JP2018005563A - Processing device, processing method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technology for planning a bargain sale content.SOLUTION: The present invention provides a processing device 10 comprising a calculation unit 11 that calculates at least one of an optimum solution of a selling price of one commodity, the optimum solution of a combination of selling prices of a plurality of commodities, the optimum solution of the commodity sold by the different price from the normal price, and the optimum solution of the selling price of the commodity sold by the different price from the normal price, using an estimation model obtained based on sales data of a commodity and selling time condition data indicating a condition in selling time, and an output unit 12 that outputs a calculation result of the calculation unit 11.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a processing apparatus, a processing method, and a program.

  Patent Document 1 discloses a technique for calculating a predicted number of sales during a special sale period from days of the week, special events, weather, special sale information (sale price, etc.), and determining an appropriate number of orders from the result.

  Patent Document 2 discloses a technique for quantitatively analyzing the degree of influence that a sale price setting plan has on peripheral products after a sale period.

  Non-Patent Document 1 discloses a technique (heterogeneous mixed learning technique) related to a model (heterogeneous mixed prediction model) in which data is classified according to a decision tree format rule and is predicted by a linear model in which different explanatory variables are combined in each case. Has been.

JP 2002-24350 A Japanese Patent Laying-Open No. 2015-95062

Junpei Fujimaki, 4 others, “Research and development of heterogeneous mixed learning technology and big data analysis solution”, [online], 29th (2015) Advanced Technology Grand Prize winner introduction page, [2016 June Search 10th of March], Internet <URL, http://www.fbi-award.jp/sentan/jusyou/2015/7.pdf>

  Properly planning the contents of special sales (eg, special sales products, special sales prices, etc.) is expected to increase sales during the special sales period. Neither Patent Literature 1 nor 2 is a technique for appropriately planning the contents of special sales. This invention makes it a subject to provide the technique for planning the content of special sale.

According to the present invention,
Using the estimation model obtained based on the sales data of the product and the sales condition data indicating the conditions at the time of sales, the optimal solution for the sales price of one product, the optimal solution for the combination of the sales prices of multiple products, the normal price A calculation means for calculating at least one of an optimal solution for a product sold at a price different from the above and an optimal solution for a sales price of a product sold at a price different from the normal price;
Output means for outputting a calculation result of the calculation means;
Is provided.

Moreover, according to the present invention,
Computer
Using the estimation model obtained based on the sales data of the product and the sales condition data indicating the conditions at the time of sales, the optimal solution for the sales price of one product, the optimal solution for the combination of the sales prices of multiple products, the normal price A calculation step for calculating at least one of an optimal solution for a product sold at a price different from the above and an optimal solution for a sales price of a product sold at a price different from the normal price;
An output step of outputting a calculation result in the calculation step;
Is provided.

Moreover, according to the present invention,
Computer
Using the estimation model obtained based on the sales data of the product and the sales condition data indicating the conditions at the time of sales, the optimal solution for the sales price of one product, the optimal solution for the combination of the sales prices of multiple products, the normal price A calculation means for calculating at least one of an optimal solution for a product sold at a price different from the above and an optimal solution for a sales price of a product sold at a price different from the normal price;
Output means for outputting a calculation result of the calculation means;
A program is provided that functions as:

  According to the present invention, a technique for planning the contents of a sale is realized.

It is a figure which shows notionally an example of the hardware constitutions of the apparatus of this embodiment. It is an example of the functional block diagram of the processing apparatus of this embodiment. It is a figure for demonstrating an example of the estimation model of this embodiment. It is a figure for demonstrating an example of the prediction formula of this embodiment. It is a flowchart which shows an example of the flow of a process of the processing apparatus of this embodiment.

  First, an example of the hardware configuration of the apparatus (processing apparatus) of this embodiment will be described. Each unit included in the apparatus according to the present embodiment is stored in a CPU (Central Processing Unit) of any computer, a memory, a program loaded into the memory, a storage unit such as a hard disk for storing the program (from the stage of shipping the apparatus in advance). In addition to existing programs, it can store programs downloaded from CDs (Compact Discs) and other recording media, servers on the Internet, etc.) and realized by any combination of hardware and software, centering on network connection interfaces Is done. It will be understood by those skilled in the art that there are various modifications to the implementation method and apparatus.

  FIG. 1 is a block diagram illustrating a hardware configuration of an apparatus according to this embodiment. As shown in FIG. 1, the apparatus includes a processor 1A, a memory 2A, an input / output interface 3A, a peripheral circuit 4A, and a bus 5A. The peripheral circuit 4A includes various modules. Note that the device may not have the peripheral circuit 4A.

  The bus 5A is a data transmission path through which the processor 1A, the memory 2A, the peripheral circuit 4A, and the input / output interface 3A transmit / receive data to / from each other. The processor 1A is an arithmetic processing device such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit). The memory 2A is a memory such as a RAM (Random Access Memory) or a ROM (Read Only Memory). The input / output interface 3A is an interface for acquiring information from an input device (eg, keyboard, mouse, microphone, etc.), external device, external server, external sensor, etc., and an output device (eg, display, speaker, printer, mailer). Etc.), an interface for outputting information to an external device, an external server, etc. The processor 1A can issue a command to each module and perform a calculation based on the calculation result.

  Hereinafter, this embodiment will be described. Note that the functional block diagram used in the following description of the embodiment shows functional unit blocks rather than hardware unit configurations. In these drawings, each device is described as being realized by one device, but the means for realizing it is not limited to this. That is, it may be a physically separated configuration or a logically separated configuration. In addition, the same code | symbol is attached | subjected to the same component and description is abbreviate | omitted suitably.

  First, the outline | summary of the processing apparatus of this embodiment is demonstrated. The processing apparatus according to the present embodiment uses a presumed evaluation value obtained by machine learning based on sales condition data indicating sales data of a product and sales conditions (eg, sales price, climate, etc.). Calculate and output the optimal solution that maximizes. The sales condition data will be described later in detail in the present embodiment.

  The evaluation value is the first evaluation value indicated based on the sales and / or profit of the entire store that sells the product, and the sales of a part of the product group sold at the store (for example, the product group of a certain manufacturer) And the 2nd evaluation value shown based on profit and / or the 3rd evaluation value calculated based on the 1st evaluation value and the 2nd evaluation value, etc. are illustrated.

  The optimal solution calculated is the optimal solution for the sales price of one product, the optimal solution for the combination of the sales prices of multiple products, the optimal solution for products sold at a price different from the normal price, and the price different from the normal price. The optimal solution of the selling price of the product to be sold is exemplified.

  Based on the content of the optimum solution output from the processing device, the operator can select a product for sale, set a sale price, and the like.

  Next, the configuration of the processing apparatus of this embodiment will be described in detail. FIG. 2 shows an example of a functional block diagram of the processing apparatus 10 of the present embodiment. As illustrated, the processing apparatus 10 includes a calculation unit 11, an output unit 12, and an input reception unit 13.

  The input receiving unit 13 receives various inputs. The input receiving unit 13 inputs data stored in an external device, inputs values made by an operator operating various input devices (eg, keyboard, mouse, microphone, touch panel display, operation buttons, etc.), etc. Accept. Details of the input received by the input receiving unit 13 will be described below as appropriate.

  The calculation unit 11 includes an estimated model obtained by machine learning based on sales data and sales condition data indicating sales conditions, sales condition data and / or predicted values of sales condition data in a target period, and Is used to calculate an optimal solution that maximizes a predetermined evaluation value. Details will be described below.

  First, the “evaluation value” will be described. The evaluation value is the first evaluation value indicated based on the sales and / or profit (eg, gross profit) of the entire store that sells the product, the sales and / or the entire product group that is sold in the store The second evaluation value shown based on the profit (eg, gross profit) or the third evaluation value calculated based on the first evaluation value and the second evaluation value is exemplified.

  The first evaluation value is an index indicating sales in the entire store where the product is sold. The higher the value, the better the sales. The first evaluation value is calculated based on sales and / or profits in the entire store where the product is sold. The first evaluation value may be, for example, sales or profits in the entire store during the target period. Alternatively, the first evaluation value is a value obtained by converting the sales and profit of the entire store during the target period into points based on the respective values and integrating them (for example, adding together). Also good. Or, the first evaluation value is the sales or profit of the entire store during the comparison period (eg, the week immediately before the period, the month immediately before, the period of the month corresponding to the period last year, etc.) (Example: Daily average value) Increase rate of sales or profits in the entire store during the target period (Example: Daily average value) and the above integrated value (Example: Daily average value) It may be an increase rate. In addition, the specific example of the 1st evaluation value illustrated here is an example to the last, and is not limited to these.

  The second evaluation value is an index indicating the sales of a part of the entire product group sold in the store. The higher the value, the better the sales. The second evaluation value is calculated based on sales and / or profits of a part of the entire product group sold at the store. The second evaluation value may be, for example, the sales or profit itself of “a part of the entire product group” during the target period. Alternatively, the second evaluation value is obtained by converting each of sales and profits of “a part of the entire product group” during the target period into points on the basis of each value and integrating them (eg, adding them together) ). Or, the second evaluation value is “a part of the entire product group during the comparison target period (eg, one week immediately before the target period, one month immediately before, the period corresponding to the target period of the previous year, etc.)” ”Sales or profit (example: average value for one day), increase rate of sales or profit (example: average value for one day) in“ partial product group ”during the target period, or the above integration It may be an increase rate of the value (for example: average value of one day). In addition, the specific example of the 2nd evaluation value illustrated here is an example to the last, and is not limited to these.

  The third evaluation value is an index indicating the sales in the entire store where the product is sold and the sales in the entire partial product group sold in the store. The higher the value, the better the sales. The third evaluation value is calculated based on the first evaluation value and the second evaluation value. As the first evaluation value increases, the third evaluation value increases. In addition, the third evaluation value increases as the second evaluation value increases.

  For example, each of the first evaluation value and the second evaluation value is converted into a point of 0 to 5 according to the value (the larger the value, the larger the point), and the first evaluation value and the second evaluation value A total point obtained by adding the points of the values may be set as the third evaluation value. In addition, the sum of a value obtained by multiplying the point of the first evaluation value by the weighting factor of the first evaluation value and a value obtained by multiplying the point of the second evaluation value by the weighting factor of the second evaluation value. The point may be a third evaluation value.

  The weighting coefficient for each of the first evaluation value and the second evaluation value may be determined in advance or may be determined by the user. The conversion rule for converting the first evaluation value into points and the conversion rule for converting the second evaluation value into points may be the same or different. The above-described 6-stage points of 0 to 5 are merely examples, and the present invention is not limited to this. Further, the specific example of the third evaluation value illustrated here is merely an example, and the present invention is not limited to this.

  For example, the input reception unit 13 may receive a selection from the user as to which of the first to third evaluation values to calculate the optimum solution that maximizes. Then, the calculation unit 11 may calculate an optimal solution that maximizes the evaluation value selected by the user.

  Further, when the input receiving unit 13 receives an input for selecting the second evaluation value, the input receiving unit 13 further receives an input (second input) for selecting a part of the product group from the product group sold in the store. May be accepted. Then, the calculation unit 11 may calculate an optimal solution that maximizes the second evaluation value indicated based on the sales and / or profits of the entire partial product group selected by the second input. Or the input reception part 13 may receive further the input which selects a specific goods from the goods group sold at a store. More specifically, the calculation unit 11 calculates an optimal solution that maximizes the second evaluation value indicated based on the sales and / or profit of the entire product related to the specific product received by the input receiving unit 13. May be. The product related to the specific product is, for example, another product different from the specific product (belonging to the product category) included in the product category to which the specific product belongs, or a product related to the specific product (specific A product similar to a product, another product handled by a manufacturer that sells (manufactures) a specific product, a product purchased together with the specific product, and the like are represented. In that case, the calculation part 11 may hold | maintain previously information, such as the information regarding the goods category and the goods which belong to the goods category as an example.

  In addition, when receiving an input for selecting the third evaluation value, the input receiving unit 13 further selects an evaluation value to be emphasized from the first evaluation value and the second evaluation value, or the first evaluation value An input (third input) for designating the ratio to be emphasized in the evaluation value and the second evaluation value may be received. And the calculation part 11 may calculate a 3rd evaluation value based on the content selected or designated by the 3rd input. In this case, for example, the calculation unit 11 uses the weight ratio of the first evaluation value and the second evaluation value as the weighted ratio specified for each of the first evaluation value and the second evaluation value. A third evaluation value may be calculated as a weighting coefficient for the evaluation value.

  In addition, the calculation unit 11 may hold in advance a weighting factor that is important and a weighting factor that is not important. Then, the calculation unit 11 applies the weighting factor of the emphasis that is held in advance as the weighting factor of the one (first evaluation value or the second evaluation value) selected to be emphasizing, and is selected when emphasizing. The third evaluation value may be calculated by applying a weighting factor that is not emphasized in advance as the weighting factor of the one that does not exist (first evaluation value or second evaluation value).

  Next, the “optimal solution” will be described. The optimal solution is a solution that maximizes the evaluation value described above. The optimal solution calculated by the calculation unit 11 is the optimal solution for the sales price of one product, the optimal solution for the combination of the sales prices of a plurality of products, the optimal solution for products sold at a price different from the normal price, and the normal price The optimal solution etc. of the sales price of the goods sold at different prices are exemplified.

  For example, the input receiving unit 13 may receive an input (fourth input) that specifies a part (one or a plurality) or all of the products sold in the store. And the calculation part 11 may calculate the optimal solution of the sales price of one goods designated by the 4th input. In addition, the calculation unit 11 may calculate an optimal solution for a combination of sales prices of a plurality of products specified by the fourth input. In addition, the calculation unit 11 may determine an optimal solution for a product to be sold at a price different from the normal price from among a plurality of products specified by the fourth input. In addition, the calculation unit 11 may determine a product to be sold at a price different from the normal price from among a plurality of products specified in the fourth input, and may calculate an optimal solution for the determined sales price of the product. .

  Next, “sales condition data” will be described. The sales condition data is data associated with the sales data of the product (including information such as the sales date, sales quantity, sales price, etc.), and indicates the sales conditions. Sales condition data includes, for example, product information, calendar data, weather data, information indicating past sales trends, event information of stores selling products, event information held in the vicinity of stores, and competing stores related to competing stores Contains at least one of the information. However, the sales condition data is not limited to the above-described example. The sales condition data may include various information as long as the information affects the evaluation value or the optimum solution.

  The merchandise information includes, for example, a sales price and a purchase price at the time of sale, a manufacturer, a product category, a normal price (a normal sales price), a normal purchase price (a normal purchase price), and the like. The calendar data includes, for example, information such as whether or not it falls on a sales day or a public holiday. The weather data includes, for example, the sales day temperature (average temperature, maximum temperature, minimum temperature, etc.), weather (sunny, rain, cloudy, snow, etc.), humidity, precipitation probability, and the like. The information indicating the past sales trend includes, for example, information such as a sales trend accompanying seasonal variation. The event information of the store includes, for example, information on events (for example, “Wine Ban Fair”, “Winter / Nabe Festival”, etc.) that are performed at the store on the sales day. The competitive store information regarding the competitive store includes, for example, the sales price of the product of the competitive store on the sales date, advertisement information (advertised product and its price), event information and the like. The event information held in the vicinity of the store includes, for example, information on events (for example, “concert”, “athletic meet”, etc.) held around the store on the sales day. Note that the specific examples of the product information, calendar data, weather data, past sales tendency information, event information, and competitor store information illustrated here are merely examples, and are not limited thereto.

  Next, the “estimated model” will be described. The estimation model is a model for estimating a predetermined value obtained by machine learning based on the sales data of the product and the condition data at the time of sale. The estimation target (predetermined value to be estimated) may be any of the first to third evaluation values described above, or a value for calculating the first to third evaluation values. Also good. As an example of a value for calculating the first to third evaluation values, for example, the number of sales of each product during the target period can be cited. The first to third evaluation values can be calculated based on the estimated sales volume of each product.

  For example, the target value (past performance) obtained from sales data is used as an objective variable, and the sales condition data (the product and sales price, calendar data, weather data, An estimation model can be generated by machine learning using information indicating past sales trends, event information, competitive store information, and the like) as explanatory variables.

  The details of machine learning are design matters, but it is preferable to generate a heterogeneous mixture prediction model (estimated model) using the heterogeneous mixture learning technique disclosed in Non-Patent Document 1. According to the heterogeneous mixed learning technique, it is possible to generate a simple and highly accurate estimation model suitable for each condition from a huge pattern of combinations of data cases and explanatory variables.

  FIG. 3 shows the concept of the estimation model obtained by the heterogeneous mixed learning technique. As shown in the figure, a case is classified based on the value of the condition data at the time of sale, and a prediction formula for estimating a predetermined value is generated by combining optimal explanatory variables corresponding to each case. The combination of explanatory variables can be different in each case.

  FIG. 4 shows an example of the prediction formula obtained by the heterogeneous mixed learning technique. As shown in the figure, the prediction formula is expressed as a sum of products of various sales time condition data values and various sales time condition data coefficients.

  According to the prediction formula, it is possible to grasp the condition data at the time of sale that affects the value of the estimation target (eg, first to third evaluation values, values for calculating the first to third evaluation values). it can. That is, it is understood that the sales condition data included in the prediction formula (in the case of FIG. 4, temperature, precipitation, etc.) affects the estimation target value.

  Further, according to the prediction formula, it is possible to grasp the influence contents of the sales condition data. That is, it can be seen that the condition data at the time of sale having a positive coefficient has a larger estimation target value as the value increases. Further, it can be seen that the condition data at the time of sale having a negative coefficient becomes smaller as the value increases.

  The calculation unit 11 may generate an estimation model. In addition, the calculation unit 11 may acquire an estimation model generated by an external system.

  Next, the “sales condition data during the target period and / or the predicted value of the sales condition data” will be described. The target period is, for example, a period during which special sales of products are performed. The target period is, for example, a future period. The input receiving unit 13 can receive an input specifying a target period (start date and end date) from the user. However, the input reception part 13 may receive the input which designates a specific day as a target period from a user.

  Moreover, the input reception part 13 can receive the input of the forecast value of the sales time condition data and / or the sales time condition data in the target period.

  For example, the input reception unit 13 may request a predicted value of weather data in a target period from a predetermined server that provides weather information, and acquire data transmitted in response thereto. In addition, the user may input the predicted value of the weather data in the target period to the processing device 10.

  In addition, the input reception unit 13 may request calendar data for a target period from a predetermined server and acquire data transmitted in response thereto. In addition, the user may input calendar data in the target period to the processing device 10.

  In addition, the input reception part 13 may request | require the event information of the store in a target period from a predetermined | prescribed server (Example: management server which memorize | stores event information), and may acquire the data transmitted according to it. In addition, the user may input store event information in the processing period 10 in the target period.

  In addition, the input reception unit 13 may request competing store information in a target period from a predetermined server (eg, a management server that stores competing store information), and may acquire data transmitted in response thereto. In addition, the user may input competitive store information in the target period into the processing device 10.

  In addition, the input reception unit 13 requests event information in a target period from a predetermined server (eg, a management server that stores event information held in the vicinity of a store), and acquires data transmitted in response thereto. May be. In addition, the user may input event information around the store in the target period to the processing device 10.

  In addition, the input receiving unit 13 may receive from the user an input (first input) that specifies candidates for selling prices of a plurality of products in the target period. For example, an input designating a plurality of values such as “90 yen, 95 yen, 100 yen” may be accepted, or an input designating a range of values such as “90 yen or more and 95 yen or less” may be accepted. May be. In the latter case, all integers included in the range are specified. That is, when “90 yen or more and 95 yen or less” is designated, “90 yen, 91 yen, 92 yen, 93 yen, 94 yen, 95 yen” is designated.

  The input receiving unit 13 may accept an input for designating a selling price candidate corresponding to all of a plurality of products sold in the store, or designating a selling price candidate corresponding to some products. Input may be accepted. In the latter case, the calculation unit 11 performs processing for calculating an optimal solution on the assumption that a product for which a sales price candidate is not designated is sold at a normal price. For example, the input receiving unit 13 requests product information including the normal price of the product from a predetermined server (eg, a management server that stores product information including the normal price of the product), and data transmitted in response thereto. May be obtained.

  Next, “a process in which the calculation unit 11 calculates an optimal solution” will be described. First, the calculation unit 11 generates a plurality of condition patterns having different sales condition data and / or prediction values of the sales condition data for the target period. Specifically, the calculation unit 11 generates a plurality of condition patterns in which the sales price of at least one product is different from each other based on the sales price candidates of the one or more products specified in the first input. For example, when five selling price candidates are specified corresponding to only one product, five condition patterns are generated. In addition, when five and three selling price candidates are specified corresponding to two products, 15 condition patterns are generated. Note that the sales condition data other than the sales price of a product for which a plurality of sales price candidates are designated and / or the predicted values of the sales condition data may be the same.

  Then, the calculation unit 11 uses the estimation model and each of the plurality of condition patterns to calculate an estimation target in the target period (eg, first to third evaluation values, first to third evaluation values). Value) is calculated. Thereafter, the calculation unit 11 specifies a condition pattern that maximizes the evaluation value (eg, first to third evaluation values) specified based on the estimated value, and sets the specified condition pattern (hereinafter referred to as an optimal condition pattern). The optimal solution is determined based on this.

  For example, the calculation unit 11 sells a predetermined product (for example, a product for which a plurality of sales price candidates are specified in the first input) indicated by the optimum condition pattern, or a predetermined plurality of products (for example, : A combination of sales prices of a plurality of products for which a plurality of sales price candidates have been designated in the first input may be determined as an optimal solution. Alternatively, the calculation unit 11 may determine one or a plurality of products whose sales price indicated by the optimum condition pattern is different from the normal price as an optimal solution for the product sold at a price different from the normal price. Or the calculation part 11 may determine the sales price shown by the optimal condition pattern of the goods sold with the price different from normal price as the optimal solution of the sales price of the goods.

  Returning to FIG. 2, the output unit 12 outputs the calculation result of the calculation unit 11. Specifically, the output unit 12 outputs the optimal solution calculated by the calculation unit 11. In addition, the output unit 12 may further output the first evaluation value, the second evaluation value, or the third evaluation value calculated based on the optimum condition pattern. Moreover, when the heterogeneous mixed prediction model mentioned above is used as an estimation model, the output part 12 may further output the content grasped | ascertained based on a prediction formula (refer FIG. 4). The contents grasped based on the prediction formula (see FIG. 4) are as described above.

  The output method by the output unit 12 is not limited, and can be realized via any output device such as a display, a projection device, a printer, and a mailer.

  The first input described above and the fourth input described above may be the same input. In other words, the input receiving unit 13 may receive, as the first input and the fourth input, an input that specifies sales price candidates corresponding to some or all of the products. The input receiving unit 13 receives a product for which a selling price candidate is designated as a product designated by the fourth input.

  The first input described above and the second input described above may be the same input. That is, the input receiving unit 13 may receive an input for designating a selling price candidate corresponding to a part of the products as the first input and the second input. The input receiving unit 13 receives a product for which a sales price candidate is designated as a product designated by the second input.

  Further, the second input described above and the fourth input described above may be the same input. That is, the input receiving unit 13 may receive an input for designating a part of products sold at the store as the second input and the fourth input.

  Next, an example of a processing flow of the processing apparatus 10 according to the present embodiment will be described with reference to the flowchart of FIG.

  In the input receiving step S10, the input receiving unit 13 receives various inputs. For example, the input receiving unit 13 receives input of the sales condition data and / or the predicted value of the sales condition data for the target period. The input receiving unit 13 may receive at least one of the first to fourth inputs described above.

  In the calculation step S11, the calculation unit 11 calculates an optimal solution that maximizes a predetermined evaluation value using the above-described estimation model and the sales condition data and / or the predicted value of the sales condition data for the target period. To do.

  In the output step S12, the output unit 12 outputs the calculation result of the calculation unit 11.

  Next, the effect of this embodiment is demonstrated. The processing apparatus 10 of the present embodiment can calculate and output an optimal solution that maximizes a predetermined evaluation value.

  For example, the processing apparatus 10 of the present embodiment can calculate and output an optimal solution that maximizes the first evaluation value indicated based on the sales and / or profits of the entire store that sells the product. The person in charge of the store planning the sale at the store selects the product for sale and sets the sale price based on the optimal solution that maximizes the first evaluation value. / Or profits can be improved.

  In addition, the processing apparatus 10 of the present embodiment calculates and outputs an optimal solution that maximizes the second evaluation value indicated based on the sales and / or profits of a part of the entire product group sold in the store. be able to. The person in charge of the manufacturer proposing the special sale at the store proposes to the store the selection of the special sale product or the setting of the special sale price based on the optimal solution that maximizes the second evaluation value regarding the company's product group. By this, sales and / or profits of the entire product can be improved. In addition, the person in charge of a store who wants to increase sales of some product groups (for example, a product group of a certain category) in the products handled in the store strategically obtains a second evaluation value for the product group. By selecting a product for sale or setting a sale price based on the optimal solution to be maximized, sales and / or profits of the entire partial product group can be improved by the sale.

  In addition, the processing apparatus 10 of the present embodiment can calculate and output an optimal solution that maximizes the third evaluation value calculated based on the first evaluation value and the second evaluation value. The person in charge of the manufacturer proposing the special sale at the store selects or sells the product for sale based on the optimum solution that maximizes the third evaluation value based on the first evaluation value and the second evaluation value relating to the company's own product group. By proposing the setting of prices to the store, sales and / or profit in the entire store and sales and / or profit in the entire company product can be improved in a balanced manner through special sales.

  Thus, according to the processing apparatus 10 of this embodiment, it becomes possible to determine the optimal sale content.

  Moreover, according to the processing apparatus 10 of this embodiment, the 1st input which designates the sales price candidate of each of several goods can be received. By specifying a plurality of sales price candidates only for the sale candidate product group, it is possible to narrow down the products for sale from the sale candidate product group. That is, it is possible to avoid the inconvenience of obtaining an unfavorable solution of selling a product other than a sale candidate. In addition, by designating the selling price candidates in advance, it is possible to avoid the inconvenience that a difficult selling price (eg, cheaper than the purchase price) is calculated as an optimal solution.

  Moreover, according to the processing apparatus 10 of this embodiment, the 2nd input which designates a part among goods sold in a store can be received. Then, the second evaluation value can be indicated based on the sales and / or profits of the entire part of the product group designated by the second input. Thus, according to the present embodiment in which the user can specify a product group related to the second evaluation value, the range of contents that can be simulated is widened, which is preferable.

  In addition, according to the processing device 10 of the present embodiment, in the process of calculating the optimum solution that maximizes the third evaluation value, the evaluation that is emphasized among the first evaluation value and the second evaluation value. A third input can be accepted that selects a value or specifies a ratio to be emphasized in the first evaluation value and the second evaluation value. The third evaluation value can be calculated based on the content selected or designated by the third input. As described above, according to the present embodiment in which the user can specify the design contents in the third evaluation value, the range of contents that can be simulated is widened.

  Moreover, according to the processing apparatus 10 of this embodiment, the 4th input which designates a part or all out of the goods sold in a store can be received. Then, the optimal solution for the sales price of the one product specified by the fourth input, the optimal solution for the combination of the sales prices of the plurality of products specified by the fourth input, and the plurality of solutions specified by the fourth input The optimal solution for the process of selecting a product to be sold at a price different from the normal price from among the products, and the product sold at a price different from the normal price selected from the plurality of products specified in the fourth input At least one of the optimal solutions for the selling price can be calculated. For example, the user can obtain an optimal solution related to the special sale candidate by designating the special sale candidate with the fourth input. That is, it is possible to avoid the inconvenience of obtaining an unfavorable solution of selling a product other than a sale candidate.

  Moreover, according to the processing apparatus 10 of this embodiment, the opportunity loss in a store can be reduced further. The reason is that the processing apparatus 10 can present to the user the product to be sold and the number of products sold. Therefore, the user can place an order for a product from the manufacturer based on the product and the number sold. That is, the user can prepare a product in advance based on the given information.

1A processor 2A memory 3A I / O I / F
4A peripheral circuit 5A bus 10 processor 11 calculation unit 12 output unit 13 input reception unit

Claims (10)

Using the estimation model obtained based on the sales data of the product and the sales condition data indicating the conditions at the time of sales, the optimal solution for the sales price of one product, the optimal solution for the combination of the sales prices of multiple products, the normal price A calculation means for calculating at least one of an optimal solution for a product sold at a price different from the above and an optimal solution for a sales price of a product sold at a price different from the normal price;
Output means for outputting a calculation result of the calculation means;
A processing apparatus. The processing apparatus according to claim 1,
The processing means for calculating the optimum solution by further using the sales time condition data and / or the predicted value of the sales time condition data for a target period. The processing apparatus according to claim 1 or 2,
Input receiving means for receiving a first input designating a selling price candidate for each of a plurality of products;
The processing means for calculating the optimum solution by further using the selling price designated by the first input. In the processing apparatus of any one of Claim 1 to 3,
The calculation means is based on the first evaluation value indicated based on the sales and / or profit in the entire store that sells the product, and on the sales and / or profit in the entire partial product group sold in the store. The processing apparatus which calculates the said optimal solution which maximizes the 2nd evaluation value or the 3rd evaluation value calculated based on the said 1st evaluation value and the said 2nd evaluation value. The processing apparatus according to claim 4, wherein
An input receiving means for receiving a second input for designating a part of the products sold in the store;
The processing apparatus in which the second evaluation value is indicated based on sales and / or profits of a part of the entire product group designated by the second input. The processing apparatus according to claim 4, wherein
The calculating means calculates the optimal solution that maximizes the third evaluation value;
A third input for selecting an evaluation value to be emphasized from among the first evaluation value and the second evaluation value or designating a ratio to be emphasized in the first evaluation value and the second evaluation value Further having an input receiving means for receiving
The processing unit is configured to calculate the third evaluation value based on content selected or designated by the third input. In the processing apparatus of any one of Claim 1 to 6,
An input receiving means for receiving a fourth input designating part or all of the products sold in the store;
The calculation means includes an optimal solution for the sales price of the one product specified by the fourth input, an optimal solution for a combination of the sales prices of the plurality of products specified by the fourth input, and the fourth input. An optimal solution for processing to select a product to be sold at a price different from the normal price from among the plurality of products specified in the above, and a normal price selected from the plurality of products specified in the fourth input A processing device that calculates at least one of the optimal solutions for selling prices of products sold at different prices. In the processing apparatus of any one of Claim 1 to 7,
The sales time condition data includes product information, calendar data, weather data, information indicating past sales trends, event information of stores that sell products, event information held in the vicinity of the stores, and competing stores related to competing stores. A processing device including at least one of the information. Computer
Using the estimation model obtained based on the sales data of the product and the sales condition data indicating the conditions at the time of sales, the optimal solution for the sales price of one product, the optimal solution for the combination of the sales prices of multiple products, the normal price A calculation step for calculating at least one of an optimal solution for a product sold at a price different from the above and an optimal solution for a sales price of a product sold at a price different from the normal price;
An output step of outputting a calculation result in the calculation step;
Processing method to execute. Computer
Using the estimation model obtained based on the sales data of the product and the sales condition data indicating the conditions at the time of sales, the optimal solution for the sales price of one product, the optimal solution for the combination of the sales prices of multiple products, the normal price A calculation means for calculating at least one of an optimal solution for a product sold at a price different from the above and an optimal solution for a sales price of a product sold at a price different from the normal price;
Output means for outputting a calculation result of the calculation means;
Program to function as.

Images