JP2020057164A - Analysis device - Google Patents

Abstract

To provide an analysis device that further accurately grasps product sales.SOLUTION: In proposal services, an analysis device 12 comprises: means for calculating the degree of product sales depending on a location on shelves at a store from information held in a database obtained by matching the placement of products on the shelves at the store and sales information of the store; and means for calculating shelf potential for the shelves at the store by adjusting the calculated degree so that the contribution of the sales of the product itself to the calculated degree can be suppressed.SELECTED DRAWING: Figure 6

Description

  The present invention relates to an analysis device.

  In sales management at retail stores such as convenience stores and supermarkets, POS (Point Of Sale) information is an important input. However, if only the POS information is used, it is difficult to obtain the knowledge of why a certain product has been sold, even if it is known how much the product has been sold. In view of the above, Japanese Patent Application Laid-Open No. 2003-139,055 discloses a technique for photographing a product shelf, analyzing the photographed image to identify the position of the product on the product shelf, and merging the identification result with the sales data. Have been.

JP-A-2006-048409

  By using the technology described in Patent Literature 1, it is possible to link a position on a product shelf with sales, and for example, it is possible to specify a position with the highest sales. However, in the technology described in Patent Literature 1, since the contribution to the sales due to the attractiveness of the product itself is ignored, the relationship between the derived position in the shelf and the sales is lacking in accuracy. In particular, there is a risk that a position where a product that sells better than other products is located may be erroneously determined to be a position with high sales.

  The present invention has been made in view of these problems, and its purpose is to separately explain the sales of products by the attractiveness of the product itself and the attractiveness of the position in the shelf where the product is placed. Therefore, an object of the present invention is to provide a technology that enables more accurate grasp of the sales of a product.

  One embodiment of the present invention relates to an analysis device. This analysis device calculates the degree of sales of a product at a position within the shelf of the store from information held in a database obtained by matching the arrangement of the products on the shelf of the store with the sales information of the store. Means for calculating the shelf potential of the store shelves by adjusting the calculated degree so that the contribution of the sale of the product itself to the calculated degree is suppressed.

  It should be noted that any combination of the above-described components, and any replacement of the components and expressions of the present invention between an apparatus, a method, a system, a computer program, a recording medium storing a computer program, and the like are also included in the present invention. This is effective as an embodiment.

  ADVANTAGE OF THE INVENTION According to this invention, the sales of goods can be grasped | acquired more accurately by dividing into the contribution by the charm of the goods themselves, and the contribution by the attractiveness of the position in the shelf where the goods are arrange | positioned.

It is an explanatory view of a proposal service concerning a 1st embodiment. It is explanatory drawing which shows an example of the calculation method of a shelf potential. It is explanatory drawing which shows another example of the calculation method of a shelf potential. FIG. 1 is a schematic diagram illustrating a configuration of a proposal system according to a first embodiment. FIG. 5 is a hardware configuration diagram of the analysis server of FIG. 4. FIG. 5 is a block diagram illustrating functions and configurations of the analysis server in FIG. 4. FIG. 7 is a data structure diagram illustrating an example of a master database in FIG. 6. FIG. 7 is a data structure diagram illustrating an example of a product potential holding unit in FIG. 6. FIG. 7 is a data structure diagram illustrating an example of a shelf potential holding unit in FIG. 6. It is a typical screen figure of a display position proposal screen. 5 is a flowchart illustrating a flow of a series of processes in the analysis server of FIG. 4. It is an explanatory view of a proposal service concerning a 2nd embodiment. It is a block diagram showing a function and composition of an analysis server concerning a 2nd embodiment. It is a typical screen figure of a shelf arrangement proposal screen. It is a typical screen figure of a goods arrangement proposal screen. It is a typical screen figure of an order permission / refusal screen. It is a typical screen figure of an automatic order management screen. It is a typical screen figure of an item management screen. 13 is a flowchart illustrating a flow of a series of processes in an analysis server according to a first modification. 13 is a flowchart illustrating a flow of a series of processes in an analysis server according to a second modification.

  Hereinafter, the same or equivalent components, members, and processes shown in each drawing are denoted by the same reference numerals, and the repeated description will be omitted as appropriate. In each drawing, some of the members that are not important for the description are omitted.

(First Embodiment)
FIG. 1 is an explanatory diagram of the proposed service according to the first embodiment. In this proposed service, we propose a model that determines the purchasing potential of retail stores such as convenience stores and supermarkets by the sum of product potential and shelf potential.
Purchase potential: an index indicating the degree of sales of a certain product displayed (or arranged) at a certain position on a certain shelf in a certain store. It is represented by the number sold, the amount sold, the number of unsold items, the number of discards, (the number sold) / (the number of displays), and the like.
-Commodity potential: An index indicating the degree of sales of a certain product itself. There is little dependence on whether the goods are displayed on which shelves and in which positions. Shows the original appeal of the product.
Shelf potential: an index indicating the degree of sales at a certain position in a certain shelf. It does not substantially depend on which product is displayed at that position, or is less dependent. Represents the attractiveness of shelves (each position inside).

  In addition, contribution by a sale (sales promotion), a sales floor, a store itself, an event around the store, or the like may be considered.

  In the proposed service, a mechanism that can analyze and evaluate product potential and shelf potential separately using digital technologies such as image analysis is realized. This proposal service discloses the shelf potential of each analyzed position in the shelf to manufacturers and suppliers, and gives the right to display on the shelf with the higher shelf potential to manufacturers and suppliers who have presented lower purchase costs. Provide a preferential bidding system. Thereby, the purchase cost can be reduced.

  The stores 2a, 2b, and 2c are provided with imaging devices 20a, 20b, and 20c such as a monitoring camera and a clerk's portable terminal. An analysis server (not shown in FIG. 1) that implements the proposed service collects images of shelves of the stores 2a, 2b, and 2c from the imaging devices 20a, 20b, and 20c via a network. At the same time, the analysis server collects POS data from the POS devices of the stores 2a, 2b, 2c via the network. The analysis server analyzes the images of the shelves in the store for each unit time to generate display state information indicating which products are displayed at which positions on the shelves. The analysis server matches the display state information with the POS data to generate matching data indicating which products displayed at which positions on the shelves have been sold, and stores the matching data in a master database (not shown in FIG. 1). Store. The analysis server analyzes the matching data held in the master database by using AI (Artificial Intelligence) or the like, and measures under what conditions which category of product can be sold at which position in which shelf, to what extent. Calculate the potential.

  By referring to the calculated shelf potential, for example, for the shelf 22a of the store 2a in FIG. 1, one of the shelf potentials “high”, “normal”, and “low” is set at each position in the shelf 22a. A retail store using the proposed service discloses the shelf potential of the shelf 22a of the store 2a in negotiations with the manufacturers 30a, 30b, and 30c of the product suppliers. The retailer negotiates the purchase cost discount with the leverage of “the right to preferentially display the product at a position on the shelf with high potential”. The retail store displays the product (drink C in FIG. 1) 32c purchased from the maker 30c that has presented the largest discount amount or discount rate at the position 24a having the highest shelf potential among the positions in the shelf 22a of the store 2a. .

  As described above, a retail store that uses the proposed service can reduce the purchase cost of a product by using the sponsorship money of the manufacturer or the like as a source of capital by using the shelf potential. Manufacturers can differentiate their products from competitors and increase sales by having their products displayed at positions with high shelf potential.

  FIG. 2 is an explanatory diagram illustrating an example of a method of calculating a shelf potential. In the present calculation method, a shelf potential is calculated by focusing on a specific product 42 and tracking sales of the product 42 displayed at various positions in the shelf 40. The store shelf 40 has four display positions: upper left (1), upper right (2), lower left (3), and lower right (4). On the first day, a product 42 is displayed on the upper left (1), a product 44 is displayed on the upper right (2), a product 46 is displayed on the lower left (3), and a product 48 is displayed on the lower right (4). Was done. On this day, the number of products sold was 50 for product 42 (upper left (1)), 60 for product 44 (upper right (2)), 15 for product 46 (lower left (3)), and product 48 (lower left). The right (4)) was 10. On the second day, the merchandise 42 is displayed on the lower left (3), and the merchandise 42 (lower left (3)) is sold at other positions on the shelf 40 as shown in FIG. there were. On the third day, the merchandise 42 is displayed on the upper right (2), and the merchandise 42 (the upper right (2)) is sold at another position on the shelf 40 as shown in FIG. there were. On the fourth day, the merchandise 42 is displayed at the lower right (4), and the merchandise 42 is displayed at other positions of the shelf 40 as shown in FIG. there were.

  From this result, the shelf potential of the shelf 40 is calculated as: upper left (1): 50, upper right (2): 70, lower left (3): 20, and lower right (4): 10. Thus, in the present calculation method, the shelf potential is calculated by comparing the degree of sales when the same product 42 is displayed at each position of the shelf 40. In this case, since the focus is on the sales of the same product 42, the contribution due to the sales difference between the products is reduced or eliminated.

  FIG. 3 is an explanatory diagram showing another example of the method of calculating the shelf potential. In this calculation method, the shelf potential is calculated by removing the contribution of the product potential of each product from the actual sales of each product displayed at each position on the shelf 40. The store shelf 40 has four display positions: upper left (1), upper right (2), lower left (3), and lower right (4). One day, a product 42 is displayed on the upper left (1), a product 44 is displayed on the upper right (2), a product 46 is displayed on the lower left (3), and a product 48 is displayed on the lower right (4). Was. On this day, the number of products sold was 50 for product 42 (upper left (1)), 70 for product 44 (upper right (2)), 30 for product 46 (lower left (3)), and product 48 (lower left). The number on the right (4) was 80. Since this sales quantity includes the contribution of the attractiveness of the product itself, the sales quantity cannot be directly used as the shelf potential.

  Therefore, the merchandise potential of the merchandise 42, 44, 46, 48 is calculated from the POS data. In the example of FIG. 3, the product potential of the product 44 is the same as the product potential of the product 42, the product potential of the product 46 is 1.5 times (1.5 times sold) the product potential of the product 42, and the product 48. It was found that the product potential of the product was eight times (sold eight times) the product potential of the product. The product potential and the actual sales quantity are analyzed together. In the example of FIG. 3, the sales quantity of each of the products 44, 46, and 48 is converted into the sales number of the product 42. Since the merchandise potential of the merchandise 44 is the same as the merchandise potential of the merchandise 42, the sales number “70” of the merchandise 42 can be obtained by multiplying the sales number “70” of the merchandise 44 by the conversion coefficient “1/1”. Since the product potential of the product 46 is 1.5 times the product potential of the product 42, the sales number “20” of the product 42 is obtained by multiplying the sales number “30” of the product 46 by the conversion coefficient “1 / 1.5”. Is obtained. Since the merchandise potential of the merchandise 48 is eight times the merchandise potential of the merchandise 42, the sales quantity “10” of the merchandise 42 can be obtained by multiplying the sales quantity “80” of the merchandise 48 by the conversion coefficient “1/8”. In this way, the sales quantity at each position of the shelf 40 is made equal to the sales quantity of the same product 42.

  From this result, the shelf potential of the shelf 40 is calculated as: upper left (1): 50, upper right (2): 70, lower left (3): 20, and lower right (4): 10. As described above, in the present calculation method, the shelf potential is calculated by comparing the degree of sales when it is assumed that the same product 42 is displayed at each position of the shelf 40. In this case, the contribution by the product potential is reduced or eliminated, so that the influence of the difference in sales between the products is reduced or eliminated.

  The calculation method of the shelf potential described with reference to FIGS. 2 and 3 is an example, and another calculation method may be used. For example, calculating the product potential by analyzing the difference in sales between products while fixing the shelf position and weather / promotion conditions to some extent, and fixing the product category and weather / promotion conditions to some extent The shelf potential may be calculated by analyzing the difference in sales for each shelf position. A publicly known statistical technique may be used for the analysis. A model in which the purchase potential is the sum of the product potential and the shelf potential is also an example, and another model may be used. For example, a product may be used instead of a sum, and other factors such as a sales promotion potential may be introduced as described above. Alternatively, the technical idea of the present embodiment can be applied to any model that uses a product potential in which the attractiveness of the product itself is dominant and a shelf potential in which the attractiveness of the shelf position is dominant.

  The reason why the weather conditions (including the seasons) are fixed above is that the weather conditions are one of the major factors that influence sales. For example, even at a convenience store near the same ski resort, shelf sales differ greatly between winter and summer. Even at convenience stores in the same office district, the sales of rice ball shelves differ greatly between rainy weather and fine weather.

  FIG. 4 is a schematic diagram illustrating a configuration of the proposal system 10 according to the first embodiment. The proposal system 10 includes an analysis server 12, a POS system 16, an imaging device 14, and a POS (Point Of Sale) terminal 15. The analysis server 12, the POS system 16, the imaging device 14, and the POS terminal 15 are communicably connected to each other via the network 11. The imaging device 14 and the POS terminal 15 are provided in the store 13. The imaging device 14 includes a surveillance camera installed in a store and a mobile terminal (a smartphone or the like) possessed by a store clerk.

  In the store 13, one or more shelves for displaying products are installed. The imaging device 14 photographs this shelf. The imaging device 14 transmits the data of the captured image to the analysis server 12 via the network 11. This image data includes shelf information indicating which shelf is the image that was shot, and time information indicating the shooting time. The shelf information includes, for example, information indicating the store 13 where the shelf is installed and information indicating the position of the shelf. Alternatively, the shelf information may include, for example, information indicating the imaging device 14 and information indicating the orientation of the imaging device 14, or information indicating the position of the imaging device 14 that is measured using, for example, GPS. May be included. The shelf information may be any information as long as it is possible to determine which shelf of which store.

  The POS system 16 communicates with the POS terminal 15 installed in the store 13, and from the POS terminal 15, for example, POS data indicating the sales and the number of products sold in the store 13 in which the POS terminal 15 is installed To receive. The POS system 16 is a system that manages the received POS data for each product and each store. The POS data may be, for example, general POS data indicating the sales amount or the number of sales of a certain product. The POS data sent from the POS terminal 15 to the POS system 16 may include, for example, information such as a category of a product.

  In the example of FIG. 4, the POS system 16 is provided separately from the store 13 where the POS terminal 15 is installed. However, the present invention is not limited to this, and the POS system 16 may be provided in the store 13. Alternatively, the POS system 16 may be integrated with the POS terminal 15.

The POS system 16 transmits POS data to be managed to the analysis server 12.
The analysis server 12 receives image data from the imaging device 14 and receives POS data from the POS system 16. The analysis server 12 performs data analysis using the received image data and POS data.

  FIG. 5 is a hardware configuration diagram of the analysis server 12 of FIG. Each of the POS terminal 15 and the POS system 16 may have the same hardware configuration as the hardware configuration of the analysis server 12 shown in FIG. The analysis server 12 includes a memory 121, a processor 122, a communication interface 123, a display 124, and an input interface 125. These components are each connected to a bus 126 and communicate with each other via the bus 126.

  The memory 121 is a storage area for storing data and programs. The data and the program may be permanently stored in the memory 121 or may be temporarily stored. The processor 122 implements various functions in the analysis server 12 by executing a program stored in the memory 121. The communication interface 123 is an interface for transmitting and receiving data to and from the outside of the analysis server 12. For example, the communication interface 123 includes an interface for accessing a wireless communication network, an interface for accessing a wireless LAN (Local Area Network), and the like. The communication interface 123 may include a wired network interface such as a USB (Universal Serial Bus), for example. The display 124 is a device for displaying various information, and is, for example, a liquid crystal display or an organic EL (Electroluminescence) display. The input interface 125 is a device for receiving an input from a user. The input interface 125 includes, for example, a touch panel provided on the display 124, various input keys, and the like.

  FIG. 6 is a block diagram showing functions and configuration of the analysis server 12 of FIG. Each block shown here can be realized in terms of hardware by elements such as a computer CPU and other mechanical devices, and can be realized in terms of software by a computer program or the like. The functional blocks are drawn. Therefore, it will be understood by those skilled in the art referred to in this specification that these functional blocks can be realized in various forms by a combination of hardware and software.

  The analysis server 12 includes an image processing unit 131, a POS data acquisition unit 132, a matching unit 133, a product potential calculation unit 134, a shelf potential calculation unit 135, a display control unit 136, a master database 137, a product potential A holding unit 138 and a shelf potential holding unit 139 are provided.

  The image processing unit 131 acquires image data indicating an image of a predetermined shelf of the store 13 from the imaging device 14 via the network 11. The image processing unit 131 analyzes the acquired image and generates display state information indicating which product is displayed at which position on the shelf. The display state information generated by the image processing unit 131 indicates the arrangement of the products on the shelves of the store 13. The analysis of the image in the image processing unit 131 may be realized by a known image analysis technique.

  The POS data acquisition unit 132 acquires POS data indicating sales at the store 13 from the POS system 16 via the network 11.

  The matching unit 133 matches the display state information generated by the image processing unit 131 with the POS data obtained by the POS data obtaining unit 132 to determine which product displayed at which position on the shelf and how much. Is generated. This matching is performed, for example, by matching the photographing time corresponding to the display state information with the data collection period corresponding to the POS data. For example, if the display state information is based on an image of a shelf taken at 8:00 in the morning on May 21, the display state information is matched with POS data indicating sales on May 21. The matching unit 133 stores the matching data in the master database 137.

  FIG. 7 is a data structure diagram showing an example of the master database 137 of FIG. The master database 137 includes a date, weather conditions such as weather and temperature of the day, sales promotion conditions indicating a sale or an event performed on the day, a shelf ID for specifying a shelf, and a position ID for specifying a position on the shelf. And a displayed product ID that identifies the product displayed at that position on the shelf, the category of the displayed product, the number of displayed products sold on that day, the number of displayed products discarded on that day, Are stored in association with each other.

  Returning to FIG. 6, the product potential calculation unit 134 calculates the product potential indicating the degree of sales of the product itself from the information stored in the master database 137. The merchandise potential calculating unit 134 may calculate the merchandise potential of the merchandise by comparing the sales quantity of the merchandise displayed at a certain position of the shelf with the weather conditions and the sales promotion conditions fixed. For example, in the example of FIG. 7, when the weather condition is fixed to “sunny, 24 degrees” and the promotion condition is fixed to “none”, the product “XX1” is displayed at the position “(1)” on the shelf “A001”. It can be understood that the sales quantity at the time of selling is “50”, and the sales quantity when the product “YY1” is displayed at the same position is “75”. The product potential calculation unit 134 calculates the product potential of the product “YY1” as 1.5 times the product potential of the product “XX1” based on the result. Alternatively, the product potential calculation unit 134 may calculate the product potential of each product by obtaining an average over the entire data held in the master database 137.

  The product potential calculation unit 134 stores the product potential calculated for each product in the product potential holding unit 138.

  FIG. 8 is a data structure diagram showing an example of the commodity potential holding unit 138 of FIG. The product potential holding unit 138 holds the product potential of each product for each weather condition and each category. Specifically, the product potential holding unit 138 holds the product category, the product ID, the weather condition, and the product potential in association with each other. In the example of FIG. 8, the product potential of the product category “cold drink” and the product potential of the product category “Onigiri” are set independently, but a conversion formula may be set between the two.

  Returning to FIG. 6, the shelf potential calculation unit 135 calculates the sales degree of the product at the position in the shelf of the store 13, that is, the number of units sold, from the information held in the master database 137. The shelf potential calculation unit 135 calculates the shelf potential of the shelf of the store 13 by adjusting the calculated sales quantity so that the contribution of the sales of the displayed product itself to the calculated sales quantity is suppressed. The shelf potential calculator 135 calculates the shelf potential using the calculation method described with reference to FIGS. 2 and 3, for example. The shelf potential calculation unit 135 stores the shelf potential calculated for each position in the shelf in the shelf potential holding unit 139.

  In one example, the product potential calculated by the product potential calculation unit 134 and the shelf potential calculated by the shelf potential calculation unit 135 are independent of each other. In another example, the contribution of the display position to the product potential is suppressed, if not independent, and the contribution of the product itself to the shelf potential is suppressed.

  FIG. 9 is a data structure diagram showing an example of the shelf potential holding unit 139 of FIG. The shelf potential holding unit 139 holds the shelf potential at each position of the shelf for each weather condition and for each category. Specifically, the shelf potential holding unit 139 holds the shelf ID of the shelf, the position ID of the position of the shelf, the weather condition, the category of the product, and the shelf potential in association with each other. In the example of FIG. 9, even at the same position “(1)” of the same shelf “A001”, the shelf potential of the product category “cold drink” and the shelf potential of the product category “hot drink” are set independently. You. A conversion formula may be set between the two. Similarly, even at the same position on the same shelf, a different shelf potential is set for each weather condition.

  Returning to FIG. 6, the display control unit 136 generates a screen visually representing the shelf potential calculated by the shelf potential calculation unit 135, and causes the display of the analysis server 12 and the display of another device to display the screen. The display control unit 136 generates the display position proposal screen 200, for example, and causes the display to display it.

  FIG. 10 is a representative screen diagram of the display position proposal screen 200. The display position proposal screen 200 is generated by the display control unit 136 with reference to the shelf potential holding unit 139 and visually represents the shelf potential calculated by the shelf potential calculation unit 135. The display position suggestion screen 200 includes a store name display area 202 that displays the name of the store, a shelf position display area 204 that displays the position of the target shelf, a shelf image 206, a first suggestion area 208, A second proposal area 210. The first suggestion area 208 and the second suggestion area 210 are displayed so as to be visually distinguishable from other areas by visual effects such as coloring, blurring, and decoration. The first proposal area 208 and the second proposal area 210 may be highlighted. The first proposal area 208 may be displayed in a different manner from the second proposal area 210.

  The first proposal region 208 corresponds to a position range where the shelf potential is the highest in the shelf. The first proposal area 208 displays a numerical value of the shelf potential, and a purchase amount and a discount amount necessary for obtaining a right to display a product there. The second proposal area 210 corresponds to a position range where the shelf potential is the second highest on the shelf. The second proposal area 210 displays the numerical value of the shelf potential and the purchase amount and the discount amount necessary to obtain the right to display the product there. The person in charge of the retail store may negotiate while showing the display position proposal screen 200 shown in FIG. 10 to the person in charge of the supplier. Alternatively, the analysis server 12 transmits the screen data of the display position proposal screen 200 to the terminal of the person in charge of the supplier via the network 11 so that the display position proposal screen 200 is displayed on the display of the terminal. Is also good.

The operation of the analysis server 12 having the above configuration will be described.
FIG. 11 is a flowchart showing a flow of a series of processes in the analysis server 12 of FIG. The analysis server 12 acquires an image of a shelf from the imaging device 14 of the store 13 via the network 11 (S302). The analysis server 12 analyzes the obtained image to specify the arrangement of the products on the shelf when the image was obtained (S304). The analysis server 12 acquires POS data from the POS terminal 15 or the POS system 16 of the same store 13 (S306). The analysis server 12 matches the arrangement of the products on the shelf specified in step S304 with the POS data obtained in step S306, and registers the result in the master database 137 (S308). The analysis server 12 determines whether the timing for calculating the potential has arrived (S310). The timing may be, for example, a fixed time of the day, once a week, or the presence of an instruction from the user. If the timing has not arrived (NO in S310), the analysis server 12 returns the process to step S302. When the timing has come (YES in S310), the analysis server 12 calculates the product potential with reference to the master database 137, and registers it in the product potential holding unit 138 (S312). The analysis server 12 calculates the shelf potential with reference to the master database 137 and the product potential calculated in step S312, and registers the shelf potential in the shelf potential holding unit 139 (S314).

  In the above embodiment, examples of the holding unit are a hard disk and a semiconductor memory. Further, based on the description in this specification, each unit is constituted by a CPU (not shown), a module of an installed application program, a module of a system program, and a semiconductor memory for temporarily storing the contents of data read from a hard disk. It will be appreciated by those skilled in the art referred to herein that this can be achieved.

  According to the analysis server 12 according to the present embodiment, it is possible to calculate the shelf potential indicating the attractiveness of the display position on the shelf. This shelf potential can be used in various situations. For example, by using an input such as AI, it is possible to more accurately predict supply and demand and optimize product arrangement. Shelf potential also helps store managers consider better product placement.

  Also, as described with reference to FIGS. 1 and 10, by visualizing the shelf potential, it can be used as leverage during purchase negotiation. As a result, the store manager has the advantage of increasing profits as shelves are more attractive, and also has the effect of improving motivation for store management.

(Second embodiment)
In the first embodiment, the case where the purchase price is dynamically auctioned using the shelf potential has been described. In the second embodiment, the optimal state of the shelf at a certain point in the future is predicted using the shelf potential. The state of the shelf to be predicted is at a level of how many products of which category should be placed on which shelf. Here, the product is not specified. This is because noise increases when trying to specify. Based on that prediction, the next step is to use the product potential to predict which product should fill the shelf.

  FIG. 12 is an explanatory diagram of the proposed service according to the second embodiment. The shelf potential is calculated for each product category by the method described in the first embodiment, and is stored in the shelf potential holding unit 139. In particular, for the product category “cold drink”, the shelf potential of the shelf 50 is as follows: upper left (1): 50, upper right (2): 70, lower left (3): 20, lower right (4): 10, Is calculated. On the other hand, regarding the product category “hot drink”, the shelf potential of the shelf 50 is as follows: upper left (1): 68, upper right (2): 45, lower left (3): 15, lower right (4): 17, Is calculated.

  The analysis server of the proposed service according to the second embodiment refers to the shelf potential of the category “cold drink” and the shelf potential of the category “hot drink” and optimizes the arrangement of the two categories on the shelf 50. I do. For example, the upper left (1) has the category “hot drink”, the upper right (2) has the category “cold drink”, the lower left (3) has the category “cold drink”, and the lower right (4) has the category “hot drink”. By assigning each "drink", the shelf potential of the entire shelf 50 can be maximized. The analysis server proposes to the user the arrangement of the categories on the shelf 50 obtained as a result of the optimization.

  FIG. 13 is a block diagram illustrating functions and configurations of the analysis server 52 according to the second embodiment. Each block shown here can be realized in terms of hardware by elements such as a computer CPU and other mechanical devices, and can be realized in terms of software by a computer program or the like. The functional blocks are drawn. Therefore, it will be understood by those skilled in the art referred to in this specification that these functional blocks can be realized in various forms by a combination of hardware and software.

  The analysis server 52 includes an image processing unit 131, a POS data acquisition unit 132, a matching unit 133, a product potential calculation unit 134, a shelf potential calculation unit 135, a display control unit 136, a master database 137, a product potential A storage unit 138, a shelf potential storage unit 139, a proposal generation unit 152, and an order placement unit 154 are provided.

  The proposal generation unit 152 generates proposal information indicating which category of product should be placed at which position on the shelf based on the shelf potential calculated for each product category. The proposal generation unit 152 generates the shelf arrangement proposal screen 220 using, for example, the method described with reference to FIG. The display control unit 136 causes the display to display the generated shelf arrangement proposal screen 220.

  FIG. 14 is a representative screen diagram of the shelf arrangement proposal screen 220. The shelf arrangement proposal screen 220 includes a date display area 222 indicating a future date, a weather forecast display area 224 displaying a weather forecast for the day, a sale information display area 226 displaying sale information for the day, and a day It has an event information display area 228 for displaying event information, a proposal arrangement display area 230 for suggesting a product category arrangement on a target shelf, and an OK button 232.

  In the proposed arrangement display area 230, a plurality of category objects 240, 242, 244, 246, 248, 250 that visually indicate where and which category should be arranged are arranged together with the image of the target shelf. For example, the second tier from the top of the shelf is occupied by an "Onigiri" object 244, which suggests that the second tier from the top of the shelf be filled with products of the category "Onigiri". In addition, the bottom of the shelf is occupied by a “lunch” object 248 on the left and the “noodles” object 250 on the right. It is proposed to arrange products of the category "noodles" on the right side. Note that each category object may be displayed in a different manner from other category objects. Alternatively, the category objects corresponding to the same category (in the example of FIG. 14, the uppermost “rice ball” object 242 and the second uppermost “rice ball” object 244) may be displayed in the same manner.

  When the user is not satisfied with the category arrangement suggested on the shelf arrangement proposal screen 220, the category may be rearranged by dragging and dropping the category object. Each category object may be configured to be able to be enlarged or reduced. An interface may be provided that allows for the placement of new categories. If the user specifies the OK button 232 after customizing the category layout as necessary on the shelf layout proposal screen 220, the proposal generator 152 acquires the category layout in the proposal layout display area 230 at that time.

  Returning to FIG. 13, the order placement unit 154 determines which product on the shelf should be placed at which position based on the proposal information generated by the proposal generation unit 152 and the product potential held in the product potential holding unit 138. And a process for ordering a product based on the result of the determination. When the user specifies the OK button 232 in the example of FIG. 14, the proposal generation unit 152 acquires the category arrangement in the proposal arrangement display area 230 at that time. The ordering unit 154 assigns an appropriate product to each category of the acquired category arrangement. For example, the ordering unit 154 determines the allocation of the product to each category in consideration of the product potential, the stock quantity, and the purchase price. The ordering unit 154 generates a product placement proposal screen 234 indicating a state in which products are allocated to the designated category via the shelf placement proposal screen 220 in FIG. 14, and displays the generated product placement proposal screen 234 on a display. .

  FIG. 15 is a representative screen diagram of the product arrangement proposal screen 234. The product placement proposal screen 234 includes a date display area 222, a weather forecast display area 224, a sale information display area 226, an event information display area 228, and a product placement display for suggesting product placement on a target shelf. An area 236 and an OK button 238 are provided.

  In the merchandise arrangement display area 236, a plurality of merchandise objects 252 that visually indicate where and which merchandise should be arranged are arranged together with the target shelf image and category object. Each product object may be displayed in a different manner from other product objects. Alternatively, the product objects corresponding to the same product may be displayed in the same manner. The operation of the product object 252 is the same as the operation of the category object.

  When the user designates the OK button 238 in the example of FIG. 15, the ordering unit 154 acquires the product arrangement in the product arrangement display area 236 at that time. The ordering unit 154 compares the number of commodities required for realizing the acquired product arrangement with the number of stocks of commodities held in an unillustrated inventory database. When it is predicted that the stock will run short, the order placement unit 154 determines the number of ordered products to solve the shortage of stock. The order placement unit 154 generates an order permission / refusal screen 260 for asking the user for permission to place an order for the determined number of products, and displays the generated order permission / refusal screen 260 on the display.

  FIG. 16 is a representative screen diagram of the order permission / refusal screen 260. The order permission / refusal screen 260 includes a date display area 222, a weather forecast display area 224, a sale information display area 226, an event information display area 228, an order content display area 262 for displaying order details, an OK button 264, And a button 266 later. When the user designates the OK button 264, the ordering unit 154 generates order data including the determined order number of the product, and transmits the order data to the server (not shown) of the supplier via the network 11. When the user specifies the button 266 later, the order placement unit 154 performs a process for suspending the order.

  The ordering unit 154 may automatically perform the ordering process without inquiring the user of the ordering. In this case, the ordering unit 154 acquires the conditions for automatic ordering from the user via the automatic ordering management screen 400 in advance.

  FIG. 17 is a representative screen diagram of the automatic order management screen 400. The automatic order management screen 400 includes an application toggle switch 402 for specifying whether or not to apply the automatic order, a store name input area 404 for accepting the input of the store name or displaying the store name by default, A clerk name input area 406 in which the name of the person in charge is received or the name of the person in charge is displayed by default, an order timing designation area 408 in which the user can specify the order timing, and an input of business hours is accepted or default. , A store implementation campaign display area 412 in which information on a campaign implemented by the store is displayed, an individual measure designation area 414 in which individual measures can be designated, and an item management screen to be described later. And an item management screen link 416 that is a link to

  In the order timing specification area 408, the user checks a check box corresponding to a desired automatic order timing. In the store implementation campaign display area 412, information on campaigns implemented in the store during the period designated by the user among the year, the current month, and the designated period is displayed. When the store is a chain store or a franchise, the store implementation campaign display area 412 is configured to be able to cooperate with the data of the headquarters. In the example of FIG. 17, when the “head office data link” link in the store execution campaign display area 412 is specified, the campaign information stored in the head office server related to the store is displayed in the store execution campaign display. It is displayed in area 412.

  The user can select an individual measure in the store in the individual measure designation area 414. For example, the user can set the presence or absence of an individual measure, the setting of the tolerance of the discard rate, the setting of the sales enhancement product, the setting of the sales suppression product, and the setting of the operation policy.

  When the item management screen link 416 is designated on the automatic order management screen 400 in FIG. 17, the ordering unit 154 generates an item management screen 420 and causes the display to display the generated item management screen 420.

  FIG. 18 is a representative screen diagram of the item management screen 420. The item management screen 420 displays a list of items (variables) that can be considered when performing demand forecasting and automatic ordering. The item management screen 420 is configured so that it is possible to specify whether each item is used for prediction or not. In particular, on the item management screen 420, it is possible to specify whether to use the shelf potential and whether to use the product potential.

  According to the analysis server 52 according to the present embodiment, it is possible to propose an optimal category arrangement of the products in the shelf based on the shelf potential, and further, it is possible to automatically place an automatic order based on the product potential. Becomes Therefore, the sales of the store can be increased while the ordering work in the store is made more efficient.

  Further, according to the analysis server 52 according to the present embodiment, it is possible to propose a category arrangement on a shelf based on a shelf potential. Since the shelf potential is a parameter whose contribution by the attractiveness of the product itself is suppressed, the proposed category arrangement is also less affected by the attractiveness of the product itself. Therefore, the user (manager of the store) can know the category arrangement that brings out the appeal of the shelf by looking at the proposed category arrangement.

  The configuration and operation of the analysis server according to the embodiment have been described above. These embodiments are exemplifications, and it is understood by those skilled in the art that various modifications can be made to the combination of each component and each processing, and such modifications are also within the scope of the present invention.

  In the first and second embodiments, the case where the arrangement of the products on the shelf is acquired by analyzing the image of the shelf has been described. However, the present invention is not limited to this. Alternatively, the arrangement of the products may be acquired by an automatic means such as attaching a barcode to each position on the shelf or attaching and reading an IC tag at each position on the product / shelf.

  In the first and second embodiments, the master database 137 is created and maintained for each store. Therefore, the information held in the master database 137 is data reflecting characteristics, habits, trends, and the like of the corresponding store. In the second embodiment, since the proposal of the optimal category arrangement and the automatic ordering process are performed based on the information held in the master database 137, the contents of the proposal reflect the characteristics of the store and the like. It will be more accurate.

  Alternatively, in the first embodiment, an integrated database may be created by integrating the master databases 137 of all stores under the umbrella of the headquarters, and the product potential and the shelf potential may be calculated based on the integrated database. . Further, for a group of stores having similar location environments, the master database 137 may be integrated to create an integrated database, and the merchandise potential and shelf potential may be calculated based on the integrated database. In this case, the proposal of the display position on the shelf displayed on the display position proposal screen 200 of FIG. 10 is not for each store but for each store group.

  In the second embodiment, when it is predicted that products will run out, stock may be exchanged from another store of the same system or the same brand instead of ordering from a supplier.

  In the second embodiment, the merchandise arrangement display area 236 of the merchandise arrangement proposal screen 234 may be displayed on the display of the clerk's portable terminal or the POS terminal 15 at the time of stocking. In this case, the stocking can be made more efficient.

  In the first embodiment, the image processing unit 131 acquires image data indicating an image of a predetermined shelf of the store 13 from the imaging device 14 via the network 11, analyzes the acquired image, Although a case has been described where display state information indicating which product is displayed at which position is described, the present invention is not limited to this. For example, in order to further improve the accuracy of the generated display state information, the image processing unit according to the first modification generates the display state information using the POS data acquired by the POS data acquisition unit 132 as auxiliary data. You may. For example, the image processing unit can use POS data having a data collection period that matches the shooting time of the image data indicating the image of the predetermined shelf of the store 13 as the auxiliary data. As a result, compared to the case where only a known image analysis technique is applied, since there is information on the product sold at that time, it can be expected that a highly accurate analysis result can be obtained.

  FIG. 19 is a flowchart illustrating a flow of a series of processes in the analysis server according to the first modification. The analysis server acquires an image of a shelf from the imaging device 14 of the store 13 via the network 11 (S302). The analysis server acquires POS data from the POS terminal 15 or the POS system 16 of the same store 13 (S322). The analysis server specifies the arrangement of the products on the shelf when the image was acquired based on the shelf image acquired in step S302 and the POS data acquired in step S322, and generates display state information (S324). ). The analysis server matches the display state information generated in step S324 with the POS data obtained in step S322, and registers the obtained matching data in the master database 137 (S326). Subsequent steps S310, S312, and S314 are the same as those in FIG.

  Note that the generation of the display state information and the matching process may be integrated. FIG. 20 is a flowchart illustrating a flow of a series of processes in the analysis server according to the second modification. The analysis server acquires an image of a shelf from the imaging device 14 of the store 13 via the network 11 (S302). The analysis server acquires POS data from the POS terminal 15 or the POS system 16 of the same store 13 (S322). The analysis server specifies the arrangement of the products on the shelf when the image was obtained based on the image of the shelf obtained in step S302 and the POS data obtained in step S322, and displayed at which position on the shelf. Matching data indicating which merchandise has been sold and to what extent are generated (S328). The matching data is registered in the master database 137. Subsequent steps S310, S312, and S314 are the same as those in FIG.

  10 proposal system, 11 network, 12 analysis server, 13 store, 14 imaging device, 15 POS terminal, 16 POS system.

Claims (5)

Means for calculating the degree of sales of a product at a position in the shelf of the store from information held in a database obtained by reconciling the arrangement of the products on the store shelf with the sales information of the store,
Means for calculating a shelf potential of store shelves by adjusting the calculated degree so that the contribution of the sale of the product itself to the calculated degree is suppressed. It further comprises means for calculating the degree of sales of the product itself as product potential,
The analysis device according to claim 1, wherein the shelf potential and the commodity potential are calculated so as to be independent of each other.   The analysis device according to claim 1, further comprising: a unit configured to generate a screen that visually indicates the calculated shelf potential. Shelf potential is calculated for each product category,
The said analysis apparatus is further provided with the means which produces | generates the proposal information which shows which category of goods should be arrange | positioned on the shelf based on the calculated shelf potential in any one of Claims 1-3. Analysis equipment.   3. The method according to claim 2, further comprising: means for determining which product should be placed at which position on the shelf based on the proposal information and the product potential, and performing a process for ordering the product based on a result of the determination. The analysis device according to claim 4.

Images