Product packaging suggestion device

Abstract

It is possible to create product packaging with high accuracy that will lead to the intended sales trends. [Solution] The system comprises: a means for generating a machine learning model that outputs a product package by inputting features related to the product type and sales of the target product; a means for inputting a dataset of features related to one product type and sales of the target product into the machine learning model and performing a process for manufacturing a product having the output product package; a means for determining whether the sales or number of sales for the product having the product package is below a predetermined target value; a means for causing the machine learning model to perform additional learning; and, if it is determined that the sales or number of sales is below a predetermined target value, a means for inputting a dataset of features related to one product type and sales of the target product into the machine learning model after the additional learning and performing a process for manufacturing a product having the output other product package.

Description

[Technical Field]

[0001] This relates to technology for automatically creating product packaging. [Background technology]

[0002] Product packaging is an important factor that determines the sales characteristics of a product, such as whether it will be a long-selling product, whether it will have high initial sales, and whether it will promote new ingredients and lead to increased sales.

[0003] Against this background, attempts have been made to analyze product packaging using computers. For example, Patent Document 1 proposes a technology for evaluating the conspicuousness of changes in product packaging design. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] International Publication No. 2021 / 095152 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the above-mentioned conventional technology has a problem in that it is not possible to create with high accuracy product packages that demonstrate the sales characteristics that a business operator intends.

[0006] In view of the above problems, the present invention aims to provide a product package suggestion device that can accurately create product packages that will lead to desired sales trends. [Means for solving the problem]

[0007] One embodiment of the disclosed product package proposal device includes: a model generation means for performing machine learning on a data set of product types, characteristics related to product sales, and product packages, and generating a machine learning model that outputs the product package by inputting the product type and characteristics related to sales of the target product; a model storage means for storing parameters that define input / output characteristics of the machine learning model; and a product manufacturing means for inputting a data set of characteristics related to one of the product types and sales of the target product into the machine learning model, and manufacturing a product that includes the output one of the product packages. By acquiring and aggregating the sales status of products having the one product package sold in the store, verification information calculation means for calculating sales of products including the one product package; After the process of manufacturing a product having the one product package, The sales Achieve the intended sales trend a target achievement determination means for determining whether the target value is below a predetermined target value; To change the input / output characteristics of the machine learning model, additional learning means for causing the machine learning model to perform additional learning; and if the goal achievement determination means determines that the target achievement is below the target achievement, determining whether the target achievement is below the target achievement after the additional learning. Input / output characteristics were changed The method is characterized by having a product remanufacturing means that inputs a data set of characteristics related to one of the product types and the sales of the target product into the machine learning model, and processes the output product to manufacture products that include the other product packages.

[0008] Furthermore, one embodiment of the disclosed product package proposal device includes: a model generation means for performing machine learning on a data set of a product type, characteristics related to product sales, and a product package, and generating a machine learning model that outputs the product package by inputting the product type and characteristics related to sales of the target product; a model storage means for storing parameters that define input / output characteristics of the machine learning model; and a product manufacturing means for inputting a data set of characteristics related to one of the product types and sales of the target product into the machine learning model, and performing processing to manufacture a product that includes the output one of the product packages. By acquiring and aggregating the sales status of products having the one product package sold in the store, verification information calculation means for calculating the number of sales of a product having the one product package; After the process of manufacturing a product having the one product package, The sales figures Achieve the intended sales trenda target achievement determination means for determining whether the target value is below a predetermined target value; To change the input / output characteristics of the machine learning model, additional learning means for causing the machine learning model to perform additional learning; and if the goal achievement determination means determines that the target achievement is below the target achievement, determining whether the target achievement is below the target achievement after the additional learning. Input / output characteristics were changed The method is characterized by having a product remanufacturing means that inputs a data set of characteristics related to one of the product types and the sales of the target product into the machine learning model, and processes the output product to manufacture products that include the other product packages. [Effects of the Invention]

[0009] The disclosed product packaging proposal device can accurately create product packaging that will lead to intended sales trends. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a diagram showing an overview of a product package proposal device according to an embodiment of the present invention; [Figure 2] 1 is a functional block diagram of a product package proposal device according to an embodiment of the present invention. [Figure 3] FIG. 1 is a diagram illustrating an example of a training dataset of a machine learning model according to an embodiment of the present invention. [Figure 4] 1 is a diagram illustrating an example of a hardware configuration of a product package proposal device according to an embodiment of the present invention. [Figure 5] 10 is a flowchart showing a flow of an example of processing by the product package proposal device according to the present embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0011] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described with reference to the drawings. (Operation principle of the product package suggestion device according to this embodiment)

[0012] The operating principle of a product package proposal device 100 according to this embodiment (hereinafter simply referred to as "the device") will be described with reference to Figures 1 and 2. Figure 1 is a diagram showing the connection relationship between the device 100 and other devices, and Figure 2 is a functional block diagram of the device 100.

[0013] 1, the device 100 is connected to a store terminal 270 via a communication network 280. The communication network 280 may be either a wired or wireless network. The store terminal 270 is a device that notifies the device 100 of the sales status and inventory status of products sold in a store (including a virtual store), and may be, for example, a POS (Point of Sales) system.

[0014] 2, the present device 100 includes a training data storage means 110, a model storage means 120, a model generation means 130, a product manufacturing means 140, a verification information calculation means 150, a goal achievement determination means 160, an additional learning means 170, and a product remanufacturing means 180. Note that the present device 100 does not necessarily have to include the storage means 110 and 120, and the present device 100 may use the storage means 110 and 120 that are included in an external device.

[0015] The training data storage means 110 stores a training dataset 240 of product types 210, product sales characteristics 220, and product packages (including images) 230. The training dataset 240 is added and updated over time.

[0016] An example of the learning data storage means 110 is shown in Fig. 3. As shown in Fig. 3, the learning data storage means 110 stores, for example, product type 210: laundry detergent, characteristics 220: long-selling product, promoted new ingredients, and product packaging 230: packaging of product A, in association with each other. The learning data storage means 110 also stores, for example, product type 210: room detergent, characteristics 220: high initial sales, high new customer acquisition rate, and product packaging 230: packaging of product D, in association with each other. The characteristics 220 related to product sales are, for example, information such as long-selling product, promoted new ingredients, high new customer acquisition rate, and high initial sales, but are not limited to these. The model storage means 120 stores parameters that define the input / output characteristics of the machine learning model 250, which will be described later.

[0017] The model generation means 130 trains the machine learning model 250 on the learning dataset 240, and generates the machine learning model 250 that outputs the product package 230 by inputting the product type 210 and the characteristics 220 related to the sales of the target product. Note that the learning algorithm is not particularly limited.

[0018] The product manufacturing means 140 inputs a data set of one product type 210 and features 220 related to sales of the target product into the machine learning model 250, and performs processing to manufacture a product 260 having one output product package 230. The product 260 is then offered for sale in a store.

[0019] The verification information calculation means 150 calculates the sales or sales quantity of the product 260 that includes one product package. The verification information calculation means 150 acquires information on the sales or sales quantity of the product 260 from the store terminal 270 and performs the work of tallying up the information.

[0020] The target achievement determination means 160 determines whether the sales or sales volume calculated by the verification information calculation means 150 is below a predetermined target value. The target value can be set as appropriate.

[0021] The additional learning means 170 causes the machine learning model 250 to perform additional learning using the added / updated learning dataset 240. Note that, before and after the additional learning, the input / output characteristics of the machine learning model 250 generally change. Furthermore, when processing is performed by the additional learning means 170, the information stored in the model storage means 120 is also updated.

[0022] If the target achievement determination means 160 determines that the target is "below the target," the product remanufacturing means 180 inputs a data set of one product type 210 and characteristics 220 related to the sales of the target product into the machine learning model 250 after additional learning, and performs a process of manufacturing a product 260 that includes the output other product package 230. Based on the above-described operating principle, the device 100 can accurately create product packages 230 that will lead to the intended sales trends. (Hardware configuration of the product package proposal device according to this embodiment)

[0023] An example of the hardware configuration of the present device 100 will be described using Fig. 4. Fig. 4 is a diagram showing an example of the hardware configuration of the present device 100. As shown in Fig. 4, the present device 100 has a CPU (Central Processing Unit) 510, a ROM (Read-Only Memory) 520, a RAM (Random Access Memory) 530, an auxiliary storage device 540, a communication I / F 550, an input device 560, a display device 570, and a storage medium I / F 580.

[0024] CPU 510 is a device that executes programs stored in ROM 520, performs arithmetic processing on data loaded into RAM 530 in accordance with program instructions, and controls the entire device 100. ROM 520 stores programs and data to be executed by CPU 510. When CPU 510 executes a program stored in ROM 520, the programs and data to be executed are loaded into RAM 530, and RAM 530 temporarily holds the arithmetic data during the calculation.

[0025] The auxiliary storage device 540 is a device that stores the OS (Operating System), which is basic software, the application program according to this embodiment, and the like, together with related data, and may include, for example, the training data storage means 110 and the model storage means 120. The auxiliary storage device 540 is, for example, a hard disk drive (HDD) or a flash memory.

[0026] The communication I / F 550 is an interface for connecting to a communication network 280 such as a wired or wireless LAN (Local Area Network) or the Internet, and transmitting and receiving data to and from another device (such as a POS system) 270 that provides a communication function.

[0027] The input device 560 is a device such as a keyboard for inputting data to the device 100. The display device (output device) 570 is a device formed of an LCD (Liquid Crystal Display) or the like, and functions as a user interface when the user uses the functions of the device 100 or when making various settings. The storage medium I / F 580 is an interface for sending and receiving data to and from a storage medium 590 such as a CD-ROM, DVD-ROM, or USB memory.

[0028] Each of the means included in device 100 may be realized by CPU 510 executing a program corresponding to each of the means stored in ROM 520 or auxiliary storage device 540. Each of the means included in device 100 may also be realized by hardware that performs the processing associated with the means. Alternatively, device 100 may be caused to execute the program by loading the program according to the present invention from an external server device via communication I / F 550 or from storage medium 590 via storage medium I / F 580. (Example of processing by the product package proposal device according to this embodiment) An example of processing by the device 100 will be described with reference to Fig. 5. Fig. 5 is a flowchart showing the flow of an example of processing by the device 100.

[0029] In S10, the model generation means 130 trains the machine learning model 250 on the learning dataset 240, and generates the machine learning model 250 that outputs the product package 230 by inputting the product type 210 and the characteristics 220 related to the sales of the target product. Note that the learning algorithm is not particularly limited. Furthermore, the parameters that define the input / output characteristics of the machine learning model 250 generated by the process in S10 are stored in the model storage means 120.

[0030] In S20, the product manufacturing means 140 inputs a data set of one product type 210 and features 220 related to sales of the target product into the machine learning model 250 generated in S10, and performs processing to manufacture a product 260 having one output product package 230. The product 260 is then offered for sale in a store.

[0031] Furthermore, after the processing of S20, the additional learning means 170 performs processing to cause the machine learning model 250 to perform additional learning using the added / updated learning dataset 240. Note that, before and after the additional learning, the input / output characteristics of the machine learning model 250 generally change. Furthermore, when processing is performed by the additional learning means 170, the information stored in the model storage means 120 is also updated.

[0032] In S30, the verification information calculation means 150 calculates the sales or sales quantity of the product 260 that includes one product package. The verification information calculation means 150 acquires information related to the sales or sales quantity from the store terminal 270 and performs the work of tallying up the information.

[0033] Then, in S30, the target achievement determination means 160 determines whether the sales or sales volume calculated by the verification information calculation means 150 is below a predetermined target value. The target value can be set as appropriate.

[0034] If it is determined in S30 that the result is "below the predetermined target value," in S40 the product remanufacturing means 180 inputs the data set of the one product type 210 and the features 220 related to the sales of the target product into the machine learning model 250 after additional learning, and performs processing to manufacture the product 260 that includes the outputted other product package 230. Then, the product 260 that includes the other product package 230 is offered for sale in the store. By performing the above-described processing, the device 100 can accurately create product packages 230 that will result in the intended sales trends.

[0035] Although the embodiments of the present invention have been described in detail above, the present invention is not limited to such specific embodiments, and various modifications and variations are possible within the scope of the gist of the present invention as set forth in the claims. [Explanation of symbols]

[0036] 100 Product packaging suggestion device 110 Learning data storage means 120 Model storage means 130 Model Generation Method 140 Means of manufacturing goods 150 Verification information calculation means 160 Goal achievement determination means 170 Additional Learning Tools 180 Means of remanufacturing goods 210 Product types 220 Characteristics of product sales 230 Product Package 240 training datasets 250 machine learning models 260 Products with product packaging output by machine learning models 270 Store terminals (POS systems) 280 Communication Network 510 CPU 520 ROM 530 RAM 540 Auxiliary storage 550 Communication Interface 560 Input Device 570 Output Device 580 Storage Media Interface 590 Storage medium

Claims

1. a model generation means for performing machine learning on a data set of a product type, characteristics related to product sales, and product packaging, and generating a machine learning model that outputs the product package by inputting the product type and characteristics related to sales of the product that is a sales target; a model storage means for storing parameters that define the input / output characteristics of the machine learning model; a product manufacturing means for inputting a data set of characteristics related to the product type and the sales of the target product into the machine learning model and manufacturing a product having the output product package; verification information calculation means for calculating sales of products including the one product package by acquiring and aggregating sales information of products including the one product package sold in a store; a target achievement determination means for determining whether or not the sales amount falls below a predetermined target value that represents an intended sales trend after a process of manufacturing a product having the one product package; additional learning means for causing the machine learning model to perform additional learning in order to change the input / output characteristics of the machine learning model; and a product remanufacturing means for inputting a data set of characteristics related to the sales of one of the product types and the target product into the machine learning model in which the input / output characteristics after the additional learning have been changed, if the target achievement determination means determines that the sales are below the target, and for manufacturing a product that includes the output other product package.

2. a model generation means for performing machine learning on a data set of a product type, characteristics related to product sales, and product packaging, and generating a machine learning model that outputs the product package by inputting the product type and characteristics related to sales of the product that is a sales target; a model storage means for storing parameters that define the input / output characteristics of the machine learning model; a product manufacturing means for inputting a data set of characteristics related to the product type and the sales of the target product into the machine learning model and manufacturing a product having the output product package; verification information calculation means for calculating the number of sales of products having the one product package by acquiring and aggregating sales status of products having the one product package sold in a store; a target achievement determination means for determining whether or not the sales volume is below a predetermined target value that represents an intended sales trend after a process of manufacturing a product having the one product package; additional learning means for causing the machine learning model to perform additional learning in order to change the input / output characteristics of the machine learning model; and a product remanufacturing means for inputting a data set of characteristics related to the sales of one of the product types and the target product into the machine learning model in which the input / output characteristics after the additional learning have been changed, if the target achievement determination means determines that the sales are below the target, and for manufacturing a product that includes the output other product package.

3. A computer having a model storage means for storing parameters that define input / output characteristics of a machine learning model that performs machine learning on a data set of a product type, a product sales characteristic, and a product package, and outputs the product package by inputting the product type and the sales characteristic of the product that is a sales target, a step in which a model generation means performs machine learning on a data set of a product type, characteristics related to product sales, and a product package, and generates a machine learning model that outputs the product package by inputting the product type and characteristics related to sales of the product that is a sales target; a step in which a product manufacturing means inputs a data set of characteristics related to one of the product types and the sales of the target product into the machine learning model, and performs processing to manufacture a product having the one of the product packages output; a step in which verification information calculation means acquires and aggregates sales statuses of products that include the one product package and are sold in a store, thereby calculating sales of the products that include the one product package; a step in which a target achievement determination means determines whether or not the sales amount falls below a predetermined target value that represents an intended sales trend after a process of manufacturing a product having the one product package; an additional learning step of causing the machine learning model to perform additional learning in order to change input / output characteristics of the machine learning model; When the product remanufacturing means is determined to be below the target by the goal achievement determination means, a data set of characteristics related to one of the product types and the sales of the target product is input into the machine learning model in which the input / output characteristics after the additional learning have been changed, and a process of manufacturing a product having the other product package output is performed.

4. A computer having a model storage means for storing parameters that define input / output characteristics of a machine learning model that performs machine learning on a data set of a product type, a product sales characteristic, and a product package, and outputs the product package by inputting the product type and the sales characteristic of the product that is a sales target, a step in which a model generation means performs machine learning on a data set of a product type, characteristics related to product sales, and a product package, and generates a machine learning model that outputs the product package by inputting the product type and characteristics related to sales of the product that is a sales target; a step in which a product manufacturing means inputs a data set of characteristics related to one of the product types and the sales of the target product into the machine learning model, and performs processing to manufacture a product having the one of the product packages output; a step in which verification information calculation means acquires and aggregates sales statuses of products including the one product package sold in a store, thereby calculating the number of sales of the products including the one product package; a step in which a target achievement determination means determines whether or not the sales volume is below a predetermined target value that represents an intended sales trend after a process of manufacturing a product having the one product package; an additional learning step of causing the machine learning model to perform additional learning in order to change input / output characteristics of the machine learning model; When the product remanufacturing means is determined to be below the target by the goal achievement determination means, a data set of characteristics related to one of the product types and the sales of the target product is input into the machine learning model in which the input / output characteristics after the additional learning have been changed, and a process of manufacturing a product having the other product package output is performed.

5. A product packaging suggestion program for causing a computer to execute the method according to claim 3 or 4.

Images