Information processing device, information processing method, and information processing program

Abstract

To provide an information processing device, an information processing method, and an information processing program that can improve the profitability of stores. [Solution] According to the present disclosure, an information processing device for generating discount information relating to discounts on store products is provided, comprising: a target determination unit and a generation unit, wherein the target determination unit determines the target to which the discount information is to be provided based on classification information relating to customer classification and consumption information relating to the consumption of the product, and the generation unit generates the discount information based on the target to be provided.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a system for issuing coupons that can be used in a store to a customer's mobile terminal. In this system, it is determined whether to issue a coupon based on the mobile terminal position information, the store position information, and the inventory position information.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

[10] [9], wherein the quality evaluation unit, when classifying the quality of the discount information into the second quality, further classifies the quality of the discount information into either a predetermined third quality or a fourth quality which is lower than the third quality; when classifying the quality of the discount information into the third quality, outputs a correction request requesting correction of the discount information as information relating to the correction of the discount information; when classifying the quality of the discount information into the fourth quality, corrects the generated discount information based on the quality information, and outputs the corrected discount information as information relating to the correction of the discount information.

[11] A method for generating discount information relating to discounts on store products, comprising: a target determination step and a generation step, wherein in the target determination step, a target is determined to be the recipient of the discount information based on classification information relating to customer classification and consumption information relating to the consumption of the product, and in the generation step, the discount information is generated based on the target.

[12] Information processing program that causes a computer to function as an information processing device for generating discount information relating to discounts on store goods, comprising a target determination unit and a generation unit, wherein the target determination unit determines the target to which the discount information is to be provided based on classification information relating to customer classification and consumption information relating to the consumption of the goods, and the generation unit generates the discount information based on the target to be provided. [Effects of the Invention]

[0007] The information processing device disclosed herein can determine which customers can be targeted for discounts to effectively increase their order volume by providing discount information based on information that influences product consumption (e.g., order volume) (consumption information) and customer classification according to customer attributes (classification information). Furthermore, by providing discount information to the determined target customers, it is possible to effectively attract customers. Therefore, the store's profits can be improved. [Brief explanation of the drawing]

[0008] [Figure 1] This is a diagram showing the configuration of system 10 according to one embodiment. [Figure 2] This block diagram shows the hardware configuration of Server 1 as shown in Figure 1. [Figure 3] This figure shows an example of time-series data on product order volume at a store. [Figure 4] Figure 1 is a block diagram showing the functional configuration of Server 1. [Figure 5] This table shows an example of the information stored by the memory unit 11 shown in Figure 4. [Figure 6] This is a flowchart showing the operation of Server 1 according to one embodiment. [Figure 7] Figure 6 is a flowchart showing an example of the quality evaluation process. [Modes for carrying out the invention]

[0009] Embodiments of this disclosure will be described below with reference to the drawings. The various features shown in the embodiments below can be combined with each other. Furthermore, each feature can constitute an invention independently. In addition, elements not specified in the claims of the embodiments below are optional and can be omitted.

[0010] 1. Configuration of System 10 System 10 according to this embodiment will be explained using Figure 1. System 10 is used for a store that sells goods. The goods are not limited to tangible objects, but may also be intangible things such as services. Goods are, for example, the provision of food and beverages within the store, and food and beverages consumed outside the store. The store sells goods to customers. The store is, for example, a restaurant. In this case, the customer consumes the goods inside the store (e.g., eat-in) or consumes the goods outside the store. When a customer consumes goods outside the store, the customer takes the goods out of the store and consumes them outside the store (e.g., takeout), or requests delivery from the store and consumes the goods outside the store (e.g., delivery). In other words, consumption of goods outside the store includes takeout and delivery. In this embodiment, the store is a restaurant that provides food and beverages, and food and beverages themselves are the goods.

[0011] System 10 includes a server 1 (information processing device), a customer terminal 2, and a store system 3. Server 1, customer terminal 2, and store system 3 are configured to communicate with each other via a network N. Network N is a mobile communication network, a public wireless communication LAN (Local Area Network), etc. The communication method between Server 1, customer terminal 2, and store system 3 is typically wireless communication.

[0012] Server 1 acquires information from at least one of customer terminal 2 and store system 3. Further, Server 1 generates discount information related to price discounts of products in the store. The discount information is information for notifying customers of price discounts of products. The discount information may be, for example, a coupon usable in the store, or information indicating that a product is discounted during a predetermined time period (e.g., sale information). In this embodiment, the discount information is a coupon. Server 1 outputs the generated discount information to customer terminal 2. As will be described later, Server 1 improves the profit of the store by adjusting the demand for products consumed outside the store. Details of Server 1 will be described later.

[0013] Customer terminal 2 is a terminal operated by a customer. Customer terminal 2 may be, for example, a mobile terminal such as a smartphone, a tablet terminal, and a notebook computer. Customer terminal 2 may also be a desktop personal computer or the like.

[0014] The customer inputs his / her own information via customer terminal 2. Specifically, the customer accesses a website via customer terminal 2. The website is, for example, a site used for the use of the store. The customer inputs membership information such as his / her name, address, nearest station, occupation, place of work, income, household composition, hobbies, membership ID (Identification), and password via customer terminal 2, for example, and performs membership registration. Customer terminal 2 outputs the input membership information to Server 1.

[0015] Hobbies mean preferences related to products. Hobbies include, for example, at least one of favorite food and drink, requirements for the store, frequency of eating out, eating-out expenses for a predetermined period (e.g., one month), desired price range, and personality. Requirements for the store are, for example, the atmosphere of the store, whether smoking is allowed, types of seats, and usage scenarios of the store. Personality is a characteristic of a customer's behavior and way of thinking. Personality is, for example, liking to eat with a large group, liking to talk to store clerks, and liking to plan drinking parties.

[0016] In addition, the website is used when customers consume products outside the store. For example, the customer inputs the membership ID and password via the customer terminal 2 and logs in to the website. The customer inputs at least one of the membership ID, product details, amount, quantity of products, and whether a coupon is used into the customer terminal 2. After executing the settlement process based on the input information, the customer terminal 2 outputs the product details, amount, quantity of products, whether a coupon is used, order date, and order time as off-premises order information to the store system 3. In this embodiment, the customer terminal 2 outputs all of this information as order information to the store system 3.

[0017] The store system 3 outputs the order information to the store staff (e.g., store employees). The store staff manufactures the products based on the order information output from the store system 3. When the customer takes out the products, the customer goes to the store to pick them up. When the customer requests delivery of the products, the store staff delivers the manufactured products to the customer's address.

[0018] The store system 3 is composed of terminals installed in the store. The store system 3 is composed of, for example, a plurality of terminals. The plurality of terminals includes, for example, POS (Point Of Sales) terminals.

[0019] The store system 3 outputs various information to the server 1. The store system 3 outputs at least one of service level information, store staff information, purchase information, cooking information, order information, coupon information, product information, store location information, discount limit amount, and order placement information.

[0020] Service level information indicates the level of service provided by the store. Service level information includes the maximum acceptable waiting time and the acceptable violation rate. The maximum acceptable waiting time is the maximum allowable waiting time from the time of order placement until the product is completed. The violation rate is the probability that the waiting time exceeds the maximum acceptable waiting time, for example, the ratio of orders where the waiting time exceeds the maximum acceptable waiting time to the total number of orders. The acceptable violation rate is the maximum acceptable probability of a violation rate. The maximum acceptable waiting time and the acceptable violation rate are selected, for example, by a store employee and entered into the store system 3.

[0021] The store system 3 receives service level information for each channel. Here, the channel indicates a classification according to the content of the service. In this embodiment, the channel includes eat-in, take-out, and delivery. The acceptable waiting time and violation rate differ depending on the content of the service. For example, the acceptable waiting time for delivery is considered to be longer than the acceptable waiting time for eat-in. In this way, by receiving service level information for each channel in the store system 3, it becomes possible to output information according to the content of the service.

[0022] Employee information is information relating to the store's employees. Employee information includes, for example, the number of employees, personnel, skill level, and past turnover rate. The number of employees is the number of employees working in the store per predetermined unit of time (e.g., 1 hour). Personnel are the characteristics of the employees working in the store. Employee characteristics include, for example, at least one of job title and gender. Skill level is the technical level of the employees. Skill level includes, for example, at least one of years of work experience, qualifications, and performance evaluations. Past turnover rate includes the store's past turnover rate and employee composition per predetermined unit of time (e.g., 1 hour). "Past" means a time before the server 1 starts the operation described later. The number of employees, personnel, and skill level are selected, for example, by the employees and entered into the store system 3. The store system 3 calculates the past turnover rate by known means based on the aforementioned order information.

[0023] Purchase information is information related to the purchase of goods. Purchase information includes, for example, at least one of the following: product details, quantity of goods, order time, and preparation completion time. Product details are information used to identify the goods, and are typically the product name or ID. Order time is the time the goods were ordered. Preparation completion time is the time the goods were ready. Store system 3 retrieves the product details, quantity of goods, and order time included in the order information. Preparation completion time is entered into store system 3, for example, by a store employee.

[0024] Cooking information is information related to the manufacturing of a product (cooking in this embodiment). Cooking information includes, for example, the contents of the product, the number of products, the cook, and at least one of the execution parallelism. The cook is the store employee in charge of manufacturing the product. The cook may be the name of the store employee in charge of manufacturing the product, or the employee's ID. Execution parallelism is the degree to which multiple products can be manufactured in parallel. Execution parallelism is, for example, the number of products that can be manufactured simultaneously. Store system 3 obtains the contents of the products and the number of products included in the order information. The cook and execution parallelism are selected, for example, by the store employee and entered into store system 3.

[0025] Order information is information relating to a customer's order. Order information includes, for example, off-site order information and in-store order information. As described above, off-site order information includes at least one of the following: product, product details, price, quantity of product, order date, order time, and whether or not a coupon was used. The store system 3 acquires off-site order information output by the customer terminal 2. Also, when a customer purchases a product at the store, the store system 3 generates in-store order information. The contents of the in-store order information are the same as the contents of the off-site order information. The storage unit 11 stores the in-store order information and off-site order information as order information.

[0026] Coupon information is information relating to coupons, which are discount information. Coupon information includes, for example, information indicating coupons that may be issued by the store. Coupon information includes, for example, whether or not the coupon is available and at least one of the discount amount. Coupon information is selected, for example, by a store employee and entered into the store system 3.

[0027] Product information is information relating to the characteristics of a product. Product characteristics include, for example, the benefits that customers can gain from the product. Product information includes, for example, at least one of the following: a product newly sold by a store (so-called new product) and a product sold in limited quantities (limited product). Sold in limited quantities includes, for example, being sold in a limited area or for a limited time.

[0028] Store location information is information indicating the location of a store. Store location information is, for example, the location information of a terminal included in store system 3. Store location information includes, for example, longitude and latitude. Store system 3 acquires store location information using, for example, a known means such as GPS (Global Positioning System).

[0029] The discount limit is the maximum discount amount for the product related to the discount information. The discount limit may be, for example, the maximum amount discounted, or the maximum percentage discount. The discount limit may be set for each individual product. The discount limit is selected, for example, by a store employee and entered into the store system 3.

[0030] Order information is information related to orders placed by the store. The store places orders with suppliers other than the store for materials necessary to manufacture goods (e.g., ingredients). The order information includes at least one of the ordered materials, the source of the materials (e.g., place of origin, producer, and brand name), the amount, the quantity, the order date, and the order time. The order information is entered into the store system 3 by the store staff. In addition, if orders are placed using a terminal included in the store system 3, the order information may be generated by that terminal.

[0031] The above describes examples of information that the store system 3 outputs to server 1, but the information output to server 1 is not limited to the information described above. The type and content of the information can be changed as appropriate without departing from the spirit of the invention.

[0032] 2. Hardware configuration of Server 1 The hardware configuration of Server 1 according to this embodiment will be explained using Figure 2. As shown in Figure 2, Server 1 may be physically configured as a computer system including one or more CPUs (Central Processing Units) 101 which are processors, one or more ROMs (Read Only Memory) 102, one or more RAMs (Random Access Memory) 103 which are main memory, an HDD (Hard Disk Drive) 104 which is auxiliary memory, and a communication unit 105 such as a communication interface for communicating with a network. For example, when the CPU 101 reads a program P (information processing program) that is pre-stored in the HDD 104 into the RAM 103 and executes it, each piece of hardware operates under the control of the CPU 101, and data is read from and written to the ROMs 102, RAM 103, and HDD 104. This realizes the various functions of Server 1 that will be described later. In addition to the hardware configuration shown in Figure 2, Server 1 may also be equipped with input devices such as a mouse and keyboard, output devices such as a display and speakers, etc.

[0033] 3. Challenges For example, if a store is a restaurant, the store provides food and beverages to customers inside the store (so-called eat-in). Alternatively, the store sells food and beverages to customers, who consume them outside the store. For example, a customer goes to the store to pick up the goods and consumes them outside the store (so-called takeout). Another example is a customer requesting delivery from the store and consuming the delivered goods at the delivery location (so-called delivery).

[0034] In Figure 3, the predicted quantity of orders for goods consumed within the store (in-store orders) is shown by a plain bar graph, and the predicted quantity of orders for goods consumed outside the store (out-of-store orders) is shown by a hatched bar graph. The predicted quantity of in-store orders changes depending on the time of day. If the predicted quantity of in-store orders is small compared to the store's manpower (number of employees and staff), it means that unnecessary labor costs are being incurred. Therefore, it is desirable to increase at least one of the in-store and out-of-store orders during such times. Conversely, if the predicted quantity of in-store orders is large compared to the store's manpower, it means that there is a shortage of staff. Therefore, it is desirable to decrease at least one of the in-store and out-of-store orders, or increase the number of staff (the number of employees during that time period), during such times.

[0035] The server 1 according to this embodiment has a unique configuration in order to solve the above-mentioned problems. In the following description, an example will be described in which the server 1 adjusts the volume of orders placed outside the store by providing discount information. The server 1 will be described in detail below.

[0036] 4. Functional configuration of Server 1 As shown in Figure 4, Server 1 comprises a storage unit 11, a maximum supply quantity estimation unit 12, an order quantity estimation unit 13, a supply target determination unit 14, a generation unit 15, a quality evaluation unit 16, a supply quantity determination unit 17, and a supply unit 18.

[0037] The memory unit 11 stores various types of information. The information stored by the memory unit 11 is classified into operational information, demand information, consumption information, classification information, quality information, and past effectiveness information. The structure and content of each type of information will be explained in detail below using Figure 5.

[0038] Operational information refers to information related to the operation of a store. Operational information includes service level information for each channel, staff information, purchase information, and cooking information. Server 1 acquires service level information for each channel, staff information, purchase information, and cooking information from the store system 3. Storage unit 11 stores this acquired information as operational information.

[0039] Demand information is information related to the demand for a product. Demand information includes information that is thought to influence the demand for a product and thus the order volume of a store. Demand information includes order information, coupon information, product information, store location information, environmental information, and external information. Server 1 acquires order information, coupon information, product information, and store location information from the store system 3. Storage unit 11 stores this acquired information as demand information.

[0040] Environmental information refers to information about the environment surrounding the store. Environmental information includes, for example, weather and temperature. Server 1 accesses, for example, a weather forecast website. Based on the acquired store location information, Server 1 acquires the weather and temperature around the store as environmental information. Storage unit 11 stores the acquired environmental information as demand information.

[0041] External information is information relating to the area outside the store. External information includes, for example, information indicating factors outside the store that influence the volume of orders received outside the store. External information includes, for example, at least one of the presence or absence of an event and SNS (Social Network Service) information. An event is an event held in the vicinity of the store. Server 1 determines the presence or absence of an event in the vicinity of the store based on the acquired store location information. The vicinity of the store is, for example, an area located within a predetermined distance from the store's location indicated by the store location information. SNS information includes, for example, SNS trends. Server 1 accesses the SNS website and acquires SNS information using known means. Storage unit 11 stores the acquired external information as demand information.

[0042] Consumption information is information related to the consumption of goods. Consumption information includes information indicating factors that influence the in-store order volume and the out-of-store order volume of goods. In this embodiment, consumption information is out-of-store consumption information. Out-of-store consumption information is information related to the consumption of goods that takes place outside the store. Out-of-store consumption information includes information indicating factors that influence the out-of-store order volume of goods. Out-of-store consumption information includes order information, customer information, product information, store location information, environmental information, external information, and discount limit. Server 1 obtains the discount limit from store system 3. Storage unit 11 stores the obtained order information, product information, store location information, environmental information, external information, and discount limit as out-of-store consumption information.

[0043] Customer information is information relating to a customer. For example, customer information includes information relating to the actions taken by the customer at the store. Customer information includes at least one of the following: member ID, frequency of use, past purchase history, coupon usage history, and LTV (Life Time Value). Server 1 obtains order information from customer terminal 2. Storage unit 11 stores the member ID included in the obtained order information as customer information.

[0044] Usage frequency refers to how often a customer uses the store. For example, it refers to how often a customer uses the store within a predetermined period (e.g., one month). Server 1 calculates the customer's usage frequency based on order information. Storage unit 11 stores the calculated usage frequency as customer information.

[0045] Past purchase history is a record of products a customer has purchased in the past. Server 1 generates past purchase history based on order information obtained from customer terminal 2. Storage unit 11 stores the generated past purchase history as customer information.

[0046] The coupon usage history is a record of coupons used by a customer in the past. Server 1 generates the coupon usage history based on order information obtained from customer terminal 2. Storage unit 11 stores the generated coupon usage history as customer information.

[0047] LTV is the expected value of the profit to be obtained from a customer over a predetermined period (e.g., the period during which a customer uses the store). Server 1 calculates LTV based on order information obtained from customer terminal 2. Server 1 may calculate LTV using known means. Storage unit 11 stores the calculated LTV as customer information.

[0048] Classification information is information relating to the classification of a store's customers. Classification information includes, for example, the classification of customers based on predetermined criteria. Classification information includes at least one of the following: classification name, classification description, number of people, and profile. The classification name is information used to identify the customer classification. The classification description is information that indicates the criteria for classifying customers. Classification descriptions include, for example, customers who mainly order on specific days of the week, customers who mainly order in the afternoon, and customers who only pick up their goods in the evening. The number of people is the number of customers classified. The profile is information that shows the results of the analysis of the classified customers. Profiles include, for example, the customer's age group, gender, income group, and preference trends.

[0049] In this embodiment, Server 1 generates classification information based on order information. Server 1 generates classification information using a classification information generation model. The classification information generation model is trained to take order information as input and output classification information. The training method for the classification information generation model is not particularly limited. Server 1 inputs order information into the classification information generation model and generates classification information. Alternatively, the classification information may be generated by a store employee. In this case, the employee analyzes the order information and generates the classification information. The storage unit 11 stores the generated classification information.

[0050] Quality information is information related to the quality of discount information. As described later, when Server 1 generates discount information, the discount information may contain inappropriate information. Inappropriate information includes, for example, wording that gives customers a bad impression, wording that misleads customers, and wording that is contrary to the facts. Quality information is information used to correct discount information when it contains inappropriate information. Quality information includes, for example, laws and regulations, intellectual property information, store information, order information, purchase order information, environmental information, excluded keywords, inappropriate cases, professional ethics, and at least one of the following:

[0051] The laws and regulations refer to the laws and regulations relating to the operation of the store. In this embodiment, the laws and regulations include the Food Sanitation Act, the Food Labeling Act, and the Entertainment Business Act, etc. Intellectual property information refers to information relating to intellectual property. Intellectual property information includes at least one of a design, trademark, and copyrighted work. Store information refers to information relating to the display at the store. Store information includes, for example, at least one of the store's brand, the product's brand, and the product's material brand.

[0052] Excluded keywords are words that should be excluded from discount information. Excluded keywords include, for example, words that give customers a bad impression. Inappropriate incidents are incidents that have occurred in the past regarding the operation of the store. Inappropriate incidents include, for example, at least one of the following: incidents of violating laws and regulations, and incidents that caused a public outcry (incidents that incurred displeasure on social media, etc.). Professional ethics are ethical standards of conduct required in the operation of the store. Professional ethics include, for example, the prevention of food misconduct. Emotional parameters are parameters related to customer emotions. Emotional parameters include, for example, parameters that show positive and negative emotions for each customer classification based on predetermined criteria. Emotional parameters include, for example, information related to general customer trends such as "families tend to choose stores that have children's menus" and "customers with high annual incomes tend to prefer high-end menus." Quality information is selected by, for example, the administrator of server 1 and entered into server 1. The administrator of server 1 is, for example, the person who provides services to the store using server 1. Storage unit 11 stores the entered quality information.

[0053] Past effect information is information relating to the discount effect of past products. Past effect information includes at least one of the past discount amount and the CVR (Conversion Rate) at the time of the discount. The storage unit 11 stores past effect information for each discount information provided in the past. The past discount amount is the amount discounted by providing the past discount information. The CVR at the time of the discount is an indicator of the discount effect obtained by providing the discount information. Server 1 generates past effect information based on order information. Storage unit 11 stores the generated past effect information.

[0054] The above describes examples of information stored in the memory unit 11, but the information stored in the memory unit 11 is not limited to the information described above. The type and content of the information can be changed as appropriate without departing from the spirit of the invention.

[0055] The classification of information stored in the memory unit 11 has been explained above. Multiple classifications may include common information. For example, demand information and off-store consumption information include product information as a common piece of information. In other words, product information is classified as both demand information and off-store consumption information. Note that the classification of information stored in the memory unit 11 is not limited to the classifications described above.

[0056] The maximum supply quantity estimation unit 12 estimates the maximum supply quantity based on the operational information stored by the storage unit 11. The maximum supply quantity is the maximum amount of goods that a store can supply. The maximum supply quantity estimation unit 12 estimates the maximum supply quantity for each predetermined unit of time (for example, 1 hour).

[0057] The order volume estimation unit 13 estimates the order volume based on the demand information stored by the storage unit 11. The product order volume is the quantity of goods ordered by the store. The order volume estimation unit 13 estimates the product order volume for each predetermined unit of time.

[0058] In this embodiment, the order volume estimation unit 13 estimates the order volume for orders placed outside the store and the order volume for orders placed inside the store. The order volume estimation unit 13 estimates the sum of the order volume for orders placed outside the store and the order volume for orders placed inside the store as the order volume for products.

[0059] The order volume estimation unit 13 estimates the amount of orders placed outside the store using an off-store order volume estimation model. The off-store order volume estimation model is trained with demand information as the explanatory variable and off-store order volume as the dependent variable. The order volume estimation unit 13 inputs the demand information into the off-store order volume estimation model and estimates the amount of orders placed outside the store.

[0060] Furthermore, the order volume estimation unit 13 estimates the in-store order volume using an in-store order volume estimation model. The in-store order volume estimation model is trained with demand information as the explanatory variable and in-store order volume as the dependent variable. The order volume estimation unit 13 inputs the demand information into the in-store order volume estimation model and estimates the in-store order volume.

[0061] The service target determination unit 14 determines the service targets to which discount information will be provided, based on the classification information and consumption information stored by the storage unit 11. The service target determination unit 14 determines the service targets at predetermined time intervals. "Providing discount information" means that the server 1 outputs discount information to the customer terminal 2. The customer terminal 2 displays the discount information output from the server 1 on its display unit (e.g., a display).

[0062] In this embodiment, the product selection unit 14 estimates the sensitivity based on the consumption information stored by the storage unit 11. The product selection unit 14 estimates the sensitivity for each predetermined unit of time. Sensitivity is the degree of customer response to price reductions on products. In this embodiment, sensitivity is the degree of customer response to price reductions on products consumed outside of stores. Sensitivity is, for example, the percentage increase in the number of orders when the price is reduced by a predetermined percentage (e.g., 1%). For example, "the sensitivity for the 1-2 o'clock time slot is 10" means that when the price of a product is reduced by 1% during the 1-2 o'clock time slot, the number of orders increases by 10%. The definition of sensitivity is not particularly limited.

[0063] The target selection unit 14 estimates sensitivity using a sensitivity estimation model. The sensitivity estimation model is trained with consumption information as the explanatory variable and sensitivity as the dependent variable. The target selection unit 14 inputs consumption information into the sensitivity estimation model and estimates sensitivity.

[0064] In this embodiment, the target selection unit 14 determines the target product based on the estimated sensitivity and the classification information stored by the memory unit 11. The target selection unit 14 determines the target product using a target selection model. The target selection model is trained with sensitivity and classification information as explanatory variables and the target product as the dependent variable. The target selection unit 14 inputs the estimated sensitivity and classification information into the target selection model and determines the target product.

[0065] In this embodiment, the target product determination unit 14 determines the target product if the difference between the maximum supply quantity estimated by the maximum supply quantity estimation unit 12 and the order quantity estimated by the order quantity estimation unit 13 is a positive value.

[0066] The generation unit 15 generates discount information based on the target product determined by the target product determination unit 14. The generation unit 15 generates discount information based, for example, on external information and classification information stored by the storage unit 11 in addition to the target product. The discount information is, for example, the content provided to the customer. The discount information includes, for example, at least one of a message title, body text, image, and link. The discount information may include information relating to discounts on products during time periods corresponding to the unit time for which the target product was determined. For example, the discount information may include information indicating that a specific product will be discounted during a specific time period (for example, between 3 PM and 4 PM). In this embodiment, the generation unit 15 generates discount information relating to discounts on products consumed outside the store.

[0067] The generation unit 15 generates discount information, for example, using a discount information generation model. The discount information generation model is trained with the target product as the explanatory variable and the discount information as the dependent variable. The generation unit 15 inputs the target product into the discount information generation model and generates discount information.

[0068] The quality evaluation unit 16 evaluates the quality of the discount information generated by the generation unit 15 based on the quality information stored by the storage unit 11. In this embodiment, the quality evaluation unit 16 outputs information related to the correction of the discount information. The quality evaluation unit 16 outputs the information related to the correction of the discount information to the store system 3.

[0069] In this embodiment, the quality evaluation unit 16 classifies the quality of the evaluated discount information into either a first quality, where no correction is required to the discount information, or a second quality, where correction is required to the discount information. For example, if the quality evaluation unit 16 finds no problems with the discount information, it classifies the quality of the discount information as first quality. If the quality evaluation unit 16 determines that there may be defects in the discount information, or that the discount information is inappropriate and quality cannot be guaranteed, it classifies the quality of the discount information as second quality.

[0070] The quality evaluation unit 16 evaluates the quality of discount information using a quality evaluation model. The quality evaluation model is trained with the generated discount information and quality information as explanatory variables and the classification result of the discount information as the dependent variable. The quality evaluation unit 16 inputs the discount information and quality information into the quality evaluation model and classifies the quality of the discount information into either first quality or second quality.

[0071] If the quality evaluation unit 16 classifies the discount information as first quality, it outputs the discount information. The quality evaluation unit 16 outputs the discount information to the store system 3. If the quality evaluation unit 16 classifies the discount information as second quality, it outputs information related to the correction of the discount information. The quality evaluation unit 16 outputs information related to the correction of the discount information to the store system 3.

[0072] If the quality evaluation unit 16 classifies the quality of the discount information as second quality, it further classifies the quality of the discount information into either a predetermined third quality or a fourth quality which is lower than the third quality. Third quality is, for example, a quality where the content of the discount information is not necessarily inappropriate, but the provision of the discount information may create risks. Fourth quality is, for example, a quality where the content of the discount information is inappropriate. If the quality evaluation unit 16 classifies the quality of the discount information as third quality, it outputs a correction request requesting correction of the discount information as information related to the correction of the discount information. If the quality evaluation unit 16 classifies the quality of the discount information as fourth quality, it corrects the discount information generated by the generation unit 15 based on the quality information and outputs the corrected discount information as information related to the correction of the discount information.

[0073] The quality evaluation unit 16 modifies the discount information using a modified model. The modified model is trained with discount information classified as quality 3 as the explanatory variable and the modified discount information as the dependent variable. The quality evaluation unit 16 inputs discount information classified as quality 4 into the modified model and outputs the modified discount information.

[0074] The provision quantity determination unit 17 determines the expected effect based on the past effect information stored by the storage unit 11 and the target of provision determined by the provision target determination unit 14. The expected effect is the effect expected from providing discount information. The expected effect is, for example, the expected CVR expected from providing discount information. The expected CVR is, for example, the ratio of the expected number of discount information uses to the number of discount information provided.

[0075] The distribution quantity determination unit 17 determines the expected effect using an expected effect determination model. The expected effect determination model is trained with past effect information and the target recipient as explanatory variables and the expected effect as the dependent variable. The distribution quantity determination unit 17 inputs the past effect information and the target recipient into the expected effect determination model and determines the expected effect.

[0076] The provision quantity determination unit 17 determines the number of discount information to be provided based on the determined expected effect and the difference value obtained by subtracting the product order quantity from the maximum supply quantity. The provision quantity is, for example, the number of customer terminals 2 to which the discount information is output. The provision quantity determination unit 17 determines the number of discount effects to be provided by, for example, the value obtained by dividing the difference value by the expected effect.

[0077] The provisioning unit 18 provides discount information based on the discount information generated by the generation unit 15 and the number of units to be provided determined by the number of units to be provided unit 17. The provisioning unit 18 outputs the discount information generated by the generation unit 15 to the customer terminal 2 in the number of units to be provided determined by the number of units to be provided unit 17. The provisioning unit 18 outputs the discount information to the customer terminal 2 that is the target of the provision determined by the target of the provision determination unit 14.

[0078] 5. Operation of Server 1 The operation of server 1 (including the information processing method according to this embodiment) will be explained using Figure 6. Before server 1 starts operation, the storage unit 11 stores operational information, demand information, consumption information (off-site consumption information in this embodiment), classification information, quality information, and past effect information.

[0079] First, the maximum supply estimation unit 12 estimates the maximum supply based on operational information (step S1). For example, in step S1, the maximum supply estimation unit 12 estimates the waiting time for each unit of time. The maximum supply estimation unit 12 compares the estimated waiting time with the maximum allowable waiting time. Based on the comparison result, the maximum supply estimation unit 12 calculates the violation rate and compares the calculated violation rate with the allowable value of the violation rate. In this way, the maximum supply estimation unit 12 estimates the maximum supply for each unit of time. Note that in step S1, the maximum supply estimation unit 12 may also use LLM (Large Language Models) to detect outliers.

[0080] Next, the order volume estimation unit 13 estimates the quantity of product orders based on the demand information (step S2). In step S2, the order volume estimation unit 13 estimates the quantity of product orders as the sum of the quantity of orders placed outside the store and the quantity of orders placed inside the store.

[0081] Next, the supply target determination unit 14 determines whether the difference value obtained by subtracting the product order quantity estimated in step S2 from the maximum supply quantity estimated in step S1 is a positive value (step S3). If the difference value is a positive value, it means that the maximum supply quantity exceeds the product order quantity, and there is room to increase the product order quantity. If the difference value is a negative value, it means that the maximum supply quantity is less than the product order quantity, and it is necessary to either decrease the product order quantity or increase the maximum supply quantity.

[0082] If it is determined that the difference value is not positive (Step S3: NO), Server 1 executes a corrective action (Step S4). The corrective action is a process to reduce the quantity of goods ordered or to increase the maximum supply. For example, the corrective action may include reducing the quantity of goods ordered by stopping the ordering of goods consumed outside the store. In this case, Server 1 executes a process to stop ordering on the website used for using the store. Alternatively, the corrective action may include increasing the maximum supply by increasing the number of store employees. In this case, Server 1 executes a process to post a job advertisement for the store on a website used for recruitment. The job advertisement may include, for example, employment conditions for the time period in which the difference value was determined to be not positive. The content of the corrective action described above is merely an example and is not particularly limited. If Server 1 determines that the difference value is not positive, it may not execute a corrective action and may execute Step S1 again.

[0083] If the difference value is determined to be positive (Step S3: YES), the product selection unit 14 estimates the sensitivity based on the consumption information stored in the storage unit 11 (Step S5). The product selection unit 14 inputs the consumption information stored in the storage unit 11 into the sensitivity estimation model and estimates the sensitivity. In Step S5, the product selection unit 14 may, after estimating the sensitivity for each unit of time, check whether the discount amount in each short time period conforms to the content of the discount limit.

[0084] Next, the target selection unit 14 determines the target to be provided based on the sensitivity estimated in step S5 and the classification information stored in the memory unit 11 (step S6, target selection step). In step S6, the target selection unit 14 may also use LLM (Large Language Models) to detect outliers.

[0085] Classification information is, for example, information that classifies customers according to their behavioral tendencies. In step S6, for example, based on sensitivity, it is possible to estimate the characteristics of customers who are likely to use the store when discount information is provided. Then, customers in the classification that match the estimated customer characteristics can be determined as the target customers. In this way, by determining the target customers based on sensitivity and classification information, it is possible to determine the target customers who will benefit most from being provided with discount information.

[0086] Next, the generation unit 15 generates discount information based on the items to be offered determined in step S6 (step S7, generation step). In step S7, the generation unit 15 inputs the determined items to be offered, store location information stored by the storage unit 11, environmental information, and external information into the discount information generation model and generates discount information.

[0087] Next, the quality evaluation unit 16 performs a quality evaluation process (step S8). The quality evaluation process evaluates the quality of the discount information generated in step S7. Details of the quality evaluation process will be described later.

[0088] Next, the provision quantity determination unit 17 determines the number of discount information to be provided, which was generated in step S7 (step S9). In step S9, the provision quantity determination unit 17 inputs the past effect information stored by the storage unit 11 and the target of provision determined in step S6 into the expected effect determination model and determines the expected effect. Next, the provision quantity determination unit 17 determines the number of discount effects to be provided by dividing the difference value calculated in step S3 by the determined expected effect.

[0089] Next, the supply unit 18 provides discount information based on the discount information generated in step S7 and the number of units to be provided determined in step S9 (step S10). In step S10, the supply unit 18 outputs the generated discount information to the customer terminal 2 in the determined number of units to be provided. The supply unit 18 outputs the discount information to the customer terminal 2 to be provided, as determined in step S6. After the above processing, the server 1 terminates its operation.

[0090] Next, the quality evaluation process (step S8) will be explained in detail using Figure 7. In the quality evaluation process, the quality evaluation unit 16 evaluates the quality of the discount information generated in step S7 based on the quality information stored by the storage unit 11.

[0091] First, the quality evaluation unit 16 classifies the quality of the discount information generated in step S7 (step S81). In step S81, the quality evaluation unit 16 inputs the generated discount information and the quality information stored by the storage unit 11 into the quality evaluation model and classifies the quality of the discount information into either first quality or second quality.

[0092] Next, the quality evaluation unit 16 determines whether the quality of the discount information was classified as first quality in step S81 (step S82). If it determines that the quality of the discount information was classified as first quality (step S82: YES), the quality evaluation unit 16 outputs the discount information generated in step S7 to the store system 3 (step S83). After executing step S83, the quality evaluation unit 16 terminates the quality evaluation process. If it determines that the quality of the discount information was not classified as first quality (step S82: NO), the quality evaluation unit 16 determines whether the quality of the discount information was classified as third quality in step S81 (step S84). Note that not classifying the discount information as first quality in step S82 means that the quality of the discount information was classified as second quality.

[0093] If the quality of the discount information is determined to be classified as third quality (step S84: YES), the quality evaluation unit 16 outputs a correction request to the store system 3 (step S85). The correction request is information requesting correction of the discount information. After executing step S85, the quality evaluation unit 16 terminates the quality evaluation process. If the quality of the discount information is not determined to be classified as third quality (step S84: NO), the quality evaluation unit 16 corrects the discount information generated in step S7 (step S86). Note that not classifying the quality of the discount information as third quality in step S84 means that the quality of the discount information has been classified as fourth quality.

[0094] In step S86, the quality evaluation unit 16 modifies the discount information generated in step S7 based on the quality information stored by the storage unit 11. The quality evaluation unit 16 inputs the discount information generated in step S7 into the modification model and generates the modified discount information.

[0095] For example, in step S86, the quality evaluation unit 16 may perform modifications to remove expressions that violate laws and regulations, intellectual property of others, and excluded keywords from the discount information. The quality evaluation unit 16 may perform modifications to insert disclaimers (e.g., strings such as "Image is for illustrative purposes only") into the discount information. The quality evaluation unit 16 may perform modifications to change adjectives and definitive expressions contained in the discount information to other expressions (e.g., modifications to change expressions such as "always," "dirty," "bad," and "absolutely"). The quality evaluation unit 16 may perform modifications to correct inaccurate information contained in the discount information based on the order information contained in the quality information. For example, if the discount information states that beef from a specific origin is used in the ingredients ordered by the store, even though it is not included in the ingredients ordered by the store, the quality evaluation unit 16 will correct the statement. If the quality evaluation unit 16 determines, based on the sentiment parameters contained in the quality information, that the target customer has a negative sentiment towards the discount information, it may perform modifications to remove the factors that cause that negative sentiment from the discount information.

[0096] Next, the quality evaluation unit 16 outputs the corrected discount information generated in step S86 (step S87). In step S87, the quality evaluation unit 16 outputs the corrected discount information to the store system 3. After these processes, the server 1 terminates the quality evaluation process.

[0097] In this embodiment, the quality evaluation unit 16 classifies the quality of the discount information and issues a warning or corrects the discount information if the quality is low. This reduces the possibility that discount information containing risks may be provided to customers.

[0098] The above describes an example of the operation of Server 1, but the content and order of each step can be modified as appropriate without departing from the spirit of the invention.

[0099] 6. Effects In this embodiment, the target audience for discount information is determined based on consumption information and classification information. Specifically, for example, by providing discount information to target audiences that can effectively increase product order volume, based on information that influences product order volume and classification according to customer attributes, it is possible to determine which target audiences can be effectively targeted. Furthermore, by providing discount information to the determined target audience, it is possible to effectively attract customers. Consequently, the store's profits can be improved.

[0100] 7. Variations In the above embodiment, Server 1 adjusted the volume of orders placed outside the store by providing discount information. The consumption information was information on consumption outside the store, and the sensitivity was the degree of customer response to discounts on products consumed outside the store. The generation unit 15 generated information related to discounts on products consumed outside the store as discount information. This makes it possible to increase the volume of orders placed outside the store by providing discount information to customers, for example, when the predicted volume of orders placed inside the store is insufficient compared to the store's manpower.

[0101] In response to this, Server 1 may adjust the in-store order volume by providing discount information. In this case, the consumption information may be in-store consumption information relating to the consumption of goods within the store. The in-store consumption information includes information indicating factors that influence the in-store order volume of goods. Sensitivity may be the degree to which customers respond to discounts on goods consumed within the store. The generation unit 15 may generate information relating to discounts on goods consumed within the store as discount information. This allows, for example, the in-store order volume to be increased by providing discount information to customers when the predicted in-store order volume is insufficient relative to the store's manpower.

[0102] Furthermore, Server 1 may adjust both the volume of orders placed outside the store and the volume of orders placed inside the store. In this case, for example, the consumption information includes both the volume of orders placed outside the store and the volume of orders placed inside the store. The target determination unit 14 estimates at least one of the sensitivity based on the volume of orders placed outside the store and the sensitivity based on the volume of orders placed inside the store. The generation unit 15 generates at least one of the discount information, which includes information relating to discounts on products consumed outside the store and information relating to discounts on products consumed inside the store. According to this modified example, since both the volume of orders placed outside the store and the volume of orders placed inside the store can be adjusted, the store's profits can be improved more effectively.

[0103] Furthermore, it is considered easier to adjust the volume of orders placed outside the store by providing discount information compared to the volume of orders placed inside the store. Generally, when customers consume goods outside the store, they are more likely to do so alone (e.g., by themselves) than when they consume goods inside the store. For example, when customers use takeout or delivery services, they often use the service alone, whereas when customers use dine-in services, they often use the service with multiple people. In other words, providing discount information is more likely to attract customers to consume goods outside the store than to consume goods inside the store. Therefore, by including consumption information about consumption outside the store and having Server 1 adjust the volume of orders placed outside the store, it is easier to improve the store's profits.

[0104] In the above embodiment, the target product determination unit 14 estimates sensitivity based on consumption information and determines the target product based on the estimated sensitivity and classification information. However, the target product determination unit 14 does not need to estimate sensitivity. In this case, for example, the target product determination unit 14 may input consumption information and classification information into a target product determination model and determine the target product. The target product determination model is trained with consumption information and classification information as explanatory variables and the target product as the dependent variable.

[0105] In the above embodiment, the order quantity estimation unit 13 estimates the quantity of orders placed outside the store and the quantity of orders placed inside the store based on demand information, and estimates the sum of the quantity of orders placed outside the store and the quantity of orders placed inside the store as the quantity of orders placed. However, the order quantity estimation unit 13 does not have to estimate the quantity of orders placed outside the store and the quantity of orders placed inside the store. In this case, the order quantity estimation unit 13 may estimate the quantity of orders placed using a product order quantity estimation model. For example, the product order quantity estimation model is trained with demand information as the explanatory variable and the quantity of orders placed as the dependent variable. The order quantity estimation unit 13 may input the demand information into the product order quantity estimation model and estimate the quantity of orders placed.

[0106] In the above embodiment, the provision quantity determination unit 17 determined the expected effect based on past effect information and the target of provision, and determined the number of discount information to be provided based on the expected effect and the difference value. However, the method for determining the number of discount information to be provided is not limited to the method described above. Also, the provision quantity determination unit 17 does not have to determine the number of discount information to be provided. In this case, the provision unit 18 provides the discount information generated by the generation unit 15, or the discount information corrected by the quality evaluation unit 16, to a predetermined customer terminal 2. The predetermined customer terminal 2 is, for example, all customer terminals 2 that have registered as members via the website.

[0107] In the above embodiment, the quality evaluation unit 16 evaluates the quality of the discount information based on the quality information and outputs the evaluation result of the discount information quality. However, the quality evaluation unit 16 does not have to evaluate the quality of the discount information. In this case, the provision unit 18 provides the discount information generated by the generation unit 15 to the customer terminal 2.

[0108] In the above embodiment, the quality evaluation unit 16 classified the quality of the discount information into one of three quality levels: first quality, second quality, or third quality, and output information according to the classification result. However, the processing of the quality evaluation unit 16 is not limited to the processing described above. The quality evaluation unit 16 may detect risks included in the discount information generated by the generation unit 15 and either remove the risks from the discount information or notify the store staff of the risks.

[0109] In the above embodiments, various models (such as a sensitivity estimation model, a target product determination model, and a discount information generation model) were described. However, these models can be configured as a single model, and their functions may be implemented integrally using, for example, an LLM and an API (Application Programming Interface).

[0110] In the above embodiment, examples of information processing using various models were described. However, this information processing may also be performed using a rule-based method. Known rules may be used as the rules. [Explanation of Symbols]

[0111] 1: Server (information processing device) 2: Customer terminal 3: Store System 10: System 11: Storage section 12: Maximum supply amount estimation part 13: Order Volume Estimation Department 14: Target of Service Selection Department 15: Generation part 16: Quality Evaluation Department 17: Provided number determination section 18:Providing Department 101: CPU 102 :ROM 103: RAM 104: HDD 105: Communications Department N: Network P: Program (information processing program)

Claims

1. An information processing device that generates discount information related to discounts on store products, It comprises a maximum supply quantity estimation unit, an order quantity estimation unit, a supply target determination unit, and a generation unit. The maximum supply quantity estimation unit estimates the maximum supply quantity, which is the maximum amount of the product that the store can supply, based on operational information relating to the operation of the store. The order quantity estimation unit estimates the quantity of the product ordered by the store, based on demand information relating to the demand for the product, The aforementioned operational information includes at least one of the following: employee information, including the number or number of employees at the store; and service level information, including the maximum allowable waiting time for the provision of the aforementioned products. The maximum supply amount estimation unit estimates the maximum supply amount for each predetermined unit time based on the operational information, The aforementioned demand information includes order information for the aforementioned products, The order quantity estimation unit estimates the order quantity for each unit of time based on the order information, The aforementioned target determination unit, when the difference obtained by subtracting the product order quantity from the maximum supply quantity is a positive value, estimates the sensitivity from the consumption information based on the classification information, which is information relating to the classification of customers and is classified according to the behavioral tendencies of the customers, and consumption information relating to the consumption of the product, which includes the customer's past purchase history or coupon usage history and is used to estimate the sensitivity, which is the degree of the customer's response to the discount on the product, and determines the target recipients to whom the discount information will be provided based on the estimated sensitivity and the classification information. The generation unit is a device that generates information relating to discounts on the product during the time period associated with the unit time, based on the target product and the unit time for which the difference value is determined to be a positive value, as discount information.

2. The apparatus according to claim 1, The device comprises consumption information relating to off-site consumption of the product, which takes place outside the store.

3. The apparatus according to claim 1, The aforementioned target determination unit estimates the sensitivity using a sensitivity estimation model, The aforementioned sensitivity estimation model is trained with the consumption information as the explanatory variable and the sensitivity as the dependent variable. The aforementioned target determination unit determines the target of provision using the target of provision determination model, The aforementioned target selection model is a device that has been trained with the sensitivity and classification information as explanatory variables and the target of the provision as the dependent variable.

4. The apparatus according to claim 1, The order volume estimation unit estimates, based on the demand information, the off-store order volume, which is the order volume of the product to be consumed outside the store, and the in-store order volume, which is the order volume of the product to be consumed inside the store. The order quantity estimation unit is a device that estimates the sum of the order quantity outside the store and the order quantity inside the store as the order quantity for the product.

5. The apparatus according to claim 1, The system further includes a unit for determining the number of discount information to be provided, The aforementioned unit for determining the number of items to be provided, Based on past effect information related to discounts, including past discount amounts and conversion rates (CVR) at the time of discounts for the aforementioned products, and the target recipients, the expected effect, which is the expected CVR expected from the provision of the discount information, is determined. A device that determines the number of discount information to be provided based on the expected effect and the difference value.

6. The apparatus according to claim 1, Furthermore, a Quality Evaluation Department will be added. The aforementioned quality evaluation unit, A device that evaluates the quality of the generated discount information based on quality information relating to the quality of the discount information, which includes laws and regulations, intellectual property information of others, or exclusion keywords, and classifies the evaluated quality of the discount information into either a first quality, which does not require modification to the discount information, or a second quality, which requires modification to the discount information.

7. The apparatus according to claim 6, The device wherein, when the quality evaluation unit classifies the quality of the discount information into the second quality, it outputs a correction request requesting correction of the discount information as information related to the correction of the discount information, or corrects the generated discount information based on the quality information and outputs the corrected discount information.

8. The apparatus according to claim 7, The aforementioned quality evaluation unit, If the quality of the discount information is classified into the second quality, the quality of the discount information is further classified into either a predetermined third quality or a fourth quality which is lower than the third quality. When the quality of the discount information is classified into the third quality, a correction request requesting correction of the discount information is output as information related to the correction of the discount information. A device that, when the quality of the discount information is classified into the fourth quality category, modifies the generated discount information based on the quality information and outputs the modified discount information as information relating to the modification of the discount information.

9. An information processing method for generating discount information related to discounts on store products, The computer performs the steps of: estimating the maximum supply quantity, estimating the order quantity, determining the target product, and generating the product. In the maximum supply estimation step, the maximum supply is estimated, which is the maximum amount of the product that the store can supply per predetermined unit time, based on operational information relating to the operation of the store, which includes at least one of the following: employee information including the number of employees or personnel at the store, and service level information including the maximum allowable waiting time for the provision of the product. In the order volume estimation step, the order volume, which is the quantity of the product ordered by the store per unit time, is estimated based on demand information relating to the demand for the product, including order information for the product. In the step of determining the target recipients, if the difference obtained by subtracting the order quantity of the product from the maximum supply quantity is a positive value, the sensitivity is estimated from the consumption information based on classification information relating to customer classification, which is classified according to the customer's behavioral tendencies, and consumption information relating to the consumption of the product, which includes the customer's past purchase history or coupon usage history, and is used to estimate the sensitivity, which is the degree of the customer's response to the discount on the product. Based on the estimated sensitivity and the classification information, the target recipients to whom the discount information will be provided are determined. The generation step involves generating information relating to discounts on the product during the time period associated with the unit time, based on the target product and the unit time for which the difference value is determined to be a positive value, as discount information.

10. An information processing program for causing a computer to function as an information processing device as described in claim 1.

Images