Information processing device and information processing method

Abstract

By considering the conditions of multiple stores, we determine the electricity usage that is assumed to be standard for each store. [Solution] The information processing device includes a dataset containing store master data and outside temperature data for multiple stores, and a trained model that has been trained by machine learning to calculate the total or predetermined unit electricity usage for a store in response to an input dataset, using the total or predetermined unit electricity usage for the store obtained in response to the dataset as training data. The device accepts input of a dataset and electricity usage obtained as actual data for one store, inputs the dataset obtained as actual data for one store into the trained model, calculates the electricity usage that is assumed to be standard for the total or predetermined unit of one store as the calculated electricity usage, and outputs the calculated electricity usage and the electricity usage obtained as actual data for one store in a format that allows comparison.

Description

Technical Field

[0001] The present invention relates to an information processing apparatus and an information processing method.

Background Art

[0002] Patent Document 1 below discloses a power management server that predicts future power consumption in a store. This power management server predicts future power consumption in the store from the POS information and power consumption information sequentially acquired based on the relevance between the past POS information sales information and inventory information in the store and the power consumption information.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] [ The power management server described in Patent Document 1 focuses on the fact that the power consumption of a store varies depending on the sales information and inventory information of the store, and predicts future power consumption while sequentially acquiring the sales information and inventory information. However, even if future power consumption is predicted, it is difficult to grasp how efficient the power consumption is for that store.

[0005] An object of the present invention is to provide an information processing apparatus and an information processing method capable of grasping the amount of electricity consumption assumed to be standard for each store's situation in consideration of the situations of a plurality of stores.

Means for Solving the Problems

[0006] An information processing device according to one aspect of the present invention includes a trained model generated by machine learning to calculate the total or predetermined electricity usage for a store in response to an input dataset, using as training data a dataset for each of a plurality of stores, including at least store master data and outside temperature data, and the total or predetermined unit electricity usage for the store obtained in correspondence with the dataset; an input unit that accepts input of a dataset obtained as actual data for at least one store and the electricity usage obtained as actual data for that store; a calculation unit that inputs the dataset obtained as actual data for at least one store into the trained model and calculates the electricity usage for the total or predetermined unit of the store as a calculated electricity usage; and an output unit that outputs the calculated electricity usage and the electricity usage obtained as actual data for one store in a format that allows comparison.

[0007] Another aspect of the present invention relates to an information processing method, which is performed by a processor and includes the steps of: generating a trained model by machine learning to calculate the total or predetermined electricity usage for a store in response to an input dataset, using as training data a dataset for each of a plurality of stores, which includes at least store master data and outside temperature data, and the total or predetermined unit electricity usage for the stores obtained in correspondence with the dataset; receiving input a dataset obtained as actual data for at least one store, and the electricity usage obtained as actual data for one store; inputting the dataset obtained as actual data for at least one store into the trained model and calculating the electricity usage for the total or predetermined unit of one store as a calculated electricity usage; and outputting the calculated electricity usage and the electricity usage obtained as actual data for one store in a format that allows for comparison.

[0008] According to these embodiments, a dataset containing store master data and outside temperature data for multiple stores, as well as the total or predetermined unit electricity consumption for each store obtained in accordance with the dataset, is used as training data. A trained model, which has been trained to calculate the total or predetermined unit electricity consumption for each store in accordance with the input dataset, is then input with a dataset obtained as actual data for one store. The model calculates the estimated standard electricity consumption for the total or predetermined unit of that store as the calculated electricity consumption, and outputs the calculated electricity consumption in a format that allows comparison between the calculated electricity consumption and the actual value of electricity consumption obtained as actual data for that store.

[0009] This makes it possible to understand, based on the conditions of multiple stores, whether the actual electricity consumption at each store is more or less energy-efficient than the standard expected level of electricity consumption in light of the store's circumstances. [Effects of the Invention]

[0010] According to the present invention, it is possible to provide an information processing device and an information processing method that can take into account the circumstances of multiple stores and determine the amount of electricity consumption that is assumed to be standard for each store. [Brief explanation of the drawing]

[0011] [Figure 1] This is a block diagram illustrating the configuration of an information processing device according to the embodiment. [Figure 2] This figure shows an example of outputting theoretical and actual standard electricity usage values ​​in a format that allows for comparison. [Figure 3] This figure shows an example of outputting the difference between the theoretical value of standard electricity consumption and the actual value of electricity consumption in a format that allows for comparison. [Figure 4] This is a flowchart illustrating the operation of the information processing device according to the embodiment. [Modes for carrying out the invention]

[0012] A preferred embodiment of the present invention will be described with reference to the attached drawings. In each drawing, components denoted by the same reference numerals have the same or similar configuration.

[0013] [Configuration of the information processing device] Referring to Figure 1, an example of the configuration of the information processing device 1 according to the embodiment will be described. The information processing device 1 includes, for example, a processor 11, a storage device 12, and a communication interface 13 as its physical configuration.

[0014] The processor 11 is, for example, a CPU (Central Processing Unit) or an MPU (Micro Processing Unit). The processor 11 executes the program 121 stored in the memory device 12 to realize the various functions described later.

[0015] The storage device 12 is, for example, a computer-readable recording medium such as a disk drive or semiconductor memory. The storage device 12 functions as a memory unit that stores programs 121 for realizing various functions of the information processing device 1, various data 122 used in the programs 121, and learned models 123. Note that the learned models 123 are not limited to being stored in the information processing device 1, but may also be stored in an external device or external system.

[0016] The various data sets 122 include, for example, data on multiple stores, which will be described later.

[0017] The communication interface 13 connects to a network and functions as a communication unit that communicates with other devices on the network.

[0018] As shown in Figure 1, the information processing device 1 has a functional configuration for realizing various functions, such as an input unit 111, a calculation unit 112, and an output unit 113. Each of these will be described below.

[0019] The input unit 111 receives the input of data related to a plurality of stores stored in the storage device 12, for example, according to an input device such as a keyboard or a read instruction in an external terminal device. Note that the input of data related to a plurality of stores may be received from an external terminal device via the communication interface 13.

[0020] The data related to a plurality of stores can include, for example, electricity consumption data for each store, store master data for each store, and outside air temperature data for each store. Examples of stores include retail stores such as convenience stores, drugstores, and supermarkets, and food and beverage stores such as cafes, restaurants, and fast food restaurants. Details of the electricity consumption data and the store master data will be described below.

[0021] The electricity consumption data is data for managing the actual value of the electricity consumption (power consumption) in the store. In addition to the actual value of the electricity consumption, the electricity consumption data can include, for example, a store identification code, an electrical equipment identification code, a switchboard identification code, and an electrical system identification code for identifying an arbitrarily set electrical system. This makes it possible to aggregate and manage the electricity consumption in a predetermined unit such as the entire store or in units of electrical equipment, switchboard, and electrical system. The unit of the switchboard can include, for example, a power unit, a lighting unit, and an air conditioning unit.

[0022] The store master data is data for managing the basic data of the store. In addition to the data related to the differences between stores (data that can affect the electricity consumption of the store), the store master data can include, for example, basic store information such as a store identification code and a store name.

[0023] The data related to the differences between stores can include, for example, data related to the structure of the store, data related to the location of the store, data related to the size of the store, data related to the business hours of the store, and data related to the location of the store.

[0024] The structure of a store can include, for example, the building structure of reinforced concrete and wooden structures.

[0025] Store locations can include, for example, stores located above ground (inside buildings), stores in underground shopping malls, and street-level stores.

[0026] The size of a store can include, for example, the total area of ​​the store, the number of floors, the number of tables, and the number of seats.

[0027] Store operating hours can include, for example, the number of days of the week the store is open, and the hours of operation.

[0028] The store's location can include, for example, the prefecture, city / ward / town / village, building name, and address.

[0029] The trained model 123 is a model generated by machine learning that uses a dataset containing store master data and outside temperature data for multiple stores, and electricity usage data for each store, either overall or in a predetermined unit, obtained in response to that dataset, as training data to calculate the electricity usage that is assumed to be typical for the store as a whole or in a predetermined unit, corresponding to the input dataset, as the calculated electricity usage (also called "standard electricity usage").

[0030] The datasets input to the trained model 123 are not limited to store master data and outside temperature data. For example, POS (Point of Sale) data for each store, data on electrical equipment for each store, and humidity data for each store may also be input to the trained model 123 as datasets. POS data includes product information collected by the point of sale management system. It may also include sales price, region, time, season, customer attributes, and product information, as well as sales performance information such as sales information and inventory information. Data on electrical equipment includes the model name and number of electrical equipment units.

[0031] The calculation unit 112 inputs the dataset obtained as actual store data into the trained model 123 and calculates a calculated electricity consumption (standard electricity consumption) that is assumed to be typical for each store, taking into account the differences in conditions among multiple stores, for the store as a whole or for a predetermined unit.

[0032] The output unit 113 outputs the calculated standard electricity consumption value and the actual electricity consumption value obtained as store performance data in a format that allows for comparison. The output method may include, for example, displaying the output content on a display, printing the output content from a printer, or transmitting the output content externally.

[0033] Referring to Figure 2, we will explain an example of outputting calculated standard electricity usage values ​​and actual electricity usage values ​​in a format that allows for comparison.

[0034] Figure 2 shows, as an example, tabs for selecting the unit of electricity usage aggregation, namely the "Entire Store" tab Ta, the "Distribution Panel 1" tab Tb, and the "Distribution Panel 2" tab Tb. Of these tabs, the "Entire Store" tab Ta is selected. Therefore, Figure 2 displays a graph Ga that shows a comparative representation of the calculated standard electricity usage and actual electricity usage for the entire store.

[0035] The horizontal axis of Graph Ga shows the actual total electricity consumption for one month [kWh], and the vertical axis shows the calculated standard electricity consumption [kWh]. Graph Ga plots dots representing data for each store. It may be possible to display the name of the store corresponding to each dot by hovering a cursor over it.

[0036] As shown in Graph Ga, store A's actual electricity consumption is about half of the calculated standard electricity consumption, meaning the actual value is considerably lower than the calculated value. Therefore, it can be concluded that store A uses electrical energy efficiently and is saving energy more than the standard level of electricity consumption expected given the circumstances of store A.

[0037] On the other hand, Store B's actual electricity consumption is about twice the calculated standard electricity consumption, meaning the actual consumption is considerably higher than the calculated value. Therefore, it can be concluded that Store B is not using electrical energy efficiently and is not saving energy at a level that would be expected under normal circumstances.

[0038] Furthermore, the format for comparing calculated standard electricity usage values ​​and actual electricity usage values ​​is not limited to graph Ga. For example, the store name, calculated standard electricity usage value, and actual electricity usage value may be displayed in a list format. In addition, it may be possible to specify any of the items displayed in the list to rearrange (sort) the display order.

[0039] Here, the calculation unit 112 may calculate the difference between the calculated value of standard electricity usage output from the trained model 123 and the actual value of electricity usage. In this case, it is preferable that the output unit 113 outputs the calculated difference between the calculated value of standard electricity usage and the actual value of electricity usage in a format that allows for comparison.

[0040] Referring to Figure 3, we will explain an example of outputting the difference between the calculated standard electricity consumption and the actual electricity consumption in a format that allows for comparison.

[0041] In Figure 3, the "Entire Store" tab Ta is selected, similar to Figure 2. Therefore, Figure 3 displays Graph Gb, which shows the difference between the calculated standard electricity consumption and the actual electricity consumption for the entire store, allowing for comparison.

[0042] The horizontal and vertical axes of graph Gb in Figure 3 are the same as those of graph Ga in Figure 2. The data for each store plotted in graph Gb is also the same as the data for each store plotted in graph Ga.

[0043] The difference between graph Gb in Figure 3 and graph Ga in Figure 2 is that graph Gb includes a line L representing standard electricity usage. This line L is illustratively represented by a straight line connecting points where calculated values ​​and actual values ​​coincide.

[0044] By displaying this line L, it becomes possible to visually determine that stores plotted above line L in graph Gb (for example, store A) are using electrical energy efficiently and are saving energy at a level that would be expected to be standard given the store's circumstances.

[0045] On the other hand, stores plotted below line L in graph Gb (for example, store B) can be visually identified as not using electrical energy efficiently and not achieving energy savings below the level of electricity consumption that would be expected in light of the store's circumstances.

[0046] Furthermore, in graph Gb, the greater the difference between the calculated standard electricity usage and the actual electricity usage, the further away from line L the data point will be plotted.

[0047] Therefore, for stores plotted above line L on graph Gb, the longer the length (distance) of the perpendicular line drawn from the store's dot to line L, the more energy-efficient the store can be considered to be, compared to the standard level of electricity consumption expected in light of its circumstances.

[0048] On the other hand, for stores plotted below line L in graph Gb, the longer the perpendicular line drawn from the store's dot to line L, the less energy-efficient the store is compared to the standard level of electricity consumption expected in light of its circumstances.

[0049] Furthermore, the format for comparing the calculated standard electricity usage with the actual electricity usage is not limited to graph Gb. For example, the store name and the difference between the calculated standard electricity usage and the actual electricity usage could be displayed in a list format. The magnitude of the difference could be expressed numerically or visually, such as by drawing a bar graph. Additionally, it may be possible to specify any of the items displayed in the list to rearrange the display order.

[0050] [Operation of the information processing device] Referring to Figure 4, an example of the operation of the information processing device 1 according to the embodiment will be described. In this example of operation, the case in which the datasets of each of the multiple stores are input to the trained model 123 and the standard electricity consumption of each of the multiple stores is calculated is described, but it is also possible to input the dataset of one store to the trained model 123 and calculate the standard electricity consumption of one store. In this case, it is preferable to output the calculated value of the standard electricity consumption of one store and the actual value of the electricity consumption of one store in a format that allows for comparison.

[0051] First, the input unit 111 of the information processing device 1 receives input of a dataset including outside temperature data and store master data obtained as actual data for each of the multiple stores, and electricity usage data obtained as actual data for each of the multiple stores (step S101).

[0052] Next, the calculation unit 112 of the information processing device 1 inputs the dataset obtained as actual data for each of the multiple stores into the trained model 123 and calculates the standard electricity consumption for each of the multiple stores (step S102).

[0053] Next, the output unit 113 of the information processing device 1 outputs the calculated value of the standard electricity consumption calculated in step S102 and the actual value of the electricity consumption received in step S101 in a format that allows for comparison (step S103).

[0054] As described above, according to the information processing device 1 of the embodiment, a dataset including store master data and outside temperature data for each of the multiple stores, and the total or predetermined unit electricity consumption for each store obtained in correspondence with the dataset, are used as training data. A trained model, which has been trained to calculate the total or predetermined unit standard electricity consumption for each store in correspondence with the input dataset, is given a dataset obtained as actual data for each store. The system then calculates the total or predetermined unit standard electricity consumption for each store as the calculated electricity consumption and outputs the calculated electricity consumption and the actual value of electricity consumption obtained as actual data for each store in a format that allows comparison.

[0055] This makes it possible to understand, based on the conditions of multiple stores, whether the actual electricity consumption at each store is more or less energy-efficient than the standard expected level of electricity consumption in light of the store's circumstances.

[0056] Therefore, according to the information processing device 1 of this embodiment, it is possible to take into account the circumstances of multiple stores and determine the amount of electricity usage that is assumed to be standard for the circumstances of each store.

[0057] It should be noted that the present invention is not limited to the embodiments described above, and can be implemented in various other forms without departing from the spirit of the invention. For this reason, the above embodiments are merely illustrative in all respects and should not be interpreted restrictively. For example, the order of each processing step described above can be arbitrarily changed or executed in parallel, as long as there is no inconsistency in the processing content.

[0058] [Note] The embodiments of this embodiment include the following disclosures.

[0059] (Note 1) A trained model (123) generated by machine learning uses, as training data, a dataset for each of several stores including at least store master data and outside temperature data, and the total or predetermined unit electricity consumption of the stores obtained in relation to the dataset, to calculate the total or predetermined unit electricity consumption of the stores in relation to the input dataset. An input unit (111) that accepts input of the data set obtained as performance data for at least one store, and the electricity usage obtained as performance data for the said one store, A calculation unit (112) inputs the dataset obtained as actual data for at least one store into the trained model (123) and calculates the electricity usage amount that is assumed to be standard for the entire store or a predetermined unit as the electricity usage amount, An output unit (113) outputs the calculated electricity consumption and the electricity consumption obtained as actual data for store 1 in a format that allows for comparison, An information processing device (1) equipped with the following.

[0060] (Note 2) The input unit (111) receives input for at least the data set obtained as performance data for one store, and the electricity usage obtained as performance data for one store, for multiple stores. The calculation unit (112) inputs the dataset obtained as actual data for at least one store into the trained model (123) and calculates the calculated electricity consumption for each of the multiple stores. The output unit (113) outputs the calculated electricity consumption for each of the multiple stores and the electricity consumption obtained as actual data for the one store in a format that allows for simultaneous comparison. The information processing device (1) described in Appendix 1.

[0061] (Note 3) The calculation unit (112) inputs the dataset obtained as actual data for at least one store into the trained model (123), and calculates the difference between the calculated electricity consumption and the electricity consumption obtained as actual data for one store. The output unit (113) outputs the difference between the calculated electricity consumption and the electricity consumption obtained as actual data for store 1 in a format that allows for comparison. Information processing device (1) as described in Appendix 1 or 2.

[0062] (Note 4) The aforementioned store master data includes data relating to the differences between the stores. An information processing device (1) as described in any of the appendices 1 to 3.

[0063] (Note 5) The data relating to the differences between the stores includes data relating to the structure of the stores. Information processing device (1) as described in Appendix 4.

[0064] (Note 6) The data relating to the differences between the aforementioned stores includes data relating to the location of the aforementioned stores. Information processing device (1) as described in Appendix 4 or 5.

[0065] (Note 7) The data relating to the differences between the stores includes data relating to the size of the stores. Information processing device (1) as described in any of Appendix 4 to 6.

[0066] (Note 8) The data relating to the differences between the aforementioned stores includes data relating to the operating hours of the aforementioned stores. An information processing device (1) as described in any of Appendix 4 to 7.

[0067] (Note 9) The data relating to the differences between the stores includes data relating to the locations of the stores. An information processing device (1) as described in any of the appendices 4 to 8.

[0068] (Note 10) The dataset further includes POS data for each store, An information processing device (1) as described in any of the appendices 1 to 9.

[0069] (Note 11) The dataset further includes data on electrical equipment for each store, An information processing device (1) as described in any of the appendices 1 to 10.

[0070] (Note 12) The dataset further includes humidity data for each store. An information processing device (1) as described in any of the appendices 1 to 11.

[0071] (Note 13) The aforementioned predetermined unit is one of the following: an electrical equipment unit, a switchboard unit, or an electrical system unit. An information processing device (1) as described in any of the appendices 1 to 12.

[0072] (Note 14) A method performed by the processor (11), The steps include: generating a trained model (123) by machine learning to calculate the total or predetermined electricity usage for each of several stores, using as training data a dataset including at least store master data and outside temperature data for each of the stores, and the total or predetermined unit electricity usage for each of the stores obtained in relation to the dataset; The process includes receiving input of at least the dataset obtained as performance data for one store, and the electricity usage obtained as performance data for one store, The steps include inputting the dataset obtained as actual data for at least one store into the trained model (123) and calculating the electricity usage that is assumed to be standard for the entire store or a predetermined unit as the electricity usage, The steps include outputting the calculated electricity consumption and the electricity consumption obtained as actual data for store 1 in a format that allows for comparison, Information processing methods including [Explanation of Symbols]

[0073] 1...Information processing device, 11...Processor, 12...Storage device, 13...Communication interface, 111...Input unit, 112...Calculation unit, 113...Output unit, 121...Program, 122...Data, 123...Trained model

Claims

1. A trained model generated by machine learning uses, as training data, a dataset for each of several stores including at least store master data and outside temperature data, and the total or predetermined unit electricity consumption of the stores obtained in relation to the dataset, to calculate the total or predetermined unit electricity consumption of the stores in relation to the input dataset. An input unit that accepts input of at least the data set obtained as performance data for one store, and the electricity usage obtained as performance data for one store, A calculation unit inputs the dataset obtained as performance data for at least one of the stores into the trained model and calculates the electricity usage that is assumed to be standard for the entire store or a predetermined unit as the electricity usage; An output unit that outputs the calculated electricity consumption and the electricity consumption obtained as actual data for store 1 in a format that allows for comparison, An information processing device equipped with the following features.

2. The input unit receives input for at least the data set obtained as performance data for one store, and the electricity usage obtained as performance data for one store, for multiple stores. The calculation unit inputs the dataset obtained as actual data for at least one store into the trained model and calculates the calculated electricity consumption for each of the multiple stores. The output unit outputs the calculated electricity consumption for each of the multiple stores and the electricity consumption obtained as actual data for store 1 in a format that allows for simultaneous comparison. The information processing apparatus according to claim 1.

3. The calculation unit inputs the dataset obtained as actual data for at least one store into the trained model and calculates the difference between the calculated electricity consumption and the electricity consumption obtained as actual data for one store. The output unit outputs the difference between the calculated electricity consumption and the actual electricity consumption data obtained for store 1 in a format that allows for comparison. The information processing apparatus according to claim 1.

4. The aforementioned store master data includes data relating to the differences between the stores. The information processing apparatus according to claim 1.

5. The data relating to the differences between the stores includes data relating to the structure of the stores. The information processing apparatus according to claim 4.

6. The data relating to the differences between the aforementioned stores includes data relating to the location of the aforementioned stores. The information processing apparatus according to claim 4.

7. The data relating to the differences between the stores includes data relating to the size of the stores. The information processing apparatus according to claim 4.

8. The data relating to the differences between the aforementioned stores includes data relating to the operating hours of the aforementioned stores. The information processing apparatus according to claim 4.

9. The data relating to the differences between the stores includes data relating to the locations of the stores. The information processing apparatus according to claim 4.

10. The aforementioned dataset further includes POS data for each store, The information processing apparatus according to claim 1.

11. The dataset further includes data on electrical equipment for each store, The information processing apparatus according to claim 1.

12. The dataset further includes humidity data for each store. The information processing apparatus according to claim 1.

13. The aforementioned predetermined unit is one of the following: an electrical equipment unit, a switchboard unit, or an electrical system unit. The information processing apparatus according to claim 1.

14. A method executed by a processor, A step of generating a trained model by machine learning that uses, as training data, a dataset for each of several stores including at least store master data and outside temperature data, and the total or predetermined unit electricity consumption of the stores obtained in relation to the dataset, and calculates the total or predetermined unit electricity consumption of the stores in relation to the input dataset. The process includes receiving input of at least the data set obtained as performance data for one store, and the electricity usage obtained as performance data for the one store, The steps include inputting the dataset obtained as actual data for at least one store into the trained model and calculating the electricity consumption as the standard amount for the entire store or a predetermined unit, and The steps include outputting the calculated electricity consumption and the electricity consumption obtained as actual data for store 1 in a format that allows for comparison, Information processing methods including

Images