Subscription system

Abstract

A subscription system capable of ensuring necessary food materials at appropriate timing is provided. The subscription system of an embodiment has a presumption unit and a subscription unit. The presumption unit presumes a future inventory state of at least one food material stored in a user's refrigerator or home, based on inventory information indicating the inventory state of the food material and presumption information reflecting the user's past consumption trend with respect to the food material. The subscription unit subscribes to the food material, based on the future inventory state of the food material presumed by the presumption unit and a delivery period required for delivery of the food material.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to an ordering system. BACKGROUND

[0002] Systems have been proposed to manage foodstuffs stored in a refrigerator. In such systems, it is desirable to ensure necessary foodstuffs at an appropriate timing.

[0003] PRIOR ART DOCUMENTS

[0004] PATENT DOCUMENTS

[0005] Patent Document 1: Japanese Patent Application Publication No. 2018-146120 SUMMARY

[0006] PROBLEMS TO BE SOLVED BY THE INVENTION

[0007] An object of the present application is to provide an ordering system capable of ensuring necessary foodstuffs at an appropriate timing.

[0008] MEANS FOR SOLVING THE PROBLEMS

[0009] An ordering system of an embodiment has a prediction unit and an ordering unit. The prediction unit predicts a future inventory state of at least one foodstuff stored in a refrigerator or a house of a user, based on inventory information indicating the inventory state of the foodstuff and prediction information reflecting a past consumption trend of the user with respect to the foodstuff. The ordering unit orders the foodstuff, based on the future inventory state of the foodstuff predicted by the prediction unit and a delivery period required for delivery of the foodstuff.

[0010] EFFECTS OF THE INVENTION

[0011] According to the present application, it is possible to provide an ordering system capable of ensuring necessary foodstuffs at an appropriate timing. BRIEF DESCRIPTION OF DRAWINGS

[0012] Figure 1 is a diagram showing an example of the structure of an ordering system of one embodiment.

[0013] Figure 2 is a diagram showing an example of the structure of a refrigerator of one embodiment.

[0014] Figure 3 is a diagram showing another example of the structure of a refrigerator of one embodiment.

[0015] Figure 4 is a diagram showing an example of the structure of a portable terminal of one embodiment.

[0016] Figure 5 is a diagram showing an example of the structure of a server of one embodiment.

[0017] Figure 6 FIG. 1 is a diagram illustrating an example of information stored in a storage section of one embodiment.

[0018] Figure 7 FIG. 2 is a diagram illustrating an example of a structure of a server of one embodiment.

[0019] Figure 8 FIG. 3 is a diagram illustrating a relationship between a date and time in a past predetermined period and an inventory amount of a food material at the date and time of one embodiment.

[0020] Figure 9 FIG. 4 is a first diagram illustrating an example of teacher data of one embodiment.

[0021] Figure 10 FIG. 5 is a second diagram illustrating an example of teacher data of one embodiment.

[0022] Figure 11 FIG. 6 is a third diagram illustrating an example of teacher data of one embodiment.

[0023] Figure 12 FIG. 7 is a fourth diagram illustrating an example of teacher data of one embodiment.

[0024] Figure 13 FIG. 8 is a fifth diagram illustrating an example of teacher data of one embodiment.

[0025] Figure 14 FIG. 9 is a sixth diagram illustrating an example of teacher data of one embodiment.

[0026] Figure 15 FIG. 10 is a seventh diagram illustrating an example of teacher data of one embodiment.

[0027] Figure 16 FIG. 11 is a diagram illustrating an example of a structure of a server of one embodiment.

[0028] Figure 17 FIG. 12 is a diagram illustrating an example of a processing flow of a subscription system of one embodiment. DETAILED DESCRIPTION

[0029] The following describes an order system of an embodiment with reference to the drawings. In the following description, the same reference numbers are assigned to structures having the same or similar functions. Also, sometimes repeated description of the structures is omitted. "Based on XX" means "at least based on XX" and can include a case where based on other elements in addition to XX. "Based on XX" is not limited to a case where XX is used directly and can include a case where based on an element obtained by performing an operation or processing on XX. "XX or YY" is not limited to a case where either one of XX and YY is selected and can include a case where both of XX and YY are selected. This is the same in a case where the number of choices is three or more. "XX" and "YY" are arbitrary elements (for example, arbitrary information). Furthermore, "detecting" is not limited to a case where a physical quantity of an object is directly sensed and can include a case where another physical quantity associated with the physical quantity of the object is directly or indirectly acquired, the physical quantity of the object is estimated or determined based on the acquired another physical quantity. Furthermore, "acquiring" is not limited to a case where the object itself is directly received and includes a case where the object is obtained by performing an operation or processing on the directly received object. Furthermore, "determining" can include deciding in correspondence with a determination result based on an operation result using a physical quantity of an object. Furthermore, "determining" can include deciding in correspondence with contents represented by output data output at the time of input of input data, as in a learned model described later.

[0030] The following describes several embodiments. The order system of an embodiment is a system that orders food materials in consideration of a stock state of the food materials and a period from ordering the food materials (that is, from ordering) to completion of delivery. In addition, delivery here includes, for example, the entire process from shipping of the food materials to delivery of the food materials to a destination. Furthermore, the period taken for delivery can be obtained from an order point of the food materials when the order point is contacted via a communication network such as a network or a telephone network. For example, in a case where a WEB site of the order point of the food materials is accessed via the network, the date of completion of delivery is displayed on the WEB site. Furthermore, for example, the order point of the food materials can be called to obtain information of the date of completion of delivery of the food materials.

[0031] <Embodiment>

[0032] (Overall structure)

[0033] Figure 1 is a diagram that shows an example of a structure of an order system 1 of an embodiment. The order system 1 has, as shown in Figure 1 , a refrigerator 10, a portable terminal 20, servers 30, 40, and 50.

[0034] Figure 2 is a diagram that shows an example of a structure of the refrigerator 10. The refrigerator 10 has, as shown in Figure 2As shown, it has a photographing section 101, a communication section 102, and a processing section 103. The photographing section 101 photographs foodstuffs. Specifically, the photographing section 101 photographs foodstuffs in the interior of the refrigerator 10 and foodstuffs that are management targets stored in the house. The photographing section 101 is, for example, a camera. The communication section 102 communicates with external devices. As examples of the external devices, there are the portable terminal 20, the server 30, and the like. The processing section 103 determines the kinds and inventory of foodstuffs. For example, the processing section 103 determines the kinds and inventory of foodstuffs by determining whether or not a pattern of an image that contains features of each foodstuff prepared in advance is contained in an image captured by the photographing section 101 using an image matching technique. In addition, the inventory includes weight and quantity (inventory number).

[0035] Figure 3 is a view that shows an example of other structures of the refrigerator 10. The refrigerator 10 can have a reception section 104 such as a touch panel that receives information of the kinds and inventory of foodstuffs, and the user manually inputs information of the kinds and inventory of foodstuffs via the reception section 104, and the processing section 103 determines the kinds and inventory of foodstuffs by determining the input information.

[0036] Figure 4 is a view that shows an example of the structure of the portable terminal 20. The portable terminal 20 has a communication section 201, a processing section 202, and a notification section 203 (an example of a display section, an example of an input section), as shown. Figure 4 The communication section 201 communicates with external devices. As examples of the external devices, there are the refrigerator 10, the server 40, and the like. The processing section 202 acquires information of an order reservation of foodstuffs. In addition, the processing section 202 changes information of the order reservation of foodstuffs (an example of information indicating contents of an order) in correspondence with, for example, an operation by the user to change the information indicating the contents of the order with respect to the notification section 203 described later. Furthermore, the processing section 202 acquires information of an order result of foodstuffs. The notification section 203 notifies information received via the communication section 201. For example, the notification section 203 notifies information of the order reservation of foodstuffs (an example of information indicating contents of an order) received via the communication section 201. Furthermore, the notification section 203 notifies information of the order result of foodstuffs (an example of information indicating contents of an order). The notification section 203 is, for example, a display section such as a touch panel having a liquid crystal panel. The notification section 203 receives an operation to change the information indicating the contents of the order by the user.

[0037] Figure 5 is a view that shows an example of the structure of the server 30. The server 30 manages the kinds and inventory of foodstuffs. The server 30 has a communication section 301, a processing section 302, and a storage section 303, as shown. Figure 5

[0038] ​The communication section 301 communicates with an external device. As an example of the external device, the refrigerator 10, the server 40, and the like can be given. The processing section 302 acquires information indicating the inventory amount of each food material from the refrigerator 10 via the communication section 301. The processing section 302 records the acquired information indicating the inventory amount in association with information indicating the date and time at which the information was acquired, to the storage section 303. The date and time includes the year, the month, the day, and the hour. The storage section 303 stores, in a predetermined period in the past, information indicating the inventory amount in association with information indicating the date and time, for each kind of food material. The predetermined period in the past is a period in which data indicating the relationship between the food material and the inventory number of the food material is stored. The data indicating the relationship between the food material and the inventory number of the food material is data used to determine parameters of a plurality of learning models described later, and data used to predict the future using the plurality of learning models having the determined parameters. Figure 6 is a diagram indicating an example of the information stored in the storage section 303. Figure 6 The example illustrated in Figure 6 is an example in which an egg and pork are recorded in the storage section 303 as food materials, and for each food material, the date and time is associated with the inventory amount of the food material at the date and time (inventory number in Figure 6 indicates an example in which the inventory amount is indicated by the number, but the inventory amount is not limited to the number, and the inventory amount can be indicated by the weight.

[0039] Figure 7 is a diagram indicating an example of the structure of the server 40. The server 40 determines the consumption frequency of each food material, estimates the future inventory state of each food material, and estimates the ordering day of each food material. The server 40 has, as illustrated in Figure 7 , a communication section 401, a processing section 402 (an example of a determination section, an example of an estimation section, an example of an ordering section, an example of an inventory estimation learning section, and an example of an ordering day learning section), and a storage section 403.

[0040] The communication section 401 communicates with an external device. As an example of the external device, the portable terminal 20, the server 30, the server 50, and the like can be given.

[0041] The processing section 402 classifies a plurality of food materials consumed by a user in association with the consumption frequency, based on the past consumption behavior of the user, and determines a special food material having a relatively high consumption frequency in each classification. For example, the processing section 402 determines the consumption frequency of each of a plurality of food materials, based on the past consumption behavior of the user in a first period and the past consumption behavior of the user in a second period longer than the first period. That is, the processing section 402 determines the consumption frequency using data different in period.

[0042] Figure 8 is a diagram indicating the relationship between the date and time in a predetermined period in the past and the inventory amount of the food material A at the date and time. Here, reference is made toFigure 8 The processing section 402 that determines the frequency of consumption using the data that differs between the periods of use will be described. In addition, Figure 8 The past prescribed period in the above is between 90 days ago and the present.

[0043] First, the processing section 402 determines whether the food material A is a food material with a high frequency of consumption (for example, consumed every day or consumed once every 2 to 3 days) using the data of the inventory amount of the food material A in the past 1 to 14 days (between two weeks ago and the present) in the recent short period. Figure 8 For example, the processing section 402 determines the frequency of consumption by setting the operation result of dividing the number of times of consumption of the food material A in the recent short period by the number of days indicated by the period as the frequency of consumption. Also, the processing section 402 determines the food material A as a food material that is consumed every day, for example, in the case where the operation result is 1 or more. In addition, the processing section 402 determines the food material A as a food material that is consumed once every 2 to 3 days, for example, in the case where the operation result is 1 / 3 or more and less than 1.

[0044] In addition, in the case where the food material A is not determined as a food material with a high frequency of consumption (in the above example, in the case where the operation result is less than 1 / 3), the processing section 402 lengthens the recent period (for example, changes between two weeks ago and the present to between 30 days ago and the present) and determines a food material with a second high frequency of consumption (for example, a food material that is consumed once every 1 week or a food material that is consumed once every two weeks). That is, the processing section 402 determines the frequency of consumption of the food material A using the data of the inventory amount of the food material A in the past 1 to 30 days as data that differs between the periods. In addition, the determination of the frequency of consumption by the processing section 402 can set the operation result of dividing the number of times of consumption of the food material A by the number of days indicated by the period as the frequency of consumption, as described above. Also, the processing section 402 determines the food material A as a food material that is consumed once every 1 week, for example, in the case where the operation result is 1 / 7 or more and less than 1 / 3. In addition, the processing section 402 determines the food material A as a food material that is consumed once every two weeks, for example, in the case where the operation result is 1 / 14 or more and less than 1 / 7.

[0045] Furthermore, if ingredient A is not determined to be the second most frequently consumed ingredient (in the example above, when the calculation result is less than 1 / 14), processing unit 402 further extends the most recent period (for example, changing 30 days to 90 days) to determine the ingredient with the second most frequently consumed ingredient (for example, consumed once every 30 days). That is, processing unit 402 uses the inventory data of ingredient A from the past 1 day to 90 days as data for different periods to determine the consumption frequency of ingredient A. Alternatively, the consumption frequency determination performed by processing unit 402 can also be as described above, using the calculation result obtained by dividing the number of times ingredient A is consumed by the period as the consumption frequency. Furthermore, processing unit 402, for example, if the calculation result is more than 1 / 30 and less than 1 / 14, determines ingredient A as an ingredient consumed once every 30 days. Furthermore, processing unit 402, for example, if the calculation result is less than 1 / 30, determines ingredient A as an ingredient with a consumption frequency of less than once every 30 days.

[0046] In practice, the processing unit 402 determines the consumption frequency of each ingredient by processing each ingredient A as described above.

[0047] In addition to determining the consumption frequency through the above-described calculation method, the processing unit 402 can also use a learned model for determining the consumption frequency to determine the consumption frequency. The parameters in the learned model are determined, for example, as follows.

[0048] The teacher data used in determining the parameters of the learning model for determining consumption frequency includes input data and output data corresponding to that input data. Figures 9-11 This is a graph representing an example of teacher data.

[0049] First, the determination of the parameters for the learning model used to identify ingredients that are consumed frequently (e.g., consumed daily or once every 2-3 days) will be explained. Figure 9 This is a graph representing teacher data used by the processing unit 402 to determine the parameters of the learning model for ingredients with high consumption frequency (e.g., consumed daily or once every 2-3 days). For example, the storage unit 403 stores this teacher data. Teacher data represents actual result data regarding the past consumption frequency of each ingredient. For each ingredient, the teacher data uses the date and time (e.g., two weeks in the case of ingredients consumed daily or once every 2-3 days) as input data, and the consumption frequency corresponding to the input data as output data. It includes multiple data points obtained by associating input data with output data corresponding to the input data. Figure 9 In the example shown, the data obtained by associating the input data with the corresponding output data is 10,000 sets of data.

[0050] For example, consider using Figure 9 The example shown illustrates how teacher data, consisting of 10,000 data points, determines the parameters of a learning model for ingredient A. In this case, the teacher data is divided, for example, into training data, evaluation data, and test data. Examples of the proportions of training, evaluation, and test data could be 70%, 15%, 15%, or 95%, 2.5%, 2.5%, etc. For instance, suppose the teacher data #1 to #10000 is divided into data #1 to #7000 as training data, data #7001 to #8500 as evaluation data, and data #8501 to #10000 as test data. In this case, data #1, serving as training data, is input into the neural network that serves as the learning model. Whenever the input data of the training data is input into the neural network, and the frequency of consumption is output from the neural network (in this case, whenever each data point from #1 to #7000 is input into the neural network), backpropagation is performed corresponding to this output, thereby changing the parameters representing the weights of the data combinations between nodes (i.e., changing the model of the neural network). In this way, training data is input into the neural network, and the parameters are adjusted.

[0051] Next, the input data (data #7001 to #8500) of the evaluation data is sequentially input into the neural network whose parameters have been modified based on the training data. The neural network outputs the consumption frequency corresponding to the input evaluation data. Here, the data output by the neural network and the frequency of consumption corresponding to the input evaluation data are compared. Figure 9 In cases where the input data establishes a correlation with the output data, and the output data differs from the input data, this allows the neural network's output to become... Figure 9 The parameters are changed by establishing a correlation between the output data and the input data. In this way, the neural network (i.e., the learning model) that determines the parameters is a learned model (hereinafter referred to as the first learned model) used to determine ingredients that are consumed frequently (e.g., consumed daily or once every 2-3 days).

[0052] Next, as final confirmation, the test data (data #8501 to #10000) is sequentially input into the neural network of the first learned model. The frequency of consumption corresponding to the input test data is then compared to the output of the neural network of the first learned model. For all test data, the frequency of consumption of the output of the neural network of the first learned model is compared to the frequency of consumption of the input test data. Figure 9 If the frequency of consumption associated with the input data is consistent, then the neural network of the first learned model is the desired model. Furthermore, if in the test data there is even one instance where the frequency of consumption of the neural network output of the first learned model is consistent with that of the input data, then the neural network is the desired model. Figure 9In a case where the consumption frequency associated with the input data is not consistent, the parameters of the learning model are decided using new teacher data. The decision of the parameters of the learning model is repeated until a first learned model having desired parameters is obtained. In a case where the first learned model having desired parameters is obtained, the first learned model is recorded in, for example, the storage section 403.

[0053] Next, the decision of the parameters of the learning model for determining a food material having a second-highest consumption frequency (for example, consumed once in one week or consumed once in two weeks) will be described. Figure 10 is a graph showing teacher data for determining the parameters of the learning model for determining a food material having a second-highest consumption frequency (for example, consumed once in one week or consumed once in two weeks) used by the processing section 402. For example, the storage section 403 stores the teacher data. The teacher data is actual result data showing past consumption frequencies of each food material. The teacher data, for each food material, sets a date and time in a prescribed period (for example, 30 days in a case of determining a food material consumed once in one week or consumed once in two weeks) for determining a food material having a second-highest consumption frequency and the inventory amount at the date and time as input data, sets the consumption frequency corresponding to the input data as output data, and includes a plurality of data obtained by associating the input data with the output data corresponding to the input data. In Figure 10 In the example shown, the data obtained by associating the input data with the output data corresponding to the input data is 10,000 sets of data.

[0054] In addition, in the case of deciding the parameters of the learning model for food material A using the teacher data composed of 10,000 sets of data shown in Figure 10 In a case where the parameters of the learning model for food material A are decided using the teacher data composed of 10,000 sets of data shown in

[0055] Next, the decision of the parameters of the learning model for determining a food material having a second-highest consumption frequency (for example, consumed once in one week or consumed once in two weeks) will be described. Figure 11This is a graph representing teacher data used by the processing unit 402 to determine the parameters of the learning model for identifying the second most frequently consumed ingredient (e.g., consumed once every 30 days). For example, the storage unit 403 stores this teacher data. The teacher data represents actual result data regarding the past consumption frequency of each ingredient. For each ingredient, the teacher data uses the date and time (e.g., 90 days in the case of identifying an ingredient consumed once every 30 days) within a specified period for which the second most frequently consumed ingredient is determined, and the inventory level at that date and time, as input data. The consumption frequency corresponding to the input data is set as output data, and includes multiple data obtained by associating the input data with the output data corresponding to the input data. Figure 11 In the example shown, the data obtained by associating the input data with the corresponding output data is 10,000 sets of data.

[0056] In addition, in use Figure 11 In the case where the teacher data, consisting of 10,000 sets of data, determines the parameters of the learning model for ingredient A, the teacher data is also divided into training data, evaluation data, and test data as described above, and this process is repeated until a learned model with the desired parameters is obtained (if the consumption frequency associated with the input data is inconsistent, new teacher data is used). Thus, the neural network (i.e., the learning model) that determines the parameters is a learned model used to determine the ingredient with the second highest consumption frequency (e.g., consumed once every 30 days) (hereinafter referred to as the third learned model). Having obtained the third learned model with the desired parameters, this third learned model is recorded, for example, in storage unit 403.

[0057] In addition, for ingredients other than ingredient A, the same method for determining the parameters of the learning model described above is used to determine the parameters of multiple learning models used to determine different frequencies of use, just as for ingredient A.

[0058] Further, the processing section 402 classifies the food materials based on the determined consumption frequencies, and determines the food materials in the top ranks of the consumption frequencies among the classified food materials (examples of the special food materials with relatively high consumption frequencies). For example, the processing section 402 classifies the food materials with a consumption frequency of every day, the food materials with a consumption frequency of once in 2 to 3 days, the food materials with a consumption frequency of once in a week, the food materials with a consumption frequency of once in two weeks, the food materials with a consumption frequency of once in a month, and the food materials with a consumption frequency of less than once in a month. Further, the processing section 402 determines the food materials in the top ranks of the consumption frequencies in the order of the food materials with a consumption frequency of every day, the food materials with a consumption frequency of once in 2 to 3 days, the food materials with a consumption frequency of once in a week, the food materials with a consumption frequency of once in two weeks, the food materials with a consumption frequency of once in a month, and the food materials with a consumption frequency of less than once in a month. Further, the processing section 402 determines the food materials in the top ranks of the consumption frequencies as long as the food materials are determined from the determined consumption frequencies. In addition, the processing section 402 can determine the food materials in the top ranks of the consumption frequencies (for example, the top 5 in the ranks of the consumption frequencies, the food materials with a consumption frequency of once in a month or more, and the like) using a learned model different from the learned model used to determine the above-described consumption frequencies. In addition, for the parameters of the learned model used to determine the food materials in the top ranks of the consumption frequencies, for example, assuming that the prescribed period is two weeks, the date and time in the two weeks and the inventory amount at the date and time are set as input data, the food materials in the top ranks of the consumption frequencies corresponding to the input data are set as output data, and the teacher data including a plurality of data obtained by associating the input data with the output data is used, the parameters of the learned model can be determined in the same manner as described above.

[0059] Further, the processing section 402 estimates a future inventory state of the food material based on inventory information indicating an inventory state of at least one food material stored in the user's refrigerator 10 or home and estimation information reflecting the user's past consumption trend with respect to the food material. The future inventory state of the food material can be a future inventory amount of the food material. As an example of the estimation information, a learned model or the like can be given. As an example of the user's past consumption trend, a relationship between the user's past different multiple times and the inventory amount of the food material at each of these times, or the like can be given. However, the future inventory state of the food material is not limited to the future inventory amount of the food material, and can be, for example, a period in which the inventory amount of the food material is below a prescribed amount (e.g., zero). In addition, in a case where there are multiple inventory registrations with respect to the same or different food material, which differ in purchase date or intake deadline, the inventory information can include information indicating a correspondence relationship between the inventory amount of each of the multiple inventory registrations and the purchase date or intake deadline of each of the multiple inventory registrations. In addition, the purchase date can include any one of the day on which the food material is ordered and the day on which the delivery of the food material is completed. In addition, the purchase date preferably unifies the day on which the food material is ordered or the day on which the delivery of the food material is completed, or the like, as a reference day. Further, the intake deadline can include at least one of a shelf life or a tasting deadline. Further, in a case where there are multiple inventory registrations with respect to the same or different food material, which differ in storage compartment of the refrigerator, the inventory information can include information indicating a correspondence relationship between the inventory amount of each of the multiple inventory registrations and the storage compartment of each of the multiple inventory registrations. Further, in a case where each food material is registered in association with a food material category indicating a category of the food material, the inventory information can include information indicating a correspondence relationship between the inventory amount of each of the multiple inventory registrations and the food material category of each of the multiple inventory registrations. For example, as an example of the food material category indicating a category of the food material, meat, fish, vegetables, or the like can be given. The processing section 402 estimates the future inventory state of a particular food material (e.g., a food material in the top several of the consumption frequency among the classified food materials). Further, the processing section 402 estimates the future inventory state of the food material based on the latest inventory state of the food material in a case where the inventory state of the food material has changed. In addition, the processing section 402 can estimate the future inventory state of the food material using a learned model that estimates the future inventory state of the food material. That is, the processing section 402 can determine parameters of the learning model as the estimation information based on teacher data with respect to the user, which establishes a correspondence relationship between inventory information indicating an inventory state of a food material at a past time and information indicating an inventory state of the food material at a future time observed from the past time, includes the inventory information indicating the inventory state of the food material at the past time as input data, and includes the information indicating the inventory state of the food material at the future time observed from the past time as output data. Further, the teacher data can also include information indicating a control pattern of the refrigerator at the past time. The parameters of the learning model are determined, for example, as follows.

[0060] Figures 12-14 is a diagram showing an example of the teacher data used in the determination of the parameters of the learning model that predicts the future inventory state of the food material. First, the determination of the parameters of the learning model that predicts the future inventory state of the food material with a high consumption frequency (e.g., consumed every day or consumed once in 2 to 3 days) will be described. Figure 12 is the teacher data used in the determination of the parameters of the learning model that predicts the future inventory state of each food material in the case of a high consumption frequency (e.g., consumed every day or consumed once in 2 to 3 days). For example, the storage section 403 stores this teacher data. The teacher data is actual result data showing the past consumption frequency of each food material. The teacher data, with respect to each food material, sets the date and time in a prescribed period (e.g., two weeks in the case of determining the food material consumed every day or consumed once in 2 to 3 days) in which the consumption frequency is high and the inventory amount at that date and time as input data, and a period in which the inventory amount of the food material becomes zero as output data, and includes a plurality of data obtained by associating the input data with the output data corresponding to the input data. In the example shown in Figure 12 In the example shown in

[0061] For example, consider the case where the parameters of the learning model with respect to the food material A are determined using the teacher data composed of 10,000 pieces of data shown in Figure 12 In this case, the teacher data is divided into, for example, training data, evaluation data, and test data. As an example of the proportions of the training data, the evaluation data, and the test data, 70%, 15%, and 15% or 95%, 2.5%, and 2.5% can be given. For example, assume that the teacher data of data #1 to #10,000 is divided into data #1 to #7,000 as training data, data #7,001 to #8,500 as evaluation data, and data #8,501 to #10,000 as test data. In this case, data #1 as the training data is input to the neural network as the learning model. Each time the input data of the training data is input to the neural network and a period in which the inventory amount becomes zero is output from the neural network (in this case, each time the respective data of data #1 to #7,000 is input to the neural network), the parameters representing the weights of the combination of data between nodes are changed (i.e., the model of the neural network is changed) in correspondence with the output, for example, by back propagation. In this way, the training data is input to the neural network and the parameters are adjusted.

[0062] Next, the input data (data #7,001 to #8,500) of the evaluation data is sequentially input to the neural network whose parameters have been changed based on the training data. The neural network outputs a period in which the inventory amount corresponding to the input evaluation data becomes zero. Here, the data output from the neural network is compared with the data in the teacher data, and the parameters of the neural network are adjusted so that the data output from the neural network matches the data in the teacher data. Figure 12The input data is neutralized to establish a correlation with the output data in a different manner so that the output of the neural network becomes Figure 12 The parameter is changed in a manner that the output data is correlated with the input data. In this way, the neural network whose parameter is decided (i.e., the learning model) is a learned model (hereinafter referred to as the fourth learned model) that is used to estimate the inventory state of the food material A in a case where the food material A is determined to be a food material with a high consumption frequency (e.g., consumed every day or consumed once in 2 to 3 days).

[0063] Next, as a final confirmation, the input data of the test data (data #8501 to #10000) is sequentially input to the neural network of the fourth learned model. The neural network of the fourth learned model outputs a period in which the inventory amount corresponding to the input test data becomes zero. For all of the test data, the period in which the inventory amount output by the neural network of the fourth learned model becomes zero coincides with the period in which the inventory amount corresponding to the input test data becomes zero. Figure 12 The neural network of the fourth learned model is a desired model in a case where the period in which the inventory amount corresponding to the input data becomes zero coincides with the period in which the inventory amount output by the neural network of the fourth learned model becomes zero. Further, in a case where the period in which the inventory amount corresponding to the input data becomes zero does not coincide with the period in which the inventory amount output by the neural network of the fourth learned model becomes zero, the parameter of the learning model is decided using new teacher data. The decision of the parameter of the learning model described above is repeated until the fourth learned model having a desired parameter is obtained. In a case where the fourth learned model having a desired parameter is obtained, the fourth learned model is recorded in the storage section 403. Figure 12

[0064] Next, the decision of the parameter of the learning model used to estimate the future inventory state of a food material with a second high consumption frequency (e.g., consumed once in 1 week or consumed once in 2 weeks) will be described. Figure 13 The teacher data used in the decision of the parameter of the learning model that represents the future inventory state of each food material in a case where the consumption frequency is second high (e.g., consumed once in 1 week or consumed once in 2 weeks). For example, the storage section 403 stores the teacher data. The teacher data is actual result data that represents the past consumption frequency of the food material. The teacher data includes a plurality of data in which the date and time in a prescribed period (e.g., 30 days in a case where a food material consumed once in 1 week or consumed once in 2 weeks is determined) in which the consumption frequency is second high and the inventory amount at the date and time are set as input data, and the period in which the inventory amount of the food material becomes zero is set as output data, for each food material. In Figure 13 In the example shown in FIG. 10, the data in which the input data and the output data corresponding to the input data are correlated is 10000 sets of data.

[0065] ​In addition, in a case where the parameters of the learning model regarding the food material A are decided using the teacher data composed of 10,000 sets of data as shown in Figure 13

[0066] Next, the decision of the parameters of the learning model for predicting the future inventory state of the food material with the second highest consumption frequency (for example, consumed once in 30 days) will be described. Figure 14 The teacher data used in the decision of the parameters of the learning model for predicting the future inventory state of each food material in a case where the consumption frequency is the second highest (for example, consumed once in 30 days) is stored in the storage section 403, for example. The teacher data is actual result data indicating the past consumption frequency of the food material. The teacher data sets the date and time in a prescribed period (for example, 90 days in a case where the food material consumed once in 30 days is determined) in which the consumption frequency is the second highest and the inventory amount at the date and time as input data, and the period in which the inventory amount of the food material becomes zero as output data, and includes a plurality of data obtained by associating the input data with the output data corresponding to the input data. In the example shown in Figure 14 In the example shown in

[0067] In addition, in a case where the parameters of the learning model regarding the food material A are decided using the teacher data composed of 10,000 sets of data as shown in Figure 14

[0068] ​​In addition, with respect to the food materials other than the food material A, the parameters of the plurality of learning models for estimating the future inventory state are determined using the above-described determination method of the parameters of the learning model, similarly to the food material A.

[0069] Further, the processing section 402 acquires information associated with the delivery period of each ordered food material (an example of the delivery period after the order of the food material). For example, the processing section 402 accesses the server 50 described later as a server of an ordering point of food materials via a communication network such as a network or a telephone network, thereby acquiring information of the delivery completion date of the food material. Specifically, in the case of a business that performs business cooperation or the like and is able to import a mechanism on a WEB site that easily acquires information of the delivery completion date, the processing section 402, for example, accesses the WEB site, and in the case where a prescribed ID (Identification) and a password are input, a signal indicating the delivery completion date is transmitted from the server 50, and the information of the transmission completion date is acquired. Further, specifically, in the case where the delivery completion date is simply displayed on the WEB site, the processing section 402, for example, performs language analysis of HTML (Hypertext Markup Language) or the like, or image analysis of the screen of the WEB site, and acquires information of the delivery completion date. In addition, the user can make a telephone call or the like to the ordering point, and record the acquired information of the delivery completion date of the food material in the storage section 403 of the server 40 or the like, and the processing section 402 acquires the information of the delivery period by accessing the storage section. Further, the user can record the information of the delivery completion date displayed on the WEB site in the storage section 403 of the server 40 or the like, and the processing section 402 acquires the information of the delivery period by accessing the storage section. Furthermore, the processing section 402 calculates the period from the order date to the delivery completion date, and acquires the calculation result as the delivery period.

[0070] In addition, in the case where the delivery completion date is expressed in units other than days, such as 1 week, 1 month, or the like, the processing section 402 can convert 1 week to 7 days, and convert 1 month to 30 days or the same day of the next month, taking into account the difference in the number of days of each month, to calculate the delivery period.

[0071] Further, the processing section 402 orders the food material based on the future inventory state of the food material estimated and the delivery period required for the delivery of the food material. For example, the processing section 402 orders a special food material (e.g., a food material whose consumption frequency is among the top in the classified food materials) based on the future inventory state of the special food material. The processing section 402 orders the food material in an amount that will not be consumed until the delivery of the ordered food material is completed from the time of the next order of the food material before the time period from the time when the inventory amount of the food material becomes zero back by the time period of the delivery of the food material. As an example of the order of the food material in an amount that will not be consumed until the delivery of the ordered food material is completed from the time of the next order of the food material, there are an order in a large amount as in the past with respect to the food material, an order in an average amount as in the past with respect to the food material, and the like. Further, the amount that will not be consumed until the delivery of the ordered food material is completed from the time of the next order of the food material can be determined in advance. Furthermore, the processing section 402 can update the order timing or the order amount of the food material determined in the past based on the future inventory state of the food material newly estimated. For example, the processing section 402 estimates the inventory state from the inventory amount of the food material each time the food material is consumed. Assume that the food material becomes zero in three days in the last estimation and it takes one day from the order to the completion of the delivery. Then, the processing section 402 determines, for example, that it is possible to order the food material in an average amount of the order amount of the past order two days later. However, the processing section 402 can order the food material immediately in a case where the food material is consumed completely the next day, for example. Furthermore, since the food material is consumed drastically, the consumption frequency also becomes high. Thus, the processing section 402 can determine that the state where the consumption frequency is high will continue in the future and increase the order amount. Furthermore, the processing section 402 can notify the user of a query about the order of the food material in a case where the inventory state of the food material does not change or the decrease in the inventory is below a threshold value in a period from the time when the order of the food material is determined to the time when the food material is actually ordered, or in a case where the inventory of the food material is increased, or in a case where the actual inventory is more than the future inventory estimated last time. Through the notification, the user can avoid the order of the food material that leads to waste.

[0072] Further, the processing section 402 can have a learned model that determines the order timing and the order amount of the food material, and determine the order timing and the order amount of the food material using the learned model. That is, the processing section 402 can establish a correspondence relationship between the future inventory state of the food material, the actual delivery period of the food material in the past, and the actual order day of the food material, determine the parameters of a learning model that estimates the order day of the food material based on teacher data that contains the future inventory state of the food material estimated by the processing section 402 and the actual delivery period of the food material in the past as input data and contains the actual order day of the food material as output data. Further, the processing section 402 can determine the order day of the food material to be ordered using the learned model whose parameters are determined. The parameters of the learning model are determined, for example, as follows.

[0073] Figure 15 is a diagram showing an example of teacher data used in deciding parameters of a learning model that estimates an ordering day of a food material. For example, the storage section 403 stores this teacher data. The teacher data is actual result data showing a relationship between an actual delivery period of an order of each food material and an actual ordering day of the food material with respect to a future inventory state of a certain time in the past of the food material. The teacher data sets the future inventory state of a certain time in the past of each food material and the actual delivery period of an order of the food material as input data, sets the actual ordering day of the food material as output data, and includes a plurality of data obtained by associating the input data with the output data corresponding to the input data. In Figure 15 the example shown, the data obtained by associating the input data with the output data corresponding to the input data is 10,000 sets of data.

[0074] For example, consider a case where the teacher data composed of 10,000 sets of data shown in Figure 15 is used to decide parameters of a learning model with respect to the food material A. In this case, the teacher data is divided into, for example, training data, evaluation data, and test data. As an example of a ratio of the training data, the evaluation data, and the test data, 70%, 15%, 15%, or 95%, 2.5%, 2.5%, and the like can be given. For example, assume that the teacher data of data #1 to #10,000 is divided into data #1 to #7,000 as training data, data #7,001 to #8,500 as evaluation data, and data #8,501 to #10,000 as test data. In this case, data #1 as the training data is input to the neural network as the learning model. Each time the input data of the training data is input to the neural network, and an ordering day is output from the neural network (in this case, each time the respective data of data #1 to #7,000 is input to the neural network), a parameter representing a weight of a combination of data between nodes, that is, the model of the neural network, is changed, for example, by back propagation, thereby changing the parameter. In this way, the training data is input to the neural network, and the parameter is adjusted.

[0075] Next, the input data of the evaluation data (data #7,001 to #8,500) is sequentially input to the neural network whose parameter is changed based on the training data. The neural network outputs an ordering day corresponding to the input evaluation data. Here, in a case where the data output from the neural network is different from the output data associated with the input data in Figure 15 , the parameter is changed in such a manner that the output of the neural network becomes the output data associated with the input data in Figure 15 . In this way, the neural network whose parameter is decided, that is, the learning model, is a learned model for estimating an ordering day (hereinafter, referred to as the 7th learned model).

[0076] Next, as a final confirmation, the input data of the test data (data #8501 to #10000) is sequentially input to the neural network of the 7th learned model. The neural network of the 7th learned model outputs the order date corresponding to the input test data. For all of the test data, the order date output by the neural network of the 7th learned model is identical to the order date associated with the input data in the Figure 15 Figure 15 If the order date output by the neural network of the 7th learned model is identical to the order date associated with the input data in the Figure 15 If the order date output by the neural network of the 7th learned model is identical to the order date associated with the input data in the Figure 15 The above determination of the parameters of the learned model is repeated until the 7th learned model having desired parameters is obtained. In a case where the 7th learned model having desired parameters is obtained, the 7th learned model is recorded in the storage section 403.

[0077] In addition, with respect to the ingredients other than the ingredient A, the parameters of the respective learned models for estimating the order date are determined using the above-described determination method of the parameters of the learned model, similarly to the ingredient A.

[0078] In addition, the processing section 402 can order the ingredient in a case where the inventory of the ingredient is changed and the inventory of the ingredient is estimated to be zero. The processing section 402 records the order information indicating the order content in a case where the ingredient is ordered to the server 50 in the storage section 403. In the order information, the information of the order date and time, the kind of the ingredient, the order amount of the ingredient, the price of the ingredient, the delivery date of the ingredient, and the like are included. In addition, the processing section 402 transmits the order information to the portable terminal 20. The storage section 403 stores the order information ordered by the processing section 402.

[0079] Figure 16 is a diagram illustrating an example of the configuration of the server 50. The server 50 has a communication section 501, a processing section 502, and a storage section 503 as illustrated in Figure 16

[0080] The communication section 501 communicates with an external device. As an example of the external device, the server 40 or the like can be given. The processing section 502 receives the order of the ingredient from the server 40 via the communication section 501. In the order of the ingredient, the information of the order date and time, the kind of the ingredient, the order amount of the ingredient, the price of the ingredient, the delivery date of the ingredient, and the like are included. In addition, the processing section 502 transmits the information indicating the delivery period of each ingredient to the server 40 via the communication section 501.

[0081] The processing section 502 records information indicating the contents of the order of the food material that has been accepted to the storage section 503. In the information indicating the contents of the order of the food material that has been accepted, information of the date and time of the order, the kind of the food material, the order quantity of the food material, the price of the food material, the delivery date of the food material, and the like are included. Further, the processing section 502 arranges the delivery of the food material for which the order has been accepted. By this arrangement, the ordered food material is delivered to the user. The storage section 503 stores the information indicating the contents of the order of the food material that has been accepted.

[0082] (Process by the order system)

[0083] Next, the process by the order system 1 of one embodiment will be described. Figure 17 is a diagram indicating an example of the flow of the process by the order system 1 of one embodiment. Here, the flow of the process by the order system 1 illustrated in Figure 17 will be described. Note that it is assumed that a learned model in which the consumption frequency of each food material is determined using teacher data prepared with data of past actual results and the like, a learned model in which the future inventory state is estimated in accordance with the consumption frequency of each food material, and a learned model in which the order date of a food material with a high consumption frequency is estimated have been prepared before the order.

[0084] The imaging section 101 images the food material. Specifically, the imaging section 101 images the food material in the interior of the refrigerator 10 and the food material that is an object of management stored in the house (step S1). The processing section 103 determines the kind and the inventory of the food material (step S2). For example, the processing section 103 determines the kind of the food material and determines the inventory thereof by using an image matching technique to determine whether or not a pattern of an image including a feature of each food material prepared in advance is included in an image captured by the imaging section 101. The processing section 103 transmits the determined kind of the food material, the inventory of the food material, and the date and time to the server 30 via the communication section 102.

[0085] The processing section 302 of the server 30 receives the kind of the food material, the inventory of the food material, and the date and time from the refrigerator 10 via the communication section 301. The processing section 302 records information indicating the received inventory in association with information indicating the date and time at which the information was received to the storage section 303. The storage section 303 stores the information indicating the inventory in association with the information indicating the date and time for each kind of food material.

[0086] The processing section 402 acquires information indicating the past consumption trend of the user from the server 30 via the communication section 401. The processing section 402 inputs the acquired information indicating the past consumption trend of the user to the learned model that determines the consumption frequency. The learned model that determines the consumption frequency outputs the consumption frequency of each food material corresponding to the input information. The consumption frequency output by the learned model is the consumption frequency of the food material. The processing section 402 classifies the food materials in accordance with the determined consumption frequencies. The processing section 402 determines the food materials with the top consumption frequencies among the classified food materials (step S3). For example, the processing section 402 classifies the food materials with the consumption frequency of every day, the food materials with the consumption frequency of once every 2 to 3 days, the food materials with the consumption frequency of once every week, the food materials with the consumption frequency of once every two weeks, the food materials with the consumption frequency of once every month, and the food materials with the consumption frequency of less than once every month. Also, the processing section 402 determines the food materials with the top consumption frequencies. For example, the processing section 402 determines the food materials with the top consumption frequencies in the order of the food materials with the consumption frequency of every day, the food materials with the consumption frequency of once every 2 to 3 days, the food materials with the consumption frequency of once every week, the food materials with the consumption frequency of once every two weeks, and the food materials with the consumption frequency of once every month (for example, the top 5 food materials with the high consumption frequency, the food materials with the consumption frequency of once every month or more, and the like). In addition, the processing section 402 can determine the food materials with the top consumption frequencies using a learned model different from the first to third learned models, which outputs only the food materials with the top consumption frequencies among the consumption frequencies determined by the learned model that determines the consumption frequency.

[0087] The processing section 402 acquires the inventory information on the food materials with the top consumption frequencies determined from the server 30 via the communication section 401. Also, the processing section 402 reads the information for estimation on the food materials with the top consumption frequencies determined from the storage section 403. The processing section 402 inputs the acquired inventory information to the read information for estimation (i.e., the learned model that estimates the future inventory state of the food material). The learned model that estimates the future inventory state of the food material outputs the future inventory state (e.g., the period when the inventory amount becomes zero) of each food material corresponding to the input information.

[0088] The processing section 402 acquires, from the server 50 via the communication section 401, information indicating the delivery period in the case where each food material of which the consumption frequency is in the top ranks is ordered. For example, the processing section 402 acquires information of the delivery completion date of the food material by accessing the server 50 of the server that is the ordering point of the food material via a communication network such as a network or a telephone network (for example, by reading the delivery completion date displayed on a WEB site). Note that the acquisition of the information of the delivery completion date can be performed at any timing as long as it is before the order. Further, the processing section 402 acquires, from the server 30 via the communication section 401, information indicating the large order amount in the past recent prescribed period for each food material of which the consumption frequency is in the top ranks. The processing section 402 inputs, to each learned model that estimates the ordering date of the food material, the delivery period of each food material indicated by the acquired information, the large order amount for each food material of which the consumption frequency is in the top ranks in the past recent prescribed period indicated by the acquired information, and the future inventory state of each food material estimated by the learned model that estimates the future inventory state of the food material. Each learned model that estimates the ordering date of the food material outputs the ordering date of the food material corresponding to the input information. The processing section 402 determines the ordering date output by the learned model that estimates the ordering date of the food material (step S4). The processing section 402 orders the food material of which the large order amount is ordered at the determined ordering date for each food material of which the consumption frequency is in the top ranks in the past recent prescribed period (step S5).

[0089] (Advantages)

[0090] The above describes the ordering system 1 of one embodiment. With the ordering system 1, the processing section 402 (one example of an estimation section) estimates the future inventory state of the food material on the basis of inventory information indicating the inventory state of at least one food material stored in the user's refrigerator 10 or in the user's house and estimation information reflecting the past consumption trend of the user with respect to the food material. Further, the processing section 402 (one example of an ordering section) orders the food material on the basis of the estimated future inventory state of the food material and the delivery period required for the delivery of the food material. With the ordering system 1, the required food material can be ensured at an appropriate timing.

[0091] <First Modification of Embodiment>

[0092] In the first modification of the embodiment, the inventory information indicating the inventory state of the food material at the past time can include information indicating the correspondence between the inventory amount of each of a plurality of inventory registrations and the purchase date or the intake deadline of each of the plurality of inventory registrations in the case where a plurality of inventory registrations exist with respect to the same or different food material with different purchase dates or intake deadlines. As examples of the intake deadline, the tasting deadline or the consumption deadline can be given. This is based on the consideration that there is a tendency to preferentially consume the food material of which the purchase date is earlier or the food material of which the intake deadline is closer, which affects the consumption of the food material, in the case where the purchase date or the intake deadline is different.

[0093] Thus, in the first modification of the embodiment, in a case where the processing section 402 determines the parameter of the learned model that estimates the future inventory state of the food material on the basis of the above consideration, the purchase date or the intake deadline is used as input of the teacher data. Further, the processing section 402 inputs the purchase date or the intake deadline to the learned model that estimates the future inventory state of the food material, and estimates the future consumption of the food material. By so doing, the ordering system 1 of the first modification of the embodiment can estimate the future inventory state of the food material with higher precision than the ordering system 1 of one embodiment, by using more factors that have an influence on the future inventory state of the food material.

[0094] <Second Modification of the Embodiment>

[0095] In the second modification of the embodiment, the inventory information that indicates the inventory state of the food material at the past time can include information that indicates a correspondence relationship between the inventory amount of each of a plurality of inventory registrations and the storage compartment of each of the plurality of inventory registrations, in a case where the plurality of inventory registrations exist with respect to the same or different food materials in different storage compartments of the refrigerator 10. As examples of the storage compartment, a refrigerating compartment, a freezing compartment, and the like can be given. It can be considered that the possibility that the food material stored in the freezing compartment is not consumed for a long period of time is high. That is, this is based on the tendency that the food material of the refrigerating compartment has a shorter period until consumption than the food material of the freezing compartment, and the consideration that this has an influence on the consumption of the food material.

[0096] Thus, in the second modification of the embodiment, on the basis of the above consideration, the processing section 402 uses the storage compartment as input of the teacher data in a case where the parameter of the learned model that estimates the future inventory state of the food material is determined. Further, the processing section 402 inputs the storage compartment to the learned model that estimates the future inventory state of the food material, and estimates the future inventory state of the food material. By so doing, the ordering system 1 of the second modification of the embodiment can estimate the future inventory state of the food material with higher precision than the ordering system 1 of one embodiment, by using more factors that have an influence on the future inventory state of the food material.

[0097] <Third Modification of the Embodiment>

[0098] In the third modification of the embodiment, the inventory information indicating the inventory state of the food material at the past time can include information indicating a correspondence relationship between the inventory amount of each of the plurality of inventory registrations and the food material category of each of the plurality of inventory registrations, in a case where each food material is registered in association with a food material category indicating a category of the food material. As an example of the food material category indicating a category of the food material, there can be mentioned meat, fish, vegetables, and the like. It can be considered that the possibility that both meat and fish are used as a food material for one meal is low. That is, this is based on the consideration that there is a tendency that the possibility that fish is not used in a case where meat is used, and the possibility that meat is not used in a case where fish is used, have an influence on the consumption of the food material. Further, in a case where the inventory of vegetables is large, it can be possible that the vegetables are consumed without consuming the meat or the fish. Further, in a case where the inventory of meat or fish is large, it can be possible that the meat or the fish is consumed without consuming the vegetables.

[0099] Thus, in the third modification of the embodiment, based on the above consideration, the processing portion 402 uses the food material category indicating a category of the food material as an input of the teacher data in a case where the parameter of the learned model that estimates the future inventory state of the food material is decided. Further, the processing portion 402 inputs the food material category indicating a category of the food material to the learned model that estimates the future inventory state of the food material, and estimates the future inventory state of the food material. Thus, the ordering system 1 of the third modification of the embodiment can estimate the future inventory state of the food material with higher accuracy than the ordering system 1 of one embodiment, by using more factors that have an influence on the future inventory state of the food material.

[0100] <Fourth Modification of the Embodiment>

[0101] In the fourth modification of the embodiment, the processing portion 402 can use information indicating a control mode of the refrigerator 10 at the past time as an input of the teacher data in a case where the parameter of the learned model that estimates the future inventory state of the food material is decided. Further, the processing portion 402 inputs the control mode to the learned model that estimates the future inventory state of the food material, and estimates the future inventory state of the food material. As an example of the control mode of the refrigerator 10, there can be mentioned a thawing mode, a refrigeration mode, and the like. In addition, since the thawing mode is a mode set for the consumption of the food material, the possibility that the food material is immediately consumed is high. Further, since the refrigeration mode is also a mode set for the consumption of the food material without a long interval, the possibility that the food material is immediately consumed is high. Based on such a consideration, the control mode is used. Thus, the ordering system 1 of the fourth modification of the embodiment can estimate the future inventory state of the food material with higher accuracy than the ordering system 1 of one embodiment, by using more factors that have an influence on the future inventory state of the food material.

[0102] <Fifth Modification of the Embodiment>

[0103] The ordering system 1 of the fifth modification example of the embodiment can have at least the refrigerator 10 and the server 50. Further, the ordering system 1 of the fifth modification example of the embodiment can have some or all of the refrigerator 10, the portable terminal 20, the server 30, the server 40, and the server 50 described above with respect to each of the ordering systems 1 of the embodiments distributed or concentrated in one of the refrigerator 10, the portable terminal 20, the server 30, the server 40, and the server 50 possessed by the ordering system 1. For example, in a case where the ordering system 1 has the refrigerator 10, the portable terminal 20, the server 30, the server 40, and the server 50, some or all of the photographing section 101, the communication section 102, the processing section 103, and the reception section 104 possessed by the refrigerator 10 can be distributed or concentrated in one of the portable terminal 20, the server 30, the server 40, and the server 50. Further, some or all of the communication section 201, the processing section 202, and the notification section 203 possessed by the portable terminal 20 can be distributed or concentrated in one of the refrigerator 10, the server 30, the server 40, and the server 50. Further, some or all of the communication section 301, the processing section 302, and the storage section 303 possessed by the server 30 can be distributed or concentrated in one of the refrigerator 10, the portable terminal 20, the server 40, and the server 50. Further, some or all of the communication section 401, the processing section 402, and the storage section 403 possessed by the server 40 can be distributed or concentrated in one of the refrigerator 10, the portable terminal 20, the server 30, and the server 50. Further, some or all of the communication section 501, the processing section 502, and the storage section 503 possessed by the server 50 can be distributed or concentrated in one of the refrigerator 10, the portable terminal 20, the server 30, and the server 40.

[0104] <Embodiment 6>

[0105] With respect to the ordering system 1 of one embodiment, the reference of the period such as the delivery period is described in units of days. However, with respect to the ordering system 1 of the sixth modification example of the embodiment, the delivery period can be referenced in units of hours, minutes, or the like. For example, in a case where the units are hours, the processing section 402 can obtain the period such as the delivery period by converting 1 day into 24 hours. Further, in a case where the units are minutes, the processing section 402 can obtain the period such as the delivery period by converting 1 day into 1440 minutes.

[0106] The embodiments of the present application are explained, but these embodiments are suggested as examples, and are not intended to limit the scope of the application. These embodiments can be implemented in other various forms, and various omissions, substitutions, and changes can be made without departing from the scope of the application. These embodiments and modifications thereof are included in the scope or spirit of the application, and are included in the scope of the application and its equivalents as recited in the claims.

[0107] In addition, the processing of the embodiments of the present application can replace the order of processing within the range where appropriate processing is performed.

[0108] The embodiments of the present application are explained, but the above-described refrigerator 10, portable terminal 20, server 30, server 40, server 50, and other control devices can also have a computer device inside. Also, the processes of the above-described processing are stored in a computer-readable recording medium in the form of a program, and the program is read and executed by a computer to perform the above-described processing. The following shows a specific example of a computer.

[0109] The computer 5 has a CPU 6 (including a vector processor), a main memory 7, a storage device 8, and an interface 9.

[0110] For example, the above-described refrigerator 10, portable terminal 20, server 30, server 40, server 50, and other control devices are respectively installed in the computer 5. Also, the actions of the above-described processing sections are stored in the storage device 8 in the form of a program. The CPU 6 reads the program from the storage device 8 and expands it to the main memory 7, and performs the above-described processing according to the program. In addition, the CPU 6 secures a storage area corresponding to each of the above-described storage sections in the main memory 7 according to the program.

[0111] As examples of the storage device 8, a HDD (Hard Disk Drive), an SSD (Solid State Drive), a magnetic disk, an optical disk, a CD-ROM (Compact Disc Read Only Memory), a DVD-ROM (Digital Versatile Disc Read Only Memory), a semiconductor memory, and the like can be given. The storage device 8 can be an internal medium directly connected to the bus of the computer 5, or can be an external medium connected to the computer 5 via the interface 9 or a communication line. In addition, in the case where the program is transmitted to the computer 5 via a communication line, the computer 5 that receives the transmission can expand the program into the main memory 7 to perform the above-described processing. In at least one embodiment, the storage device 8 is a non-transitory tangible storage medium.

[0112] Further, the program can also realize a part of the above-described functions. Furthermore, the program can also be a file, i.e., a difference file (difference program), which can realize the above-described functions by being combined with a program already recorded in a computer device.

[0113] Explanation of Reference Signs

[0114] 1…ordering system; 5…computer; 6…CPU; 7…main memory; 8…storage device; 9…interface; 10…refrigerator; 20…portable terminal; 30, 40, 50…server; 101…imaging section; 102, 201, 301, 401…communication section; 103, 202, 302, 402…processing section; 104…reception section; 203…notification section; 303, 403…storage section.

Claims

1. An ordering system, comprising: The estimation department estimates the future inventory status of a food item based on inventory information indicating the inventory status of at least one food item stored in a user's refrigerator or residence and estimation information reflecting the user's past consumption trends regarding the food item. The ordering department orders the ingredients based on the future inventory status of the ingredients predicted by the estimation department and the delivery period required for the delivery of the ingredients; as well as The determination department, based on the user's past consumption patterns, categorizes multiple food items consumed by the user according to consumption frequency, and identifies special food items within each category. These special food items are those consumed more than once a month, or are among the most frequently consumed items in the categorized food items. The prediction unit predicts the future inventory status of the special ingredient based on the inventory information and the prediction information. The ordering department orders the special ingredients based on the future inventory status of the special ingredients.

2. The ordering system as described in claim 1, wherein, The determining unit determines the consumption frequency of each of the plurality of ingredients based on the user's past consumption trends during the first period and the user's past consumption trends during the second period, which is longer than the first period.

3. The ordering system as described in claim 1, wherein, The ordering system also includes an order date learning unit, which determines the parameters of a learning model for predicting the order date of the ingredients based on the following teacher data. This teacher data establishes a correspondence between the future inventory status of the ingredients predicted by the prediction unit, the past actual delivery period of the ingredients, and the actual order date of the ingredients. The system includes the future inventory status of the ingredients predicted by the prediction unit and the past actual delivery period of the ingredients as input data, and the actual order date of the ingredients as output data. The ordering department uses the learning model, whose parameters are determined by the ordering day learning department, to determine the ordering day for the ingredients.

4. The ordering system as described in any one of claims 1 to 3, wherein, The estimation unit estimates the future inventory status of the ingredients based on the latest inventory status when the inventory status of the ingredients changes. The ordering department updates the previously determined ordering time or quantity of the ingredients based on the future inventory status of the ingredients newly predicted by the prediction department.

5. An ordering system, comprising: The estimation unit, based on inventory information indicating the inventory status of at least one food item stored in a user's refrigerator or residence and estimation information reflecting the user's past consumption trends regarding said food item, estimates the future inventory status of said food item; and The ordering department, based on the future inventory status of the ingredients predicted by the estimation department and the delivery period required for the distribution of the ingredients, orders the ingredients. During the period from the moment the order for the ingredient is decided until the actual order is placed, if the inventory status of the ingredient does not change or the decrease in inventory is less than a threshold, or if the inventory of the ingredient increases, or if the actual inventory is greater than the future inventory status previously predicted by the estimation department, the ordering department will notify the user of an inquiry regarding the order for the ingredient.

6. The ordering system as described in claim 5, wherein, The ordering system also includes an order date learning unit, which determines the parameters of a learning model for predicting the order date of the ingredients based on the following teacher data. This teacher data establishes a correspondence between the future inventory status of the ingredients predicted by the prediction unit, the past actual delivery period of the ingredients, and the actual order date of the ingredients. The system includes the future inventory status of the ingredients predicted by the prediction unit and the past actual delivery period of the ingredients as input data, and the actual order date of the ingredients as output data. The ordering department uses the learning model, whose parameters are determined by the ordering day learning department, to determine the ordering day for the ingredients.

7. The ordering system as described in claim 5 or 6, wherein, The estimation unit estimates the future inventory status of the ingredients based on the latest inventory status when the inventory status of the ingredients changes. The ordering department updates the previously determined ordering time or quantity of the ingredients based on the future inventory status of the ingredients newly predicted by the prediction department.

Images