Store business support system, store business support method, and store business support program

Abstract

To provide a store business support system which detects an event which has been difficult to estimate due to entanglement with many factors, and performs store business according to the event on the system side.SOLUTION: The store business support system comprises an AI processing unit and a robot. The AI processing unit receives input of commercial information being information related to a store, products, or visitors and analyzes the received commercial information by AI processing to estimate demand information indicating a demand for each product. The robot replenishes products on the basis of the demand information estimated by the AI processing unit. Alternatively, the AI processing unit receives input of commercial information including an analysis result of an image generated by capturing an indoor or outdoor environment with an imaging apparatus and analyzes the received commercial information by AI processing to estimate visitor reception information which is based on at least the indoor or outdoor environment and is information to be presented to visitors. The robot answers to inquiries from visitors on the basis of the visitor reception information estimated by the AI processing unit.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present application mainly relates to a system in which a robot assists store operations using the results of analyzing various information. [Background technology]

[0002] Patent Document 1 discloses a manners violation detection system and a remote monitoring system as a system for dealing with the reduction in the number of station staff per station, etc. The manners violation detection system detects any manners violation occurring within the station by analyzing the detection results of sensors installed within the station, and notifies station staff. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2002-232871 A Summary of the Invention [Problem to be solved by the invention]

[0004] In the etiquette violation detection system of Patent Document 1, detectable etiquette violations are limited to those that can be directly detected by a sensor. Therefore, it is difficult to estimate events that involve multiple factors. However, it is preferable to be able to detect these events. Furthermore, the main purpose of the etiquette violation detection system is to detect etiquette violations, and the actual response is carried out by station staff. However, in order to reduce the burden on station staff, it is preferable for the system to also handle the response.

[0005] This type of problem is not limited to station operations, but is similar in store operations. That is, in store operations, it is preferable to estimate events that have been difficult to estimate in the past because of the involvement of many factors, and to perform store operations accordingly on the system side.

[0006] This application has been made in consideration of the above circumstances, and its main objective is to provide a store operations support system that detects events that have traditionally been difficult to estimate because they involve multiple factors, and performs store operations on the system side accordingly. [Means for solving the problem]

[0007] The problem to be solved by the present application is as described above. Next, the means for solving this problem and the effects thereof will be described.

[0008] According to a first aspect of the present application, there is provided a store operation support system having the following configuration. That is, the store operation support system includes an AI processing unit and a robot. The AI ​​processing unit receives commercial information related to a store, products, or customers, analyzes the input commercial information using AI processing, and estimates demand information indicating demand for each product. The robot replenishes products based on the demand information estimated by the AI ​​processing unit.

[0009] According to a second aspect of the present application, there is provided a store operation support system having the following configuration. That is, the store operation support system includes an AI processing unit and a robot. The AI ​​processing unit receives commercial information, which is information related to a store, a product, or a customer, and includes an image generated by capturing an image of an indoor or outdoor area with an imaging device or an analysis result of the image, and analyzes the input commercial information by AI processing to estimate visitor handling information, which is based on at least an indoor or outdoor environment and is information to be presented to a customer. The robot responds to inquiries from customers based on the visitor handling information estimated by the AI ​​processing unit.

[0010] According to a third aspect of the present application, there is provided a store operation support method as follows. That is, in the store operation support method, commercial information related to a store, products, or customers is input, the input commercial information is analyzed by AI processing, and demand information indicating demand for each product is estimated. A robot is caused to replenish products based on the demand information estimated by AI processing.

[0011] According to a fourth aspect of the present application, there is provided a store operation support program that causes a computer to execute the following processes: That is, commercial information related to a store, products, or customers is input, the input commercial information is analyzed by AI processing, and demand information indicating demand for each product is estimated. Based on the demand information estimated by AI processing, a command is generated to cause a robot to replenish products.

[0012] According to a fifth aspect of the present application, there is provided a store operation support method having the following configuration. That is, in the store operation support method, commercial information is input, which is information related to a store, a product, or a customer, and includes an image generated by capturing an indoor or outdoor image with an imaging device, or an analysis result of the image, and the input commercial information is analyzed by AI processing to estimate visitor response information that is based on at least the indoor or outdoor environment and is information to be presented to a customer. A robot is caused to answer inquiries from customers based on the visitor response information estimated by the AI ​​processing.

[0013] According to a sixth aspect of the present application, there is provided a store operation support program for causing a computer to execute the following processes. That is, commercial information is input, which is information related to a store, a product, or a customer, and includes an image generated by capturing an indoor or outdoor image with an imaging device, or an analysis result of the image, and the input commercial information is analyzed by AI processing to estimate visitor response information that is based on at least an indoor or outdoor environment and is information to be presented to a customer. The visitor response information estimated by the AI ​​processing is transmitted to a robot as information for the robot to respond to inquiries from customers. Effect of the Invention

[0014] According to the present application, it is possible to estimate events that have traditionally been difficult to estimate because they involve multiple factors, and to operate the store accordingly on the system side. [Brief description of the drawings]

[0015] [Figure 1]1 is a block diagram of a store operation support system and related devices according to an embodiment of the present application. [Diagram 2] A diagram showing the information handled by the AI ​​processing server and the store operations performed by robots. [Diagram 3] 1 is a flowchart showing the processing performed by the image analysis server and the AI ​​processing server. [Figure 4] 1 is a flowchart showing the processing performed by a robot or an AI processing server. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] Next, an embodiment of the present application will be described with reference to the drawings.

[0017] The store operations support method of this embodiment is realized by using a store operations support system 1. The store operations support system 1 analyzes various information that affects store operations using AI processing, and estimates new information that cannot be directly detected by sensors. The store operations support system 1 performs store operations using a robot 30 based on the estimated information. This makes it possible to realize various services that were not possible in the past.

[0018] In this embodiment, the stores targeted by the store operations support system 1 are located within facilities such as train stations, shopping malls, commercial buildings, and department stores. These types of stores are preferable as application targets because in addition to information acquired independently by the store, there is information that can be acquired from the facility. The store targeted by the store operations support system 1 may be one store within a facility, or multiple stores within a facility. Note that applying the store operations support system 1 to a store within a facility is just one example, and the store operations support system 1 can also be applied to a store that is set up independently of the facility, for example.

[0019] As shown in FIG. 1, the store operation support system 1 includes an image analysis server 21, an AI processing server 22, and a robot 30.

[0020] The image analysis server 21 and the AI ​​processing server 22 are provided on the cloud. The cloud is an environment constructed by a server or the like that is physically separated from the target facility and can communicate with devices in the facility via the Internet. The image analysis server 21 and the AI ​​processing server 22 may be provided in the facility, not limited to being provided on the cloud. The image analysis server 21 and the AI ​​processing server 22 are each a server device including a CPU, storage, memory, a communication device, and the like. The CPU executes the store operation support program stored in the storage to perform processing related to the store operation support of this embodiment. The image analysis server 21 may be realized by one server device, or may be realized by a plurality of server devices in cooperation with each other. The same applies to the AI ​​processing server 22. Alternatively, one or more server devices having the functions of both the image analysis server 21 and the AI ​​processing server 22 may be provided without separating the server devices for the image analysis server 21 and the AI ​​processing server 22.

[0021] The image analysis server 21 receives images captured and generated by the imaging devices 11. The imaging range of at least one imaging device 11 is indoors, and an indoor image is generated. The imaging range of at least one imaging device 11 is outdoors, and an outdoor image is generated. The imaging device 11 captures images at a predetermined time interval to generate images. The imaging device 11 may be a camera whose main function is to capture still images, or a video camera whose main function is to capture moving images. The imaging device 11 may be a camera fixed to a structure of the facility, or a camera module provided on the robot 30 or the like. The imaging device 11 has a communication function and transmits images by wired communication or wireless communication. The images transmitted by the imaging device 11 are uploaded to the image analysis server 21 via a PC or a router in the facility. In addition, it is preferable that the imaging device 11 transmits information for identifying the imaging position in association with the generated image. The information for identifying the imaging position is, for example, the position where the imaging device 11 is fixed, or the position of the robot 30 when the robot 30 captures the image. When the identification information of the imaging device 11 and the position at which the imaging device 11 is fixed are registered in association with each other, the identification information of the imaging device 11 corresponds to information for specifying the imaging position.

[0022] The image analysis server 21 analyzes the image received from the imaging device 11 to generate image analysis information shown in FIG. 2. The image analysis server 21 of this embodiment analyzes the image to generate information related to congestion and weather. For example, the image analysis server 21 identifies people by performing known matching processing or the like from indoor and outdoor images to identify the number of visitors at each location. This identifies the current congestion status of indoor and outdoor rest areas, meeting places, event venues, etc. Furthermore, the image analysis server 21 can identify the current weather conditions by performing known matching processing or the like from outdoor images to identify rain, snow, clouds, etc. The image analysis server 21 may also identify the clothing or belongings of people outdoors and estimate the current weather conditions based on the identified information. The clothing is, for example, a short-sleeved shirt, a long-sleeved shirt, or a coat. The belongings are, for example, a hat, an umbrella, or a raincoat. The image analysis server 21 transmits the above image analysis information to the AI ​​processing server 22.

[0023] As shown in FIG. 1, the AI ​​processing server 22 acquires general information, visitor information, and store information in addition to the image analysis information received from the image analysis server 21. As shown in FIG. 2, the general information, visitor information, store information, and image analysis information are collectively referred to as commercial information. Commercial information is information related to a store, a product, or a visitor. In other words, commercial information is information that may affect store operations. Here, among the above-mentioned image analysis information, the congestion situation is information related to visitors, since it indicates the number and distribution of visitors. Among the image analysis information, the weather situation is information related to visitors, since it affects the number of visitors visiting the facility and obviously affects the behavior of visitors within the facility.

[0024] General information is not information specific to a store or facility, but information related to various other matters. General information includes season, weather, and time period. The weather conditions indicated by the image analysis information described above indicate the current actual weather conditions near the facility, particularly the degree of rain or snow. In contrast, the weather in general information is information indicated by a weather forecast, for example, the weather every few hours in a wide area. The weather may include not only information such as sunny, cloudy, rain, etc., but also temperature. Time period is information indicating whether the current time is morning, daytime, evening, or night, or whether it is a weekday or a holiday. Season, weather, and time period are all information related to visitors, since they all affect the number of visitors visiting the facility.

[0025] The general information is posted on a website on the Internet. The AI ​​processing server 22 accesses the website via the Internet to acquire the general information.

[0026] Visitor information is information related to customers' visits to a facility or store. Visitor information includes the number of visitors, the timing of visit, and the number of people leaving. The number of visitors is the number of visitors to a facility or store according to date and time. The timing of visit is, for example, a time period in a day when the number of visitors is clearly higher or lower, or a day of the week when the number of visitors is clearly higher or lower. The number of people leaving is the number of people leaving a facility or store according to date and time.

[0027] For example, if the facility is a station, the number of visitors and the number of people leaving the facility according to date and time can be easily collected by using a ticket gate. If the facility is a shopping mall or a department store, the number of visitors and the number of people leaving the facility can be detected and collected using a sensor installed at the entrance. The sensor can be, for example, a photoelectric sensor, a camera, or a LiDAR. In this case, a computer having a function of analyzing the detection results of the sensor is installed in the facility or on the cloud. The collected visitor information is stored in the visitor information database 12. The visitor information database 12 is stored in a computer in the facility or in a server device installed on the cloud, and the contents are updated as needed. The AI ​​processing server 22 periodically accesses the visitor information database 12 to acquire the visitor information.

[0028] Store information is information related to a store. Store information includes product inventory, product sales, advertisements, events, and store maps. Product inventory is the current number of items in stock for each item. The number of items in stock is the sum of the number of items displayed on store fixtures and the number of items stored in the back yard, and these are registered separately. Note that the number of items in stock may include not only the current number of items in stock, but also past numbers of items in stock. Product sales is the number of sales points for each item according to the date and time. Advertisements are information that associates the type of advertisement with the date and time the advertisement was placed. Events are information that associates the type of event with the date and time the event was held. Store maps are information that shows the aisles, store fixtures, and facilities within the store.

[0029] Product inventory and product sales can be obtained from information registered in the POS system, inventory management system, or ordering system introduced by the store. Advertisements and events can be obtained from information recorded by the store operator, etc. Store maps are information used daily in stores. These store information are stored in the store information database 13. The store information database 13 is stored in a computer within the facility or in a server device located on the cloud, and the contents are updated as needed. The AI ​​processing server 22 periodically accesses the store information database 13 to obtain visitor information.

[0030] The above-mentioned commercial information is an example, and some of it may be omitted, or other information may be added. Other information includes sensor information detected by a sensor. The sensor information includes, for example, heat information detected by a heat sensor, vibration information detected by a vibration sensor, three-dimensional detection information detected by a three-dimensional sensor, and sound information detected by a sound sensor. For example, from the heat information, it is possible to detect, for example, the outdoor or indoor temperature. From the vibration information or three-dimensional detection information, it is possible to detect, for example, the crowded state. From the sound information, it is possible to detect, for example, the bustle of an event venue.

[0031] The AI ​​processing server 22 analyzes commercial information using AI processing, which will be described later, to estimate new information and generate commands for causing the robot 30 to perform work.

[0032] The robot 30 is capable of autonomous travel. For example, the robot 30 can carry out the task of transporting merchandise and the task of serving visitors. The robot 30 includes a detection sensor 31, a control device 32, a transport device 33, a travel drive unit 34, a display unit 35, and a speaker 36.

[0033] The detection sensor 31 detects the surroundings of the robot 30 to obtain various information. The detection sensor 31 is a camera, a microphone, a radar, a LiDAR, or an ultrasonic sensor. By traveling based on the detection result of the detection sensor 31, the robot 30 does not collide with passengers or pillars. The detection sensor 31 is also used to recognize a target product when transporting the product, and to recognize a visitor's voice inquiry when responding to the visitor. The control device 32 is a microcomputer having a CPU, storage, memory, communication device, etc., and controls the transport device 33, the travel drive unit 34, the display unit 35, and the speaker 36 based on the received command.

[0034] The transport device 33 is a tool used by the robot 30 to transport goods. The transport device 33 includes, for example, a tool for picking up goods and a platform for placing the picked up goods. The travel drive unit 34 is a motor that drives wheels, crawlers, legs, etc. for moving the robot. The control device 32 controls the travel drive unit 34 according to a destination position such as a destination of the goods.

[0035] The display unit 35 is a liquid crystal display or an organic EL display and is capable of displaying information. The display unit 35 is a touch panel and can be operated by a store clerk or a customer. The control device 32 generates an image showing a response to the customer's inquiry and causes the display unit 35 to display it. The speaker 36 emits sound based on the voice data. The control device 32 generates voice data showing a response to the customer's inquiry and causes the speaker 36 to output the sound.

[0036] Next, with reference to FIG. 2, information estimated by the AI ​​processing server 22 through AI processing will be described.

[0037] The information estimated by AI processing is divided into store clerk assistance information and customer response information. Store operation assistance information is information for assisting store operations. Specifically, the store operation assistance information includes information for the robot 30 to perform operations and information to be presented to store clerks. Customer response information is information that is mainly used by the robot 30 to respond to inquiries from customers.

[0038] Store operation support information includes "demand information for each product" and "replenishment amount based on demand information." Demand information for each product is information that estimates changes in demand for each product. The estimation time varies, for example, estimating changes in demand in a few hours, within the same day, within the next day, one week from now, etc. The replenishment amount based on demand information is a recommended value for the number of products to replenish in store fixtures based on changes in demand.

[0039] Next, an example of the relationship between commercial information, demand information, and replenishment amount will be described as a basis for the fact that appropriate demand information and replenishment amount can be estimated by AI processing. The demand for a product is related to, for example, the appeal of the product itself and the number of visitors. Here, the appeal of the product itself is related to the past sales of the product. However, since the appeal of the product itself may decrease over time, it is necessary to consider time. The number of visitors is related to the visitor information. Note that the number of visitors also changes depending on the season, weather, and time of day, so it is preferable to further consider general information. Furthermore, when demand information for a specific location is required, such as sales at an event venue, it is preferable to consider the congestion situation for each location obtained from image analysis information. Furthermore, the replenishment amount is related to the demand information and the current inventory. As shown above, since commercial information, demand information, and replenishment amount are related, it is possible to estimate the demand information and replenishment amount based on the commercial information.

[0040] The store operation support information includes "timing of employee breaks based on predicted congestion." The timing of employee breaks may be determined in advance. In this case, a situation may occur in which the number of employees working in the store is small even though the store is very crowded. Even if the employee decides the timing of breaks based on the current congestion situation, the store may become very crowded after the employee goes on break. As described above, when the timing of breaks is determined using these methods, the balance between the congestion situation and the number of employees working in the store becomes poor. In contrast, if the congestion situation can be predicted, the employee can take a break according to the predicted congestion situation. As a result, a balance between the congestion situation and the number of employees working in the store can be achieved.

[0041] Next, an example of the relationship between commercial information and break timing will be described as a basis for the fact that the AI ​​processing can estimate the appropriate break timing. As a premise, it will be described that there is a relationship between commercial information and the crowded situation. That is, it is clear that the number of visitors is related to the crowded situation. However, as described above, it is preferable to consider not only the number of visitors but also general information. Furthermore, since there may be cases where the number of visitors to the facility is large but the number of visitors to the store is small, it is preferable to further consider the crowded situation for each location included in the image analysis information. In addition, since the crowded situation may be influenced by advertisements or events, the crowded situation is also related to the store information. As shown above, commercial information and crowded situation are related. Furthermore, if the crowded situation can be estimated, the balance can be achieved by allocating the number of store clerks according to the level of the crowded situation. Therefore, it is easy to determine the break timing of the store clerk based on the estimated result of the crowded situation. As shown above, since commercial information and the break timing of the store clerk are related, it is possible to estimate the break timing of the store clerk based on the commercial information.

[0042] Store operation support information includes "products or equipment according to season, weather, events, etc." Some products sell well in certain seasons or weather, and some products sell well when certain events are held. It is preferable to display a large number of these products on store fixtures in advance. The same goes for equipment; for example, if it suddenly starts raining, it is preferable to quickly set up equipment such as umbrella stands or umbrella bags.

[0043] Next, an example of the association between commercial information and products or equipment according to the season, etc. will be described as a basis for the fact that AI processing can estimate products or equipment according to the season, etc. For example, by using commercial information, it is possible to identify the correspondence between sales of a specific product and the season, weather, time period, events, etc. Then, by analyzing the correspondence between the timing when sales of a specific product were higher than usual and the season, weather, time period, events, etc., it is possible to identify appropriate products according to the season, etc. Furthermore, sudden rain can be identified from the current weather conditions in the image analysis information. As described above, since commercial information and products or equipment according to the season, etc. are associated, it is possible to estimate products or equipment according to the season, etc. based on commercial information.

[0044] The visitor handling information includes "appropriate waiting locations based on congestion and weather conditions." A waiting location is, for example, a location where a visitor waits for a break, to meet someone, to kill time until a reservation time, or the like. If there are multiple waiting locations in a facility, the congestion status of each waiting location changes from time to time. Therefore, a situation may occur where one waiting location is very crowded while another waiting location is empty. Also, a situation may occur where the guide sign for an outdoor waiting location remains on even in heavy rain or very cold weather conditions. In other words, since an appropriate waiting location differs depending on congestion and weather conditions, the currently appropriate waiting location can be useful information for a visitor.

[0045] Next, an example of the association between commercial information and an appropriate waiting place will be described as a basis for the fact that an appropriate waiting place can be estimated by AI processing. The image analysis information described above includes the congestion state for each indoor and outdoor location. Therefore, the image analysis information includes the congestion state for each indoor or outdoor waiting place. The image analysis information also includes the current weather conditions. Note that it is preferable to further consider general information in order to take into account future weather changes and temperature, etc. As shown above, since commercial information and an appropriate waiting place are associated with each other, it is possible to estimate an appropriate waiting place based on the congestion state and weather conditions based on the commercial information.

[0046] The customer response information includes "queue waiting time." Customers may decide whether or not to stand in line by taking the waiting time into consideration, and knowing the waiting time may reduce the discomfort of standing in line. Therefore, the waiting time of the line can be useful information for customers.

[0047] Next, an example of the relationship between commercial information and queue waiting time will be described as a basis for the fact that the queue waiting time can be estimated by AI processing. The image analysis information described above includes the congestion status for each indoor and outdoor location. Therefore, the image analysis information includes the congestion status of the queue. In addition, since the image analysis information is updated at a predetermined time interval, the progress of the queue per unit time can be obtained based on multiple images. The queue progress can also be estimated based on past cases. That is, the image analysis information includes the progress of the queue when a queue occurred in the past. In addition, the progress of the queue differs depending on the type of queue, but since the store information includes event information, it is possible to obtain the past progress of similar queues by using it. As shown above, since commercial information and appropriate waiting places are related to each other, the queue waiting time can be estimated based on commercial information.

[0048] The visitor response information includes "recommended products based on season, weather, events, etc." As described above, there are needs for products based on season, weather, events, etc., and recommending products that satisfy those needs is useful information. Also, as described above, it is possible to estimate products based on season, etc., based on commercial information.

[0049] As described above, the commercial information and the information estimated by the AI ​​are related. However, since many factors are involved, it is difficult to estimate these pieces of information using a uniform rule. In this regard, the AI ​​processing server 22 estimates information using the judgment model shown below. The judgment model is created by machine learning the commercial information and the store work auxiliary information or the visitor handling information in past examples or virtual examples for learning. The machine learning may be deep learning. As a result, the above-mentioned relationship between the commercial information and the store work auxiliary information or the visitor handling information is learned. Then, at the time of estimation, the AI ​​processing server 22 inputs the acquired commercial information into the judgment model, and can obtain the store work auxiliary information and the visitor handling information as the output of the judgment model. The judgment model uses the past examples or virtual examples for learning as teacher data and outputs an estimation result based on the teacher data. In this way, the AI ​​processing server 22 estimates information using the identification system AI. However, the AI ​​processing server 22 may generate information using the generation system AI. Even when the generation system AI is used, the data to be machine-learned is the same as that of the identification system AI. By using generative AI, new information can be estimated, not just the store operation support information or customer service information that has been learned. This makes it possible to make new suggestions to store clerks or customers based on the new information.

[0050] Next, the AI ​​processing server 22 generates commands based on the estimated store operation auxiliary information and transmits them to the robot 30, and transmits the estimated customer handling information to the robot 30, causing the robot 30 to perform store operations. As shown in Fig. 2, the store operations performed by the robot are divided into product transportation, information presentation to store clerks, and information presentation to customers.

[0051] Product transportation includes "replenishment of products according to demand information." Specifically, instructions including an identification number for identifying the product to be replenished, a location where the product to be replenished is stored, a location where the product to be replenished is arranged, the number of products to be replenished, etc. are sent from the AI ​​processing server 22 to the robot 30. The robot 30 picks up the products and arranges them on the store fixtures based on the instructions received from the AI ​​processing server 22. This allows products to be replenished based on predicted demand information, making it less likely that products will not be arranged on the shelves. In addition, when the products are heavy or when there are many products to be replenished, the burden on the store staff can be particularly reduced. Note that when it is determined that there are no products to be replenished, the AI ​​processing server 22 or the robot 30 may order the corresponding products.

[0052] The product transportation includes "transporting the product selected by the customer." The robot 30 waits in the store unless there is a special command from the AI ​​processing server 22. The customer instructs the robot 30 to transport the product by voice or touch panel operation. As a result, the robot 30 tracks the customer, picks up the product instructed by the customer, places it on a stand, and transports it. The robot 30 may present recommended products, which will be described later, while transporting the product selected by the customer. In addition, if a fee is charged for transporting the product or presenting information, which will be described later, the robot 30 may be provided with a payment terminal. For example, it is preferable to provide the robot 30 with a payment terminal that can handle payment by electronic money or IC tickets.

[0053] The information presented to the store clerk includes "presentation of break timing" and "preparation of merchandise or equipment according to the season, etc." When the robot 30 receives an inquiry from the store clerk about the break timing, the robot 30 presents the break timing estimated by AI processing using the display unit 35 or the speaker 36. When the robot 30 receives an inquiry about merchandise or equipment according to the season, etc., the robot 30 presents the merchandise or equipment according to the season, etc., estimated by AI processing using the display unit 35 or the speaker 36. The robot 30 may present information to the store clerk on its own initiative.

[0054] The information presented to the customer includes "presentation of an appropriate waiting place," "queue waiting time," "recommended products based on the season, etc.", and "guidance of products and sales locations." When the robot 30 receives an inquiry from a customer, it presents, using the display unit 35 or the speaker 36, an appropriate waiting place, queue waiting time, or recommended products based on the season, etc., estimated by AI processing, depending on the content of the inquiry. The robot 30 may present information to the customer on its own initiative.

[0055] Moreover, information may be presented to visitors by the AI ​​processing server 22 in addition to the robot 30. In this case, the visitor uses a communication device such as a smartphone to access a website provided by the AI ​​processing server 22 and inputs an inquiry. The AI ​​processing server 22 analyzes the content of the inquiry, generates a response to the inquiry, and displays it on the website.

[0056] The information presented to the customer further includes "guidelines for products and sales locations." When the robot 30 receives an inquiry from a customer, the robot 30 accesses a store map, an inventory management system, etc. according to the content of the inquiry to identify the location of the relevant product or sales location, and presents it to the customer using the display unit 35 or the speaker 36. The robot 30 may also guide the customer to the relevant location.

[0057] As shown in Fig. 2, the AI ​​processing server 22 estimates information to be estimated by AI based on commercial information, and generates commands or information based on the information. Alternatively, the AI ​​processing server 22 may directly generate commands or information based on the commercial information. In this case, during learning, machine learning can be performed by associating the commercial information with the store operations to be performed based on the commercial information. Furthermore, during estimation, the store operations to be performed are output by inputting commercial information.

[0058] Next, the flow of processing performed by the store operations support system 1 will be described with reference to the flowcharts of FIG. 3 and FIG.

[0059] First, the image analysis server 21 acquires the image transmitted by the imaging device 11 (S101). Next, the image analysis server 21 analyzes the acquired image and generates image analysis information (S102).

[0060] Next, the AI ​​processing server 22 acquires commercial information (S103). Specifically, the AI ​​processing server 22 acquires image analysis information from the image analysis server 21, acquires general information from websites on the Internet, acquires visitor information from the visitor information database 12, and acquires store information from the store information database 13.

[0061] Next, the AI ​​processing server 22 analyzes the commercial information by AI processing to generate store operation support information and visitor handling information (S104). Furthermore, the AI ​​processing server 22 generates a command based on the store operation support information (S105) and transmits the visitor handling information to the robot 30 (S106). The robot 30 stores the visitor handling information received from the AI ​​processing server 22 and uses it to respond to inquiries from visitors, etc.

[0062] Next, the AI ​​processing server 22 judges whether the robot 30 is working or not (S107). For example, if the robot 30 is replenishing merchandise or transporting merchandise selected by a customer, the robot 30 is working and cannot perform another task. Therefore, in this case, the AI ​​processing server 22 does not transmit the generated command at the present time, and transmits the command to the robot 30 when the robot 30 is no longer working (S108).

[0063] These processes are performed at regular intervals, so that the robot 30 can replenish products based on the latest commercial information, and the latest information can be presented to customers.

[0064] The robot 30 performs the process shown in Fig. 4. First, the robot 30 determines whether or not it has received a command to transport a product from the AI ​​processing server 22 (S201). If it determines that it has received a command to transport a product, the robot 30 transports the product based on the command (S202).

[0065] The robot 30 determines whether or not a visitor has made an inquiry (S203). When the robot 30 determines that a visitor has made an inquiry, it responds to the visitor based on visitor response information and store information (S204). The store information is used to provide information about products and sales locations. In addition, the robot 30 may directly respond to questions that it can answer by itself, and transmit questions that it cannot answer to the AI ​​processing server 22, and present the answers received from the AI ​​processing server 22 on the display unit 35 or speaker 36.

[0066] The robot 30 determines whether or not an inquiry has been received from a store clerk (S205). If the robot 30 determines that an inquiry has been received from a store clerk, it responds to the store clerk based on the store work assistance information (S206). In this case, too, the robot 30 may directly respond to questions that it can answer itself to the store clerk, and may transmit questions that it cannot answer itself to the AI ​​processing server 22 and present the answers received from the AI ​​processing server 22 on the display unit 35 or speaker 36. Also, as described above, inquiries from customers or store clerks can be directly handled by the AI ​​processing server 22.

[0067] (Feature 1) The store operation support system 1 of this embodiment includes an AI processing server 22 and a robot 30. The AI ​​processing server 22 receives commercial information related to the store, products, or customers, analyzes the input commercial information using AI processing, and estimates demand information indicating the demand for each product. The robot 30 replenishes products based on the demand information estimated by the AI ​​processing server 22.

[0068] This allows AI to be used to analyze commercial information, making it possible to estimate events that were previously difficult to estimate because many factors are involved. In addition, the robot 30 performs store operations based on the estimated information, reducing the burden on store staff. In particular, it allows for early and accurate response to increases in demand.

[0069] (Feature 2) The store operation support system 1 of this embodiment includes an AI processing server 22 and a robot 30. The AI ​​processing server 22 receives commercial information related to the store, products, or customers, including images captured indoors or outdoors by the imaging device 11 or analysis results of the images, analyzes the input commercial information by AI processing, and estimates visitor handling information that is based on at least the indoor or outdoor environment and is information to be presented to visitors. The robot 30 responds to inquiries from visitors based on the visitor handling information estimated by the AI ​​processing server 22.

[0070] This allows AI to be used to analyze commercial information, making it possible to estimate events that were previously difficult to estimate because many factors are involved. In addition, the robot 30 can perform store operations based on the estimated information, reducing the burden on store staff. In particular, it can provide accurate answers to inquiries from customers.

[0071] (Feature 3) In the store operation support system 1 of this embodiment, the robot 30 includes a transport device 33 that transports the products selected by the customers.

[0072] This allows the conveying device to be used to provide services to customers even when product replenishment is not required.

[0073] (Feature 4) In the store operation support system 1 of this embodiment, the AI ​​processing server 22 analyzes commercial information by AI processing and estimates appropriate break timings for store clerks. The robot 30 presents the appropriate break timings for store clerks estimated by the AI ​​processing server 22 to the store clerk.

[0074] This allows, for example, store staff to be informed of break times based on the level of congestion.

[0075] (Feature 5) In the store operation support system 1 of this embodiment, the AI ​​processing server 22 analyzes commercial information by AI processing and estimates products based on the season, weather, or event. The robot 30 presents the products estimated by the AI ​​processing server 22 to the store clerk or customer.

[0076] This makes it possible to display products appropriate to the situation on store fixtures and to present products appropriate to the situation to customers.

[0077] (Feature 6) In the store operation support system 1 of this embodiment, the AI ​​processing server 22 analyzes commercial information by AI processing and estimates an appropriate waiting place based on the congestion situation and weather conditions. The robot 30 presents the product estimated by the AI ​​processing server 22 to the store clerk or customer.

[0078] This makes it possible to display products appropriate to the situation on store fixtures and to present products appropriate to the situation to customers.

[0079] (Feature 7) In the store operation support system 1 of this embodiment, the AI ​​processing server 22 analyzes commercial information by AI processing and estimates the queue waiting time. The robot 30 presents the queue waiting time estimated by the AI ​​processing server 22 to the customer.

[0080] This allows visitors to be presented with information to help them decide whether or not to join the queue.

[0081] (Feature 8) In the store business support system 1 of this embodiment, the commercial information used by the AI ​​processing server 22 when performing AI processing further includes at least one of temperature information, sound information, three-dimensional detection information, and vibration information detected using a sensor.

[0082] This allows the situation to be estimated more accurately.

[0083] (Feature 9) In the store operations support system 1 of this embodiment, the commercial information used by the AI ​​processing server 22 when performing AI processing includes visitor information, which is information about visitors, and store information, which is information about the store.

[0084] This allows the situation to be estimated more accurately.

[0085] The above-mentioned features 1 to 9 can be combined as follows to realize a store operation support system. The same applies to the method or program. [Configuration 1] Store operation support system 1 having feature 1 [Configuration 2] Store operation support system 1 with feature 2 [Configuration 3] In addition to configuration 1 or 2, a store operation support system 1 further having feature 3 [Configuration 4] In addition to any one of configurations 1 to 3, a store operation support system 1 further having feature 4 [Configuration 5] In addition to any one of configurations 1 to 4, a store operation support system 1 further having feature 5 [Configuration 6] In addition to any one of configurations 1 to 5, a store operation support system 1 further having feature 6 [Configuration 7] In addition to any one of configurations 1 to 6, a store operation support system 1 further having feature 7 [Configuration 8] In addition to any one of configurations 1 to 7, a store operation support system 1 further having feature 8 [Configuration 9] In addition to any one of configurations 1 to 8, a store operation support system 1 further having feature 9

[0086] The functions of the elements disclosed herein can be performed using circuits or processing circuits, including general purpose processors, special purpose processors, integrated circuits, Application Specific Integrated Circuits (ASICs), conventional circuits, and / or combinations thereof, configured or programmed to perform the disclosed functions. Processors are considered processing circuits or circuits because they include transistors and other circuits. In this disclosure, a circuit, unit, or means is hardware that performs the recited functions or hardware that is programmed to perform the recited functions. The hardware may be hardware disclosed herein or other known hardware that is programmed or configured to perform the recited functions. Where the hardware is a processor, which is considered a type of circuit, the circuit, means, or unit is a combination of hardware and software, and the software is used to configure the hardware and / or the processor. [Explanation of symbols]

[0087] 1. Store operation support system 21 Image analysis server 22 AI processing server (AI processing section) 30. Robot

Claims

1. an AI processing unit that receives commercial information related to a store, a product, or a customer, analyzes the input commercial information by AI processing, and estimates demand information indicating demand for each product; A robot that replenishes products based on the demand information estimated by the AI ​​processing unit; A store operations support system equipped with:

2. an AI processing unit that receives commercial information related to a store, a product, or a customer, and that includes an image captured indoors or outdoors by an imaging device or an analysis result of the image, analyzes the input commercial information by AI processing, and estimates customer response information that is based on at least the indoor or outdoor environment and is to be presented to the customer; A robot that responds to inquiries from visitors based on the visitor response information estimated by the AI ​​processing unit; A store operations support system equipped with:

3. The store operation support system according to claim 1, The store operations support system includes a transport device that transports the products selected by the customers.

4. The store operation support system according to claim 2, The AI ​​processing unit analyzes the commercial information by AI processing to estimate appropriate break timings for store clerks, The robot is a store operations support system that presents appropriate break times for store employees estimated by the AI ​​processing unit to the store employees.

5. The store operation support system according to claim 2, The AI ​​processing unit analyzes the commercial information by AI processing to estimate products based on seasons, weather, or events, The robot is a store operations support system that presents the products estimated by the AI ​​processing unit to store clerks or customers.

6. The store operation support system according to claim 2, The AI ​​processing unit analyzes the commercial information by AI processing and estimates an appropriate waiting location based on congestion and weather conditions, The robot is a store operations support system that presents the products estimated by the AI ​​processing unit to store clerks or customers.

7. The store operation support system according to claim 2, The AI ​​processing unit analyzes the commercial information by AI processing to estimate a queue waiting time, The robot is a store operations support system that presents customers with the queue waiting time estimated by the AI ​​processing unit.

8. The store operation support system according to claim 1 or 2, A store operations support system, wherein the commercial information used by the AI ​​processing unit when performing the AI ​​processing further includes at least one of temperature information, sound information, three-dimensional detection information, and vibration information detected using a sensor.

9. The store operation support system according to claim 1 or 2, A store operations support system, wherein the commercial information used by the AI ​​processing unit when performing the AI ​​processing includes visitor information, which is information about visitors, and store information, which is information about the store.

10. Commercial information relating to a store, a product, or a customer is input, and the input commercial information is analyzed using AI processing to estimate demand information indicating demand for each product; A store operation support method that causes a robot to replenish products based on the demand information estimated by the AI ​​processing.

11. Commercial information relating to a store, a product, or a customer is input, and the input commercial information is analyzed using AI processing to estimate demand information indicating demand for each product; A store operations support program that causes a computer to execute a process that generates instructions for a robot to replenish products based on the demand information estimated by the AI ​​processing.

12. commercial information is input, the commercial information being information related to a store, a product, or a customer, and including an image generated by capturing an image of an indoor or outdoor area using an imaging device or an analysis result of the image; the input commercial information is analyzed using AI processing to estimate customer response information that is based on at least the indoor or outdoor environment and is information to be presented to the customer; A store operation support method in which a robot answers inquiries from customers based on the customer response information estimated by the AI ​​processing.

13. commercial information is input, the commercial information being information related to a store, a product, or a customer, and including an image generated by capturing an image of an indoor or outdoor area using an imaging device or an analysis result of the image; the input commercial information is analyzed using AI processing to estimate customer response information that is based on at least the indoor or outdoor environment and is information to be presented to the customer; A store business support program that transmits the customer response information estimated by the AI ​​processing to a robot as information for the robot to respond to inquiries from customers.

Images