Information processing device, information processing method, and program

Abstract

An information processing device (1) for implementing an artificial intelligence (AI) that solves the problem of determining price information of each of a plurality of products with time information so as to greater increase a total profit obtained over the future from another product in accordance with the demand for each of said products is provided with: a first acquisition unit (11) that, upon input of the price information and presentation time of each among first products with time information which are to be presented, acquires a demand function for outputting an estimated value for the demand for each of the first products; a second acquisition unit (12) that, upon input of the price information and presentation time of each of the first products to be presented, acquires an average gross profit function for outputting an estimated value for the average gross profit obtained from the first products and a second product in accordance with the demand for each of the first products; and a searching unit (13) that uses the demand function and the average gross profit function to search for an optimal solution for the price information of each of the first products to be presented so as to greater increase the total profit based on the estimated value for the average gross profit and the estimated value for the demand for each of the first products at each of a plurality of times at or later than a current time.

Description

Information processing device, information processing method, and program

[0001] The present invention relates to an information processing device that determines the price of a product.

[0002] Dynamic pricing, which changes the price of a product, is known. Patent Literature 1 describes a technology for determining a recommended price for a product based on demand forecasts, inventory data, and the like, so as to maximize profits, etc. This technology determines the recommended price by taking into account how price affects demand. For example, this technology determines a recommended price for future hotel stays for all days starting from today through to future dates.

[0003] Japan Special Table Publication No. 2012-524340

[0004] Some products are set with time information such as a deadline, date, period, or time period. For example, a product such as a hotel stay is set with time information such as the date of stay. The profit obtained from a hotel stay is determined by the price (the accommodation fee for each stay date) and the demand (the number of guests staying on each stay date). Therefore, by using the technology described in Patent Literature 1, it is possible to search for a price for each stay date that will maximize the total profit into the future, based on an estimated value of demand corresponding to the price for each of multiple stay dates.

[0005] On the other hand, for example, time information such as a time slot is set for a product such as a delivery slot in online shopping. However, unlike hotel accommodation, the profits obtained from such a delivery slot are not determined only by price (shipping fee for each delivery slot) and demand (the number of people using each delivery slot), but are also affected by the purchase amount of the goods to be delivered when the delivery slot is accepted.

[0006] However, the technology described in Patent Document 1 does not take into consideration the case where profits can be obtained from other products in response to demand for the product with time information set. As a result, there is a problem in that it is difficult to determine the price of each of multiple products with time information set so as to maximize the total future profits that can be obtained from the product and other products in response to demand for each product.

[0007] One aspect of the present invention has been made in consideration of the above-mentioned problems, and aims to provide a technology that can determine the price of each of multiple products with time information set so as to maximize the total future profits that can be obtained from that product and other products as demand for each product increases.

[0008] An information processing device according to one aspect of the present invention comprises: a first acquisition means for acquiring a demand function that, when price information for each of a plurality of first products to be presented, with time information set for each first product, and a presentation time when this price information is presented simultaneously, outputs an estimated value of demand for each of the plurality of first products to be presented that will be generated in response to the presentation of this price information; a second acquisition means for acquiring an average gross profit function that, when price information for each of the plurality of first products to be presented and the presentation time are input, outputs an estimated value of average gross profit to be obtained from the first product and a second product related to the first product, in accordance with the demand for each of the plurality of first products to be presented that will be generated in response to the presentation of this price information; and a search means for determining price information for each of the first products to be presented using the demand function and the average gross profit function to search for price information for each of the first products to be presented that will increase the estimated value of demand for each of the first products to be presented and the total profit based on the estimated value of the average gross profit when each of a plurality of times after the current time is set as the presentation time.

[0009] An information processing method according to one aspect of the present invention is an information processing method executed by at least one processor, and includes: when price information for each of a plurality of first products to be presented, with time information set for each first product, and a presentation time when the price information is to be presented simultaneously, obtaining a demand function that outputs an estimate of demand for each of the plurality of first products to be presented that will be generated in response to the presentation of this price information; when price information for each of the plurality of first products to be presented and the presentation time are input, obtaining an average gross profit function that outputs an estimate of average gross profit to be obtained from the first products and second products related to the first products, in accordance with the demand for each of the plurality of first products to be presented that will be generated in response to the presentation of this price information; and using the demand function and the average gross profit function, determining price information for each of the first products to be presented that will increase the total profit based on the estimated demand and the estimated average gross profit when each of a plurality of times after the current time is set as the presentation time.

[0010] A program according to one aspect of the present invention is a program for causing a computer to function as an information processing device, and causes the computer to function as: a first acquisition means for acquiring a demand function that, when price information for each of a plurality of first products to be presented, with time information set for each first product, and a presentation time when this price information is presented simultaneously, outputs an estimated value of demand for each of the plurality of first products to be presented that will arise in response to the presentation of this price information; a second acquisition means for acquiring an average gross profit function that, when price information for each of the plurality of first products to be presented and the presentation time are input, outputs an estimated value of average gross profit to be obtained from the first product and second products related to the first products, in accordance with the demand for each of the plurality of first products to be presented that will arise in response to the presentation of this price information; and a search means for determining price information for the first product of each of the first products to be presented, by using the demand function and the average gross profit function to search for price information for each of the first products to be presented that will increase the estimated value of demand for each of the first products to be presented and the total profit based on the estimated value of the average gross profit when each of a plurality of times from the current time onwards is set as the presentation time.

[0011] According to one aspect of the present invention, the price of each of multiple products with time information set can be determined so as to maximize the total future profits that can be obtained from that product and other products as demand for each product increases.

[0012] FIG. 1 is a block diagram showing the configuration of an information processing device according to exemplary embodiment 1 of the present invention. FIG. 2 is a flow diagram showing the flow of an information processing method according to exemplary embodiment 1 of the present invention. FIG. 3 is a block diagram showing the configuration of an information processing system according to exemplary embodiment 2 of the present invention. FIG. 4 is a block diagram showing the functional configuration of an information processing device according to exemplary embodiment 2 of the present invention. FIG. 5 is a flow diagram explaining the flow of an information processing method according to exemplary embodiment 2 of the present invention. FIG. 6 is a flow diagram showing a detailed flow of a sampling process of a demand function in exemplary embodiment 2 of the present invention. FIG. 7 is a flow diagram showing a detailed flow of a sampling process of an average gross profit function in exemplary embodiment 2 of the present invention. FIG. 8 is a flow diagram showing a detailed flow of a search process in exemplary embodiment 2 of the present invention. FIG. 9 is a schematic diagram showing an example of a screen in application example 1 of exemplary embodiment 2 of the present invention. FIG. 10 is a schematic diagram explaining an example of the operation of the information processing system according to application example 1 described above at a certain point in time. FIG. 11 is a schematic diagram explaining example of the operation of the information processing system in application example 1 described above at a plurality of points in time. FIG. 12 is a schematic diagram showing an example of a screen in application example 2 of a modified example of exemplary embodiment 2 of the present invention. FIG. 13 is a diagram showing an example of the hardware configuration of each device according to each exemplary embodiment of the present invention.

[0013] [First Exemplary Embodiment] A first exemplary embodiment of the present invention will be described in detail with reference to the drawings. This exemplary embodiment is a basic form of the exemplary embodiments described below.

[0014] (Overview of Information Processing Device 1) The information processing device 1 according to a first exemplary embodiment of the present invention is a device that determines price information for multiple first products to be presented among multiple first products with time information. The price information is information including the price of the first product. Here, the multiple first products to be presented are products for which price information is presented to the user simultaneously. Furthermore, the first product with time information refers to a product for which time information is set. The time information is information related to time, and examples include a deadline, date, date and time, period, or time period. Furthermore, the first product is a product from which a profit can be obtained from a second product related to the first product due to demand for the first product. Note that the number of second products from which a profit can be obtained in relation to the first product may be one or multiple.

[0015] For example, when a delivery slot in online shopping is a first product, time information such as a time period may be set for the first product, and a shipping fee may be set according to the delivery slot. When a demand for such a delivery slot arises, not only can the shipping fee be earned, but also profits can be earned from the second product, i.e., the purchased item, that is delivered in response to the demand.

[0016] Furthermore, in an online reservation service for a hair salon, when a reservation slot for a stylist is defined as a first product, time information, i.e., a time slot, may be set for the first product, and a nomination fee may be set according to the reservation slot. When demand for such a reservation slot arises, not only will a nomination fee be earned but profits will also be earned from a second product, i.e., a styling menu offered in response to the demand. Furthermore, in an online reservation service for a restaurant, time information, i.e., a time slot, may be set for the first product, i.e., a table charge according to the time slot. When demand for such a reservation slot arises, not only will a table charge be earned but profits will also be earned from a second product, i.e., a dish ordered in response to the demand. Furthermore, in an online advance ticket service for a theme park, time information, i.e., a ticket, may be set for the first product, i.e., a date of use, and a ticket price may be set according to the date of use. When demand for such a ticket arises, not only will a ticket price be earned but profits will also be earned from a second product, i.e., an attraction used within the theme park in response to the demand. Note that the first and second products are not limited to the examples described above.

[0017] Demand for such a first product with time information depends on price information and the passage of time. Furthermore, demand for multiple first products with time information is correlated with each other. The information processing device 1 determines price information for multiple first products based on price information, demand over time, demand correlation, and profits from the first and second products, to maximize total future profits. The maximized total profit can be calculated from estimated demand for each first product when price information is presented at multiple points in time from the current time onward, and estimated average gross profits from the first and second products due to the demand for the first products. Here, the average gross profit is calculated by dividing the total profits from the first and second products due to the demand for a certain first product by the demand for the first product. For example, if the number of users accepted for a certain delivery slot (first product) in online shopping is five, the average gross profit for the delivery slot can be calculated by dividing the total profits from the five users by five.

[0018] (Configuration of information processing device 1) The configuration of the information processing device 1 according to this exemplary embodiment will be described with reference to Fig. 1. Fig. 1 is a block diagram showing the configuration of the information processing device 1. As shown in Fig. 1, the information processing device 1 includes a first acquisition unit 11, a second acquisition unit 12, and a search unit 13.

[0019] The first acquisition unit 11 acquires a demand function that outputs an estimated value of the demand for each of the multiple first products to be presented that will arise in response to the presentation of this price information when price information for each of the multiple first products to be presented, with time information set for each first product, and the presentation time at which this price information will be presented simultaneously, are input.

[0020] When the price information of each of the multiple first products to be presented and the time of presentation are input, the second acquisition unit 12 acquires an average gross profit function that outputs an estimated value of the average gross profit to be obtained from the first products and second products related to the first products, in accordance with the demand for each of the multiple first products to be presented that arises in response to the presentation of this price information.

[0021] The search unit 13 determines price information for the first product of each of the presented items by using a demand function and an average gross profit function to search for price information for the first product of each of the presented items that will increase the total profit based on the estimated demand and estimated average gross profit for the first product of each of the presented items when each of multiple points in time after the current point in time is the presentation point in time.

[0022] (Example of Implementation by Program) When the information processing device 1 is configured by a computer including at least one processor and a memory, a program for causing the computer to function as the information processing device 1 is stored in the memory. The program causes a computer to function as a first acquisition unit 11 that acquires a demand function that, when price information for each of a plurality of first products to be presented, with time information set for each first product, and the presentation time at which this price information is presented simultaneously, outputs an estimated value of demand for each of the plurality of first products to be presented that will arise in response to the presentation of this price information; a second acquisition unit 12 that, when price information for each of the plurality of first products to be presented and the presentation time are input, acquires an average gross profit function that, in accordance with demand for each of the plurality of first products to be presented that will arise in response to the presentation of this price information, outputs an estimated value of average gross profit to be obtained from the first products and second products related to the first products; and a search unit 13 that determines price information for each of the first products to be presented by using the demand function and the average gross profit function to search for price information for each of the first products to be presented that will increase the total profit based on the estimated value of demand and the estimated average gross profit for each of the first products to be presented when each of a plurality of times from the current time onwards is set as the presentation time.

[0023] (Flow of information processing method S1) The information processing device 1 configured as described above has at least one processor, and the at least one processor executes the information processing method S1. Figure 2 is a flow diagram showing the flow of the information processing method S1. As shown in Figure 2, the information processing method S1 includes steps S11 to S13.

[0024] In step S11, when price information for each of the multiple first products to be presented, with time information set for each first product, and the presentation time at which these price information are presented simultaneously are input, at least one processor obtains a demand function that outputs an estimated value of demand for each of the multiple first products to be presented that will arise in response to the presentation of this price information.

[0025] In step S12, when price information for each of the multiple first products to be presented and the time of presentation are input, at least one processor obtains an average gross profit function that outputs an estimate of the average gross profit to be obtained from the first products and second products related to the first products, in response to demand for each of the multiple first products to be presented that arises in response to the presentation of this price information.

[0026] In step S13, at least one processor determines price information for the first product of each of the presented items by using a demand function and an average gross profit function to search for price information for the first product of each of the presented items that will increase the total profit based on the estimated demand and estimated average gross profit for the first product of each of the presented items when each of multiple points in time from the current time onwards is the presentation point in time.

[0027] (Effects of this exemplary embodiment) According to this exemplary embodiment, when price information for each of a plurality of first products to be presented, with time information set for each first product, and the presentation time at which this price information is presented simultaneously is input, a demand function is obtained that outputs an estimate of the demand for each of the plurality of first products to be presented that will arise in response to the presentation of this price information, and when price information for each of the plurality of first products to be presented and the presentation time are input, an average gross profit function is obtained that outputs an estimate of the average gross profit to be obtained from the first product and second products related to the first product, in accordance with the demand for each of the plurality of first products to be presented that will arise in response to the presentation of this price information, and the demand function and average gross profit function are used to search for price information for each of the first products to be presented that will increase the total profit based on the estimated demand and estimated average gross profit for each of the first products to be presented when each of a plurality of times from the current time onwards is set as the presentation time, thereby determining the price information for each of the first products to be presented.

[0028]

[0033] A second exemplary embodiment of the present invention will be described in detail with reference to the drawings. Note that components having the same functions as those described in the first exemplary embodiment are denoted by the same reference numerals, and their description will be omitted as appropriate.

[0029] In this exemplary embodiment, the price of the first product is set to be discounted or free if the purchase amount of the second product is equal to or greater than a threshold. For example, the shipping fee for a delivery slot in an online supermarket may be set to be free if the purchase amount is equal to or greater than a free shipping price (an example of a threshold). Furthermore, the shipping fee for a delivery slot may be set to be discounted if the purchase amount is equal to or greater than a shipping discount price (an example of a threshold). Hereinafter, the combination of the price of the first product and the threshold value of the purchase amount of the second product for discounting or freeing the price of the first product will also be referred to as "price information." The information processing system 10 according to this exemplary embodiment is a system for selling multiple first products by dynamically determining and simultaneously presenting price information for the first products to be presented with time information. Note that the information processing system 10 may, but does not necessarily, include a function for selling second products. Specific examples of the first product and the second product are as described in the first exemplary embodiment, and detailed description thereof will not be repeated.

[0030] In addition, in this exemplary embodiment, an example is described in which, as the first products to be presented, multiple first products that can accept demand are applied from among multiple first products whose set time information falls within a specified period.

[0031] "The set time information is included within a predetermined period" means that the deadline, date, period, or time slot indicated by the time information is included within the predetermined period. Note that some first products have time information set that includes a time slot (or date, period, etc.) and a deadline for accepting demand, such as a delivery slot in an online supermarket. In such cases, the deadline may be set before the time slot. For example, a delivery slot may be set with a delivery time slot from 10:00 to 12:00 and a cutoff time of 9:00. In this case, "the set time information is included within a predetermined period" may mean that the time slot indicated by the time information is included within the predetermined period, or that the deadline indicated by the time information is included within the predetermined period. Furthermore, "demand can be accepted" means that the deadline for accepting demand for the first product has not yet arrived. If the deadline for accepting demand is not included in the time information, the deadline for accepting demand may be, for example, the start point or end point of the time slot included in the time information.

[0032] In this exemplary embodiment, the price information for each of the multiple first products to be offered is repeatedly determined using a demand function and an average gross profit function trained based on performance data observed by offering the price information. Note that, since the first products cannot accept demand after the set time information has passed, the number of first products to be offered decreases over time. For example, it is considered that the delivery slots of an online supermarket tend to have a trend in the demand generation pattern on a daily basis. Therefore, for the multiple delivery slots included in the target day, the determination of price information for each remaining delivery slot on the target day may be repeated until the closing time of the final delivery slot arrives.

[0033] <Configuration of Information Processing System 10> The configuration of the information processing system 10 will be described with reference to Fig. 3. Fig. 3 is a block diagram showing the configuration of the information processing system 10. As shown in Fig. 3, the information processing system 10 includes an information processing device 1A, a sales system 2, and a performance record storage device 3. These devices are communicably connected via a network. The network may be configured to include part or all of a wireless LAN (Local Area Network), a wired LAN, the Internet, a WAN (Wide Area Network), or a mobile data communication network.

[0034] The information processing device 1A searches for and determines price information for multiple first products to be presented and second products related to each of the first products so as to maximize the total future profits from the first products. The total future profits are the sum of profits to be obtained at multiple points in time from the present time. The profits to be obtained at each point in time can be calculated from an estimated demand for each of the multiple first products when price information for the first products is presented at that point in time and an estimated average gross profit to be obtained from the first products and second products due to the demand for the first products. The sales system 2 sells each of the first products to be presented by presenting price information determined by the information processing device 1A to users. In this exemplary embodiment, the sales system 2 also has a function of selling second products. The sales record storage device 3 stores sales records of the first products and second products sold in the sales system 2. The information processing device 1A, the sales system 2, and the sales record storage device 3 are each configured by a computer including one or more processors and memories.

[0035] <Configuration of Information Processing Device 1A> The functional configuration of the information processing device 1A will be described with reference to FIG. 4. FIG. 4 is a block diagram showing the functional configuration of the information processing device 1A. As shown in FIG. 4, the information processing device 1A includes a control unit 110, a storage unit 120, and a communication unit 130. The control unit 110 controls each unit of the information processing device 1A. For example, the control unit 110 includes a first acquisition unit 11, a second acquisition unit 12, a search unit 13, and a presentation unit 14. The storage unit 120 stores various data used by the control unit 110. For example, the storage unit 120 stores a demand function model ML1 and an average gross profit function model ML2. The communication unit 130 connects to a network and communicates with other devices under the control of the control unit 110.

[0036] The first acquirer 11 is configured similarly to the first exemplary embodiment, and is also configured as follows: The first acquirer 11 acquires a demand function learned based on the first performance data. The first acquirer 11 also calculates a posterior distribution of parameters included in the demand function based on the first performance data, and acquires the demand function based on parameters sampled by the posterior distribution.

[0037] The first performance data is information that associates price information for each of the multiple first products to be presented, the presentation time when the price information was presented simultaneously, and the performance value of demand for each of the first products observed in response to the presentation of the price information. The first performance data may also include price information for each of the multiple first products to be presented determined by the search unit 13, the presentation time when the price information was presented simultaneously, and the performance value of demand for each of the first products observed in response to the presentation of the price information. The first performance data is observed by the sales system 2 and stored in the performance storage device 3.

[0038] The second acquisition unit 12 is configured similarly to the first exemplary embodiment, but is also configured as follows: The second acquisition unit 12 acquires an average gross profit function learned based on the second performance data. The second acquisition unit 12 also calculates a posterior distribution of parameters included in the average gross profit function based on the second performance data, and acquires the average gross profit function based on parameters sampled by the posterior distribution.

[0039] The second performance data is information associating price information for each of the multiple first products to be presented, the time when the price information was simultaneously presented, and the actual average gross profit values ​​obtained from the first products and the second products, observed in response to the demand for each of the first products that arose in response to the presentation of the price information. The second performance data may also be information associating price information for each of the multiple first products to be presented, determined by the search unit 13, the time when the price information was simultaneously presented, and the actual average gross profit values ​​obtained from the first products and the second products, observed in response to the demand for each of the first products that arose in response to the presentation of the price information. The second performance data is observed by the sales system 2 and stored in the performance storage device 3.

[0040] The search unit 13 is configured similarly to the first exemplary embodiment, but is also configured as follows. That is, the search unit 13 determines price information for each of a plurality of first products to be presented by searching for a set of price and threshold for each of the first products to be presented. The search is performed to maximize the total profit based on the estimated demand and estimated average gross profit when each of a plurality of points in time from the current point in time is set as the presentation point in time. Furthermore, the search unit 13 performs the search under constraints regarding the inventory quantity of the first product.

[0041] The presentation unit 14 simultaneously presents the price information of each of the first products to be presented determined by the search unit 13 to the user via the sales system 2 that sells each of the first products to be presented.

[0042] <Flow of information processing method S10> The information processing device 1A configured as described above executes the information processing method S10 according to this exemplary embodiment. The flow of the information processing method S10 will be described with reference to Fig. 5. Fig. 5 is a flow diagram illustrating the flow of the information processing method S10. As shown in Fig. 5, the information processing method S10 includes steps S101 to S111.

[0043] In the following description, in a sequence of first products in which J first products with time information whose set time information falls within a predetermined period are arranged on a time axis in order based on the time information, the jth first product from the one closest to the start of the predetermined period will be referred to as the "first product j," where J is an integer of 2 or greater, and j is a natural number between 1 and J.

[0044] Furthermore, a point in time based on the time information set for the first product j is represented as Tj. Tj may be, for example, a deadline, a due date, etc. indicated by the time information. Tj may also be the start point, end point, etc. of a period or time slot indicated by the time information. Hereinafter, for ease of explanation, "a point in time Tj based on the time information set for the first product j" will also be simply referred to as "the deadline Tj of the first product j", etc. However, Tj is not limited to a point in time called a "deadline", and may be any other point in time indicated by the time information.

[0045] In step S101, the first acquisition unit 11 sets a demand function model ML1. The demand function model ML1 is a model representing a demand function. Here, the input of the demand function includes price information for each of the multiple first products to be presented and information indicating the presentation time when the price information is presented simultaneously. As described above, the price information includes a set of the price of the first product and a threshold value for the amount of use of the second product. Furthermore, the output of the demand function includes an estimated value of the demand for each first product that will arise in response to the presentation of this price information.

[0046] Here, it is assumed that the demand for the first commodity is determined probabilistically based on the above-described inputs. In this case, the demand function may be a function that outputs a value specifying a demand distribution (e.g., the mean value of the distribution) for the above-described inputs. When the relationship between the input and output of such a demand function is defined by parameters, the demand function model ML1 may be represented by a demand function template and the distribution of the parameters. For example, in step S101, the first acquisition unit 11 sets the demand function template and the initial distribution of the parameters as the demand function model ML1.

[0047] In the following, it is assumed that the demand for the first product follows a Poisson distribution. The Poisson distribution is specified by its mean value. Therefore, the following explanation focuses on an example in which a model representing a demand function that outputs the mean value of the demand distribution is applied as the demand function model ML1. However, the distribution that the demand is assumed to follow is not limited to the Poisson distribution.

[0048] In step S102, the second acquisition unit 12 sets an average gross profit function model ML2. The average gross profit function model ML2 is a model that represents an average gross profit function. Here, the input of the average gross profit function is the same as the input of the demand function, and includes price information for each of the multiple first products to be presented and information indicating the presentation time when this price information is presented simultaneously. In addition, the output of the average gross profit function includes an estimated value of the average gross profit obtained from the first products and the second products in response to the demand for each of the first products that arises in response to the presentation of this price information.

[0049] Here, it is assumed that the average gross profit associated with the demand for the first product is determined probabilistically based on the above-mentioned inputs. In this case, the average gross profit function may be a function that outputs a value specifying the distribution of the average gross profit (e.g., the mean value of the distribution) for the above-mentioned inputs. When the relationship between the input and output of such an average gross profit function is defined by parameters, the average gross profit function model ML2 may be represented by an average gross profit function template and the distribution of the parameters. For example, in step S102, the second acquisition unit 12 sets the average gross profit function template and the initial distribution of the parameters as the average gross profit function model ML2.

[0050] In the following, we assume that the average gross profit associated with the demand for the first product follows a Gaussian distribution based on the central limit theorem. One of the values ​​that specifies a Gaussian distribution is the mean value. Therefore, we will mainly explain an example in which a model representing an average gross profit function that outputs the mean value of the distribution of average gross profit is applied as the average gross profit function model ML2. However, the distribution that the average gross profit is assumed to follow is not limited to a Gaussian distribution.

[0051] Note that steps S101 and S102 are not limited to being executed in this order, and may be executed in the reverse order, or some or all of the processes may be executed in parallel.

[0052] In step S103, the first acquisition unit 11 calculates a posterior distribution of the parameters included in the demand function model ML1 based on the first actual data, and acquires a demand function based on the parameters sampled by the posterior distribution. Acquiring a demand function based on the parameters sampled by the posterior distribution is referred to as sampling a demand function. Details of the demand function sampling process will be described with reference to FIG. 6. FIG. 6 is a flow chart showing a detailed flow of the demand function sampling process (step S103). As shown in FIG. 6, the demand function sampling process includes steps S201 to S203.

[0053] In step S201, the first acquisition unit 11 calculates the posterior distribution of the parameters included in the demand function model ML1. Here, the first acquisition unit 11 assumes a prior distribution of the parameters included in the demand function model ML1. In the following, an example in which a Gaussian process is assumed as the prior distribution of the parameters will be described, but the prior distribution is not limited to this. The first acquisition unit 11 calculates the posterior distribution of the parameters based on the first performance data acquired from the performance storage device 3 and the prior distribution of the parameters. Here, it is assumed that one or more pieces of first performance data obtained in the past have already been stored in the performance storage device 3.

[0054] The first performance data is data including, for each of a plurality of first products of a plurality of presentation targets whose price information is presented simultaneously, the plurality of price information presented simultaneously and the demand for each of the first products of the presentation targets observed in response to the presentation of the price information. For example, the first performance data is expressed by the following formula (1):

[0055] In formula (1), D1 indicates the first performance data. i is a subscript that identifies multiple first products included in one piece of first performance data. p indicates a string in which price information presented simultaneously for multiple first products, including the first product i, is arranged in order of deadline Ti. Hereinafter, the price information string will also be simply referred to as a "price information string." t indicates the presentation time when the price information string p is presented simultaneously. Ti, as described above, indicates the time (also referred to as the deadline) indicated by the time information set for the first product. ξi indicates the context of the first product. Examples of context include, but are not limited to, the day of the week on which the deadline Ti of the first product occurs, the type of campaign set for the first product, etc. Note that ξi may be omitted if the context is not taken into consideration. d_i is the demand observed for the first product i in response to the presentation of the price information string p. Ui indicates the union of this information for each product.

[0056] In step S202, the first acquiring unit 11 samples the parameters based on the posterior distribution of the parameters.

[0057] In step S203, the first acquisition unit 11 acquires a demand function by applying the sampled parameters to the demand function model ML1. The acquired demand function is an example of a sampled demand function. As a specific example, the first acquisition unit 11 acquires a demand function that constructs an average of demand using the sampled parameters and outputs an estimated value of the average of demand. For example, the sampled demand function is expressed by the following equation (2).

[0058] As shown in equation (2), the demand function λ represents a sampled demand function. The output of the demand function λ is an estimated average demand for each of the multiple first products to be presented. The input of the demand function λ includes a price information sequence p presented simultaneously for the multiple first products to be presented, a presentation time t of the price information sequence p, a deadline Ti for each first product to be presented, and a context ξi for each first product to be presented. This concludes the explanation of the specific flow of the sampling process of the demand function.

[0059] In step S104 of Fig. 5, the second acquisition unit 12 calculates a posterior distribution of parameters included in the average gross profit function based on the second performance data, and acquires the average gross profit function based on the parameters sampled using the posterior distribution. Acquiring the average gross profit function based on the parameters sampled using the posterior distribution is referred to as "sampling the average gross profit function." Details of the sampling process of the average gross profit function will be described with reference to Fig. 7. Fig. 7 is a flow chart showing a detailed flow of the sampling process of the average gross profit function (step S104). As shown in Fig. 7, the sampling process of the average gross profit function includes steps S301 to S303.

[0060] In step S301, the second acquisition unit 12 calculates the posterior distribution of the parameters included in the average gross profit function model ML2. Here, the second acquisition unit 12 assumes a prior distribution of the parameters included in the average gross profit function model ML2. In the following, an example in which a Gaussian process is assumed as the prior distribution of the parameters will be described, but the prior distribution is not limited to this. The second acquisition unit 12 calculates the posterior distribution of the parameters based on the second performance data acquired from the performance storage device 3 and the prior distribution of the parameters.

[0061] Here, it is assumed that one or more pieces of second performance data obtained in the past have already been stored in the performance storage device 3. The second performance data is data including, for each of a plurality of first products for which price information was presented simultaneously, the multiple pieces of price information presented simultaneously, and the average gross profit obtained from the first products and the second products observed in response to the presentation of this price information. For example, the second performance data is expressed by the following equation (3):

[0062] In formula (3), D2 represents the second performance data. i, p, t, Ti, and ξi are as explained in formula (1), and therefore will not be explained again. r_i represents the average gross profit obtained from the first product i and the second product, observed in response to the demand for the first product i that arose in response to the presentation of the price information sequence p.

[0063] In step S302, the second acquiring unit 12 samples the parameters based on the posterior distribution of the parameters.

[0064] In step S303, the second acquisition unit 12 acquires an average gross profit function by applying the sampled parameters to the average gross profit function model ML2. The acquired average gross profit function is an example of a sampled average gross profit function. As a specific example, the second acquisition unit 12 constructs an average gross profit using the sampled parameters and acquires an average gross profit function that outputs an estimated value of the average gross profit. For example, the average gross profit function is expressed by the following equation (4):

[0065] As shown in equation (4), the average gross profit function μ represents a sampled average gross profit function. In equation (4), the input of the average gross profit function μ is the same as the demand function shown in equation (2), and includes the price information sequence p presented simultaneously for multiple first products to be presented, the presentation time t of the price information sequence p, the expiration date Ti of each first product to be presented, and the context ξi of each first product to be presented. This concludes the explanation of the specific flow of the sampling process of the average gross profit function.

[0066] In step S105 of Fig. 5, the search unit 13 searches for price information for each of the multiple first products to be presented using the sampled demand function and the sampled average gross profit function. The search is performed to maximize the total profit based on the estimated demand and estimated average gross profit when each of multiple points in time from the current time onward is set as the presentation point in time. Details of the search process will be described with reference to Fig. 8. Fig. 8 is a flow diagram showing the detailed flow of the search process (step S105). As shown in Fig. 8, the search process includes steps S401 to S402.

[0067] In step S401, the search unit 13 identifies a first product among a plurality of presented targets to be priced. Here, the first products among the plurality of presented targets are first products whose deadline Tj has not yet arrived among J first products j whose set time information falls within a predetermined period.

[0068] In step S402, the search unit 13 searches for price information for each first product to be presented by solving an optimization problem that maximizes the total future profits under constraints on inventory quantities using the sampled demand function and the sampled average gross profit function.

[0069] Here, the optimization problem is expressed by the following equation (5) based on the demand function λ (sampled demand function) shown in equation (2) and the average gross profit function μ (sampled average gross profit function) shown in equation (4).

[0070] In addition, in equation (5), the inputs of the demand function λ and the average gross profit function μ are written as λ(p, t') and μ(p, t'), omitting the notation of Ti determined by the subscript jt' and the notation of ξi, which are not required, from the inputs shown in equations (2) and (4), respectively.

[0071] Furthermore, jt' is a subscript given to the product whose deadline arrives first after time t'. λjt' and μjt' represent the demand function λ and average gross profit function μ obtained at time t'. At time t', the subscripts of the first products of the multiple presentation targets whose deadline Tj has not yet arrived within the specified period are expressed as j = jt', jt'+1, ..., J. In other words, at time t', (J-jt'+1) first products are presentation targets for which price information is presented simultaneously.

[0072] Furthermore, P in formula (5) represents K candidates cp1 to cpK for the price of the first product, as shown in the following formula (6). Furthermore, Rd represents L candidates crd1 to crdL for the threshold value of the spending amount of the second product, as shown in the following formula (7).

[0073] Therefore, (P × Rd) in equation (5) Ｊ－ｊｔ’＋１ represents all the different combinations of (J-Jt'+1) items extracted from K x L price information candidates, each of which is a combination of any one of the K price candidates and any one of the L threshold value candidates. In other words, (P x Rd) Ｊ－ｊｔ’＋１ represents a price information sequence candidate for each of the (J-Jt'+1) first products to be presented. p represents one of these price information sequence candidates. xjt'(p, t') represents a weight assigned to the price information sequence candidate p to be presented at time t'. T represents the deadline Tj set for the first product J.

[0074] That is, the objective function included in equation (5) represents the total profit calculated based on a weighted price information sequence candidate p and the estimated demand and average gross profit corresponding to candidate p at each time point between time t' = t (the present time) and time T. In other words, the maximization problem shown in equation (5) represents a search for weights of the price information sequences of (J - Jt' + 1) first products to be presented at each of multiple times (t' = t, t + 1, ..., T) from the present time onwards, so as to maximize future profits at the multiple times.

[0075] Furthermore, the constraint on the inventory quantity in the optimization problem is expressed by the following equation (8).

[0076] In equation (8), demand [λ ｊｔ’ (p, t')] ｊ－ｊｔ’＋１ indicates the estimated demand for the first product (j-jt'+1) according to the price information sequence p presented at time t'. In other words, equation (8) indicates the constraint that for each of the (J-Jt'+1) first products (j = jt, jt+1, ..., J) to be presented at current time t, the weighted sum of the estimated demand according to the price information sequence p presented at each of multiple times after the current time t is less than or equal to the current inventory quantity nj(t-1). In other words, equation (8) indicates the constraint that there will be no shortage of inventory for each of the (J-Jt'+1) first products at each of multiple times after the current time.

[0077] Note that the weight x(p, t') to be searched for in equation (5) is further constrained by the following equation (9).

[0078] Equation (9) indicates that the weights assigned to price information candidates p at each of multiple points in time t' (t, t+1, ..., T) after the current point in time are each 0 or greater, and the sum of the weights assigned to each candidate p at each point in time t' is 1.

[0079] The search unit 13 searches for an optimal solution for the weight of each price information sequence candidate p by solving the maximization problem shown in equation (5) under the constraints shown in equations (8) and (9). This concludes the explanation of the specific flow of the search process.

[0080] 5, the search unit 13 determines price information for each of the first products of the plurality of presentation targets based on the search result of step S105. For example, the search unit 13 may determine the searched price information sequence candidate with the largest weight as the price information to be proposed for the first products of the plurality of presentation targets.

[0081] In step S107, the presentation unit 14 causes the sales system 2 to present price information for the multiple first products to be presented, determined in step S106. The presentation destination is a consumer. For example, the sales system 2 presents a sales screen for selling the multiple first products to one or more user terminals (not shown), including the price information for each first product acquired from the presentation unit 14. The sales system 2 executes a process for selling each of the multiple first products to be presented, based on user operations on each user terminal. The sales system 2 also executes a process for selling a second product to a user who purchases the first product. The sales system 2 also stores first performance data and second performance data in the performance storage device 3 based on the sales performance of the multiple first products to be presented and the sales performance of the second product. The first performance data indicates the demand for the first product observed in response to the simultaneous presentation of price information for each of the multiple first products to be presented. The second actual data indicates the average gross profit obtained from the first product and the second product observed in response to the demand for the first product that arose in response to the simultaneous presentation of price information for each of the multiple first products to be presented.

[0082] In step S108 , the first acquisition unit 11 acquires, from the result storage device 3 , the first performance data stored for each of the plurality of first products to be presented.

[0083] In step S109 , the second acquisition unit 12 acquires, from the result storage device 3 , the second performance data stored for each of the plurality of first products to be presented.

[0084] Note that steps S108 and S109 are not limited to being executed in this order, and may be executed in the reverse order, or some or all of the processes may be performed in parallel.

[0085] In step S110, the first acquirer 11 trains the demand function model ML1 based on the acquired first performance data. As an example of the training, the first acquirer 11 may update the posterior distribution of the parameters included in the demand function model ML1 based on the first performance data.

[0086] In step S111, the second acquiring unit 12 trains the average gross profit function model ML2 based on the acquired second performance data. As an example of the training, the second acquiring unit 12 may update the posterior distribution of the parameters included in the average gross profit function model ML2 based on the second performance data.

[0087] Note that steps S111 and S112 are not limited to being executed in this order, and may be executed in the reverse order, or some or all of the processes may be executed in parallel.

[0088] Next, the information processing device 1A sets the new time as the current time and repeats the processing from step S103. Here, if the expiration date of any of the first products of the multiple presented targets, whose price information was most recently presented simultaneously, has not passed at the new current time, the processing from step S103 is repeated for the first products of the same multiple presented targets. This allows the price information of the first products of the same multiple presented targets to be updated. On the other hand, at the new current time, the expiration date of at least one of the first products of the multiple presented targets, whose price information was most recently presented simultaneously, may have passed. In this case, the processing from step S103 is repeated with the remaining first products, excluding the first products whose expiration date has passed, as the first products of the new multiple presented targets. This allows price information to be determined for the first products of the new multiple presented targets, and they are sold in the sales system 2.

[0089] In the description of the information processing method S10, an example has been described in which the first acquiring unit 11 samples the demand function by sampling parameters from the posterior distribution of the parameters included in the demand function model ML1. This is not a limitation, and the first acquiring unit 11 may sample the demand function by calculating the posterior distribution from the prior distribution of the demand function indicated by the demand function model ML1 and the first actual data. Also, an example has been described in which the second acquiring unit 12 samples the average gross profit function by sampling parameters from the posterior distribution of the parameters included in the average gross profit function model ML2. This is not a limitation, and the second acquiring unit 12 may sample the average gross profit function by calculating the posterior distribution from the prior distribution of the average gross profit function indicated by the average gross profit function model ML2 and the second actual data.

[0090] Effects of the Present Exemplary Embodiment In addition to the same configuration as in the first exemplary embodiment, the present exemplary embodiment employs a configuration in which price information for each of the first products to be presented determined by the search unit 13 is simultaneously presented to the user via the sales system 2 that sells each of the first products to be presented. This configuration, in addition to the effects of the first exemplary embodiment, provides the effect of more reliably increasing the total future profits to be obtained from the first products and second products in accordance with the demand for each of the first products to be presented, based on the average demand obtained in accordance with the presented price information for the first products to be presented. Furthermore, for the product buyer, the effect of being able to purchase a desired first product at a more appropriate price while also purchasing a second product related to the first product is also provided.

[0091] In this exemplary embodiment, the price of the first product is set to be discounted or free when the amount spent on the second product is equal to or greater than a threshold, the price information for the first product includes the price of the first product and a set of the threshold presented simultaneously with the price, and the search unit 13 determines the price information for each of the first products to be presented by searching for a set of the price and the threshold for each of the first products to be presented that will increase the total profit.This configuration has the effect of determining a set of the price of the first product and the threshold that will increase the total future profits from the first product and the second product when the threshold for the amount spent on the second product affects the demand for the first product.

[0092] In this exemplary embodiment, the first acquisition unit 11 acquires a demand function trained based on first performance data correlating price information for each of the multiple first products to be presented, the time at which the price information was simultaneously presented, and actual demand values ​​for each of the multiple first products observed in response to the presentation of the price information, and the second acquisition unit 12 acquires an average gross profit function trained based on second performance data correlating price information for each of the multiple first products to be presented, the time at which the price information was presented, and actual average gross profit values ​​obtained from the first products and the second products observed in response to the demand for each of the multiple first products to be presented in response to the presentation of the price information. This configuration provides the effect of more accurately estimating the demand and average gross profit for each of the multiple first products to be presented, and more accurately determining price information for each of the multiple first products to be presented by using such estimated values.

[0093] In the present exemplary embodiment, the first performance data includes price information for each of the plurality of first products to be presented determined by the search unit 13, the presentation time when the price information was presented simultaneously, and the performance values ​​of demand for the first products for each of the plurality of products to be presented observed in response to the presentation of the price information, and the second performance data includes price information for each of the plurality of first products to be presented determined by the search unit 13, the presentation time when the price information was presented simultaneously, and the performance values ​​of average gross profits obtained from the first products and the second products observed in response to the demand for the first products for each of the plurality of products to be presented that arose in response to the presentation of the price information. With the above configuration, the demand function used to determine the price information for the plurality of first products to be presented is updated based on the performance of demand for the first products that arose in response to the presentation of the price information, thereby achieving the effect of improving the accuracy of the demand function. In addition, the average gross profit function used to determine the price information of the first products of the multiple presented items is updated based on the actual average gross profit observed in response to the demand for the first products that has arisen in response to the presentation of the price information, thereby achieving the effect of improving the accuracy of the average gross profit function.

[0094] In this exemplary embodiment, the first acquisition unit 11 calculates a posterior distribution of parameters included in the demand function based on the first performance data and acquires the demand function based on the parameters sampled using the posterior distribution. The second acquisition unit 12 calculates a posterior distribution of parameters included in the average gross profit function based on the second performance data and acquires the average gross profit function based on the parameters sampled using the posterior distribution. Therefore, according to this exemplary embodiment, even if price information that is not the optimal price information is searched for, by using a probabilistic process called sampling, it is possible to probabilistically present the optimal price information. Therefore, the presented price information can be improved to be more appropriate. As a result, it is possible to efficiently search for the first performance data and the second performance data and increase profits by using the first performance data and the second performance data.

[0095] <Application Example 1> An application example of the exemplary embodiment 2 will be described. In this application example, a delivery slot of an online supermarket is applied as the first product, and a product from the online supermarket is applied as the second product. Furthermore, a free shipping price is applied as the threshold for the amount of use of the second product. Time information including a time period and a closing time is set for the delivery slot of the first product. The number of deliveries that can be accepted for each delivery slot is applied as the inventory quantity. Furthermore, one day is applied as the specified period. In other words, the information processing system 10 dynamically determines the shipping fee and free shipping price for each delivery slot that is currently being accepted and is included in one day of a certain target date.

[0096] In this application example, the target date includes four delivery slots (J=4). When distinguishing between the delivery slots, they are referred to as delivery slot j (j=1 to 4). Delivery slot 1 has a delivery time period set to 10:00 to 12:00. Delivery slot 2 has a delivery time period set to 12:00 to 14:00. Delivery slot 3 has a delivery time period set to 14:00 to 16:00. Delivery slot 4 has a delivery time period set to 16:00 to 18:00.

[0097] 9 is a schematic diagram showing example screens of an online supermarket in this application example. Screen examples G1 to G5 shown in FIG. 9 are, as an example, displayed by the sales system 2 on a user terminal (not shown) of a user who uses the online supermarket.

[0098] Screen example G1 shows an example of a login screen for an online supermarket. The sales system 2 issues and stores IDs and passwords in advance to users of the service. The user enters their ID and password on screen example G1 and operates the login button. If the ID and password are successfully authenticated in response to this operation, the sales system 2 displays screen example G2 on the user terminal.

[0099] Screen example G2 shows an example of a screen on which a product can be added to a shopping cart. In this example, screen example G2 displays a list of products included in the vegetable category, an "Add to Cart" button corresponding to each product, and a "Check Cart" button. On screen example G2, the user operates the "Add to Cart" button corresponding to the product they wish to purchase. In response to this operation, the sales system 2 stores the corresponding product in information indicating the user's shopping cart. Although not shown, screen example G2 may be configured to enable browsing between other categories. Furthermore, screen example G2 may be configured to enable an operation to specify the number of products to purchase. Furthermore, if the user wants to check their shopping cart, they operate the "Check Cart" button. In response to this operation, the sales system 2 displays screen example G3 on the user terminal.

[0100] Screen example G3 shows an example of a screen for confirming the contents of the shopping cart. Screen example G3 displays a list of products included in the shopping cart, a subtotal of the prices of these products, and a "Select a delivery slot" button. When the user confirms the products included in the shopping cart and proceeds to place an order, they press the "Select a delivery slot" button. In response to this operation, the sales system 2 displays screen example G4 on the user terminal.

[0101] Screen example G4 shows an example of a screen on which a delivery slot can be selected. Screen example G4 displays delivery slot information G41, radio buttons for selecting a delivery slot, and a "Proceed to Order" button. The delivery slot information G41 includes the delivery time period for each delivery slot, the remaining number of acceptable orders, and shipping fee information G411. In this application example, the initial number of acceptable orders for each delivery slot is 10. FIG. 9 shows a state in which all initial acceptable orders remain for each delivery slot. The remaining number of acceptable orders may be updated in real time. The shipping fee information G411 includes the shipping fee and free shipping price for each delivery slot and is based on the price information sequence determined by the information processing device 1A. The shipping fee information G411 is updated the next time a new price information sequence is determined. The user refers to the shipping fee information G411 and selects the radio button corresponding to the desired delivery slot. In this example, delivery slot 3 (14:00-16:00) is selected. If the user wishes to place an order within the selected delivery slot, the user operates the "Order" button. In response to this operation, the sales system 2 displays a screen example G5 on the user terminal.

[0102] Screen example G5 shows an example of a screen for confirming the contents of an order. Screen example G5 displays a list of products included in the shopping cart, a subtotal of the prices of those products, the delivery time slot for the selected delivery slot, the corresponding shipping fee, the discount amount, the total of these prices, and a "Confirm Order" button. The discount amount indicates the discount on shipping costs because the subtotal of the product prices is equal to or greater than the free shipping price. If the user wants to confirm the order with these details, they press the "Confirm Order" button. In response to this operation, the sales system 2 executes a settlement process using the payment method set by the user and an output process of arrangement information for arranging the ordered products.

[0103] FIG. 10 is a schematic diagram illustrating an example of the operation of the information processing system 10 in this application example at a certain point in time. Here, the case where the "certain point in time" is before the deadline of 9:00 for delivery slot 1 is described. At this certain point in time (hereinafter also referred to as the current point in time), all four delivery slots are currently being accepted. Therefore, the information processing device 1A determines the price information sequence Offer1 by searching for the price information of each delivery slot, with delivery slots 1 to 4 currently being offered. In this example, the price information sequence Offer1 includes four pairs of shipping fees and free shipping prices: (200 yen, 1500 yen) for delivery slot 1, (200 yen, 2000 yen) for delivery slot 2, (500 yen, 3000 yen) for delivery slot 3, and (300 yen, 2500 yen) for delivery slot 4.

[0104] Specifically, the information processing device 1A solves the constrained optimization problem shown in equations (5) to (9) using a demand function sampled from the demand function model ML1 at the current time and an average gross profit function sampled from the average gross profit function model M2. The information processing device 1A then searches for weights for candidate price information sequences. The objective of the optimization problem is to maximize the total future sales profit on the target date. The total sales profit to be maximized is calculated based on the estimated number of orders accepted for each delivery slot and the estimated average gross profit when the price information sequence is offered at multiple points in time after the current time on the target date. The optimization problem is also constrained by the constraint that the estimated number of orders accepted for each delivery slot at multiple points in time after the current time on the target date does not exceed the remaining number of orders that can be accepted. The information processing device 1A then determines the price information sequence with the largest weight as the price information sequence Offer1 described above and transmits it to the sales system 2.

[0105] The sales system 2 generates delivery slot information 411 including shipping fee information G411 based on the price information column Offer1 and presents it to the user terminal of a user using the online supermarket. Assume that the time of presentation is, for example, 6:00. The delivery slot information 411 is included in the example screen G4 shown in FIG. 9 . The sales system 2 also accepts orders from one or more users by displaying the example screens G1 to G5 on one or more user terminals. The sales system 2 also calculates first actual data D1-1 and second actual data D2-1 based on orders confirmed until the next update of the shipping fee information G411, and stores them in the actual data storage device 3.

[0106] In this example, the number of orders accepted for each delivery slot (observed demand) is assumed to be 4 for delivery slot 1, 4 for delivery slot 2, 5 for delivery slot 3, and 5 for delivery slot 4. Therefore, as shown in FIG. 9 , first actual result data D1-1 is stored in the actual result storage device 3. The first actual result data D1-1 includes first actual result data for each of the four delivery slots, which associates shipping fee information for the delivery slot, the time point at which the shipping fee information was presented, the number of orders accepted for the delivery slot in response to the presentation of the shipping fee information, and the delivery time period for the delivery slot. For example, the first element of the first actual result data D1-1 indicates the first actual data for delivery slot 1. The first element includes shipping fee information for delivery slot 1 ("Shipping fee 200 yen, free shipping price 1500 yen"), the time point at which the information was presented ("6:00"), the number of orders accepted for delivery slot 1 ("4"), and the delivery time period for delivery slot 1 ("10:00-12:00"). According to formula (1), the first item of the first element is a column of shipping fee information for the four delivery slots, but in the example of Figure 10, the columns of shipping fee information for the four delivery slots are clear based on the other second to fourth elements, so illustration of only the shipping fee information for the delivery slot in the first item is omitted. The second, third, and fourth elements of the first actual data D1-1 are similarly explained.

[0107] Furthermore, the average gross profit for each delivery slot is assumed to be 1,900 yen for delivery slot 1, 2,500 yen for delivery slot 2, 3,100 yen for delivery slot 3, and 2,900 yen for delivery slot 4. Therefore, as shown in FIG. 9, second performance data D2-1 is stored in the performance storage device 3. The second performance data D2-1 includes second performance data for each of the four delivery slots, which associates shipping fee information for the delivery slot, the time at which the shipping fee information was submitted, the average gross profit for the delivery slot in response to the submission of the shipping fee information, and the delivery time slot for the delivery slot. For example, the first element of the second performance data D2-1 indicates the second performance data for delivery slot 1. The first element includes shipping fee information for delivery slot 1 ("shipping fee 200 yen, free shipping price 1,500 yen"), the time at which the delivery slot was submitted ("6:00"), the average gross profit for delivery slot 1 ("1,900 yen"), and the delivery time slot for delivery slot 1 ("10:00 to 12:00"). According to formula (3), the first item of the first element is a column of shipping fee information for the four delivery slots, but in the example of Figure 10, the columns of shipping fee information for the four delivery slots are clear based on the other second to fourth elements, so illustration of only the shipping fee information for the delivery slot in the first item is omitted. The second, third, and fourth elements of the second actual data D2-1 are similarly explained.

[0108] The information processing system 1A updates the demand function model ML1 based on the first performance data D1-1. It also updates the average gross profit function model ML2 based on the second performance data D2-1. The information processing system 10 repeats the above process until 9:00, when four delivery slots are currently being accepted.

[0109] FIG. 11 is a schematic diagram illustrating an example of the operation of the information processing system 10 at multiple points in time in this application example. In FIG. 11, the operation before the closing time of delivery slot 1, 9:00, has been described with reference to FIG. 10 , and therefore will not be described again. However, since the price information sequence determination process has been repeated since the time illustrated in FIG. 10 , a new price information sequence Offer1a different from the price information sequence Offer1 has been determined. Furthermore, based on the price information sequence Offer1a, the shipping fee information G411 has been updated to shipping fee information G411a. Furthermore, as a result of one or more users selecting delivery slots and confirming their orders, the remaining acceptable quantity included in the delivery slot information G41a has been updated. Furthermore, the first actual data D1-1a and the second actual data D2-1a are data based on sales performance observed according to the shipping fee information most recently submitted up until the closing time of delivery slot 1, 9:00.

[0110] As shown in FIG. 11 , once delivery slot 1 closes at 9:00, three currently accepted delivery slots, 2 to 4, become candidates for offer until delivery slot 2 closes at 11:00. At a certain point between 9:00 and 11:00, the information processing device 1A references the first and second performance data D1-1a and D2-1a most recently stored in the performance data storage device 3 to update the demand function model ML1 and the average gross profit function model ML2. The information processing device 1A also solves the constrained optimization problem shown in equations (5) to (9) to determine a price information sequence Offer2 for delivery slots 2 to 4 and transmits it to the sales system 2. The sales system 2 generates delivery slot information G42, including shipping fee information G421, based on the price information sequence Offer2 and presents it to the user terminal. The shipping fee information G421 differs from the shipping fee information G411 in that it does not include information related to delivery slot 1, and in that it includes the remaining acceptability numbers for each delivery slot, the free shipping price for delivery slot 2, and the shipping fee for delivery slot 4. The sales system 2 generates first performance data D1-2 and second performance data D2-2 based on the sales performance up until the next update of the shipping fee information G421, and stores them in the performance storage device 3. The information processing system 10 repeats these operations until 11:00, which is the closing time for delivery slot 2.

[0111] After delivery slot 2 closes at 11:00, delivery slots 3 and 4 that are still being accepted become candidates for offer until delivery slot 3 closes at 13:00. At some point between 11:00 and 13:00, the information processing device 1A references the first actual data D1-2 and second actual data D2-2 most recently stored in the actual data storage device 3 to update the demand function model ML1 and the average gross profit function model ML2. The information processing device 1A also solves the constrained optimization problem shown in equations (5) to (9) to determine a price information sequence Offer3 for delivery slots 3 and 4 and transmits it to the sales system 2. The sales system 2 generates delivery slot information G43, including shipping fee information G431, based on the price information sequence Offer3 and presents it to the user terminal. The shipping fee information G431 differs from the shipping fee information G421 in that it does not include information related to delivery slot 2, and in the remaining acceptability numbers for each delivery slot, and the shipping fee and free shipping price for delivery slot 3. The sales system 2 generates first performance data D1-3 and second performance data D2-3 based on the sales performance up until the next update of the shipping fee information G431, and stores them in the performance storage device 3. The information processing system 10 repeats these operations until 11:00, which is the closing time for delivery slot 3.

[0112] After delivery slot 3's closing time of 13:00 passes, only delivery slot 4 that is currently being accepted becomes an offer target until delivery slot 4's closing time of 15:00. At a certain point between 13:00 and 15:00, the information processing device 1A references the first actual data D1-3 and the second actual data D2-3 most recently stored in the actual data storage device 3 to update the demand function model ML1 and the average gross profit function model ML2. The information processing device 1A also determines a price information sequence Offer4 for delivery slot 4 by solving the constrained optimization problem shown in equations (5) to (9) and transmits it to the sales system 2. The sales system 2 generates delivery slot information G44 including shipping fee information G441 based on the price information sequence Offer4 and presents it to the user terminal. The shipping fee information G441 differs from the shipping fee information G431 in that it does not include information regarding delivery slot 3, and in the remaining number of deliveries that can be accepted for each delivery slot and the free shipping price for delivery slot 4. The sales system 2 generates first performance data D1-3 and second performance data D2-3 based on the sales performance up until the next update of the shipping fee information G441, and stores them in the performance storage device 3. The information processing system 10 repeats these operations until 15:00, which is the closing time of the delivery slot 4.

[0113] According to this application example, the shipping fee and free shipping price for each delivery slot in an online supermarket can be priced taking into account the estimated number of orders accepted for that delivery slot and the average gross profit. This allows the seller to increase the total future profit on the target date in cases where profit is not determined solely by the shipping fee and the number of orders accepted for that delivery slot, but is also determined by the free shipping price for that delivery slot and the average gross profit of sales results for that delivery slot. Furthermore, since pricing is performed under the constraint that the estimated number of orders accepted for each delivery slot does not exceed the remaining number of orders available, the possibility of each delivery slot closing early on the target date is reduced. As a result, the seller reduces opportunity losses due to the customer's desired delivery slot already closing early. Furthermore, for users, the desired delivery slot can be used at a more appropriate shipping fee, and the possibility of the desired delivery slot closing early is reduced, improving convenience.

[0114] [Modification] A modification of the above-described exemplary embodiment 2 will be described, in which a plurality of first products to be presented are selected from a plurality of first products with time information whose supply will continue into the future, in order of closest to the present time on the time axis based on the time information. Specific examples of "a plurality of first products selected in order of closest to the present time on the time axis" include "a predetermined number of first products whose order on the time axis is earliest after the present time" and "a plurality of first products whose order on the time axis is from the present time until a predetermined period from the present time".

[0115] In this modification, "first products whose order on the time axis is a predetermined number J from the current time point onwards" are applied as the first products to be presented. The subscripts of these first products are set to 1, 2, ..., J from the current time point onwards in order of earliest order. In this modification, when the deadline for accepting demand for the first product (j=1) arrives, the search unit 13 of the information processing device 1A increments the subscripts of the first products from j=2 onwards by one. In other words, the subscript j represents the relative order from the current time point. In other words, the subscript j of a certain product changes over time.

[0116] The information processing device 1A solves the constrained optimization problem shown in equations (5) to (9) while updating the subscripts as described above over time. In this way, the information processing device 1A repeatedly determines price information (a set of the price of the first product and the threshold value for the amount of the second product) for J first products selected in order from the product closest to the current time on the time axis. This allows the information processing system 10 to always dynamically determine price information for J first products from the current time onwards, so as to maximize the total future profits obtained from the first and second products.

[0117] According to this modification, among the first products with time information whose supply will continue into the future, multiple first products are selected in descending order of their order on the time axis based on the time information from the present time, and demand can actually be generated according to the presented price information. This has the effect of more reliably increasing the total future profits obtained from the first products and the second products according to the demand for each of the presented first products.

[0118] <Application Example 2> An application example of this modified example will be described. In this application example, a stylist reservation slot in a salon reservation service is applied as the first product, and a nomination fee is applied as the price of the first product. A stylist reservation slot is a product that will continue to be supplied into the future. A time period is set as time information for each reservation slot. Furthermore, the start of the time period is applied as the reservation deadline for each reservation slot. The number of customers that the stylist can handle for each reservation slot is applied as the inventory quantity. Furthermore, a styling menu provided by the stylist is applied as the second product. A nomination fee-free price, at which the nomination fee is waived, is applied as the threshold for the amount spent on the second product. The nomination fee and nomination fee-free price will hereinafter also be referred to as nomination fee information.

[0119] For example, the information processing device 1A dynamically determines the nomination fee and discounted price for each of four reservation slots per day up to three days in advance (a total of 12 reservation slots) as the presenting target. The sales system 2 also accepts reservations from users for these 12 reservation slots. The four reservation slots per day are set to the following time periods, for example, 10:00-12:00, 12:00-14:00, 14:00-16:00, and 16:00-18:00. When distinguishing between the various reservation slots, they may be referred to as reservation slot j (j = 1-12), etc.

[0120] 12 is a schematic diagram showing an example of a salon reservation screen in this application example. Screen examples G21 to G25 shown in FIG. 12 are, as an example, displayed by the sales system 2 on a user terminal (not shown) of a user who uses the salon reservation service.

[0121] Screen example G21 shows an example of a login screen for the salon reservation service. The sales system 2 issues and stores an ID and password in advance to users of the service. The user enters the ID and password on screen example G1 and operates the login button. If the ID and password are successfully authenticated in response to the operation, the sales system 2 displays screen example G22 on the user terminal.

[0122] Screen example G22 shows an example of a screen from which a stylist can be selected. In this example, screen example G2 displays stylists A, B, and C and a nomination button corresponding to each stylist. The user operates the nomination button corresponding to the desired stylist on screen example G22. In response to this operation, sales system 2 displays screen example G23 on the user terminal.

[0123] Screen example G23 is an example of a screen on which a user can select a styling menu available from a designated stylist. In this example, screen example G23 displays menus for cut, color, perm, and shampoo, check boxes for selecting each menu, and a confirm button. On screen example G23, the user operates the check boxes corresponding to one or more desired menus, and then operates the confirm button. In response to this operation, sales system 2 displays screen example G24 on the user terminal.

[0124] Screen example G24 shows an example of a screen on which a reservation slot can be selected. Screen example G24 displays reservation slot information G241, radio buttons for selecting a reservation slot, and a "Confirm Reservation" button. The reservation slot information G241 includes the remaining number of reservations available and nomination fee information (a combination of the nomination fee and the discounted price) for each of the 12 reservation slots. In this application example, the initial number of reservations available for each reservation slot is three. In FIG. 9, because some reservation slots have already been reserved, one of the remaining numbers from one to three is displayed as the number of reservations available. The nomination fee information for each reservation slot is based on the price information sequence determined by the information processing device 1A. The nomination fee information for each reservation slot is updated the next time new nomination fee information is determined. The user refers to the nomination fee information and selects the radio button corresponding to the desired reservation slot. In this example, a reservation slot from 14:00 to 16:00 on April 1 is selected. If the user wants to make a reservation for the selected reservation slot, they click the "Confirm Reservation" button. In response to this operation, the sales system 2 displays a screen example G25 on the user terminal.

[0125] Screen example G25 shows an example of a screen for confirming the details of a reservation. Screen example G25 displays the designated stylist, the time slot of the selected reservation slot, the selected menu and subtotal, the nomination fee, the discount amount, the sum of these prices, and a "Confirm reservation" button. The discount amount indicates the discount of the nomination fee due to the subtotal of the selected menu being equal to or greater than the discounted price. If the user wants to confirm the reservation with these details, they press the "Confirm reservation" button. In response to this operation, the sales system 2 executes a settlement process using the payment method set by the user and sends reservation information indicating the reservation details to the salon.

[0126] The operation of the information processing system 10 at each point in time in this modified example can be explained in much the same way as in the description of application example 1 with reference to Figures 10 and 11, by replacing the target date, delivery slot, shipping fee, product, and order with reservation slot, nomination fee, menu, and reservation, respectively, up to three days in the future. However, the difference is that each time the deadline for a reservation slot arrives, a new reservation slot for the next three days is added to the available reservation slots, dynamically pricing 12 reservation slots so that they are always available. For example, after 9:00 on April 1, reservations for the 10:00-12:00 slot on April 1 are closed, and a new reservation slot for the 10:00-12:00 slot on April 4 is made available.

[0127] According to this application example, the stylist nomination fee and nomination fee-free price for each appointment slot at a salon can be priced taking into account the estimated number of appointments accepted and the average gross profit. This allows salons or stylists to increase their total future profits up to three days into the future in cases where profits are not determined solely by the stylist nomination fee and the number of appointments accepted, but are also determined by the nomination fee-free price and the average gross profit per customer served by the stylist in each appointment slot. Furthermore, pricing is based on the constraint that the estimated number of appointments accepted for each appointment slot does not exceed the remaining number of appointments available, reducing the likelihood that each appointment slot will close early within the three-day period. As a result, salons and stylists can reduce opportunity losses due to customers' desired appointment slots already closing early. Furthermore, users can reserve their desired appointment slots with their desired stylists at a more appropriate nomination fee and reduce the likelihood that their desired appointment slots will close early, improving convenience.

[0128] [Other Modifications] This exemplary embodiment can also be applied to cases where the price of the first product is not set to be discounted or free when the amount spent on the second product is equal to or greater than a threshold. In such cases, this exemplary embodiment can be configured similarly and achieve similar effects by modifying it so that the price information is the price of the first product instead of the combination of the price of the first product and the threshold amount spent on the second product.

[0129] [Example of implementation by software] Some or all of the functions of the information processing device 1, 1A, sales system 2, and performance storage device 3 (hereinafter referred to as each device) may be implemented by hardware such as an integrated circuit (IC chip), or by software.

[0130] In the latter case, each device is realized, for example, by a computer that executes instructions of a program, which is software that realizes each function. An example of such a computer (hereinafter referred to as computer C) is shown in Figure 13. Computer C includes at least one processor C1 and at least one memory C2. Memory C2 stores a program P for operating computer C as each device. In computer C, processor C1 reads and executes program P from memory C2, thereby realizing the functions of each device.

[0131] The processor C1 may be, for example, a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), a micro processing unit (MPU), a floating point number processing unit (FPU), a physics processing unit (PPU), a tensor processing unit (TPU), a quantum processor, a microcontroller, or a combination thereof. The memory C2 may be, for example, a flash memory, a hard disk drive (HDD), a solid state drive (SSD), or a combination thereof.

[0132] The computer C may further include a RAM (Random Access Memory) for expanding the program P during execution and for temporarily storing various data. The computer C may also include a communication interface for transmitting and receiving data to and from other devices. The computer C may also include an input / output interface for connecting input / output devices such as a keyboard, a mouse, a display, and a printer.

[0133] The program P can also be recorded on a non-transitory, tangible recording medium M that can be read by the computer C. Such a recording medium M can be, for example, a tape, a disk, a card, a semiconductor memory, or a programmable logic circuit. The computer C can acquire the program P via such a recording medium M. The program P can also be transmitted via a transmission medium. Such a transmission medium can be, for example, a communication network or broadcast waves. The computer C can also acquire the program P via such a transmission medium.

[0134] [Additional Note 1] The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. For example, embodiments obtained by appropriately combining the technical means disclosed in the above-described embodiments are also included in the technical scope of the present invention.

[0135] [Additional Note 2] Part or all of the above-described embodiment can also be described as follows: However, the present invention is not limited to the following described aspects.

[0136] (Supplementary Note 1) An information processing device comprising: a first acquisition means for acquiring a demand function that, when price information for each of a plurality of first products to be presented, with time information set for each first product, and a presentation time when the price information is simultaneously presented, outputs an estimate of demand for each of the plurality of first products to be presented that will be generated in response to the presentation of this price information; a second acquisition means for acquiring an average gross profit function that, when price information for each of the plurality of first products to be presented and the presentation time are input, outputs an estimate of average gross profit to be obtained from the first product and a second product related to the first product, in accordance with demand for each of the plurality of first products to be presented that will be generated in response to the presentation of this price information; and a search means for determining price information for each of the first products to be presented, by using the demand function and the average gross profit function to search for price information for each of the first products to be presented that will increase the estimated value of demand for each of the first products to be presented and the total profit based on the estimated average gross profit when each of a plurality of times from the current time onwards is set as the presentation time.

[0137] (Supplementary Note 2) The information processing device described in Supplementary Note 1 further includes a presentation means for simultaneously presenting price information of the first product of each of the presentation targets determined by the search means to a user via a sales system that sells the first product of each of the presentation targets.

[0138] (Supplementary Note 3) An information processing device as described in Supplementary Note 1 or 2, wherein the price of the first product is set to be discounted or free when the amount spent on the second product is equal to or greater than a threshold, the price information includes a set of the price of the first product of each of the presented objects and the threshold value presented at the same time, and the search means determines the price information of the first product of each of the presented objects by searching for a set of the price and threshold value for each of the first products of the multiple presented objects that will increase the total profit.

[0139] (Supplementary Note 4) The information processing device described in any one of Supplementary Notes 1 to 3, wherein the first acquisition means acquires the demand function learned based on first performance data correlating price information for each of the plurality of first products to be presented, the time of presentation, and actual values ​​of demand for each of the first products observed in response to the presentation of this price information, and the second acquisition means acquires the average gross profit function learned based on second performance data correlating price information for each of the plurality of first products to be presented, the time of presentation, and actual values ​​of average gross profit obtained from the first product and the second product observed in response to the demand for each of the first products that arose in response to the presentation of this price information.

[0140] (Supplementary Note 5) The information processing device described in Supplementary Note 4, wherein the first performance data is data including price information for each of the plurality of first products to be presented determined by the search means, the time of presentation, and the performance value of demand for each of the first products observed in response to the presentation of this price information, and the second performance data is data including price information for each of the plurality of first products to be presented determined by the search means, the time of presentation, and the performance value of average gross profit obtained from the first product and the second product observed in response to the demand for each of the first products that arose in response to the presentation of this price information.

[0141] (Supplementary Note 6) The information processing device described in Supplementary Note 5, wherein the first acquisition means calculates a posterior distribution of parameters included in the demand function based on the first performance data, and acquires the demand function based on parameters sampled by the posterior distribution, and the second acquisition means calculates a posterior distribution of parameters included in the average gross profit function based on the second performance data, and acquires the average gross profit function based on parameters sampled by the posterior distribution.

[0142] (Supplementary Note 7) An information processing method executed by at least one processor, comprising: when price information for each of a plurality of first products to be presented, with time information set for each first product, and a presentation time when the price information is to be presented simultaneously, obtaining a demand function that outputs an estimate of demand for each of the plurality of first products to be presented that will be generated in response to the presentation of this price information; when price information for each of the plurality of first products to be presented and the presentation time are input, obtaining an average gross profit function that outputs an estimate of average gross profit to be obtained from the first product and a second product related to the first product, in accordance with the demand for each of the plurality of first products to be presented that will be generated in response to the presentation of this price information; and using the demand function and the average gross profit function, determining price information for each of the first products to be presented that will increase the total profit based on the estimated demand and the estimated average gross profit when each of a plurality of times from the current time onwards is set as the presentation time.

[0143] (Supplementary Note 8) A program for causing a computer to function as an information processing device, the program causing the computer to function as: a first acquisition means for acquiring a demand function that, when price information for each of a plurality of first products to be presented, with time information set for each first product, and a presentation time when the price information is simultaneously presented, outputs an estimate of demand for each of the plurality of first products to be presented that will be generated in response to the presentation of this price information; a second acquisition means for acquiring an average gross profit function that, when price information for each of the plurality of first products to be presented and the presentation time are input, outputs an estimate of average gross profit to be obtained from the first product and a second product related to the first product, in accordance with demand for each of the plurality of first products to be presented that will be generated in response to the presentation of this price information; and a search means for determining price information for each of the first products to be presented, by using the demand function and the average gross profit function to search for price information for each of the first products to be presented that will increase the total profit based on the estimated demand and the estimated average gross profit when each of a plurality of times from the current time onwards is set as the presentation time.

[0144] [Additional Note 3] Part or all of the above-described embodiment can also be expressed as follows.

[0145] (Supplementary Note 1) An information processing device comprising at least one processor, wherein the at least one processor executes: a first acquisition process for acquiring a demand function that, when price information for each of a plurality of first products to be presented, with time information set for each first product, and a presentation time when the price information is to be simultaneously presented, outputs an estimate of demand for each of the plurality of first products to be presented that will be generated in response to the presentation of this price information; a second acquisition process for acquiring an average gross profit function that, when price information for each of the plurality of first products to be presented and the presentation time are input, outputs an estimate of average gross profit to be obtained from the first product and a second product related to the first product, in accordance with demand for each of the plurality of first products to be presented that will be generated in response to the presentation of this price information; and a search process for determining price information for each of the first products to be presented, by using the demand function and the average gross profit function to search for price information for each of the first products to be presented that will increase the total profit based on the estimated demand and the estimated average gross profit when each of a plurality of times from the current time onwards is set as the presentation time.

[0146] (Supplementary Note 2) The information processing device described in Supplementary Note 1, wherein the at least one processor further executes a presentation process of simultaneously presenting price information of the first product of each of the presentation targets determined by the search process to a user via a sales system that sells the first product of each of the presentation targets.

[0147] (Supplementary Note 3) An information processing device as described in Supplementary Note 1 or 2, wherein the price of the first product is set to be discounted or free when the amount spent on the second product is equal to or greater than a threshold, the price information includes a set of the price of the first product of each of the presented objects and the threshold value presented at the same time, and in the search process, the at least one processor determines the price information of the first product of each of the presented objects by searching for a set of the price and threshold value for each of the first products of the multiple presented objects that increases the total profit.

[0148] (Supplementary Note 4) In the first acquisition process, the at least one processor acquires the demand function learned based on first performance data correlating price information for each of the plurality of first products to be presented, the time of presentation, and actual values ​​of demand for each of the first products observed in response to the presentation of this price information; and in the second acquisition process, the at least one processor acquires the average gross profit function learned based on second performance data correlating price information for each of the plurality of first products to be presented, the time of presentation, and actual values ​​of average gross profit obtained from the first product and the second product observed in response to the demand for each of the first products that arose in response to the presentation of this price information. An information processing device described in any one of Supplementary Notes 1 to 3.

[0149] (Supplementary Note 5) The information processing device described in Supplementary Note 4, wherein the first performance data is data including price information for each of the plurality of first products to be presented determined in the search process, the time of the presentation, and the performance value of demand for each of the first products observed in response to the presentation of this price information; and the second performance data is data including price information for each of the plurality of first products to be presented determined in the search process associated with the time of the presentation, and the performance value of average gross profit obtained from the first product and the second product observed in response to the demand for each of the first products that arose in response to the presentation of this price information.

[0150] (Supplementary Note 6) The information processing device described in Supplementary Note 5, wherein in the first acquisition process, the at least one processor calculates a posterior distribution of parameters included in the demand function based on the first performance data, and acquires the demand function based on parameters sampled by the posterior distribution; and in the second acquisition process, the at least one processor calculates a posterior distribution of parameters included in the average gross profit function based on the second performance data, and acquires the average gross profit function based on parameters sampled by the posterior distribution.

[0151] The information processing device may further include a memory that stores a program for causing the at least one processor to execute the first acquisition process, the second acquisition process, and the search process. The program may be recorded on a computer-readable, non-transitory, tangible recording medium.

[0152] REFERENCE SIGNS LIST 1, 1A Information processing device 10 Information processing system 2 Sales system 3 Performance record storage device 11 First acquisition unit 12 Second acquisition unit 13 Search unit 14 Presentation unit 110 Control unit 120 Storage unit 130 Communication unit C1 Processor C2 Memory

Claims

1. a first acquisition means for acquiring a demand function that outputs an estimated value of demand for each of the plurality of first products to be presented, which is generated in response to the presentation of price information, when price information for each of the plurality of first products to be presented, with time information set for each first product, and a presentation time when the price information is to be presented simultaneously, are input; a second acquisition means for acquiring an average gross profit function that outputs an estimated value of average gross profit to be obtained from the first products and second products related to the first products in response to the demand for each of the first products that is generated in response to the presentation of the price information when price information for each of the first products that are to be presented and the presentation time are input; a search means for determining price information of the first commodity of each of the presented objects by searching for price information of the first commodity of each of the presented objects that maximizes the total profit based on the estimated demand value of the first commodity of each of the presented objects and the estimated average gross profit value when each of a plurality of points in time from the current point in time is set as the presentation point in time using the demand function and the average gross profit function; An information processing device comprising:

2. a presentation means for simultaneously presenting price information of each of the first commodities to be presented determined by the search means to a user via a sales system that sells each of the first commodities to be presented; The information processing device according to claim 1 , further comprising:

3. The price of the first product is set to be discounted or free when the amount of use of the second product is equal to or greater than a threshold value; The price information includes a price of each of the first products to be presented and a set of the thresholds presented at the same time as the price; the search means determines price information of the first products of the plurality of presentation targets by searching for a combination of a price and a threshold value for each of the first products of the plurality of presentation targets that increases the total profit; 3. The information processing device according to claim 1.

4. The first acquisition means acquiring the demand function learned based on first performance data that associates price information of each of the plurality of first products to be presented, the presentation time, and performance values ​​of demand for each of the first products observed in response to the presentation of the price information; The second acquisition means acquire the average gross profit function learned based on second performance data that associates price information of each of the plurality of first products to be presented, the presentation time, and performance values ​​of average gross profits obtained from the first products and the second products observed in accordance with demand for each of the first products that has arisen in response to the presentation of the price information; 3. The information processing device according to claim 1.

5. the first performance data includes price information of each of the plurality of first products to be presented determined by the search means, the presentation time, and a performance value of demand for each of the first products observed in response to the presentation of the price information; The second performance data includes price information of each of the plurality of first products to be presented determined by the search means, and performance values ​​of average gross profits obtained from the first products and the second products observed at the time of presentation and in accordance with demand for each of the first products that arose in response to the presentation of the price information. The information processing device according to claim 4 .

6. the first obtaining means calculates a posterior distribution of parameters included in the demand function based on the first performance data, and obtains the demand function based on parameters sampled by the posterior distribution; the second obtaining means calculates a posterior distribution of parameters included in the average gross profit function based on the second performance data, and obtains the average gross profit function based on parameters sampled by the posterior distribution. The information processing device according to claim 5 .

7. 1. An information processing method executed by at least one processor, comprising: When price information for each of a plurality of first products to be presented, each of which has time information set for that first product, and a presentation time point at which the price information is to be simultaneously presented are input, a demand function is obtained that outputs an estimated value of demand for each of the plurality of first products to be presented that will be generated in response to the presentation of the price information; When price information for each of the plurality of first products to be presented and the presentation time are input, an average gross profit function is obtained that outputs an estimated value of average gross profit to be obtained from the first products and second products related to the first products, in accordance with demand for each of the plurality of first products to be presented that is generated in response to the presentation of the price information; determining price information for the first commodity of each of the presented objects by searching for price information for the first commodity of each of the presented objects that maximizes the total profit based on the estimated demand value of the first commodity of each of the presented objects and the estimated average gross profit value when each of a plurality of points in time from the current point in time is set as the presentation point in time using the demand function and the average gross profit function; An information processing method, including:

8. A program for causing a computer to function as an information processing device, the program comprising: a first acquisition means for acquiring a demand function that outputs an estimated value of demand for each of the plurality of first products to be presented, which is generated in response to the presentation of price information, when price information for each of the plurality of first products to be presented, with time information set for each first product, and a presentation time when the price information is to be presented simultaneously, are input; a second acquisition means for acquiring an average gross profit function that outputs an estimated value of average gross profit to be obtained from the first products and second products related to the first products in response to the demand for each of the first products that is generated in response to the presentation of the price information when price information for each of the first products that are to be presented and the presentation time are input; a search means for determining price information of the first commodity of each of the presented objects by searching for price information of the first commodity of each of the presented objects that maximizes the total profit based on the estimated demand value of the first commodity of each of the presented objects and the estimated average gross profit value when each of a plurality of points in time from the current point in time is set as the presentation point in time using the demand function and the average gross profit function; A program that functions as a