WO2023106034A1

WO2023106034A1 - Food sales promotion control device, food sales promotion system, and food sales promotion method

Info

Publication number: WO2023106034A1
Application number: PCT/JP2022/042018
Authority: WO
Inventors: 健一柿本; 涼平渡邊; 郁人 ▲高▼嵜; 賀美井上
Original assignee: 株式会社Ｊ－オイルミルズ
Priority date: 2021-12-08
Filing date: 2022-11-11
Publication date: 2023-06-15

Abstract

Provided are a food sales promotion control device, a food sales promotion system, and a food sales promotion method that can easily and accurately calculate a discount rate for precooked foods displayed on display shelves according to a remaining time until a disposal timing is reached.　A fried food sales promotion control device 4 that controls sales promotion of a fried food X displayed in a hot showcase 1 includes: a data acquisition unit 41 that acquires change-over-time data indicating a change in the fried food X over time; a remaining time prediction unit 44 that predicts, on the basis of the acquired change-over-time data and a disposal determination criterion used to determine a disposal timing for the fried food X, the remaining time until the disposal timing of the fried food X is reached; a discount rate calculation unit 46 that calculates a discount rate of the fried food X on the basis of the predicted remaining time; and a notification unit 47 that outputs a notification signal for providing, to notification devices 312, 312A, notification of information including the calculated discount rate.

Description

FOOD SALES PROMOTION CONTROL DEVICE, FOOD SALES PROMOTION SYSTEM, AND FOOD SALES PROMOTION METHOD

The present invention relates to a food sales promotion control device, a food sales promotion system, and a food sales promotion method.

In recent years, efforts have been made to promote the Sustainable Development Goals (Sustainable Development Goals, 2030 Agenda for Sustainable Development, adopted by the United Nations Summit on September 25, 2015, hereinafter referred to as "SDGs"). It is Along with this, in order to ensure sustainable production and consumption patterns, technologies aiming to prevent waste generation, reduce waste, and significantly reduce waste generation through product recycling and reuse are being developed. Are known.

In order to avoid the disposal of products as much as possible, supermarkets, convenience stores, and other retailers that handle food in particular conduct sales promotions through discounts based on the degree of product deterioration (degree of deterioration). For example, the sales promotion system described in Patent Document 1 is a system for promoting sales of products that can be discarded over time, and the closer the current time is to the point of discarding (the more the deterioration progresses), the more the price is reduced. By issuing coupons with increasing amounts, it is possible to promote sales of products that must be discarded at a predetermined disposal point before the disposal point.

Japanese Unexamined Patent Application Publication No. 2007-11852

Assume that the sales promotion system described in Patent Document 1 is applied to the sales promotion of cooked foods such as side dishes that are cooked in a store and displayed on display shelves. For example, if fried food fried in a fryer installed in a store is displayed and sold in a hot showcase, which is a display shelf equipped with a heat retention function, the discount amount of the coupon will be applied when the cooking is completed. It will be calculated based on the elapsed time from time.

In this case, the accuracy of the "elapsed time from the time of frying" is important. In general, the elapsed time after cooking is affected by the accuracy of the employee's recording and measurement because the store employee records the time when the cooking is completed and measures the elapsed time thereafter. Therefore, there is a possibility that the employee will record the time of unloading incorrectly, or make a mistake in measuring the elapsed time from the time of unloading, or make a number of mistakes. Discount amounts may be imprecise.

Also, in general, the fried foods displayed on the display shelves are of various types and fried in various ways, and the disposal point differs depending on the type. Manual time management by store employees for each of these wide variety of fried foods is cumbersome and easily leads to mistakes when calculating discount amounts.

Therefore, the object of the present invention is a food sales promotion control device that can easily and accurately calculate the discount rate according to the remaining time until the point of disposal for the cooked food displayed on the display shelf. To provide a food sales promotion system and a food sales promotion method.

In order to achieve the above object, the present invention provides a food sales promotion control device for controlling sales promotion of cooked food displayed on a display shelf, in which changes over time of the food displayed on the display shelf are controlled. Based on a data acquisition unit that acquires the time-dependent change data shown, the time-dependent change data acquired by the data acquisition unit, and a disposal criterion preset as a criterion for determining the point of disposal of the food, a remaining time prediction unit for predicting the remaining time until the point of disposal of the food displayed on the display shelf; a discount rate calculation unit that calculates a discount rate; and a notification unit that outputs a notification signal for notification of information including the discount rate calculated by the discount rate calculation unit to a notification device. Characterized by

According to the present invention, it is possible to easily and accurately calculate the discount rate according to the remaining time until the point of disposal for the cooked food displayed on the display shelf. Problems, configurations, and effects other than those described above will be clarified by the following description of each embodiment.

1 is an external perspective view showing a configuration example of a hot showcase; FIG. It is a graph which shows the transition of the deteriorated flavor with respect to time obtained by the sensory evaluation. BRIEF DESCRIPTION OF THE DRAWINGS It is a system block diagram which shows one structural example of the fried food sales promotion system which concerns on 1st Embodiment of this invention. FIG. 4 is a configuration diagram showing an example of a hardware configuration of a controller installed in a store; It is a figure which shows the example of a display on a monitor. It is a figure which shows the example of a display with a portable terminal. 1 is a plan view of the inside of a hot showcase to which the fried food sales promotion system according to the first embodiment is applied, viewed from the back side; FIG. It is a graph which shows transition of the deep-fried color with respect to the elapsed time obtained by the sensory evaluation. 10 is a graph showing the transition of the color component R value with respect to the elapsed time obtained by photographing the fried chicken displayed in the hot showcase as a still image and analyzing the image. FIG. 10 is a graph showing transition of color component G value with respect to elapsed time obtained by photographing fried chicken displayed in a hot showcase as a still image and analyzing the image; FIG. FIG. 10 is a graph showing transition of color component B value with respect to elapsed time obtained by photographing fried chicken displayed in a hot showcase as a still image and analyzing the image; FIG. 10 is a graph showing the transition of the color component R value with respect to the elapsed time obtained when the fried chicken displayed in the hot showcase is photographed as a moving image and the image is analyzed. 10 is a graph showing the transition of the color component G value with respect to the elapsed time obtained when the fried chicken displayed in the hot showcase is photographed as a moving image and the image is analyzed. 10 is a graph showing the transition of the color component B value with respect to the elapsed time obtained when the fried chicken displayed in the hot showcase is photographed as a moving image and the image is analyzed. 10 is a graph showing the transition of the color component R value with respect to the elapsed time obtained by photographing the croquettes displayed in the hot showcase as a still image and analyzing the image. 10 is a graph showing the transition of the color component G value with respect to the elapsed time obtained by photographing the croquettes displayed in the hot showcase as a still image and analyzing the image. 10 is a graph showing the transition of the color component B value with respect to the elapsed time obtained by photographing the croquettes displayed in the hot showcase as a still image and analyzing the image. 10 is a graph showing the transition of the color component R value with respect to the elapsed time obtained when the croquettes displayed in the hot showcase are photographed as a moving image and the image is analyzed. 10 is a graph showing the transition of the color component G value with respect to the elapsed time obtained when the croquettes displayed in the hot showcase are photographed as a moving image and the image is analyzed. 10 is a graph showing the transition of the color component B value with respect to the elapsed time obtained when the croquettes displayed in the hot showcase are photographed as a moving image and the image is analyzed. 10 is a graph showing the transition of the color component R value with respect to elapsed time obtained by photographing hash potatoes displayed in a hot showcase as a still image and analyzing the image. 10 is a graph showing the transition of the color component G value with respect to the elapsed time obtained by photographing hash potatoes displayed in a hot showcase as a still image and analyzing the image. 10 is a graph showing the transition of the color component B value with respect to the elapsed time obtained by photographing hash potatoes displayed in a hot showcase as a still image and analyzing the image. It is a functional block diagram which shows the function which the fried food sales promotion control apparatus which concerns on 1st Embodiment has. It is a flowchart which shows the flow of the process performed with the fried food sales promotion control apparatus which concerns on 1st Embodiment. It is a graph which shows transition of the average area with respect to elapsed time obtained when three fried chicken displayed in the hot showcase are image|photographed as a still image and the image is analyzed. 4 is a graph showing changes in weight change rate with respect to elapsed time for fried chicken displayed in a hot showcase. FIG. 4 is a top view of the bottom surface of a predetermined stage in the hot showcase, showing a plurality of weight measurement sensors laid out on each tray. It is a graph which shows the transition of the stickiness of the batter of fried food with respect to the elapsed time obtained by sensory evaluation. It is a graph which shows the time change of the water content contained in the coating of the surface side of the fried chicken displayed in the hot showcase. It is a graph which shows the time change of the water content contained in the coating of the back side of the fried chicken displayed in the hot showcase. It is a graph which shows the time change of the crispy feeling of the fried chicken displayed in the hot showcase. 4 is a graph showing changes over time in the stickiness of batter of fried chicken displayed in a hot showcase. FIG. 4 is a plan view of a portion corresponding to the top surface of a predetermined stage in the hot showcase as viewed from below, showing a plurality of near-infrared sensors attached to the top surface portion; Fig. 10 is a graph showing the transition of the intensity of the odor of fried food with respect to the elapsed time obtained by sensory evaluation. FIG. 4 is a plan view of a portion corresponding to the top surface of a predetermined stage in the hot showcase as viewed from below, showing a plurality of odor sensors attached to the top surface portion; It is a functional block diagram which shows the function which the fried food sales promotion control apparatus which concerns on 2nd Embodiment has. It is a flowchart which shows the flow of the process performed with the fried food sales promotion control apparatus which concerns on 2nd Embodiment. It is a functional block diagram which shows the function which the fried food sales promotion system which concerns on 3rd Embodiment has. It is the top view which looked at the inside of the hot showcase to which the fried food sales promotion system based on 4th Embodiment is applied from the back side. It is a functional block diagram which shows the function which the fried food sales promotion control apparatus which concerns on 4th Embodiment has. It is a flowchart which shows the flow of the process performed with the fried food sales promotion control apparatus which concerns on 4th Embodiment. It is a flowchart which shows the flow of the process performed with the fried food sales promotion control apparatus which concerns on 5th Embodiment.

The fried food sales promotion system according to each embodiment of the present invention is installed, for example, in a hot showcase installed near the accounting place of a small store such as a convenience store, or in a delicatessen section (food section) such as a supermarket. This system manages the point of disposal of cooked foods displayed on display shelves such as showcases, especially fried foods (for example, fried chicken, croquettes, French fries, etc.).

The "disposal point" of cooked food refers to a state in which the flavor of the cooked food has deteriorated due to the passage of a predetermined time from the time cooking was completed, making it unsuitable for sale to customers. means the point in time when The "disposal point" is arbitrarily set by the manufacturer or retail store according to the type of cooked food. Therefore, the "time of disposal" includes not only the so-called "expiration date" but also the "best before date".

In the present embodiment, when the expressions "manage the point of disposal" and "manage the point of disposal" are used, they mean, as described above, "determining that the expiry date has arrived", and It also includes "determining that the expiration date has come." In addition, in terms of "managing the point of disposal," etc., it is not enough to determine when a predetermined "time point" has arrived. including the treatment of food in order to

Here, for "treatment of food", for example, if deterioration of the flavor of food is caused by the temperature environment, advice is given to move the food to a temperature environment that can further reduce the tendency of deterioration of the flavor of the food. Including outputting information. In other words, "process for food" includes processing for changing the display environment of the food to be managed so as to shift the time when the food is to be discarded into the future.

<Configuration of hot showcase 1>
First, a configuration example of a hot showcase 1, which is one aspect of a display shelf, will be described with reference to FIG.

FIG. 1 is a perspective view showing one configuration example of the hot showcase 1. FIG.

The hot showcase 1 is an example of a display shelf for fried foods that is installed in a retail store such as a convenience store, and on which fried foods X cooked in the store are displayed. The internal space of the hot showcase 1, that is, the display space in which the fried food X is displayed is kept at an appropriate temperature so that the display environment of the fried food X can be maintained under suitable conditions, and the fried food X in a more suitable state is sold to the customer. managed to make it possible.

In FIG. 1, three

shelves

11, 12, 13 are provided in the hot showcase 1, and several kinds of fried foods X are displayed on each

shelf

11, 12, 13. A plurality of fried foods X are arranged on the same tray 2 by type. In FIG. 1, three trays 2 are placed on each

shelf

11, 12, and 13, respectively. In the following description, in order to distinguish the

shelves

11, 12 and 13, the upper shelf of the hot showcase 1 is the first shelf 11, the middle shelf is the second shelf 12, and the lower shelf is is the third shelf 13 .

<Regarding the deterioration of the flavor of fried food X>
Next, deterioration of the flavor of the fried foods X displayed in the hot showcase 1 will be described with reference to FIG.

Fig. 2 is a graph showing the transition of the deteriorated flavor of fried food X over time obtained by sensory evaluation.

In FIG. 2, the horizontal axis indicates the elapsed time from the time when the fried food to be evaluated is fried, and the vertical axis converts the flavor of the fried food X into a predetermined evaluation point, and the cumulative value of the points is the deterioration of the flavor. It is a "degraded flavor" that is the degree of deterioration. That is, if the value of deteriorated flavor is low, it is presumed that the fried food is less deteriorated and is maintained in a suitable state. Conversely, if the deteriorated flavor value is high, it is presumed that the fried food X has deteriorated and is nearing the point of disposal. Therefore, based on whether the value of the deteriorated flavor exceeds a predetermined threshold value, it is possible to determine whether or not the fried food X has reached the point of disposal.

As shown in FIG. 2, in general, the fried food X has an increased deteriorated flavor (the deterioration of the flavor becomes stronger) as the time elapsed from the time of completion of frying, which is the time at which cooking is completed, increases. That is, based on FIG. 2, it can be said that the flavor of fried food deteriorates with the passage of time. Therefore, among the plurality of fried foods X displayed in the hot showcase 1, those that have passed a predetermined time from the time of frying are said to have a reduced flavor and are not suitable for sale to customers. Since it can be estimated, the fried food X that has reached the predetermined elapsed time is subject to disposal (subject to suspension of sales).

In retail stores such as convenience stores and supermarkets, in order to reduce the disposal of fried food X as much as possible, sales promotions are carried out by discounting prices based on the degree of deterioration (progress of deterioration) of fried food X. Hereinafter, as one aspect of the food sales promotion system, a fried food sales promotion system that promotes sales by discounting the fried food X displayed in the hot showcase 1 according to the elapsed time will be described for each embodiment.

<First embodiment>
A fried food sales promotion system 3 according to the first embodiment of the present invention will be described with reference to FIGS.
(Hardware configuration of fried food sales promotion system 3)
First, the hardware configuration of the fried food sales promotion system 3 will be described with reference to FIGS.

FIG. 3 is a system configuration diagram showing one configuration example of the fried food sales promotion system 3 according to the first embodiment of the present invention. FIG. 4 is a configuration diagram showing an example of the hardware configuration of the controller 311 installed in the store 31. As shown in FIG. FIG. 5A is a diagram showing a display example on the monitor 312. FIG. FIG. 5B is a diagram showing a display example on the mobile terminal 312A. FIG. 6 is a plan view of the inside of the hot showcase 1 to which the fried food sales promotion system 3 according to the first embodiment is applied, viewed from the rear side.

The fried food sales promotion system 3 is, as shown in FIG. and a management server 320 configured as described above. Each controller 311 and the management server 320 are directly or indirectly connected to each other so that information can be communicated via a communication network N such as an Internet line.

Each controller 311 controls the fried food sales promotion control device 4 that controls the sales promotion of the fried food X displayed in the hot showcase 1 (that is, the food sales promotion control device that controls the sales promotion of the cooked food displayed on the display shelf). control device). Note that each controller 311 has a function related to the sales promotion of the fried food X that the fried food sales promotion control device 4 has, for example, a function for performing environmental management such as temperature and humidity in the hot showcase 1, It may have a function related to state management of various devices provided in the store 31 . On the other hand, the management server 320 executes, for example, processing related to sales management of each store 31 .

　In FIG. 3, each hot showcase 1 is communicably connected to each controller 311. In this case, the communication means may be wired or wireless. Also, the hot showcase 1 transmits detection data obtained by detecting the state of the fried food X displayed on each

shelf

11, 12, 13 (see FIG. 1) to the management server 320 via the controller 311. It is sufficient to have a function for Note that the function of transmitting detection data to the management server 320 may be implemented without the controller 311 . For example, if the hot showcase 1 has a configuration that enables direct communication with the management server 320 , detection data can be directly transmitted to the management server 320 without going through the controller 311 .

As shown in FIG. 4, each controller 311 includes a CPU (Central Processing Unit) 301, a RAM (Random Access Memory) 302, a ROM (Read Only Memory) 303, an HDD (Hard Disk Drive) 304, and an I/F (Interface) 305 . These configurations are connected to each other via a common bus 306 .

The CPU 301 is computing means and controls the operation of the controller 311 as a whole. A RAM 302 is a volatile storage medium from which information can be read and written at high speed, and is used as a work area when the CPU 301 processes image information, for example. The ROM 303 is a read-only non-volatile storage medium and stores programs such as firmware.

The HDD 304 is a non-volatile storage medium that can read and write information and has a large storage capacity, and stores an OS (Operating System), control programs and application programs for executing various types of information processing described later. . Note that the HDD 304 can be replaced by, for example, an SSD (Solid State Drive) regardless of the type of device as long as it implements the functions of storing and managing information as a non-volatile storage medium.

Connected to the I/F 305 are the camera 5 for photographing the inside of the hot showcase 1, the monitor 312 for displaying the user interface, and the communication module 313 for realizing information communication with devices other than the controller 311.

The communication module 313 configures a communication connection interface for enabling communication via the communication network N with the controller 311 . That is, the controller 311 can communicate information with the management server 320, the mobile terminal 312A, and the like via the communication module 313 connected to the I/F 305. FIG. The communication module 313 also realizes information communication between the controller 311 and other devices such as sensors installed in the hot showcase 1 .

Each controller 311 having such a hardware configuration implements a processing function of the control program stored in the ROM 303 and the control program and application program loaded from the storage medium such as the HDD 304 to the RAM 302 by the arithmetic function of the CPU 301. It is an information processing device that By executing these information processes, a software control section including various functional modules in each controller 311 is configured. A functional block that implements the function of each controller 311 is configured by a combination of the software control unit configured in this way and the hardware resources including the configuration described above.

Note that the management server 320 shown in FIG. 3 also has the same hardware configuration as each controller 311, and each configuration executes the control program and application program stored in the respective storage media to control the management server. A functional block that implements the functions of H.320 is configured.

Specific information processing for controlling the discounted sales promotion of the fried food X displayed in the hot showcase 1 is executed by the fried food sales promotion control device 4, which will be described later. All of the functions of the fried food sales promotion control device 4 may be implemented in the store software on the controller 311 side or the headquarters software on the management server 320 side, or the functions may be distributed between the store software and the headquarters software. may be

The monitor 312 is installed, for example, near the hot showcase 1 or in a conspicuous place in the store 31, and displays product information including information such as product discounts and other campaign information to customers visiting the store 31. do. Specifically, as shown in FIG. 5A, product names and quantities that are subject to discounts at the store 31, such as "50% off for limited 2 croquettes!!" and "30% off for 3 yakitori only!" , and information on products to be discounted, including the discount rate (hereinafter sometimes simply referred to as “discount information”).

Further, the discount information is displayed not only on the monitor 312 installed in the store 31, but also on the mobile terminal 312A (smartphone, tablet, etc.) possessed by the customer P who can use the store 31, as shown in FIG. 5B. You can receive it as a notification.

For example, the customer P installs a dedicated application on the mobile terminal 312A in advance, and registers the store 31 from among the multiple stores 31 at which he or she wishes to receive discount information. The customer P can receive discount information from any store 31 as a notification by performing such a setting operation in advance.

Also, for example, as shown in FIG. 5B, the customer P installs a dedicated application on the mobile terminal 312A in advance and makes settings so that discount information can be received. Then, when the customer P approaches within a radius of, for example, 100 m from a certain shop 31, if the discount target product is displayed in the hot showcase 1, the shop 31 sends the portable terminal 312A the target discount. Information is sent.

It should be noted that the discount information sent to the mobile terminal 312A does not necessarily have to be the information after the discount on the product to be discounted. product discount schedule.

Thus, the monitor 312 and the mobile terminal 312A are one aspect of a notification device that notifies information (discount information) including the discount rate of the fried foods X displayed in the hot showcase 1. It should be noted that the notification means of the discount information on the monitor 312 and the mobile terminal 312A does not necessarily have to be only characters, and may be at least one of notification means including voice, characters, color, and light. Therefore, for example, in place of the display on the monitor 312, an in-store announcement may be made by voice, or the tray 2 on which the fried food X to be discounted may be illuminated in red.

In addition to displaying product discount information and campaign information, the monitor 312 also displays setting information input to the controller 311 by an employee of the store 31 in which the hot showcase 1 is installed, and information about the hot showcase 1 . You may display the operation information etc. which were performed. In this case, the monitor 312 preferably has separate display screens for customers and employees, for example.

The angle of view and focus of the camera 5 are adjusted so that all of the plurality of fried foods X displayed on each

shelf

11, 12, 13 can be acquired as images of the individual fried foods X, as shown in FIG. A plurality of such devices are installed in the hot showcase 1 in this state. For example, a plurality of cameras 5 are installed on the top surface above the center of each tray 2 . There are no particular restrictions on the mounting positions and number of the cameras 5 as long as the angle of view of all the fried foods X displayed in the hot showcase 1 can be captured.

Further, in the present embodiment, the camera 5 uses a camera capable of capturing still images and moving images, but it may not necessarily be a camera capable of capturing both still images and moving images. , a still camera, or the like, capable of capturing only still images.

(Data indicating the state of fried food used for setting the point of disposal)
Next, in predicting the disposal point of the fried food X used in the process (fried food sales promotion process) executed by the fried food sales promotion control device 4, FIG. 12. In this embodiment, the data indicating the state of the fried food X is exemplified by "color". In addition, hereinafter, where general considerations such as trends and characteristics of changes over time related to “fried food” are described, simply “fried food” is used and the symbol X is not attached.

Fig. 7 is a graph showing the transition of the deep-fried color against the elapsed time obtained by the sensory evaluation. The horizontal axis in FIG. 7 indicates the elapsed time from the time when the fried food to be evaluated is fried, and the vertical axis in FIG.

As shown in Fig. 7, for fried foods, the index indicating the color depth of the surface generally increases as the elapsed time from the time of frying increases. That is, fried foods tend to darken in color over time. The color of the surface of the fried food is state data indicating the state of the fried food and also corresponds to temporal change data that indicates the temporal change of the fried food. Here, the value of the "degraded flavor" shown in FIG. 2 also increases as the elapsed time from the frying time increases. Therefore, there is a correlation between the color intensity of the surface of the fried food and the deterioration of the flavor of the fried food.

Therefore, in the fried food sales promotion control device 4 according to the present embodiment, the color of the fried food X displayed in the hot showcase 1 is used as an index indicating the state of the fried food X, and the fried food X is displayed based on the change in the color of the fried food X. Predict the remaining time until the point of disposal of

In this embodiment, the hot showcase 1 corresponds to a data detection device that detects temporal change data indicating the temporal change of the fried food X displayed in the hot showcase 1, and the hot showcase 1 as a state sensor that detects the state of the fried food X. It utilizes a plurality of cameras 5 (see FIG. 6) mounted within. Then, the fried food sales promotion control device 4 identifies the surface image of the individual fried food X from the still image or moving image taken by each camera 5, calculates the RGB value of the pixel (each pixel) constituting the surface image, These RGB values are analyzed as color components of each fried food X.

It should be noted that the method of analyzing the color of each fried food X does not necessarily have to be the RGB method, and as another analysis method, for example, wavelength analysis of still images or moving images captured by each camera 5 may be performed. In the case of a moving image, a still image is extracted from the moving image at a predetermined sampling time, and the color component of each fried food X corresponding to a predetermined elapsed time is analyzed using the still image as an analysis target. Furthermore, in the present embodiment, the RGB value is used as an index indicating the color of the fried food X, but it is not limited to this, for example, it is expressed by the three elements of hue (Hue), saturation (Saturation), and lightness (Value). Other color indices such as HSV may be used.

Next, the applicant uses an image similar to the image of fried food X obtained using camera 5 to determine the color component (R component, G An example in which analysis was performed for each of the components (component, B component) and the trends over time were experimentally confirmed will be described with reference to the graphs shown in FIGS.

FIG. 8A is a graph showing the transition of the color component R value with respect to the elapsed time obtained when the fried chicken displayed in the hot showcase 1 is photographed as a still image and the image is analyzed. FIG. 8B is a graph showing the transition of the color component G value with respect to the elapsed time obtained by photographing the fried chicken displayed in the hot showcase 1 as a still image and analyzing the image. FIG. 8C is a graph showing the transition of the color component B value with respect to the elapsed time obtained by photographing the fried chicken displayed in the hot showcase 1 as a still image and analyzing the image.

As shown in FIGS. 8A to 8C, when the color component is analyzed from the still image of the fried chicken displayed in the hot showcase 1, after 2 hours (=2h) from the time of frying (=0h), R The component, G component, and B component are all reduced. Then, as time passed from the time of frying to 4 hours, 6 hours, and 7 hours, all of the R component, G component, and B component tended to decrease overall.

FIG. 9A is a graph showing the transition of the color component R value with respect to the elapsed time obtained when the fried chicken displayed in the hot showcase 1 is photographed as a moving image and the image is analyzed. FIG. 9B is a graph showing the transition of the color component G value with respect to the elapsed time obtained when the fried chicken displayed in the hot showcase 1 is photographed as a moving image and the image is analyzed. FIG. 9C is a graph showing the transition of the color component B value with respect to the elapsed time obtained when the fried chicken displayed in the hot showcase 1 is photographed as a moving image and the image is analyzed.

As shown in FIGS. 9A to 9C, when analyzing the color component from the moving image of the fried chicken displayed in the hot showcase 1, after 2 hours (= 2 h) from the time of frying (= 0 h), the R component is decreasing, while the G and B components are increasing.

Then, the R component tends to decrease as a whole as 4 hours, 6 hours, and 7 hours pass from the time of frying, as in the case of analyzing the color component from the still image of fried chicken. On the other hand, the content of the G component decreased 4 hours after the frying compared to the contents at the time of the frying and after 2 hours, and further decreased 7 hours after the frying. In addition, the component B tends to slightly increase as the time passes from the time of frying to 4 hours, 6 hours, and 7 hours.

FIG. 10A is a graph showing the transition of the color component R value with respect to the elapsed time obtained when the croquettes displayed in the hot showcase 1 are photographed as still images and the images are analyzed. FIG. 10B is a graph showing the transition of the color component G value with respect to the elapsed time obtained by photographing the croquettes displayed in the hot showcase 1 as a still image and analyzing the image. FIG. 10C is a graph showing the transition of the color component B value with respect to the elapsed time obtained by photographing the croquettes displayed in the hot showcase 1 as a still image and analyzing the image.

As shown in FIGS. 10A to 10C, when analyzing the color component from the still image of the croquette displayed in the hot showcase 1, after 2 hours (= 2 h) from the time of frying (= 0 h), the R component remains unchanged, while the G and B components increase.

FIG. 11A is a graph showing the transition of the color component R value with respect to the elapsed time obtained when the croquettes displayed in the hot showcase 1 are photographed as a moving image and the image is analyzed. FIG. 11B is a graph showing the transition of the color component G value with respect to the elapsed time obtained when the croquettes displayed in the hot showcase 1 are photographed as a moving image and the image is analyzed. FIG. 11C is a graph showing the transition of the color component B value with respect to the elapsed time obtained when the croquettes displayed in the hot showcase 1 are photographed as a moving image and the image is analyzed.

As shown in FIGS. 11A to 11C, when analyzing the color component from the video of the croquettes displayed in the hot showcase 1, after 2 hours (= 2 h) from the time of frying (= 0 h), the R component is On the other hand, the G component and B component are increasing, as in the case of analyzing the color components from the still image of the croquette.

FIG. 12A is a graph showing the transition of the color component R value with respect to the elapsed time obtained when hash potatoes displayed in the hot showcase 1 are photographed as still images and the images are analyzed. FIG. 12B is a graph showing the transition of the color component G value with respect to elapsed time obtained by photographing hash potatoes displayed in the hot showcase 1 as a still image and analyzing the image. FIG. 12C is a graph showing the transition of the color component B value with respect to the elapsed time obtained by photographing hash potatoes displayed in the hot showcase 1 as a still image and analyzing the image.

As shown in FIGS. 12A to 12C, when the color components are analyzed from the still image of hash potatoes displayed in the hot showcase 1, after 2 hours (= 2 h) from the time of frying (= 0 h), R The B component and the G component are decreasing, while the B component is increasing. The R component tends to decrease as 4 hours and 6 hours pass from the time of frying, and slightly increase after 7 hours pass. On the other hand, the G component and B component tend to decrease as the time passes from the time of frying to 4 hours, 6 hours, and 7 hours, respectively.

In this way, the fried food X displayed in the hot showcase 1 tends to change color over time from the time of frying (experimentally confirmed fried chicken, croquettes, and hash potatoes, especially, It is possible to set in advance a discarding reference value (reference RGB value) that serves as a discarding criterion for judging the discarding point of the fried food X.

Note that the discard standard value may be set uniformly to a predetermined value regardless of the type of fried food X, or may be set to a different value for each type of fried food X. For example, as shown in FIGS. 9A-C and 11A-C, when the color components are analyzed from the moving images, the same trend is observed for fried chicken and croquettes (the R component decreases, and the G and B components increase. However, when analyzing the color components from the still image, the tendency is different between fried chicken and croquette (in fried chicken, all the R, G, and B components are reduced, In the croquettes, there is no change in the R component, and increases in the G and B components).

Therefore, when analyzing color components from a still image, the discard reference value may be set to a predetermined value regardless of the type of fried food X. Further, when the color component is analyzed from the moving image, by setting the disposal reference value to a different value for each type of fried food X, the fried food sales promotion control device 4 can more accurately predict the remaining time until the time of disposal. It can be done with precision.

As described above, the method of setting the discard reference value may be dynamically changed depending on whether the format of the data input to the fried food sales promotion control device 4 is a still image or a moving image. 9A to C and FIGS. 11A to 11C are mainly due to differences in shooting conditions such as the angle of view of each camera 5, exposure time, and lighting in the hot showcase 1. Therefore, the method of setting the discard reference value may be changed according to the imaging conditions.

(Functional configuration of fried food sales promotion control device 4)
Next, a specific functional configuration of the fried food sales promotion control device 4 will be described with reference to FIG.

FIG. 13 is a functional block diagram showing the functions of the fried food sales promotion control device 4. As shown in FIG.

The fried food sales promotion control device 4 includes, for example, a data acquisition unit 41, a type identification unit 42, an analysis unit 43, a remaining time prediction unit 44, a storage unit 45, a discount rate calculation unit 46, and a notification unit 47. , and a learning unit 48 .

The data acquisition unit 41 acquires surface images (still images or moving images) of a plurality of fried foods X for each tray 2 photographed by each camera 5 as data regarding the color of each fried food X. For example, the image area of each fried food X is specified by executing image processing for extracting the outline of each fried food X included in the image captured by the camera 5 . Next, an analysis region forming a predetermined pixel group is specified from the specified image region.

Here, the “predetermined pixel group” includes pixels adjacent to the pixels extracted as the outline of the fried food X, for example, includes pixels specified as the image area of the fried food X, and has a certain range around it. It refers to the part including the pixels expanded to That is, the number of pixels in the analysis area is greater than the number of pixels in the image area of the fried food X.

Then, the data acquisition unit 41 acquires the R component, the G component, and the B component of each pixel included in the specified analysis area, and transfers them to the analysis unit 43 .

Note that the R component, G component, and B component of each pixel may be acquired by the analysis unit 43 . The analysis area may not only be set according to the size of the image area of the fried food X, but may also be set with a predetermined number of pixels regardless of this size. Alternatively, the analysis area may be specified by pixels sampled at a constant rate with respect to the number of pixels included in the image area.

The type identifying unit 42 identifies the type of each fried food X from the individual image of each fried food X extracted from each surface image acquired by the data acquisition unit 41 . The type identification unit 42 identifies the type of fried food X by comparing it with a reference sample image. Note that the sample image is stored in the storage unit 45 . In addition, the identification of the type of fried food X in the type identification unit 42 is performed by, for example, an employee of the store 31 manually specifying the type of fried food X via an input terminal (for example, a touch panel, a keyboard, etc.) installed in the store 31. It is also possible to input

Note that the fried food sales promotion control device 4 does not necessarily include the type specifying unit 42, and the discard reference value for determining when to discard the fried food X is set to a predetermined value regardless of the type of the fried food X. If there is, the type identification unit 42 is unnecessary.

The analysis unit 43 analyzes the color components (RGB values) of each fried food X from each individual image.

The remaining time prediction unit 44 is based on the color component of each fried food X analyzed by the analysis unit 43, the type of each fried food X, and the disposal standard value set for each type of each fried food X. , the remaining time (hereinafter sometimes simply referred to as “remaining time”) until the time of disposal of each fried food X is predicted. The discard reference value is stored in the storage unit 45 .

In this embodiment, as an index used to set the point of disposal of the fried food X, that is, as data indicating a change in the state of the fried food X over time, the color component (color) of the surface of the fried food X is used. The reference value is set to a discard reference value for color components.

The index used to set the point of disposal should be an index that has been confirmed to function in grasping the state of fried food X. Therefore, in addition to optical indicators such as the color of the fried food X, physical indicators such as the size, weight, moisture content, and volatile content of the fried food X, and Chemical indices such as volatile component composition, acid value, anisidine value, carbonyl value, peroxide value, iodine value, and amount of polar compounds are included. Therefore, there are various indicators other than the color of the fried food X that are used for setting the point of disposal. Therefore, the remaining time prediction unit 44 predicts the remaining time using the discard reference value for each index used for setting the discard point.

Further, when the discard reference value is set to a predetermined value regardless of the type of fried food X, the remaining time prediction unit 44 calculates the color component of each fried food X analyzed by the analysis unit 43 and the storage unit Based on the discard reference value stored in 45, the remaining time until the point of discard of each fried food X is predicted.

The discount rate calculation unit 46 calculates the discount rate for the fixed price of each fried food X based on the remaining time of each fried food X predicted by the remaining time prediction unit 44 . For example, the discount rate calculation unit 46 holds in advance a reference table in which discount rates are set for a certain remaining time. In this reference table, for example, when the remaining time is 2 hours, the discount rate is 10% of the list price, when the remaining time is 1 hour, the discount rate is 20% of the list price, and when the remaining time is 30 minutes, the discount rate is set. A predetermined discount rate is associated with a predetermined remaining time, such as 50% of the list price.

If the remaining time of the individual fried food X predicted by the remaining time prediction part 44 is the same as the remaining time set in the reference table, the discount rate calculating unit 46 calculates the discount rate corresponding to the remaining time of the fried food X Decide as a discount rate to

If the remaining time predicted by the remaining time prediction unit 44 is different from the remaining time set in the reference table, the fried food X is calculated based on the discount rate associated with the approximate remaining time. Calculate the discount rate of For example, if the remaining time predicted by the remaining time prediction unit 44 is 1 hour and 30 minutes, the discount rate calculation unit 46 calculates the discount rate (10%) for the remaining time of 2 hours and the remaining time "15%" calculated by adding the discount rate (20%) for 1 hour and proportionally dividing it is calculated as the discount rate.

In this way, in the fried food sales promotion control device 4, as the remaining time predicted by the remaining time prediction unit 44 decreases, that is, as the fried food X approaches the point of disposal, the discount rate for the fixed price of the fried food X increases. do. As a result, the selling price of the fried food X decreases as the remaining time suitable for selling the fried food X decreases, and the customer P who uses the store 31 is more motivated to buy.

The notification unit 47 transmits a notification signal for notifying information including the discount rate calculated by the discount rate calculation unit 46 and the type of fried food X to which the calculated discount rate is applied to the monitor 312 or the mobile terminal 312A ( notification device).

The notification unit 47 not only outputs the notification signal to the monitor 312 and the portable terminal 312A that notify the customer P, but also displays the display terminal (for example, a business monitor or output to a tablet for business use, etc.). As a result, the employee of the store 31 can grasp which fried foods X among the plurality of fried foods X displayed in the hot showcase 1 are discounted. In addition, the notification unit 47 outputs, as a notification signal to the display terminal used by the employee of the store 31, the remaining time until the point of disposal of the fried food X, etc., in addition to the discount information related to the fried food X to be discounted. may contain information about

Furthermore, the information (notification signal) output by the notification unit 47 is not limited in its type, expression format, and notification format as long as it is information capable of prompting the customer P to purchase the fried food X to be discounted. do not have.

In this embodiment, the fried food sales promotion control device 4 creates a learned model capable of predicting the remaining time until the fried food X is discarded by machine learning or regression analysis. It includes a learning unit 48 that performs transfer learning on the finished model.

Specifically, the learning unit 48 generates a learned model by performing machine learning and regression analysis using teacher data including color component data (data indicating the tendency of change over time up to the discard reference value). do. Then, the learning unit 48 updates the discard reference value stored in the storage unit 45 based on the generated trained model, and also performs transfer learning. In this way, the discard reference value, which is the basis of the remaining time predicted by the remaining time prediction unit 44, is updated at any time by machine learning or regression analysis, thereby improving the remaining time prediction accuracy of the remaining time prediction unit 44. .

Specifically, the learning unit 48 uses, for example, linear regression, support vector machine (SVM), bagging, boosting, etc., from the data (explanatory variables) of the discard criterion value already stored in the storage unit 45. , Adaboost, decision tree, random forest, logistic regression, neural network, deep learning, convolutional neural network (CNN), Recurrent Neural Network (RNN)), LSTM (Long Short-Term Memory), etc. Create a calibration curve (model formula).

Types of linear regression (analysis) include simple regression, multiple regression, partial least squares (PLS: Partial Least Squares) regression, orthogonal projection partial least squares (OPLS: Orthogonal Partial Least Squares) regression, etc. One or more can be used.

Simple regression is a method of predicting one objective variable with one explanatory variable, and multiple regression is a method of predicting one objective variable with multiple explanatory variables. In addition, (orthogonal projection) partial least squares regression is a method of extracting principal components so that the covariance between principal components, which are a small number of feature quantities (obtained by principal component analysis of only explanatory variables) and the objective variable, is maximized. is. (Orthogonal projection) partial least squares regression is a suitable technique when the number of explanatory variables is greater than the number of samples and when the correlation between the explanatory variables is high.

The standard curve obtained by machine learning and regression analysis in the learning unit 48 is applied to the discard reference value stored in the storage unit 45 to update the discard reference value, and the update result is used to predict the remaining time. It becomes possible to provide to the unit 44.

It should be noted that the generation of a trained model in the learning unit 48 may be executed for each user who creates and inputs data. In this case, when using the learned model to predict the remaining time until the time of disposal of fried food X, each user uses only the learned model generated by providing his or her own data. This makes it possible to predict the remaining time specific to the environment within the hot showcase 1 of each user.

In addition, the generation of trained models and transfer learning in the learning unit 48 may be performed without distinguishing between users who create and input data. In this case, a trained model can be generated using a larger amount of data. Then, when using the generated learned model, the remaining time of the fried food X is predicted using characteristics (type of fried food X, etc.) and color components predetermined for each user as input data. As a result, based on the environment in the hot showcase 1 of a plurality of users, it is possible to predict the remaining time with high accuracy using a trained model that has undergone a large amount of machine learning and regression analysis.

In addition, the learning unit 48 can create not only a learned model capable of predicting the remaining time of fried food X, but also a learned model capable of specifying the type of fried food X. . In this case, the accuracy of specifying the type of fried food X in the type specifying unit 42 is further improved.

The fried food sales promotion control device 4 does not necessarily include the learning unit 48 . , or acquire the discard reference value updated by a learning device that is different from the fried food sales promotion control device 4 and executes the same processing as the processing in the learning unit 48, Predict the remaining time of fried food X.

Hereinafter, in the second embodiment, the case where the remaining time prediction unit 44 predicts the remaining time of the fried food X by continuously using the predetermined disposal reference value stored in the storage unit 45 will be described as follows in the third embodiment. A case where the remaining time prediction unit 44 predicts the remaining time of the fried food X by acquiring the discard reference value updated by a learning device different from the fried food sales promotion control device 4 will be described.

(Processing by fried food sales promotion control device 4)
Next, the flow of processing executed by the fried food sales promotion control device 4 will be described with reference to FIG.

FIG. 14 is a flowchart showing the flow of processing executed by the fried food sales promotion control device 4. FIG.

In the fried food sales promotion control device 4, first, the data acquisition unit 41 acquires the surface image of the fried food X for each tray 2 detected and output by each camera 5 in the data detection step (the image of the fried food X displayed on the tray 2). state) is acquired (step S401).

Next, the fried food sales promotion control device 4 extracts an individual image of each fried food X from each surface image acquired in step S401 (step S402). The type of X is specified (step S403; type specifying step).

Next, the analysis unit 43 analyzes the color components (RGB values) of each fried food X from each individual image extracted in step S402 (step S404). Subsequently, the remaining time prediction unit 44 compares the color components analyzed in step S404 with the discard reference value for the type specified in step S403 for each fried food X, and determines the remaining time until the time of discarding. Time is predicted (step S405; remaining time prediction step).

If it is determined that the discounted fried food X is included in the hot showcase 1 based on each remaining time predicted in step S405 (step S406/YES), the discount rate calculation unit 46 A discount rate corresponding to the remaining time is calculated for the target fried food X (step S407; discount rate calculation step).

Next, the notification unit 47 outputs a notification signal for notifying information including the discount rate calculated in step S407 to the monitor 312 or the mobile terminal 312A (step S408). As a result, the monitor 312 and the portable terminal 312A notify the discount information about the fried food X to be discounted (notification step).

On the other hand, if it is determined in step S406 that the discounted fried food X is not included in the hot showcase 1 (step S406/NO), the process returns to step S401 to repeat the process.

Further, in the present embodiment, when the remaining time is predicted for each fried food X in step S405, subsequently, the learning unit 48 uses a pre-generated learned model to calculate the color components analyzed in step S404. Transfer learning is performed using teacher data including (step S410).

Then, the learning unit 48 updates the discard criterion value stored in the storage unit 45 based on the learned model subjected to transfer learning in step S410 (step S411; discard determination criterion updating step). When the processes of step S408 and step S411 are executed respectively, the process in the fried food sales promotion control device 4 ends.

In this way, with respect to the fried food X displayed in the hot showcase 1, based on the data of the color component indicating the change over time of the fried food X and the disposal standard value as the standard for determining the time point of discarding the fried food X , by predicting the remaining time of the fried food X, the discount rate of the fried food X according to the remaining time can be easily and accurately calculated as compared with the case where the staff of the store 31 manually calculates the remaining time of the fried food X. can do.

In the present embodiment, the fried food sales promotion control device 4 acquires the surface image of the fried food X photographed by each camera 5 as data related to the color of the fried food X, but is not limited to this. You may acquire the detected data as data regarding the color of the fried food X. FIG.

Further, in the present embodiment, the color component of each fried food X is analyzed from the surface image of the fried food X, and the color component is used as the state data (time change data) of the fried food X. However, as described above, the fried food X The state data of may be other than the color component of the fried food X, and may be, for example, the area, which is an index indicating the size of the fried food X.

FIG. 15 is a graph showing the transition of the average area with respect to the elapsed time obtained when three fried chickens displayed in the hot showcase 1 are photographed as still images and the images are analyzed. In FIG. 15, the horizontal axis indicates the elapsed time from the fried chicken to be evaluated, and the vertical axis indicates the average area of each of the three fried chickens calculated by image analysis. ” is an indexed value.

According to the graph shown in FIG. 15, the fried chicken's area (average area), which is an indicator of size, decreases as the elapsed time from frying increases. That is, fried chicken tends to decrease in size over time.

Here, the value of the "degraded flavor" shown in FIG. 2 increases as the elapsed time increases. Also, as described above, the average area decreases as the elapsed time increases. Therefore, there is a correlation between the reduction in fried food area (average area) and the reduced flavor of fried food. Therefore, the fried food sales promotion control device 4 uses the area, that is, the size, of the fried food X displayed in the hot showcase 1 as an index (state data) indicating the state of the fried food X to determine the size of the fried food X. The remaining time of fried food X can also be predicted based on the change in .

In this case, in the fried food sales promotion control device 4, the data acquisition unit 41 acquires the surface image of the fried food X photographed by each camera 5, and then the analysis unit 43 acquires the individual extracted from each surface image. The area of each fried food X is calculated by analyzing the individual images of the fried food X.

<Second embodiment>
Next, a fried food sales promotion control device 4A according to a second embodiment of the present invention will be described with reference to FIGS. 16 to 25. FIG. In this embodiment, the same reference numerals are assigned to the same components as those described in the fried food sales promotion system 3 and the fried food sales promotion control device 4 according to the first embodiment, and the description thereof will be omitted. The same applies to the third to fifth embodiments below.

In the fried food sales promotion control device 4 according to the first embodiment, the color (color component) of the fried food X is mainly used as the state data to predict the remaining time of the fried food X, but the fried food sales promotion control according to the present embodiment In the device 4A, a case will be described in which an index indicating the state of the fried food X other than the color is used as the state data.

FIG. 16 is a graph showing changes in the rate of change in weight of the displayed fried chicken with respect to elapsed time when fried chicken is displayed as an example of the fried food X in the hot showcase 1 .

As shown in FIG. 16, the fried chicken being displayed in the hot showcase 1 increases in the total weight of the fried chicken as the elapsed time from the time of frying (storage time in the hot showcase 1) increases. The rate of change (%) has decreased. Note that, as shown in FIG. 16, the weight change rate (%) is a negative change rate. That is, FIG. 16 shows that the weight of the whole fried chicken decreases as time passes. That is, according to the graph of FIG. 16, the moisture contained in the fried chicken evaporates over time, reducing the total amount of moisture contained in the fried chicken, resulting in a change in the weight of the fried chicken. The rate is negative.

Here, the value of the "degraded flavor" shown in FIG. 2 increases as the elapsed time increases. Also, as described above, the weight of the fried food decreases as the elapsed time increases. Therefore, there is a correlation between the overall weight reduction of the fried food and the reduced flavor of the fried food. Therefore, the fried food sales promotion control device 4 uses the weight (total weight) of the fried food X displayed in the hot showcase 1 as the state data of the fried food X, so that based on the change in the weight of the fried food X It is also possible to predict the remaining time of the fried food X.

FIG. 17 is a top plan view of the bottom surface of a predetermined stage in the hot showcase 1, showing a plurality of weight measurement sensors 61 spread over each tray 2. FIG.

To measure the weight of the fried foods X displayed in the hot showcase 1, for example, a mat-type weight measurement sensor 61 shown in FIG. 17 is used. The weight measurement sensor 61 may be a sensor that detects the weight of the fried food X itself, or may be a sensor that detects the total amount of water contained in the fried food X by detecting a change in dielectric constant. .

A plurality of weight measurement sensors 61 are attached so as to correspond to the position of each fried food X displayed in the hot showcase 1 . In FIG. 17, nine weight measurement sensors 61 are laid out in a matrix on each tray 2, and a maximum of nine fried foods X can be displayed on one tray 2 at one time. It is possible to measure the thickness individually.

Note that even when a plurality of fried foods X are arranged within a measurable range of one weight measurement sensor 61, if the weight of each fried food X can be measured, the number and arrangement of the weight measurement sensors 61 are determined. You don't have to limit the location.

Fig. 18 is a graph showing changes in the stickiness of fried food batter versus elapsed time obtained by sensory evaluation. In FIG. 18, the horizontal axis indicates the elapsed time from the rising time of the fried food to be evaluated, and the vertical axis indicates the index value of the "stickiness of the batter" of the fried food.

As shown in FIG. 18, for fried foods, the index indicating the stickiness of the batter generally increases as the time elapsed from the time of frying increases. In other words, fried food tends to become more sticky as time passes. This "stickiness of the batter" is caused by the amount of water contained in the batter of the fried food, and when the amount of water increases, the batter becomes sticky.

Here, FIG. 19A is a graph showing temporal changes in the amount of moisture contained in the batter on the surface side of fried chicken being displayed in the hot showcase 1, and FIG. It is a graph which shows the time change of the water content contained in the coating of the back side of fried chicken. The “back side” of the fried chicken is the side that contacts the tray 2 , and the “front side” is the opposite side, that is, the upper side when displayed in the hot showcase 1 .

As shown in FIGS. 19A and 19B, the fried chicken being displayed in the hot showcase 1 is more stable than when fried (elapsed time in FIGS. 19A and 19B = 0 hours) on both the front side and the back side. As the elapsed time increases (for example, elapsed time=2 hours in FIGS. 19A and 19B), the percentage (%) of moisture content in the batter increases.

In addition, FIG. 20A is a graph showing changes over time in the crispness of the fried chicken being displayed in the hot showcase 1, and FIG. It is a graph which shows a change.

The crispness in FIG. 20A tends to be crispier as the sensory evaluation score shown on the vertical axis increases, and the stickiness of the coating in FIG. The higher the sensory evaluation score, the stronger the stickiness tends to be.

As shown in FIG. 20A, the fried chicken being displayed in the hot showcase 1 has a longer time (elapsed time = 2 hours in FIG. 20A) from the time of frying (elapsed time = 0 hours). The index indicating crispness decreases.

On the other hand, as shown in FIG. 20B, the fried chicken displayed in the hot showcase 1 is fried (elapsed time = 0 hours) as the elapsed time increases (elapsed time = 2 hours in FIG. 20B). The index indicating the stickiness of the clothes is increased. That is, the fried chicken batter becomes less crispy and more sticky as the time elapsed after frying increases.

For fried chicken, the moisture contained in the ingredients inside (chicken) moves to the batter until the specified time has passed since the time of frying. Due to this movement of water, the amount of water contained in the batter increases according to the elapsed time from the time of frying. Therefore, fried chicken is not suitable for sale as a food, such as losing a suitable texture (decreased crispness) and becoming an unsuitable texture (increased stickiness of clothes). It can be said that this is due to the amount of moisture contained in the chicken batter.

Here, the "degraded flavor" shown in FIG. 2 also increases as the elapsed time increases. Also, as described above, the moisture content of the batter increases as the elapsed time increases. Therefore, there is a correlation between an increase in the moisture content of the batter and a decrease in the flavor of the fried food. Therefore, the fried food sales promotion control device 4 uses the moisture content of the batter of the fried food X displayed in the hot showcase 1 as the state data of the fried food X, so that based on the change in the moisture content of the batter of the fried food X The remaining time of fried food X can also be predicted.

FIG. 21 is a top view of a portion corresponding to the top surface of a predetermined stage in the hot showcase 1 as seen from below, showing a plurality of near-infrared sensors 63 attached to the top surface portion. .

The amount of water contained in the batter of fried food X can be detected using, for example, the near-infrared sensor 63 shown in FIG. The near-infrared sensor 63 reflects near-infrared light on the fried food X and detects a change in absorptance at a specific wavelength corresponding to the amount of water contained in the fried food X. can be measured. Therefore, by reflecting the near-infrared light emitted from the near-infrared sensor 63 on the batter of the fried food X, the amount of moisture contained in the batter of the fried food X can be measured. It should be noted that the overall moisture content in the fried food X described above can also be measured using this near-infrared sensor 63 in addition to the weight measurement sensor 61 (see FIG. 17).

In addition, the near-infrared sensor 63 can reflect near-infrared light on the fried food X and detect changes in absorptance of a specific wavelength depending on the content of the component contained in the fried food X. In addition to the amount of water contained in the fried food X, the acid value, anisidine value, carbonyl value, peroxide value, iodine value, and polar compound amount of the fried food X can be measured.

Similar to the weight measurement sensor 61, the near-infrared sensor 63 is also installed so as to correspond to the positions of the fried foods X displayed in the hot showcase 1, and measures the moisture content of the batter of each fried food X and It is installed so that the positional relationship is such that the acid value can be detected.

FIG. 22 is a graph showing the transition of the odor intensity of fried food with respect to the elapsed time obtained by sensory evaluation. In FIG. 22, the horizontal axis indicates the elapsed time from the rising time of the fried food to be evaluated, and the vertical axis indicates the indexed value of the "strength of odor" of the fried food.

Here, "smell" includes, in a general sense, the pleasant odor of fried food and the unfavorable odor of fried food. Hereinafter, the desirable odor of fried food is referred to as "aroma", and the undesirable odor of fried food is referred to as "odor". Therefore, FIG. 22 shows the change over time of the odor containing both fragrance and odor (the odor as a whole in which the fragrance and odor are not distinguished).

As shown in FIG. 22, fried foods generally have a lower odor intensity, which is an indicator of odor strength, as the time elapsed from the time of frying increases. In other words, fried foods tend to have less odor as time passes. This means that the overall volatile content of the fried food is reduced.

As the odor of fried foods fades, the balance between aroma and odor changes, and the proportion of substances that cause odor, such as aldehydes and ketones, increases. will sense the odor from the fried food. Aldehyde-based or ketone-based substances are generated when the edible oil used to fry the food deteriorates and the fatty acids contained in the edible oil are decomposed.

Here, the value of the "degraded flavor" shown in FIG. 2 increases as the elapsed time increases. In general, when the odor intensity of the fried food decreases and the ratio of the odor to the fried food increases, it can be said that the flavor of the fried food has decreased. Therefore, there is a correlation between a decrease in the volatile components of the fried food, or an increase in the ratio of substances such as aldehydes or ketones due to changes in the volatile component composition of the fried food, and a decrease in the flavor of the fried food.

Therefore, the fried food sales promotion control device 4A uses the volatile component or the volatile component composition of the fried food X displayed in the hot showcase 1 as the state data of the fried food X to obtain the volatile component or the volatile component composition of the fried food X. The remaining time of the fried food X can also be predicted based on the change in the component composition.

FIG. 23 is a top view of a portion corresponding to the top surface of a predetermined stage in the hot showcase 1, as viewed from below, showing a plurality of odor sensors 62 attached to the top surface portion. In addition, in FIG. 23, the outer frame of the tray 2 is indicated by a two-dot chain line.

In general, odors arise from the fried food X. Therefore, as shown in FIG. It is attached to a portion corresponding to the top surface of 12 and 13 .

Like the weight measurement sensor 61 and the near-infrared sensor 63, the odor sensor 62 is installed so as to correspond to the positions of the fried foods X displayed in the hot showcase 1, and is capable of detecting the odor of each fried food X. It is attached so as to have a positional relationship.

The odor sensor 62 may be a sensor capable of detecting both the aroma and odor (odor) of the fried food X, a sensor capable of detecting the aroma of the fried food X, or a sensor capable of detecting the odor of the fried food X. There may be.

For example, if the odor sensor 62 detects the odor of the fried food X, the fried food sales promotion control device 4A discards odors containing components such as aldehydes or ketones when predicting the remaining time of the fried food X. Use the reference value. Then, the relationship between the elapsed time from the time of frying the fried food X and the strength of the odor shows that the longer the time elapsed from the time of frying the fried food X, the more the strength of the odor tends to increase. is reversed.

The sensor specifications of the odor sensor 62 are not particularly limited. A semiconductor gas sensor that detects a gas concentration by changing the resistance value of a semiconductor, an infrared gas sensor, an electrochemical gas sensor, a catalytic combustion gas sensor, a biosensor, or the like can be applied.

Next, the functions of the fried food sales promotion control device 4A according to this embodiment will be described with reference to FIGS. 24 and 25. FIG. In the following description, the fried food sales promotion control device 4A uses the moisture content of the fried food X as the state data. In the case where the composition, acid value, anisidine value, carbonyl value, peroxide value, iodine value, and amount of polar compounds) are used as state data, the explanation thereof is omitted.

FIG. 24 is a functional block diagram showing the functions of the fried food sales promotion control device 4A according to the second embodiment. FIG. 25 is a flow chart showing the flow of processing executed by the fried food sales promotion control device 4A according to the second embodiment.

As shown in FIG. 24, the fried food sales promotion control device 4A according to the present embodiment includes, for example, a data acquisition unit 41A, a type identification unit 42A, a remaining time prediction unit 44A, a storage unit 45A, and a discount rate calculation unit. 46A and a notification unit 47A.

The data acquisition unit 41A acquires surface images of a plurality of fried foods X for each tray 2 photographed by each camera 5, and acquires the weight of each fried food X detected by the weight measurement sensor 61.

The type identifying unit 42A identifies the type of each fried food X from the extracted individual image of each fried food X, similar to the type identifying unit 42 in the first embodiment.

The remaining time prediction unit 44A is based on the weight of each fried food X acquired by the data acquisition unit 41A, the type of each fried food X, and the discard standard value set for each type of each fried food X. , to predict the remaining time of each fried item X. The discard standard value in this embodiment is a discard standard value related to the weight of the fried food X, and is preset and stored in the storage unit 45A.

The discount rate calculation unit 46A calculates the discount rate for the fried food X to be discounted based on the remaining time predicted by the remaining time prediction unit 44A, like the discount rate calculation unit 46 in the first embodiment.

The notification unit 47A, like the notification unit 47 in the first embodiment, transmits a notification signal for notifying the discount information of the fried food X including the discount rate calculated by the discount rate calculation unit 46A to the monitor 312 or the mobile terminal. 312A (notification device).

As shown in FIG. 25, in the fried food sales promotion control device 4A, first, the data acquisition unit 41A acquires the surface image of the fried food X for each tray 2 output from each camera 5 (step S421).

Next, an individual image of each fried food X is extracted from each surface image acquired in step S421 (step S422), and the type specifying unit 42 determines the type of each fried food X from each individual image extracted in step S422. is specified (step S423).

Also in this embodiment, the fried food sales promotion control device 4A does not necessarily include the type specifying step (step S423) for specifying the type of each fried food X. In that case, steps S421 and The processing in step S422 may be omitted.

Subsequently, the data acquisition unit 41A acquires the weight of each fried food X output from each weight measurement sensor 61 as state data (step S424).

Then, the remaining time prediction unit 44A compares the weight acquired in step S424 with the discarding standard value related to the type specified in step S423 for each piece of fried food X, and determines the remaining time until the point of discarding. Predict (step S425).

If it is determined that the discounted fried food X is included in the hot showcase 1 based on each remaining time predicted in step S425 (step S426/YES), the discount rate calculation unit 46A A discount rate corresponding to the remaining time is calculated for the target fried food X (step S427).

Next, the notification unit 47A outputs a notification signal for notifying information including the discount rate calculated in step S427 to the monitor 312 or the mobile terminal 312A (step S428), and the processing in the fried food sales promotion control device 4A is completed. finish.

On the other hand, if it is determined in step S426 that the discounted fried food X is not included in the hot showcase 1 (step S426/NO), the process returns to step S421 to repeat the process.

Also in the fried food sales promotion control device 4A according to the present embodiment, it is possible to achieve the same actions and effects as those of the first embodiment. Furthermore, in the present embodiment, the fried food sales promotion control device 4A does not need to include an "analyzer", and can predict the remaining time of fried food X using the state data acquired by the data acquisition unit 41A. , the number of operations to be performed can be reduced.

<Third Embodiment>
Next, a fried food sales promotion system 3A according to a third embodiment of the present invention will be described with reference to FIG.

FIG. 26 is a functional block diagram showing the functions of the fried food sales promotion system 3A according to the third embodiment.

A fried food sales promotion system 3A according to the present embodiment includes a fried food sales promotion control device 4B that controls the sales promotion of fried food X, and a learning device 7 that generates a learned model capable of predicting the remaining time of fried food X. , is composed of That is, in this embodiment, unlike the configuration of the fried food sales promotion system 3 according to the first embodiment, the function of generating a learned model capable of predicting the remaining time of the fried food X is provided by the fried food sales promotion control device 4B. The learning device 7, which is a device different from the .

The fried food sales promotion control device 4B has the same configuration as the fried food sales promotion control device 4A according to the second embodiment described above, and includes a data acquisition unit 41B, a type identification unit 42B, a remaining time prediction unit 44B, and a storage unit. 45B, a discount rate calculation unit 46B, a notification unit 47B, and a communication unit 40 are further included.

The data acquisition unit 41B, the type identification unit 42B, the remaining time prediction unit 44B, the storage unit 45B, the discount rate calculation unit 46B, and the notification unit 47B all have the same functions as in the fried food sales promotion control device 4A. has the function of Therefore, the fried food sales promotion control device 4B can realize the same processing flow as the processing flow in the fried food sales promotion control device 4A.

The learning device 7 includes a communication unit 70, a data acquisition unit 71, a learned model generation unit 73, and an update unit 74.

The communication unit 40 of the fried food sales promotion control device 4B and the communication unit 70 of the learning device 7 provide functions including an interface for mutual information communication via the communication network N.

The data acquisition unit 71 of the learning device 7 obtains the remaining time of the fried food X predicted by the remaining time prediction unit 44B of the fried food sales promotion control device 4B and the data acquisition unit 41 of the fried food sales promotion control device 4B. Temporal change data (state data), that is, data indicating the tendency of temporal change until the fried food X is discarded, is acquired via the communication unit 70 .

The learned model generation unit 73 performs machine learning and regression analysis using teacher data including time-varying data acquired by the data acquisition unit 71 to generate a learned model.

Based on the learned model generated by the learned model generation unit 73, the updating unit 74 updates the fried food X disposal criteria stored in the storage unit 45B of the fried food sales promotion control device 4B.

<Fourth Embodiment>
Next, a fried food sales promotion system according to a fourth embodiment of the present invention will be described with reference to FIGS. 27-29.

FIG. 27 is a plan view of the inside of the hot showcase 1A to which the fried food sales promotion system according to the fourth embodiment is applied, viewed from the rear side.

In this embodiment, three cameras 5A are provided as photographing devices for photographing images including a plurality of fried foods X displayed on each

shelf

11, 12, 13 in the hot showcase 1A. In FIG. 27, the camera 5A is arranged on one end side of the top surface of each

shelf

11, 12, 13, but it is possible to photograph an image including all of the plurality of fried foods X displayed in the hot showcase 1A. If possible, there are no particular restrictions on the number of cameras 5A and their mounting positions.

If it is not possible to acquire an image including all of the plurality of fried foods X displayed on each

shelf

11, 12, 13 in setting the angle of view of the camera 5A, a plurality of cameras 5A are installed as shown in FIG. By doing so, it is sufficient that an overall image including all of the plurality of fried foods X and surface images of the individual fried foods X can be photographed. Also, for example, by varying the angle of view setting with one camera 5A, similarly, the entire image including all of the plurality of fried foods X and the surface image of each fried food X can be photographed. You may

A video camera capable of capturing moving images is used as the camera 5A, and images including individual movements of the fried foods X in the hot showcase 1A are captured. Individual fried foods X displayed in the hot showcase 1A are not always placed in the same position as they were placed immediately after being fried.

For example, immediately after being fried, the fried food X is placed on the tray 2 arranged on one of the

shelves

11, 12, 13 in the hot showcase 1A. The position of the fried food X may be changed in the tray 2 of the tray 2, or the fried food X may be moved to a tray 2 different from the tray 2 on which the fried food X was initially placed.

Therefore, the camera 5A shoots the state of each

shelf

11, 12, 13 as a moving image, thereby shooting an image including the movement (positional movement, etc.) of each fried food X included in the moving image as a whole image. Based on this animation, by executing the processing in the fried food sales promotion control device 8, which will be described later, even if the individual fried food X placed on each

shelf

11, 12, 13 is moved, the individual fried food X can be identified. By tracking, the passage of time can be obtained individually.

Note that the camera 5A does not necessarily have to be a video camera capable of shooting moving images, and may be any camera capable of acquiring images of the fried food X continuously over time. For example, it may be a camera such as a still camera that can take only still images. In that case, it is sufficient to be able to continuously shoot to the extent that individual movements of the fried foods X in the hot showcase 1A can be acquired as image data.

Next, the functions of the fried food sales promotion control device 8 according to this embodiment will be described with reference to FIGS. 28 and 29. FIG.

FIG. 28 is a functional block diagram showing the functions of the fried food sales promotion control device 8 according to the fourth embodiment. FIG. 29 is a flow chart showing the flow of processing executed by the fried food sales promotion control device 8 according to the fourth embodiment.

The fried food sales promotion control device 8 according to the present embodiment includes, for example, an image acquisition unit 81, an identification information generation unit 82, an individual surface image management unit 83, a time measurement unit 84, a remaining time prediction unit 85, and a storage A unit 86 , a discount rate calculation unit 87 , and a notification unit 88 are included.

The image acquisition unit 81 is a data acquisition unit (the

data acquisition units

41, 41A, 41B), an image including a plurality of deep-fried foods X displayed in the hot showcase 1A captured by the camera 5A is acquired as temporal change data.

The identification information generation unit 82 generates identification information that individually identifies the surface images of the plurality of fried foods X displayed in the hot showcase 1A. Note that the identification information generated by the identification information generating section 82 may be stored in the storage section 86 .

The individual surface image management unit 83 manages the identification information generated by the identification information generation unit 82 in association with the surface image of each fried food X included in the image acquired by the image acquisition unit 81 . In the present embodiment, the individual surface image management unit 83 associates, in addition to the identification information generated by the identification information generation unit 82, type information specifying the type of the fried food X with the surface image of each fried food X. to manage. Note that the type information of the fried food X is stored in the storage unit 86 .

The time measurement unit 84 measures the time that the surface image associated with the identification information by the individual surface image management unit 83 is included in the image acquired by the image acquisition unit 81 .

The remaining time prediction unit 85 calculates the time measured by the time measurement unit 84 and the elapsed time (hereinafter referred to as “reference time”) for discarding the fried food X set in advance as a criterion for determining the discarding of the fried food X. , the remaining time of each fried food X is predicted. Note that the reference time is stored in the storage unit 86 .

The discount rate calculation unit 87 has the same function as the discount

rate calculation units

46, 46A, and 46B already described, and based on the remaining time of each fried food X predicted by the remaining time prediction unit 85, A discount rate for the fixed price of fried food X is calculated.

The notification unit 88 has the same function as the already described

notification units

47, 47A, and 47B, and provides information including the discount rate calculated by the discount rate calculation unit 87 (discount information related to the discounted fried food X). A notification signal for notification is output to the monitor 312 and the portable terminal 312A (notification device).

As shown in FIG. 29, in the fried food sales promotion control device 8, first, the image acquisition unit 81 acquires images (displayed in the hot showcase 1) captured by each camera 5 in a capturing step (corresponding to a detecting step). image including a plurality of fried foods X of (step S801).

Subsequently, the identification information generation unit 82 generates identification information for individually identifying the surface images of the plurality of fried foods X included in the image acquired in step S801 (step S802; identification information generation step).

Next, the individual surface image management unit 83 generates identification information of each fried food X, extracts a surface image of each fried food X from the image acquired in step S801, and for the extracted surface image, The identification information generated in step S802 and the type information read from the storage unit 86 are associated and managed (step S803; individual surface image management step).

Next, the time measurement unit 84 measures the time during which each surface image associated with the identification information and type information in step S803 is included in the image acquired in step S801 (step S804; time measurement step). .

Next, the remaining time prediction unit 85 compares the time measured in step S804 with the reference time stored in the storage unit 86, and predicts the remaining time of each fried food X (step S805; remaining time prediction step).

If it is determined that the discounted fried food X is included in the hot showcase 1A based on each remaining time predicted in step S805 (step S806/YES), the discount rate calculation unit 87 A discount rate corresponding to the remaining time is calculated for the target fried food X (step S807; discount rate calculation step).

Next, the notification unit 88 outputs a notification signal for notifying information including the discount rate calculated in step S807 to the monitor 312 and the mobile terminal 312A (step S808). As a result, the monitor 312 and the portable terminal 312A notify the discount information about the fried food X to be discounted (notification step).

On the other hand, if it is determined in step S806 that the discounted fried food X is not included in the hot showcase 1 (step S806/NO), the fried food whose surface image is out of the image acquired in step S801. It is determined whether or not there is X (step S809).

For the fried food X whose surface image is out of the acquired image in step S809 (step S809/YES), the processing in the fried food sales promotion control device 8 ends. On the other hand, the fried food X whose surface image is not out of the acquired image in step S809 (step S809 / NO), that is, the fried food that is continuously displayed in the hot showcase 1 and has not reached the remaining time to be discounted For X, return to step S801 and repeat the process.

As described above, in this embodiment, a plurality of fried foods X displayed in the hot showcase 1A are continuously or intermittently photographed by the camera 5A, and individual fried foods X are identified based on the photographed images. Therefore, even if the fried foods X move within the hot showcase 1A, it is possible to accurately determine the disposal point of each fried foods X and accurately predict the remaining time until the disposal point. Become. As a result, the burden on the employees of the store 31 who have been manually recording the time is reduced, and sales promotion can be easily performed by discounting the plurality of fried foods X in the hot showcase 1A.

<Fifth Embodiment>
Next, a fried food sales promotion control device according to a fifth embodiment of the present invention will be described with reference to FIG.

FIG. 30 is a flow chart showing the flow of processing executed by the fried food sales promotion control device according to the fifth embodiment.

In the fried food sales promotion control device according to the present embodiment, unlike the fried food sales promotion control device 8 according to the fourth embodiment, the fried food X is once taken out from the hot showcase 1A and returned to the hot showcase 1A. Including case handling.

First, the image acquisition unit 81 acquires an image including a plurality of fried foods X displayed in the hot showcase 1A (step S821), similar to step S801 in the fourth embodiment.

Next, based on the surface image-related information stored in the storage unit 45A, the fried food sales promotion control device 8 already stores the surface images of the individual fried foods X included in the image acquired this time in the storage unit 86. In other words, it is determined whether it is included in the image acquired in the past step S821 (step S822).

Here, the "surface image related information" is information that can be acquired from the surface image of each fried food X extracted from the image taken by the camera 5A. Specifically, the color tone (color composition), size, and shape. Therefore, the "surface image-related information" may include the surface image itself, or may include information in which information such as feature points included in the surface image is structured. In any case, the "surface image related information" is an information group that can determine whether or not the elapsed time was measured in the past for the fried food X included in the image (acquired image) acquired in step S821. If it is

If it is determined in step S822 that the surface image to be determined is not associated with the surface image-related information stored in the storage unit 86 (step S822/NO), the identification information and type information are newly associated (step S823). Note that there may be a plurality of surface images to be determined in step S822. In that case, the determination in step S822 is performed for each of the plurality of surface images.

On the other hand, in step S822, the surface image to be determined is associated with the surface image-related information stored in the storage unit 86. If it is the surface image of the fried food X determined to be included in the acquired image in step S821 (step S822 / YES), the identification information and type information stored in the storage unit 86 for the corresponding surface image are associated (step S824).

Next, based on the identification information and type information associated in step S823 or step S824, the time measurement unit 84 measures the time during which the surface image is included in the acquired image from the measurement time serving as the starting point (step S825). ). Specifically, when the process proceeds from step S823 to step S825, the time measurement unit 84 newly measures the time during which the surface image is included in the acquired image starting from the current time. On the other hand, when the process proceeds from step S824 to step S825, the time measuring unit 84 measures the time during which the surface image is included in the acquired image, starting from the measurement time immediately before the surface image is removed from the acquired screen in the previous time. to resume. That is, for the fried food X once taken out of the hot showcase 1A and returned to the hot showcase 1A again, the measurement of the time during which the surface image is included in the acquired image is restarted.

Then, the remaining time prediction unit 85 predicts the remaining time of each fried food X based on the time measured in step S825 and the reference time stored in the storage unit 86 (step S826).

Subsequently, when it is determined that the fried food X to be discounted is included in the hot showcase 1A based on each remaining time predicted in step S826 (step S827/YES), the discount rate calculation unit 87 calculates the discount rate according to the remaining time for the target fried food X, as in S807 in the fourth embodiment (step S828).

Next, similarly to step S808 in the fourth embodiment, the notification unit 88 outputs a notification signal for notifying information including the discount rate calculated in step S828 to the monitor 312 or the portable terminal 312A (step S829). ).

On the other hand, if it is determined in step S827 that the discounted fried food X is not included in the hot showcase 1 (step S827/NO), the fried food whose surface image is out of the image acquired in step S821. It is determined whether or not there is X (step S830).

Regarding the fried food X whose surface image is out of the acquired image in step S830 (step S830/YES), the timing at which it is determined that the surface image of the fried food X is not included in the acquired image, that is, "YES" in step S830 The measurement time up to the determined timing, the identification information, the type information, and the surface image-related information assigned to the surface image of the fried food X are stored (step S831), and the process returns to step S821.

That is, the surface image-related information used in step S822, the identification information and type information associated with each surface image in step S824, and the measurement time immediately before the surface image is removed from the acquisition screen used in step S825 are all This is the information stored in the storage unit 86 in step S831.

On the other hand, the fried food X whose surface image is not out of the acquired image in step S830 (step S830 / NO), that is, the fried food that is continuously displayed in the hot showcase 1 and has not reached the remaining time to be discounted For X, the process returns to step S821 to repeat the process.

In this way, by including the processing when the fried food X is once taken out of the hot showcase 1 and returned to the hot showcase 1 again in the processing executed by the fried food sales promotion control device, the hot It is possible to accurately set the disposal times of the plurality of fried foods X displayed in the showcase 1A, and to predict the remaining time of the fried foods X in the remaining time prediction unit 85 more accurately.

In this embodiment, the fried food sales promotion control device automatically processes the case where the fried food X is taken out of the hot showcase 1 and returned to the hot showcase 1 again. For example, when the fried food X is taken out from the hot showcase 1, the store employee manually inputs the identification information of the fried food X to the input terminal or the like, and the fried food sales promotion control device It may be determined whether or not the fried food X is new using the input identification information.

The embodiment of the present invention has been described above. In addition, the present invention is not limited to the above-described embodiments, and includes various modifications. For example, the above-described embodiments have been described in detail in order to explain the present invention in an easy-to-understand manner, and are not necessarily limited to those having all the configurations described. Further, part of the configuration of this embodiment can be replaced with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of this embodiment. Furthermore, it is possible to add, delete, or replace a part of the configuration of this embodiment with another configuration.

For example, in the above embodiment, the

hot showcase

1, 1A provided with three

shelves

11, 12, 13 is described as one aspect of the display shelf on which the fried foods X are displayed. For example, a single tray is also included, and the form is not limited as long as the fried food X can be displayed.

Further, in the above embodiment, a system for promoting sales of fried foods X displayed in the

hot showcases

1 and 1A has been described as the fried foods sales promotion system. Any system that promotes sales of cooked foods, such as a system that promotes sales of steamed Chinese buns, may be used.

1, 1A: Hot showcase (display shelf)
3, 3A: Fried

food promotion system

4, 4A, 4B, 8: Fried food

promotion control device

5, 5A: Camera (data detection device, state sensor, photographing device)
7: Learning device 61: Weight measurement sensor (data detection device, state sensor)
62: Odor sensor (data detector, state sensor)
63: Near-infrared sensor (data detector, state sensor)
41, 41A, 41B:

data acquisition units

42, 42A, 42B;

type identification units

44, 44A, 44B, 85: remaining

time prediction units

46, 46A, 46B, 87: discount

rate calculation units

47, 47A, 47B, 88: Notification unit 81; image acquisition unit (data acquisition unit)
82: Identification information generation unit 83: Individual surface image management unit 84: Time measurement unit 312: Monitor (notification device)
312A: Mobile terminal (notification device)
X: Fried food

Claims

In a food sales promotion control device for controlling sales promotion of cooked food displayed on a display shelf,
a data acquisition unit that acquires time-dependent change data indicating the time-dependent change of the food displayed on the display shelf;
Disposal of the food displayed on the display shelf based on the temporal change data acquired by the data acquisition unit and a disposal criterion set in advance as a criterion for determining the point of disposal of the food. a remaining time prediction unit that predicts the remaining time up to the point in time;
a discount rate calculation unit that calculates a discount rate for the list price of the food based on the remaining time predicted by the remaining time prediction unit;
A food sales promotion control device, comprising: a notification unit for outputting a notification signal for notification of information including the discount rate calculated by the discount rate calculation unit to a notification device.
The food promotion control device according to claim 1,
The data acquisition unit
Acquiring state data output from a state sensor that detects the state of the food displayed on the display shelf as the time change data,
The state data acquired by the data acquisition unit is
Color, size, weight, moisture content, volatile component content, volatile component composition, acid value, anisidine value, carbonyl value, peroxide value, iodine value, and polarity of the food displayed on the display shelf Data relating to at least one of the indicators including the amount of the compound,
The discard criterion is
A food sales promotion control apparatus, wherein a disposal standard value indicating a standard for disposal of the food is set for each of the indexes.
In the food sales promotion control device according to claim 2,
further comprising a type identification unit that identifies the type of the food displayed on the display shelf,
The remaining time prediction unit
Based on the state data acquired by the data acquisition unit, the type of food specified by the type specifying unit, and the disposal criteria set for each type of food, the display A food sales promotion control apparatus characterized by predicting the remaining time until the time of disposal of the food displayed on the shelf.
The food promotion control device according to claim 1,
an image acquisition unit that corresponds to the data acquisition unit and acquires an image including a plurality of the foods displayed on the display shelf as the time change data;
an identification information generation unit that generates identification information that individually identifies surface images of the plurality of foods displayed on the display shelf;
an individual surface image management unit that associates and manages the identification information generated by the identification information generation unit with the surface image included in the image acquired by the image acquisition unit;
a time measurement unit that measures the time that the surface image associated with the identification information is included in the image in the individual surface image management unit,
The remaining time prediction unit
Based on the time measured by the time measurement unit and the elapsed time when discarding the food set in advance as the discard determination criterion, the number of food items displayed on the display shelf is determined. A food sales promotion control device for predicting each of the remaining times up to the point of disposal.
In the food sales promotion control device according to claim 4,
The individual surface image management unit,
In addition to the identification information, type information specifying the type of the food is managed in association with the surface image,
The remaining time prediction unit
Based on the time measured by the time measurement unit and the elapsed time based on the type information associated by the individual surface image management unit, the plurality of the displayed on the display shelf A food sales promotion control device for predicting the remaining time until the food is discarded.
The food promotion control device according to claim 1,
The remaining time prediction unit
The food displayed on the display shelf reaches the point of disposal by using a trained model generated using teacher data including data indicating the tendency of change over time until the food reaches the disposal criteria. A food sales promotion control device that predicts the remaining time until.
In a food sales promotion system for promoting sales of cooked food displayed on display shelves,
a data detection device for detecting temporal change data indicating the temporal change of the food displayed on the display shelf;
a food sales promotion control device for controlling sales promotion of the food by calculating a discount rate for the fixed price of the food based on the time-varying data output from the data detection device;
a notification device that notifies information including the discount rate calculated by the food sales promotion control device;
The food promotion control device includes:
Acquiring the time-varying data detected by the data detection device,
Based on the acquired change data over time and a disposal criterion set in advance as a criterion for determining the disposal point of the food, the remaining amount of the food displayed on the display shelf until the point of disposal is determined. predict the time
Calculate the discount rate for the list price of the food based on the predicted remaining time,
A food sales promotion system, wherein a notification signal for notifying information including the calculated discount rate is output to the notification device.
In the food sales promotion system according to claim 7,
The data detection device is
A state sensor that detects the state of the food displayed on the display shelf,
The food promotion control device includes:
Acquiring state data output from the state sensor as the time-varying data,
The state data is
Color, size, weight, moisture content, volatile component content, volatile component composition, acid value, anisidine value, carbonyl value, peroxide value, iodine value, and polarity of the food displayed on the display shelf Data relating to at least one of the indicators including the amount of the compound,
The discard criterion is
A food sales promotion system, wherein a disposal standard value indicating a standard for disposal of the food is set for each of the indicators.
In the food sales promotion system according to claim 8,
The food promotion control device includes:
Furthermore, specifying the type of the food displayed on the display shelf,
At the time of disposal of the food displayed on the display shelf based on the acquired state data, the identified food type, and the disposal criteria set for each food type A food sales promotion system characterized by predicting the remaining time until the end.
In the food sales promotion system according to claim 7,
The data detection device is
A photographing device for photographing an image including the plurality of foods displayed on the display shelf,
The food promotion control device includes:
Acquiring the image output from the imaging device as the temporal change data,
generating identification information that individually identifies surface images of the plurality of foods displayed on the display shelf;
managing the generated identification information in association with the surface image included in the acquired image;
measuring each time the surface image associated with the identification information is included in the image;
Until the point of disposal of the plurality of foods displayed on the display shelf based on each measured time and the elapsed time when discarding the food set in advance as the discard judgment criteria. A food sales promotion system characterized by predicting the remaining time of each.
In the food sales promotion system according to claim 10,
The food promotion control device includes:
In addition to the identification information, type information specifying the type of the food is managed in association with the surface image,
Based on each measured time and the elapsed time based on the associated type information, the remaining time until the point of disposal of the plurality of food items displayed on the display shelf is calculated. A food sales promotion system characterized by predicting.
In the food sales promotion system according to claim 7,
The notification device is
A food seller characterized by being installed in a store equipped with said display shelf and notifying information including said discount rate by at least one notifying means out of notifying means including voice, text, color and light. promotion system.
In the food sales promotion system according to claim 7,
The notification device is
A mobile terminal owned by a customer who can use the store equipped with the display shelf,
A food sales promotion system, wherein the discount rate is notified to the customer by at least one of notification means including voice, text, color, and light.
A food sales promotion method for promoting sales of cooked food displayed on a display shelf, comprising:
A data detection device for detecting temporal change data indicating the temporal change of the food displayed on the display shelf; and calculating a discount rate for the fixed price of the food based on the temporal change data output from the data detection device. Using a food sales promotion control device that controls the sales promotion of the food and a notification device that notifies information including the discount rate calculated by the food sales promotion control device,
a data detection step in which the data detection device detects temporal change data indicating the temporal change of the food displayed on the display shelf;
The food sales promotion control device controls the display shelf based on the time-varying data detected in the data detection step and a disposal criterion set in advance as a criterion for determining when to discard the food. a remaining time prediction step of predicting the remaining time until the displayed food is discarded;
a discount rate calculation step in which the food sales promotion control device calculates a discount rate for the fixed price of the food based on the remaining time predicted in the remaining time prediction step;
and a notification step in which the notification device notifies information including the discount rate calculated in the discount rate calculation step.
In the food sales promotion method according to claim 14,
In the data detection step,
The data detection device detects, as the temporal change data, the color, size, weight, moisture content, volatile component content, volatile component composition, acid value, anisidine value, Detecting data related to at least one of indicators including carbonyl value, peroxide value, iodine value, and amount of polar compounds,
In the remaining time prediction step,
The food sales promotion control device uses the disposal standard value indicating the standard of disposal of the food set for each of the indicators as the disposal determination standard, and controls the disposal of the food displayed on the display shelf until the point of disposal. A food sales promotion method characterized by predicting the remaining time.
In the food sales promotion method according to claim 14,
a photographing step corresponding to the data detecting step, in which a photographing device photographs an image including the plurality of foods displayed on the display shelf;
an identification information generating step in which the food sales promotion control device generates identification information for individually identifying surface images of the plurality of foods displayed on the display shelf;
An individual surface image management step in which the food sales promotion control device manages the surface image included in the image acquired in the photographing step in association with the identification information generated in the identification information generation step. and,
a time measurement step in which the food sales promotion control device measures the time that the surface image associated with the identification information is included in the image in the individual surface image management step,
In the remaining time prediction step,
The food sales promotion control device displays the food on the display shelf based on the time measured in the time measurement step and the elapsed time when the food is discarded, which is preset as the discard determination criterion. A food sales promotion method, characterized in that each of the remaining times until the point of disposal of the plurality of food items that have been sold is predicted.