13d.

[0112] The mode of the borrowing determining unit 13f is not limited to the above-mentioned one, and can be changed as appropriate. <Evaluation material data acquisition unit 13g>

[0113] The evaluation material data acquisition unit 13g is a component that acquires evaluation material data ZD, which is data used to evaluate whether the suitability prediction data YD output by the first trained model 13b is appropriate or not.

[0114] Specifically, the evaluation material data acquisition unit 13g displays a page for inputting the evaluation material data ZD on the user terminal 12, and as shown in Figure 2, has the user US provide the evaluation material data ZD, and acquires the evaluation material data ZD from outside the agricultural machinery sharing support system 1 via the network 11.

[0115] In addition, as shown in Figure 2, the evaluation material data acquisition unit 13g transmits the evaluation material data ZD to the memory unit 13j, and stores and accumulates it as data necessary to regenerate the first trained model 13b, the second trained model 13c, and the third trained model 13d.

[0116] In addition, the evaluation material data ZD acquired by the evaluation material data acquisition unit 13g is not particularly limited as long as it is data that can be used to evaluate whether the suitability prediction data YD was appropriate or not. An example of the suitability prediction data YD is shown below. <Past Weather Data KD>

[0117] As the evaluation material data ZD, past weather data KD can be used.

[0118] Here, "weather data KD" may include all data related to weather, such as weather, temperature, humidity, precipitation, wind direction, wind volume, etc. Also, "past" here means data before the evaluation material data acquisition unit 13g acquires the evaluation material data ZD. In other words, "past weather data KD" means data related to weather before the evaluation material data acquisition unit 13g acquires the evaluation material data ZD, and corresponds to, for example, data on the weather at the borrowing date and time ST and data on the weather before that.

[0119] The method of acquiring the past weather data KD is not particularly limited, and may be, for example, provided by the user US (which may be a user US other than the borrowing user SU). Note that "provision from the user US" here is not limited to the case where the user US himself measures various data and provides it, but also includes the case where the user US provides data to the agricultural machinery sharing support system 1 via a system, service, etc. that he uses (for example, "MIHARASU (registered trademark of Nishimu Electronics Co., Ltd." is known as a system that acquires weather data KD for agriculture).) This is also understood to be included in "provision from the user US".

[0120] Furthermore, the past weather data KD may be acquired from outside the agricultural machinery sharing support system 1 (for example, the Japan Meteorological Agency) via the network 11, without going through the user US. Data relating to evaluation by borrowing user SU

[0121] As the evaluation material data ZD, data relating to the evaluation by the borrowing user SU (hereinafter, referred to as "user evaluation data UH") can be used.

[0122] The user evaluation data UH is, for example, an evaluation by the borrowing user SU on whether or not the borrowing date and time ST was suitable for borrowing the borrowed agricultural equipment SN. As a method of providing the evaluation, for example, after the borrowing date and time ST, the borrowing user SU can be induced to provide an evaluation by displaying a page on the user terminal 12 for providing an answer to whether or not the borrowing date and time ST was suitable for borrowing the borrowed agricultural equipment SN. In addition, the form of the evaluation and the form of the page are not particularly limited, but for example, the borrowing user SU may be prompted to give scores to items such as "progress of farm work", "suitability of temperature", and "suitability of weather", or to input a comment in text. In addition, a form used in an existing questionnaire system may be applied.

[0123] Furthermore, the user evaluation data UH may be, for example, evaluations by the borrowing user SU on the borrowed agricultural machinery SN provided by the borrowing user SU. That is, even for the same borrowed agricultural machinery SN, if there are differences in the evaluations by the borrowing user SU due to differences in weather, for example, it is possible to improve the accuracy of the suitability prediction TY by learning the feature amount by machine learning. The method of provision, the form of evaluation, and the form of the page are the same as in the case of the evaluation by the borrowing user SU on "whether the borrowing date and time ST was a suitable date and time for borrowing the borrowed agricultural machinery SN."

[0124] The aspect of the evaluation material data acquisition unit 13g is not limited to the above and can be changed as appropriate. The evaluation material data ZD acquired by the evaluation material data acquisition unit 13g is not limited to the above, and for example, data related to the results of agricultural work such as harvest volume may be acquired as the evaluation material data ZD. <Incentive Granting Section 13h>

[0125] The incentive granting unit 13h is a component that grants an incentive to the borrowing user SU who has provided the evaluation material data ZD.

[0126] Specifically, the incentive granting unit 13h determines the incentive to be granted, and transmits data relating to the incentive (hereinafter referred to as "incentive data ID") to the user terminal 12, as shown in Fig. 2. Furthermore, if time is required between when the incentive granting unit 13h determines the incentive to be granted and when the incentive is granted, the incentive data ID is associated with the user data UD of the user US transmitting the incentive data ID, and transmitted to the storage unit 13j, where it is stored and accumulated.

[0127] The "incentive" here refers to a reward for the borrowing user SU, such as a discount on the fee, the provision of a coupon, or the granting of points that can be used within the agricultural machinery sharing support system 1. It is also understood that the incentive includes so-called "negative incentives" that create a relative benefit by imposing some kind of penalty (such as suspension of use of the agricultural machinery sharing support system 1) for not providing the evaluation material data ZD. The "incentive data ID" here refers to the aforementioned incentive in digital form, and the user US can receive the incentive within the agricultural machinery sharing support system 1 based on the incentive data ID.

[0128] The form of the incentive granting unit 13h is not limited to the above and can be changed as appropriate. For example, when the incentive is granted in the form of an item rather than data, the incentive granting unit 13h may perform a process for sending the item to the user US. In this case, the incentive granting unit 13h does not necessarily need to transmit the incentive data ID to the user terminal 12. <Model generation unit 13i>

[0129] As shown in FIG. 2, the model generation unit 13i is a component that generates a first trained model 13b, a second trained model 13c, and a third trained model 13d using the training data GD received from the memory unit 13j.

[0130] As the learning data GD, the first trained model 13b is generated using the suitability prediction data YD, the data used to output the suitability prediction data YD, and the evaluation material data ZD obtained from the borrowing user SU, the second trained model 13c is generated using the suitability prediction data YD, the data used to output the suitability prediction data YD, and the evaluation material data ZD obtained from the borrowing user SU, and the third trained model 13d is generated using the suitability prediction data YD, the data used to output the suitability prediction data YD, and the determination of the borrowing user SU in the borrowing determination unit 13f.

[0131] In addition, the model generation unit 13i can continue to output highly accurate suitability prediction data YD by regenerating the first trained model 13b, the second trained model 13c, and the third trained model 13d each time a certain amount of the above-mentioned data is accumulated in the memory unit 13j or at regular intervals.

[0132] The learning algorithm used by the model generation unit 13i is not particularly limited, and for example, reinforcement learning, deep learning, neural networks, genetic programming, inductive logic programming, and support vector machines can be used.

[0133] The mode of the model generating unit 13i is not particularly limited and can be changed as appropriate. For example, it is not necessary to use all the data stored and accumulated in the storage unit 13j as data for generating the trained model described above. For example, before the trained model generating step (see [model generating step ST08]), data to be used for generating the trained model may be carefully selected and only a portion of the carefully selected data may be used so that highly accurate suitability prediction data YD can be output. <Storage section 13j>

[0134] The storage unit 13j is a member that temporarily or long-term stores various data, and is realized by various recording media such as a hard disc drive (HDD), a solid state drive (SSD), an SD card, or a flash memory.

[0135] For example, the storage unit 13j stores and accumulates data necessary for the model generation unit 13i to regenerate the first trained model 13b, the second trained model 13c, and the third trained model 13d, as shown in Fig. 2. The data is transmitted to the model generation unit 13i as training data GD every time a certain amount of data is accumulated or at regular intervals, and is used for regenerating the first trained model 13b, the second trained model 13c, and the third trained model 13d.

[0136] The storage unit 13j also stores and accumulates the incentive data ID and the user data UD transmitted by the incentive granting unit 13h in association with each other.

[0137] The mode of the storage unit 13j is not particularly limited, and in addition to the above-mentioned mode, the storage unit 13j may be configured to store and accumulate data used by each member of the agricultural machinery sharing support system 1. The storage unit 13j does not necessarily have to be provided in the server 13, and may be provided in a server different from the server 13, or a storage unit of the user terminal 12 may be used. <Other processing by server 13>

[0138] In addition to the processing using the above-mentioned components, the server 13 can perform processing necessary to support the sharing of agricultural machinery, for example, when the borrowing user SU determines the borrowed agricultural machinery SN and borrowing date and time ST using the borrowing determination unit 13f, it can perform processing to actually lend and borrow the borrowed agricultural machinery SN and support the return of the borrowed agricultural machinery SN.

[0139] The method of these processes is not particularly limited, and for example, a method used in an existing system for supporting car sharing can be applied. [Agricultural machinery sharing support method using agricultural machinery sharing support system 1]

[0140] A method for supporting agricultural machinery sharing using the agricultural machinery sharing support system 1 according to one embodiment of the present invention will be described with reference to the flowchart of FIG.

[0141] The agricultural machinery sharing support method using the agricultural machinery sharing support system 1 is broadly composed of eight steps: a preparation step ST01, a borrowing request specification step ST02, a suitability prediction data output step ST03, a suitability prediction display step ST04, a borrowing decision step ST05, a lending step ST06, an evaluation material data acquisition step ST07, and a model generation step ST08. [Preparation process ST01]

[0142] The preparation process ST01 is a process of making the preparations necessary to use the agricultural machinery sharing support system 1, such as generating the first trained model 13b, the second trained model 13c, and the third trained model 13d, registering information related to the user US (user registration), and registering the agricultural machinery to be shared (agricultural machinery registration).

[0143] The aspect of the preparation step ST01 is not particularly limited and can be changed as appropriate. For example, the same preparation as that performed in an existing system for supporting car sharing may be performed. In addition, if the necessary preparations have been completed, the preparation step ST01 may be omitted. [Borrowing request designation process ST02]

[0144] The borrowing request specification process ST02 is a process in which the borrowing user SU specifies the temporarily borrowed agricultural implement KN and the temporary borrowing date and time KT.

[0145] Specifically, the borrowing user SU specifies one or more temporary borrowing dates and times KT from a calendar-like operation screen (see FIG. 3) displayed on the user terminal 12. After that, the borrowing user SU specifies the temporary borrowing agricultural equipment KN from a list of the temporary borrowing agricultural equipment KN displayed on the user terminal 12 (see FIG. 4).

[0146] The borrowing request specification step ST02 is not limited to the above and can be modified as appropriate. For example, the borrowing request specification step ST02 may be performed in a manner similar to that used when a user specifies a car and a borrowing date and time in an existing car sharing support system. [Property prediction data output process ST03]

[0147] The suitability prediction data output process ST03 is a process in which the first trained model 13b outputs suitability prediction data YD using the provisionally borrowed agricultural machinery KN and provisional borrowing date and time KT specified in the borrowing request designation unit 13a, as well as the weather forecast data KY for the provisional borrowing date and time KT.

[0148] The output suitability prediction data YD is transmitted to the suitability prediction display unit 13e and is used for displaying the suitability prediction TY to the borrowing user SU.

[0149] In addition, when the agricultural machinery sharing support system 1 has the second trained model 13c, in the suitability prediction data output step ST03, the second trained model 13c outputs the second correction data HD2 using the user data UD of the borrowing user SU received from the user terminal 12, and transmits it to the suitability prediction display unit 13e. Similarly, when the agricultural machinery sharing support system 1 has the third trained model 13d, in the suitability prediction data output step ST03, the third trained model 13d outputs the third correction data HD3 using the user data UD of the borrowing user SU received from the user terminal 12, and transmits it to the suitability prediction display unit 13e.

[0150] The mode of the suitability prediction data output step ST03 is not limited to the above, and can be changed as appropriate. [Property prediction display process ST04]

[0151] 5 or 6, the suitability prediction display step ST04 is a step of displaying the suitability prediction TY based on the suitability prediction data YD output by the first trained model 13b on the user terminal 12. This allows the borrowing user SU to consider whether or not to borrow the content specified in the suitability prediction TY borrowing request specification step ST02.

[0152] In addition, the suitability prediction display process ST04 is a process of displaying the suitability prediction TY based on the suitability prediction data YD corrected by the second correction data HD2 if the agricultural machinery sharing support system 1 has the second trained model 13c, and corrected by the third correction data HD3 in the third trained model 13d.

[0153] Through this process, the agricultural machinery sharing support system 1 can provide the borrowing users SU with the suitability prediction TY, which is useful information for the borrowing users SU, thereby promoting an increase in the number of borrowing users SU.

[0154] The mode of the suitability prediction display step ST04 is not limited to the above, and can be changed as appropriate. [Borrowing decision process ST05]

[0155] The borrowing decision process ST05 is a process for having the borrowing user SU decide the borrowed agricultural machinery SN and the borrowing date and time ST. When the borrowing user SU designates one temporary borrowing date and time KT, the borrowing user SU can decide the designated temporary borrowing agricultural machinery KN as the borrowed agricultural machinery SN and decide the selected temporary borrowing date and time KT as the borrowing date and time ST by pressing the decision button KB shown in Fig. 5, or when the borrowing user SU designates multiple temporary borrowing dates and times KT, by selecting one temporary borrowing date and time KT from the list and pressing the decision button KB shown in Fig. 6.

[0156] This allows the borrowing user SU to determine the borrowed agricultural machinery SN and the borrowing date and time ST after checking the suitability prediction TY, making it easy for the borrowing user SU to determine the borrowed agricultural machinery SN and the borrowing date and time ST, allowing the borrowing user SU to borrow the agricultural machinery after checking the appropriate date and time for borrowing the agricultural machinery.

[0157] The mode of the borrowing determination step ST05 is not limited to the above, and can be changed as appropriate. [Lending and Borrowing Process ST06]

[0158] The lending process ST06 is a process in which the agricultural machinery sharing support system 1 actually lends and borrows the borrowed agricultural machinery SN and supports its return.

[0159] The mode of the rental and borrowing process ST06 is not particularly limited. For example, in an existing system for supporting car sharing, a mode for supporting the rental and borrowing of a car from rental to return may be applied. [Evaluation material data acquisition process ST07]

[0160] The evaluation material data acquisition step ST07 is a step of acquiring evaluation material data ZD, which is data used to evaluate whether the suitability prediction data YD output by the first trained model 13b is appropriate or not. The evaluation material data ZD acquired in the evaluation material data acquisition step ST07 is stored and accumulated in the memory unit 13j together with the corresponding suitability prediction data YD and the data used to output the suitability prediction data YD, and is used when the model generation unit 13i generates the first trained model 13b in the model generation step ST08.

[0161] Moreover, the evaluation material data acquisition step ST07 includes a step of, when the evaluation material data ZD is acquired from the borrowing user SU, granting an incentive to the user by the incentive granting unit 13h.

[0162] The aspect of the evaluation material data acquisition step ST07 is not limited to the above-mentioned one, and can be changed as appropriate. [Model generation process ST08]

[0163] In the model generation step ST08, the model generation unit 13i regenerates the first trained model 13b, the second trained model 13c, and the third trained model 13d every time a certain amount of data used to generate the first trained model 13b, the second trained model 13c, and the third trained model 13d is accumulated, or every certain period of time. This step allows the agricultural machinery sharing support system 1 to continue providing the borrowing user SU with a highly accurate suitability prediction TY.

[0164] The model generation step ST08 is not limited to the above and can be modified as appropriate. In order to output highly accurate suitability prediction data YD, the agricultural machinery sharing support method using the agricultural machinery sharing support system 1 according to one embodiment of the present invention may include a step of carefully selecting data to be used in generating the trained model before this step. [Effects of the present invention]

[0165] As described above, the agricultural machinery sharing support system 1 (including an example of the "agricultural machinery sharing support program" in the claims) according to one embodiment of the present invention and the agricultural machinery sharing support method using the same can display a suitability prediction and suggest to the borrowing user an "appropriate farm work schedule," which is useful information, thereby promoting an increase in the number of borrowing users SU. [Explanation of symbols]

[0166] 1. Agricultural machinery sharing support system 11. Network 12...User terminal 13…Server 13a…Borrowing request specification section 13b…First trained model 13c…Second trained model 13d…Third trained model 13e…Suitability prediction display section 13f…Borrowing decision section 13g…Evaluation material data acquisition section 13h…Incentive Granting Department 13i…Model generation section 13j…Storage section AD…Administrator US…User SU: Borrowed user UD...User data KN: Temporary borrowing of agricultural machinery KT: Temporary borrowing date and time KY: Weather forecast data YD: Suitability prediction data HD2: Second correction data HD3…Third correction data TY…suitability prediction GD: Learning data ZD: Evaluation material data KD: Weather data UH: User evaluation data ID…incentive data SN…Borrowed agricultural equipment ST…Borrow date and time KB... Decision button

Claims

1. An agricultural machinery sharing support system for supporting sharing of agricultural machinery between users, A borrowing request designation unit that allows a borrowing user, who is a user who borrows agricultural machinery, among the users to designate a temporary borrowing agricultural machinery that is the agricultural machinery that the user wishes to borrow and a temporary borrowing date and time that is the date and time at which the user wishes to borrow the temporary borrowing agricultural machinery; A first trained model that uses weather forecast data for a specific date and time to output suitability prediction data, which is data related to a prediction of whether the specific date and time is a suitable date and time for borrowing a specific agricultural machine; a suitability prediction display unit that displays to the borrowing user a suitability prediction, which is a prediction of whether the provisional borrowing date and time is a suitable date and time for borrowing the provisional borrowing agricultural equipment specified in the borrowing request specification unit, based on suitability prediction data related to the provisional borrowing date and time specified in the borrowing request specification unit and the provisional borrowing date and time output by the first trained model using weather forecast data for the provisional borrowing date and time specified in the borrowing request specification unit; A borrowing determination unit that allows the borrowing user to determine a borrowed agricultural machine, which is the agricultural machine to be actually borrowed, and a borrowing date and time, which is the date and time when the borrowed agricultural machine will actually be borrowed; an evaluation material data acquisition unit that acquires evaluation material data used to evaluate whether the suitability prediction data output by the first trained model is appropriate; Equipped with The first trained model is an agricultural machinery sharing support system generated by machine learning using the suitability prediction data, the data used to output the suitability prediction data, and the evaluation material data as training data.

2. The agricultural machinery sharing support system according to claim 1, The evaluation material data is Historical weather data, Data relating to an evaluation by the borrowing user as to whether the borrowing date and time was a suitable date and time for borrowing the borrowed agricultural equipment, or Data relating to evaluation of the borrowed agricultural equipment by the borrowing user; An agricultural machinery sharing support system that includes at least one of the above data.

3. The agricultural machinery sharing support system according to claim 1, The first trained model further includes an agricultural machinery sharing support system that outputs suitability prediction data using data related to agricultural work performed in the past by the borrowing user.

4. The agricultural machinery sharing support system according to claim 1, The first trained model further outputs suitability prediction data using user data, which is data including at least one of the borrowing user's height, farming experience, age, gender, shipping form of agricultural products, or sales volume.

5. The agricultural machinery sharing support system according to claim 1, The evaluation material data acquisition unit acquires part or all of the evaluation material data from the borrowing user.

6. The agricultural machinery sharing support system according to claim 5, further comprising: a second trained model that is generated by machine learning using the suitability prediction data, the data used to output the suitability prediction data, and the evaluation material data acquired from the borrowing user as training data, and outputs second correction data that is data for correcting the suitability prediction data based on the preferences of each of the borrowing users; The suitability prediction display unit displays a suitability prediction reflecting the preferences of each borrowing user based on the second correction data.

7. The agricultural machinery sharing support system according to claim 5, further comprising: An agricultural machinery sharing support system comprising an incentive granting unit that grants incentives to borrowing users who provide the evaluation material data.

8. The agricultural machinery sharing support system according to claim 1, The borrowing request designation unit allows the borrowing user to designate a plurality of temporary borrowing dates and times; The first trained model outputs a plurality of suitability prediction data relating to the provisional borrowing agricultural machinery and a plurality of provisional borrowing dates and times specified by the borrowing request specification unit, The suitability prediction display unit displays a suitability prediction list that is a list of the provisional borrowing dates and times and suitability predictions for each of the provisional borrowing dates and times based on the suitability prediction data output by the first trained model, The borrowing decision unit is an agricultural machinery sharing support system that allows the borrowing user to decide the borrowing date and time from the suitability prediction list displayed by the suitability prediction display unit.

9. The agricultural machinery sharing support system according to claim 8, further comprising: a third trained model that is generated by machine learning using the suitability prediction data, the data used to output the suitability prediction data, and the borrowing user's decision in the borrowing decision unit as learning data, and outputs third correction data that is data for correcting the suitability prediction data based on the preferences of each of the borrowing users; The suitability prediction display unit displays a list of suitability predictions reflecting the preferences of each borrowing user based on the third correction data.

10. An agricultural machinery sharing support program for supporting sharing of agricultural machinery between users, A borrowing request designation function that allows a borrowing user, who is a user who borrows agricultural machinery, to designate a temporary borrowing agricultural machinery that is the agricultural machinery that he / she wishes to borrow and a temporary borrowing date and time that is the date and time at which he / she wishes to borrow the temporary borrowing agricultural machinery; a suitability prediction data output function for outputting suitability prediction data, which is data relating to a prediction of whether or not a specific date and time is suitable for borrowing a specific agricultural machine, to the first trained model using weather forecast data for the specific date and time; A suitability prediction display function that displays to the borrowing user a suitability prediction, which is a prediction of whether the provisional borrowing date and time is a suitable date and time for borrowing the provisional borrowing agricultural equipment specified by the borrowing request specification function based on suitability prediction data related to the provisional borrowing date and time specified by the borrowing request specification function, output by the first trained model using weather forecast data for the provisional borrowing date and time specified by the borrowing request specification function; A borrowing determination function that allows the borrowing user to determine the borrowed agricultural machinery that is the agricultural machinery to be actually borrowed and the borrowing date and time that is the date and time when the borrowed agricultural machinery will actually be borrowed; an evaluation material data acquisition function for acquiring evaluation material data used to evaluate whether the suitability prediction data output by the first trained model is appropriate; This is realized on a computer. The first trained model is an agricultural machinery sharing support program generated by machine learning using the suitability prediction data, the data used to output the suitability prediction data, and the evaluation material data as training data.

11. A method for supporting sharing of agricultural machinery between users using an agricultural machinery sharing support system, comprising: A borrowing request designation process in which a borrowing user, who is a user who borrows agricultural machinery, among the users, designates a temporary borrowing agricultural machinery that is the agricultural machinery that he / she wishes to borrow and a temporary borrowing date and time that is the date and time at which he / she wishes to borrow the temporary borrowing agricultural machinery through the agricultural machinery sharing support system; A suitability prediction data output process in which the agricultural machinery sharing support system uses weather forecast data for a specific date and time to output suitability prediction data, which is data related to a prediction of whether the specific date and time is a suitable date and time for borrowing a specific agricultural machinery, to a first trained model; A suitability prediction display process for displaying to the borrowing user a suitability prediction, which is a prediction of whether the provisional borrowing date and time is suitable for borrowing the provisional borrowing agricultural equipment specified by the borrowing request designation function based on suitability prediction data relating to the provisional borrowing date and time specified by the borrowing request designation function, output by the first trained model using weather forecast data for the provisional borrowing date and time specified by the borrowing request designation function; A borrowing determination step for having the borrowing user determine, via the agricultural machinery sharing support system, the borrowing agricultural machinery that is the agricultural machinery to be actually borrowed and the borrowing date and time that is the date and time when the borrowing agricultural machinery will actually be borrowed; An evaluation material data acquisition process in which the agricultural machinery sharing support system acquires evaluation material data which is data used to evaluate whether the suitability prediction data output by the first trained model is appropriate; Including, An agricultural machinery sharing support method in which the first trained model is generated by machine learning using the suitability prediction data, the data used to output the suitability prediction data, and the evaluation material data as training data.

Images