Information processing system, information processing method and program

The system optimizes discount rates in retail stores by predicting demand and using dynamic programming to balance sales and waste, ensuring profit maximization while minimizing waste and operational costs.

JP2024165469AActive Publication Date: 2024-11-28YORK-BENIMARU CO LTD
View PDF 13 Cites 0 Cited by

Patent Information

Application Number
JP2023081697
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-17
Publication Date
2024-11-28
Estimated Expiration
2043-05-17

AI Technical Summary

Technical Problem

In retail stores, determining discount rates based on personal experience can lead to excessive discounts and waste, undermining profit maximization.

Method used

An information processing system that utilizes a demand forecasting model to predict product demand at each discount rate, calculating an optimal discount rate using dynamic programming to maximize expected profit from the current time to closing time, incorporating penalty parameters to avoid waste and operational costs.

Benefits of technology

The system effectively calculates optimal discount rates to maximize store profits by balancing sales and waste reduction, considering operational costs and predetermined discount intervals.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024165469000001_ABST
    Figure 2024165469000001_ABST
Patent Text Reader

Abstract

To provide: an information processing system capable of calculating an optimum discount rate to maximize profits of a store; an information processing method; and a program.SOLUTION: An information processing system (1) includes: a reception part (372) for receiving a commodity of a discount object; a demand prediction model (132) for predicting a demand amount of the commodity corresponding to each discount value at each time; a calculation part (142) being a calculation part for calculating an optimum discount rate at the present time to maximize an expected profit to be predicted in accordance with the present time, a next discount time, a store closing time, and a stock amount on the basis of a demand amount predicted by the demand prediction model, and calculating the discount rate by using a penalty parameter to avoid commodity disposal; and an output part (371) for outputting the calculated discount rate.SELECTED DRAWING: Figure 7
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to an information processing system, an information processing method, and a program for processing information on products in a retail store. [Background technology]

[0002] In retail stores that handle food products, in order to reduce waste caused by products that are close to their expiration date (or expiry date), the sales price of the products is discounted. In order to maximize the store's profits, it is desirable to reduce the discount rate as much as possible while still reducing waste.

[0003] For example, Patent Document 1 discloses a handheld terminal equipped with an acquisition means for acquiring product information including the product price, a setting means for setting information regarding discounts, a calculation means for calculating the discounted price of the product acquired by the acquisition means based on the discount information set by the setting means, and a display means for displaying the product price set by the setting means and the discounted price calculated by the calculation means on the same screen of a display device.

[0004] Furthermore, Patent Document 2 discloses an automated discount decision system for determining the timing of discounting the price of a product, which makes it possible to select discount decision factors including at least time, which are necessary for determining the timing of discounting the price of a product, and sets a numerical value for the selected discount decision factor, which can be stored as set element data, and determines the timing of discounting by comparing this set set element data with actual element data, which is the actual numerical value for the discount decision factor. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2013-88908 A [Patent Document 2] JP 2007-199933 A Summary of the Invention [Problem to be solved by the invention]

[0006] In discount operations, each person in charge may decide the discount rate based on their own experience. In this case, since the discount rate depends on the experience of each person in charge, there is a risk that some people may give excessive discounts or cause waste.

[0007] Therefore, an object of the present invention is to provide an information processing system, an information processing method, and a program that are capable of calculating an optimal discount rate to maximize a store's profits. [Means for solving the problem]

[0008] An information processing system according to one embodiment of the present invention comprises a reception unit that receives products that are subject to discount, a demand forecasting model that predicts the demand for the products according to each discount rate at each time, and a calculation unit that calculates an optimal discount rate at the current time that maximizes expected profits from the current time to closing time based on the demand predicted by the demand forecasting model, the discount rate being predicted according to the current time, the next discount time, closing time and inventory quantity, the calculation unit calculating the discount rate using a penalty parameter to avoid waste of products, and an output unit that outputs the calculated discount rate.

[0009] According to this aspect, the information processing system can calculate an optimal discount rate by adjusting the tolerance for disposal using the penalty parameter.

[0010] In the above information processing system, the calculation unit According to the following formula (1), (t, c, r prev ,t') the discount rate r that maximizes Q * Select

number

number

[0011] In the above information processing system, the calculation unit may calculate the discount rate using a parameter that introduces the operation cost of a person in charge of the discount operation. According to this aspect, the information processing system can calculate a discount rate that reflects the operation cost of a person in charge of the discount operation and is more suitable for actual business.

[0012] In the above information processing system, the calculation unit According to the following formula (1), (t, c, r prev ,t') the discount rate r that maximizes Q * Select

number

number

[0013] In the information processing system, the next discount time may be a predetermined time after the current time. According to this aspect, the information processing system can calculate an optimal discount rate for a product for which a discount operation is performed at each predetermined time.

[0014] In the above information processing system, the calculation unit According to the following formula (8), (t, c, r prev ,t') the discount rate r that maximizes Q* Select

number

number

[0015] In the above information processing system, the penalty parameters may be determined by performing a random number simulation using sales information including the sales prices and discount prices or discount rates of multiple types of products. According to this aspect, the information processing system can select appropriate parameter values ​​according to the purpose.

[0016] A method according to another aspect of the present invention includes an information processing system receiving merchandise to be discounted, predicting the demand for the merchandise at each discount rate at each time using a demand prediction model, and calculating an optimal discount rate at the current time based on the demand predicted by the demand prediction model, which maximizes the expected profit from the current time to closing time, as predicted according to the current time, the next discount time, closing time, and inventory, wherein the discount rate is calculated using a penalty parameter to avoid waste of the merchandise, and outputting the calculated discount rate.

[0017] A program according to another aspect of the present invention causes one or more computers to accept merchandise to be discounted, predict the demand for the merchandise at each discount rate using a demand forecasting model, calculate an optimal discount rate for the current time based on the demand predicted by the demand forecasting model, which maximizes the expected profit from the current time to closing time, as predicted based on the current time, the next discount time, closing time, and inventory, wherein the discount rate is calculated using a penalty parameter to avoid waste of the merchandise, and output the calculated discount rate.

[0018] A program according to another aspect of the present invention causes a computer connected to an information processing device via a communications network to recognize identification information given to products that are subject to discounts and identify the products; to have the information processing device predict the demand for the products at each discount rate at each time; and based on the predicted demand, to calculate an optimal discount rate at the current time using a penalty parameter to avoid waste of the products, which maximizes the expected profit from the current time to closing time, as predicted based on the current time, the next discount time, closing time, and inventory quantity; and to present the calculated discount rate and / or a discounted price based on the calculated discount rate.

[0019] The program may cause the information processing device to calculate the discount rate by further using a parameter that introduces an operation cost of a person in charge of performing the discount operation.

[0020] In the above program, the next discount time may be a predetermined time after the current time. Effect of the Invention

[0021] According to the present invention, it is possible to provide an information processing system, an information processing method, and a program capable of calculating an optimal discount rate to maximize a store's profits. [Brief description of the drawings]

[0022] [Figure 1] 1 is a diagram showing a schematic configuration of an information processing system according to an embodiment of the present invention. [Diagram 2] 1 is a block diagram showing a schematic configuration of an information processing device according to an embodiment of the present invention; [Diagram 3] FIG. 13 is a conceptual diagram of calculating an optimal discount rate using dynamic programming. [Figure 4] FIG. 2 is a schematic diagram illustrating information stored in a database. [Diagram 5] FIG. 2 is a block diagram showing a schematic configuration of a user terminal according to an embodiment of the present invention. [Figure 6] 10 is a flowchart showing the operation of a user terminal in a product discount process. [Figure 7] 11 is a schematic diagram illustrating a screen of a user terminal in a product discount process; FIG. [Figure 8] 11 is a schematic diagram illustrating a screen of a user terminal in a product discount process; FIG. [Figure 9] 13 is a schematic diagram illustrating a setting screen for discount time and closing time. FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0023] With reference to the accompanying drawings, the embodiments of the present invention will be described. Note that the following embodiments are intended to facilitate understanding of the present invention and are not intended to limit the present invention. In addition, the present invention can be modified in various ways without departing from the gist of the invention. Furthermore, a person skilled in the art can adopt an embodiment in which each element described below is replaced with an equivalent, and such an embodiment is also included in the scope of the present invention.

[0024] <Configuration of the embodiment> FIG. 1 is a diagram showing a schematic configuration of an information processing system according to an embodiment of the present invention. As shown in FIG. 1, a sales price setting support system (information processing system) 1 according to this embodiment includes a sales price setting support device (information processing device) 10 having a database 20, and a user terminal 30 connected to the sales price setting support device 10 via a communication network N. A printer (sticker printer) 40 that outputs discount stickers for products is connected to the user terminal 30 by wire or wirelessly. Alternatively, the printer 40 may be built into the user terminal 30. Although FIG. 1 shows only one user terminal 30 and one printer 40, a plurality of user terminals 30 and a plurality of printers 40 may be provided.

[0025] The communication network N includes a communication network capable of transmitting and receiving information between the sales price setting support device 10 and the user terminal 30. The communication network N may be, for example, the Internet, a LAN, a dedicated line, a telephone line, an in-house network, a mobile communication network, Bluetooth, WiFi (Wireless Fidelity), other communication lines, or a combination thereof, and may be either wired or wireless.

[0026] FIG. 2 is a block diagram showing a schematic configuration of a sales price setting support device 10 according to an embodiment of the present invention. In FIG. 2, a single sales price setting support device 10 is assumed, and an example of a necessary configuration is shown, but the sales price setting support device 10 can also be configured as a part of a multi-functional distributed system consisting of multiple computer systems. For example, some of the functions of the sales price setting support device 10 may be included in the database 20 or another information processing device capable of communicating with the database 20. The sales price setting support device 10 is, for example, configured by a computer with high arithmetic processing power, and by executing a predetermined program, realizes a server function such as transmitting and receiving information to and from a user terminal 30, calculating discount rates for various products, and transmitting the calculated discount rates to the user terminal 30. As shown in FIG. 2, the sales price setting support device 10 includes a communication device 11, a memory 12, a storage medium 13, and a calculation device 14.

[0027] The communication device 11 is a hardware module for connecting the selling price setting support device 10 to the communication network N and communicating with other terminals on the communication network N. The communication device 11 is configured using a modulation / demodulation device such as an ISDN modem, an ADSL modem, a cable modem, an optical modem, or a soft modem.

[0028] The memory 12 is, for example, a main memory configured from a semiconductor memory such as a RAM. The storage medium 13 is a computer-readable storage device such as a hard disk drive (HDD) and / or a solid state drive (SSD). The storage medium 13 stores various programs 131 including an operating system program and a driver program, and various data used during the execution of these programs. Specifically, in addition to the programs 131, the storage medium 13 stores a demand forecast model 132 that forecasts the demand amount of each product according to each discount rate at each time until the sales deadline for each product.

[0029] The arithmetic device 14 is made up of processors such as a CPU (Central Processing Unit) and a GPU (Graphical processing unit), an arithmetic logic unit that processes arithmetic operations, logical operations, bit operations, etc., and various registers, and centrally controls each part of the selling price setting support device 10 by executing various programs stored in the storage medium 13. The various registers are, for example, a program counter, a data register, an instruction register, a general-purpose register, etc. Furthermore, the arithmetic device 14 executes the program 131 to cooperate with the user terminal 30 and realize a selling price setting support function such as discount processing. The functional parts realized by the arithmetic device 14 include a demand forecast model construction part 141 and a discount rate calculation part 142.

[0030] The demand forecast model construction unit 141 constructs a demand forecast model 132 that predicts the demand volume of each product according to each discount rate at each time until the sales deadline for each product, using sales information including the original sales prices and discount prices or discount rates of multiple types of products stored in a database 20 described below. Predicting the demand volume includes predicting the demand quantity such as the number of sales or the sales amount. The algorithm used in the demand forecast model 132 is not particularly limited, and any demand forecast algorithm can be used.

[0031] The discount rate calculation unit 142 calculates a discount rate that maximizes profits at each time based on the demand amount, the stock amount, and the remaining sales period. In detail, the discount rate calculation unit 142 calculates an optimal discount rate predicted according to the current time, the next discount time, the closing time, and the stock amount by using dynamic programming (dynamic pricing) based on the demand amount predicted by the demand prediction model 132.

[0032] Figure 3 is an image of how dynamic programming is used to calculate the discount rate that maximizes the final value (profit) Q by sequentially searching for discount rates and inventory levels in the opposite direction to the progression of time, going back from time T.

[0033] Fig. 4 is a schematic diagram illustrating information stored in the database 20. As shown in Fig. 4, the database 20 stores a product information database (DB) 21 and a sales information database (DB) 22. Although the database 20 is connected to the sales price setting support device 10 in Fig. 1, the two may be connected via a communication network N.

[0034] Product information is stored in the product information database 21. Specifically, information such as product identification information, product name, sales price, and currently set discount rate is stored. Note that, in order to initialize the discount rate set on the previous day, the currently set discount rate is set to zero at a predetermined time after closing time, and is desirably zero at opening time.

[0035] The sales information database 22 stores information on past sales performance for each of a plurality of types of products. Specifically, information such as the original sales price, the actual sales price (discounted price) or discount rate, the number of items sold, the sales amount, the discount loss amount, and the waste loss amount is stored. Such sales information for each product may be stored for each store ID, which is the identification information of the store, or for each user ID, which is the identification information of the user who performs the discount operation.

[0036] The user terminal 30 is a device used by a user who performs a discount operation, and can be configured as a dedicated or general-purpose portable terminal device (computer) such as a smartphone, a tablet terminal, or a PDA. FIG. 5 is a block diagram showing a schematic configuration of the user terminal 30. As shown in FIG. 5, the user terminal 30 includes a communication device 31, a display device 32, an input device 33, a memory 34, a storage medium 35, a reading unit 36, and a calculation device 37.

[0037] The communication device 31 is a hardware module for connecting the user terminal 30 to the communication network N and communicating with other terminals on the communication network N (e.g., the sales price setting support device 10 and the printer 40), and is configured using a modulation / demodulation device such as an ISDN modem, ADSL modem, cable modem, optical modem, or soft modem.

[0038] The display device 32 is configured using, for example, a liquid crystal display or an organic EL display, and presents various information to the user.

[0039] The input device 33 is an input device such as a keyboard, various operation buttons, a touch panel provided on the display device 32, etc., which accepts operations performed by a user and inputs a signal corresponding to the operation to the arithmetic device 37.

[0040] The memory 34 is a main memory configured by a semiconductor memory such as a RAM. The storage medium 35 is a storage configured by a semiconductor memory such as a ROM. The storage medium 35 stores various programs including an operating system program and a driver program. Specifically, the storage medium 35 stores a program 351 that causes the user terminal 30 to execute a series of discount processes, which recognizes a discounted product by a reading unit 36 ​​described later, obtains a discount rate for the product from the sales price setting support device 10, and presents the discount rate and / or discount price of the target product.

[0041] The reading unit 36 ​​is, for example, a barcode scanner or an optical camera, and reads code information (product identification code) representing product identification information attached to a product or a price tag, etc., and inputs a read signal or an image signal representing the product identification code to the calculation device 37. The product identification code may be, for example, a barcode or a two-dimensional code, or may be other codes (for example, an arrangement of numbers or symbols). If the product identification code is a barcode, a barcode scanner or an optical camera can be used as the reading unit 36. Also, if the product identification code is a two-dimensional code or other codes, an optical camera can be used as the reading unit 36. The reading unit 36 ​​may be built into the user terminal 30. Alternatively, the reading unit 36 ​​may be configured by connecting an external device to the user terminal 30.

[0042] The arithmetic device 37 is made up of a processor such as a CPU or GPU, an arithmetic logic unit that processes arithmetic operations, logical operations, bit operations, etc., and various registers, and centrally controls each part of the user terminal 30 by executing various programs stored in the storage medium 13. The various registers are, for example, a program counter, a data register, an instruction register, a general-purpose register, etc. Furthermore, the arithmetic device 37 executes the program 351 to cooperate with the selling price setting support device 10 and realize a selling price setting support function such as discount processing. The functional parts realized by the arithmetic device 37 include a discount price presentation part 371 and a product recognition part 372.

[0043] The discount price presenting unit 371 presents the discount rate and / or the discount price of the discounted product based on the discount rate calculated by the discount rate calculation unit 142.

[0044] The commodity recognition unit 372 recognizes the commodity identification code based on the read signal or image signal of the commodity identification code input from the reading unit 36, and identifies the target commodity.

[0045] The printer 40 is a sticker printer that outputs discount stickers in accordance with a control signal received from the user terminal 30 .

[0046] Next, we will explain the detailed methodology of the demand forecasting model and discount optimization algorithm. <Demand forecast model> In this embodiment, demand forecasting is performed for each product. In order to select the optimal discount rate, the transition probability of the inventory quantity when each discount rate is selected for each product and each time is required. Therefore, the demand forecasting model predicts the demand amount for each product, each time, and each discount rate.

[0047] Sales information does not necessarily contain data for all discount rates for each product. In order to estimate the demand curve at each time from limited data, in this embodiment, the demand for all products is estimated using one demand forecasting model. In addition, a constraint is imposed so that the demand increases monotonically with increasing discount rates.

[0048] In the demand forecasting model 132, it is necessary to accurately estimate the demand curve, that is, the change in demand with respect to the change in discount rate when conditioned by product and time. On the other hand, since there is a strong correlation between time and discount rate in the sales information, when building a model for predicting the demand volume using these two variables simultaneously as features, there is a concern that it may not be possible to effectively extract the effect of the change in discount rate on the change in the demand volume.

[0049] Therefore, in this embodiment, Discounting a product affects the amount of demand per visitor, not the number of visitors. The number of customers depends on the time of day Demand per customer does not vary significantly depending on the time of day Based on the above three assumptions, we used separate models to predict the number of customers and the demand per customer, and then constructed a model to predict demand by multiplying each predicted value.

[0050] <Discount Optimization Algorithm> The discount optimization algorithm is expected to output the optimal discount rate for each product when the time, stock quantity, and the next time to consider discounting are input in order to maximize the final profit.

[0051] In this embodiment, the following three business practices are taken into consideration when formulating the discount optimization algorithm. The first point is to avoid the disposal of products as much as possible. Even if it increases the discount loss slightly, it is prioritized to avoid the disposal. The second point is that the number of discount operations per day is limited to about several times. In addition to the large number of products, the discount operation is performed in parallel with other business. Furthermore, since there are operational costs for reading product codes and issuing discount stickers and attaching them to products, the number of discount operations is limited to about several times per day. The third point is that the discount rate should not be made smaller than the previous time. For example, a certain product is sold at a 50% discount because there is a large amount of inventory remaining, and even if the inventory decreases faster than expected, the discount rate is not changed to a smaller rate than the previous time, such as 30%.

[0052] <Formulation of discount optimization algorithm> The final profit is defined as Q. t is the current time, t' is the next discount time, t'' is the next next discount time, T is the closing time, c is the number of items in stock, r is the discount rate at time t, and r prev is defined as the discount rate that was set up to the current time t. Q(t,c,t',r) is the expected profit to be obtained in the period (t,T) from time t to time T when (time, stock quantity)=(t,c), the discount rate is r, and the next discount time is t'. In this case, according to formula (1), (t,c,r prev ,t') the discount rate r that maximizes Q * Select .

[0053]

number

[0054] Furthermore, Q is calculated using dynamic programming as follows. The calculation of Q starts from the closing time T, and the value of Q at each (t, c, t', r) is calculated recursively in descending order of time.

[0055]

number

[0056] At this time, p0 is the original selling price. Also, P t,s,r (c) is defined as the probability that c units will be sold when sold from time t to time s at a discount rate r. This is a function obtained by inputting the demand forecast result into the λ parameter of the Poisson distribution. φ is a hyperparameter that adjusts the magnitude of the penalty imposed when a product remains unsold by closing time in order to satisfy the requirement of avoiding product disposal. In this embodiment, the tolerance for disposal can be adjusted by introducing the disposal penalty parameter φ into the discount optimization algorithm.

[0057] Here, Q0 represents the expected profit in the period (t, t') and Q1 represents the expected profit in the period (t', T). In addition, N(t, c, t', r) is the number of units expected to be sold in the period (t, t') when (time, number in stock) = (t, c), the discount rate is r, and the next discount time is t'. V(t', c', r) is the average value of the expected profit for time t'' when the discount rate up to that point is r, the number in stock is c', and the next discount time is t'' at time t'. In formula (6), by finding the maximum value of Q under the constraint of r' ≧ r, the requirement that the discount rate should not be smaller than the previous time is met. t'' is assumed to be indefinite in the period (t', T), that is, to follow a uniform distribution. In the case of t' = T in formula (5), it represents the magnitude of the penalty when waste occurs, and is a function proportional to the number of wastes.

[0058] <How to determine discard penalty parameters> The waste penalty parameter is a parameter introduced to meet the requirement of avoiding waste of products as much as possible. Here, a method for determining the waste penalty parameter will be described. Intuitively, it is thought that the purchase price of the product should be the penalty for waste, but in reality, it is difficult to determine it easily because various factors such as labor costs for displaying the product are involved. In this embodiment, a random number simulation is performed using sales information to reduce discounts and waste losses and select parameters that can meet the requirements. The simulation is performed multiple times by changing the random number seed for each date, product, and waste penalty parameter. The detailed procedure is shown below.

[0059] 1. Determine (date, product, disposal penalty parameters) 2. Set the initial time t to "opening time" and the inventory quantity c to "initial inventory quantity" 3. When t is the "closing time", end the simulation. 4. Using t and c, determine the discount rate r from the results of the discount optimization algorithm. 5. Using t and r from "4.", sample the demand amount n from the results of the demand forecasting model. 6. c:=c-min(c,n) 7. t:=t+1 8. Return to "3."

[0060] After the simulation is completed, the discount and waste loss for each waste penalty parameter can be calculated by marginalizing the date and product. From the results, appropriate parameter values ​​are selected according to the purpose, such as values ​​that minimize the discount and waste loss. In this embodiment, parameter values ​​that achieve the target value for reducing discount loss and minimize waste loss are selected.

[0061] <Variation 1 of the discount optimization algorithm formulation> As described above, the discount operation requires the operation cost of reading the product code and the operation cost of issuing the discount sticker and attaching it to the product. Therefore, in this embodiment, these operation costs are introduced.

[0062]

number

[0063] Equation (6') is a modification of equation (6), where h(r,c',r') is defined by equation (7) below.

[0064]

number

[0065] <Variation 2 of the discount optimization algorithm formulation> In the discount optimization algorithm described above, three business practices have been taken into consideration when formulating the discount optimization algorithm. In this embodiment, the second point, that discount operations are performed only a few times per day, is changed to that discount operations are performed at a predetermined time interval, for example, every hour, and the discount optimization algorithm is formulated. For example, in the case of a limited number of products, such as prepared foods and fresh fish, it is considered to be a business practice to perform discount operations every hour.

[0066] The final profit is defined as Q. t is the current time, c is the number of items in stock, and r is the discount rate at time t. Q(t,c,r) is the expected profit when (time, number of items in stock)=(t,c) and the discount rate is r. Then, using equation (8), the discount rate r that maximizes Q for (t,c) is * Select .

[0067]

number

[0068] Furthermore, Q is calculated using dynamic programming as follows. The calculation of Q starts from the closing time T, and the value of Q at each (t, c, r) is recursively calculated in descending order of time.

[0069]

number

[0070] At this time, p0 is the original selling price. Also, P t,r (c) is defined as the probability that c units will be sold when selling at a discount rate r from time t until a certain time t0 (1 hour in this example) later. This is a function obtained by inputting the demand forecast results into the λ parameter of the Poisson distribution. φ is a hyperparameter that adjusts the size of the penalty imposed when a product remains unsold by closing time in order to meet the requirement of avoiding product waste.

[0071] Here, Q0 represents the expected profit in the period (t, t+t0) and Q1 represents the expected profit in the period (t+t0, T). Additionally, N(t, c, r) is the number of units expected to be sold in the period (t, t+t0) when (time, number in stock) = (t, c) and the discount rate is r. V(t, c) is the expected profit at closing time when the discount rate r is selected to maximize the expected profit at closing time at time t when the number in stock is c. In equation (13), finding the maximum value of Q within the constraint r' ≧ r satisfies the requirement that the discount rate not be made smaller than it was at the previous time.

[0072] <Operation of the embodiment> The operation of the sales price setting support system 1 will now be described. Fig. 6 is a flow chart showing the operation of the user terminal in the discount process of the product. Fig. 7 and Fig. 8 are schematic diagrams showing examples of the screen of the user terminal in the discount process of the product. In the following, it is assumed that the product identification code is a barcode, and the barcode is read from an image captured by a camera built into the user terminal 30.

[0073] The reading unit 36 ​​of the user terminal 30 reads the product identification code of the target product (step S101). In this embodiment, the product identification code is a barcode, and the barcode is read from an image captured by a camera built into the user terminal 30.

[0074] In detail, when the user terminal 30 detects an operation (e.g., a tap operation) on the capture button m02 provided on the product identification screen M1 as shown in (b) of Fig. 7, the user terminal 30 reads the barcode displayed in the reading area m01 set on the screen. At this time, if the reading area m01 includes multiple barcode images, the user terminal 30 may calculate the center of gravity of each barcode image and read the barcode with the smallest distance between the center of gravity of the barcode image and the center of gravity of the reading area m01.

[0075] The product recognition unit 372 of the user terminal 30 queries the selling price setting support device 10 for information on the target product based on the read product identification code (barcode) (step S102). The selling price setting support device 10 acquires the product name and selling price of the product corresponding to the product identification code from the product information database 21.

[0076] The user terminal 30 displays the product screen of the target product (step S103). The target product display screen M2 shown in FIG. 7(c) has a display field m03 for the product name of the target product and a display field m04 for the original selling price, and the user can confirm that the product identification code of the target product has been successfully read by checking the display fields m03 and m04. The user then checks the stock quantity of the target product, moves the cursor to the stock quantity input field m05, and operates (e.g., taps) the numeric keypad m06 to input the stock quantity of the target product.

[0077] When the user terminal 30 accepts the input of the stock quantity in the stock quantity input field m05 (step S104), it inquires of the selling price setting support device 10 about the discount rate of the target product (step S105). In this embodiment, the user terminal 30 inquires of the selling price setting support device 10 about the discount rate of the target product based on the product identification information, the stock quantity, the current time, the next discount time, and the closing time. In response, the selling price setting support device 10 transmits the discount rate of the target product to the user terminal 30 based on the product identification information, the stock quantity, the current time, the next discount time, the closing time, and the currently set discount rate. The discount price presentation unit 371 of the user terminal 30 displays the discount rate and / or the discount price based on the received discount rate (step S106).

[0078] The target product display screen M3 shown in FIG. 7(d) has a discount price display field m07 and a discount rate display field m08 for the target product. In these display fields, the recommended discount price and discount rate calculated by the sales price setting support device 10 are displayed. When the user terminal 30 detects manual input of the discount price or discount rate in these display fields (step S107: Yes), it displays the manually input discount price or discount rate on the target product display screen M3 (step S108). This allows the user to manually change the presented discount price or discount rate.

[0079] On the other hand, if the user terminal 30 does not detect a manual input of the discount price or discount rate (step S107: No), the process proceeds to step S109.

[0080] As shown in FIG. 8(a), when the user terminal 30 detects an operation on the print button m20 provided on the target product display screen M3 (step S109: Yes), it instructs the printer 40 to print a discount sticker (step S110). As a result, the printer 40 outputs discount stickers on which the discount price displayed in the discount price display field m07 or the discount rate displayed in the discount rate display field m08 is printed, the number of which is displayed in the stock quantity input field m05. In addition, the user terminal 30 transmits the newly set discount rate to the selling price setting support device 10. In response to this, the selling price setting support device 10 updates the currently set discount rate of the target product in the product information database 21. If the user terminal 30 does not detect an operation on the print button m20 (step S109: No), the process returns to step S107.

[0081] After that, the user terminal 30 reads the product identification code of another product (step S101), which allows the user to perform a discount operation on the other product.

[0082] In this embodiment, an example has been described in which the selling price setting support device 10 acquires the inventory quantity from the user terminal 30, but in another embodiment, the selling price setting support device 10 may acquire the inventory quantity from the database 20.

[0083] The following describes additional functions in the sales price setting support system 1 according to this embodiment.

[0084] <Setting discount time and closing time> In this embodiment, the discount time and closing time can be set on the user terminal 30. Fig. 9 is a schematic diagram illustrating a setting screen for discount time and closing time. For example, the discount time setting screen M5 in Fig. 9 is provided with the first to third discount times and the closing time.

[0085] When the user terminal 30 detects an operation on the add button m30, it displays the discount time setting screen M5 with the increased number of discounts. By the user setting the discount time and closing time in advance, the user terminal 30 can be used to inquire the sales price setting support device 10. Note that, although the example in which the user terminal 30 stores the discount time and closing time has been described in this embodiment, in another embodiment, the database 20 may store the discount time and closing time. [Explanation of symbols]

[0086] 1...Sales price setting support system (information processing system), 10...Sales price setting support device (information processing device), 11...31...Communication device, 12...34...Memory, 13...35...Storage medium, 14...37...Calculation device, 20...Database, 21...Product information database (DB), 22...Sales information database (DB), 30...User terminal, 32...Display device, 33...Input device, 36...Reading unit, 40...Printer, 131...351...Program, 132...Demand forecast model, 141...Demand forecast model construction unit, 142...Discount rate calculation unit (calculation unit), 371...Discount price presentation unit (output unit), 372...Product recognition unit (reception unit)

Claims

1. a reception unit for receiving products to be discounted; A demand forecasting model for forecasting the demand amount of the product according to each discount rate at each time; a calculation unit that calculates an optimal discount rate at the current time that maximizes an expected profit from the current time to the closing time, the discount rate being predicted according to the current time, the next discount time, the closing time, and the number of items in stock, based on the demand volume predicted by the demand prediction model, the calculation unit calculating the discount rate using a penalty parameter for avoiding the disposal of products; an output unit for outputting the calculated discount rate; An information processing system comprising:

2. The calculation unit is According to the following formula (1), (t, c, r prev , t'), the discount rate r that maximizes Q * Select [0010] Q is calculated using dynamic programming as follows: [0025] t is the current time, t' is the next discount time, t'' is the next next discount time, T is the closing time, c is the number of items in stock, r is the discount rate at time t, and r prev is the discount rate that has been set up to the current time t, Q(t, c, t', r) is the expected profit to be obtained in the period (t, T) from time t to time T when (time, stock quantity) = (t, c), the discount rate is r, and the next discount time is t', and p 0 is the original selling price, P t,s,r (c) is the probability of selling c units when selling from time t to time s at a discount rate r, φ is a penalty parameter, and Q 0 is the expected profit in period (t, t'), and Q 1 is the expected profit in the period (t', T), N(t, c, t', r) is the number of units expected to be sold in the period (t, t') when (time, number in stock) = (t, c), the discount rate is r, and the next discount time is t', and V(t', c', r) is the average expected profit for time t'' when, at time t', the discount rate up to that point is r, the number in stock is c', and the next discount time is t'', The information processing system according to claim 1 .

3. The information processing system according to claim 1 , wherein the calculation unit calculates the discount rate using a parameter that introduces an operation cost of a person in charge of performing the discount operation.

4. The calculation unit is According to the following formula (1), (t, c, r prev , t'), the discount rate r that maximizes Q * Select [0030] Q is calculated using dynamic programming as follows: [0045] t is the current time, t' is the next discount time, t'' is the next next discount time, T is the closing time, c is the number of items in stock, r is the discount rate at time t, and r prev is the discount rate that has been set up to the current time t, Q(t, c, t', r) is the expected profit to be obtained in the period (t, T) from time t to time T when (time, stock quantity) = (t, c), the discount rate is r, and the next discount time is t', and p 0 is the original selling price, P t,s,r (c) is the probability of selling c units when selling from time t to time s at a discount rate r, φ is a penalty parameter, and Q 0 is the expected profit in period (t, t'), and Q 1 is the expected profit in the period (t', T), N(t, c, t', r) is the number of units expected to be sold in the period (t, t') when (time, stock quantity) = (t, c), the discount rate is r, and the next discount time is t', V(t', c', r) is the average expected profit for time t'' when the discount rate up to that point is r, the stock quantity is c', and the next discount time is t'', and I [ conditions x] is a function that takes the value 1 when the condition x is True and takes the value 0 when the condition x is False, τ is the cost per product for reading the product identification code when there is excess stock at the time of the discount operation, and ρ is the cost per product for issuing discount stickers and attaching them to the products when there is excess stock at the time of the discount operation. The information processing system according to claim 3 .

5. The information processing system according to claim 1 , wherein the next discount time is a predetermined time after the current time.

6. The calculation unit is According to the following formula (8), (t, c, r prev , t'), the discount rate r that maximizes Q * Select [0050] Q is calculated using dynamic programming as follows: [006] t is the current time, T is the closing time, c is the number of items in stock, r is the discount rate at time t, Q(t,c,r) is the expected profit to be obtained in the period (t,T) from time t to time T when (time, number of items in stock)=(t,c) and the discount rate is r, and p 0 is the original selling price, P t,r (c) is calculated by dividing the discount rate r by the time t and the predetermined time t 0 is the probability that c units will be sold when the product is sold later, φ is the penalty parameter, and Q 0 is the period (t, t + t 0 ) is the expected profit at 1 is the period (t+t 0 , T), and N(t, c, r) is the expected profit in the period (t, t + t) when (time, stock quantity) = (t, c) and the discount rate is r. 0 ) is the expected number of sales at time t, and V(t, c) is the expected profit at closing time when the discount rate r is selected to maximize the expected profit at closing time when the inventory quantity is c at time t.

6. The information processing system according to claim 5.

7. The penalty parameters are determined by performing a random number simulation using sales information including sales prices and discount prices or discount rates of a plurality of types of products. The information processing system according to any one of claims 1 to 6.

8. The information processing system Accepting discounted items; predicting the demand for the product at each discount rate at each time using a demand forecasting model; Calculating an optimal discount rate at the current time that maximizes expected profits from the current time to the closing time, which is predicted according to the current time, the next discount time, the closing time, and the number of items in stock, based on the demand volume predicted by the demand forecast model, and calculating the discount rate using a penalty parameter for avoiding product waste; outputting the calculated discount rate; The method includes:

9. On one or more computers, Accepting discounted items; predicting the demand for the product at each discount rate at each time using a demand forecasting model; Calculating an optimal discount rate at the current time that maximizes expected profits from the current time to the closing time, which is predicted according to the current time, the next discount time, the closing time, and the number of items in stock, based on the demand volume predicted by the demand forecast model, and calculating the discount rate using a penalty parameter for avoiding product waste; outputting the calculated discount rate; A program that executes the following.

10. A computer connected to an information processing device via a communication network, Recognizing identification information given to a product to be discounted and identifying the product; having the information processing device predict the demand for the product corresponding to each discount rate at each time, and calculating, based on the predicted demand, an optimal discount rate at the current time that maximizes expected profits from the current time to the closing time, which is predicted according to the current time, the next discount time, the closing time, and the number of items in stock, using a penalty parameter to avoid the disposal of the product; presenting the calculated discount rate and / or a discounted price based on the calculated discount rate; A program that executes the following.

11. 11. The program according to claim 10, further causing the information processing device to calculate the discount rate by using a parameter that incorporates an operation cost of a person in charge of performing the discount operation.

12. 11. The program according to claim 10, wherein the next discount time is a predetermined time after the current time.

Citation Information

Patent Citations

  • Method for inventory management

    JP1995110833A

  • Commodity sale managing method and commodity sale data managing system

    JP2001109960A

  • Discount sale management apparatus, discount sale management method and its program

    JP2005174035A

  • Commodity sales data processor, and commodity sales data processing system

    JP2008139951A

  • Merchandise sales management method and system

    JP2008243236A