Product demand prediction method, device, equipment and computer storage medium

Abstract

This application provides a product demand forecasting method, apparatus, device, and computer storage medium. The product demand forecasting method includes: acquiring product information of target products; dividing the target products according to the product information to obtain a product hierarchy diagram; acquiring a baseline node in the product hierarchy diagram; performing product demand forecasting on the baseline node to obtain a baseline demand forecast value for the baseline node; processing the baseline demand forecast value of the baseline node according to the cascading relationship of the product hierarchy diagram to obtain the target demand forecast value for the product corresponding to each node in the product hierarchy diagram. In this embodiment, product demand forecasting is performed based on the baseline node in the product hierarchy diagram to obtain the target demand forecast value for the product corresponding to each node in the product hierarchy diagram, thereby achieving accurate forecasting of product demand in different dimensions.

Description

Technical Field

[0001] This application relates to the field of computer technology, specifically to a product demand forecasting method, apparatus, device, and computer storage medium. Background Technology

[0002] Generally speaking, a company's supply chain planning department or sales department will develop a demand plan for the company's product sales and plan the product sales target for the whole year to guide the allocation of production and marketing resources.

[0003] Currently, companies generally use a top-down, hierarchical approach to plan product sales targets. This involves first determining the overall annual and monthly sales targets based on management goals, then setting monthly targets for each major product category, and finally breaking them down to the SKU (Stock Keeping Unit) level. This approach, where management determines product demand planning and overall sales targets without fully considering historical data trends, can lead to unrealistic targets and significant discrepancies between the actual results and actual targets. Furthermore, when planning or sales departments break down overall targets to individual product levels, this process relies heavily on manual experience, resulting in inaccurate task allocation. Summary of the Invention

[0004] This application provides a product demand forecasting method, apparatus, device, and computer storage medium, aiming to solve the technical problem of inaccurate product demand forecasting under different dimensions in existing technical solutions.

[0005] On the one hand, this application provides a product demand forecasting method, the product demand forecasting method comprising:

[0006] Obtain product information for the target product;

[0007] Based on the product information, each target product is divided into categories to obtain a product hierarchy diagram;

[0008] Obtain the baseline node in the product hierarchy diagram, perform product demand forecasting on the baseline node, and obtain the baseline demand forecast value of the baseline node.

[0009] Based on the cascading relationships in the product hierarchy diagram, the baseline demand forecast value of the baseline node is processed to obtain the target demand forecast value of the product corresponding to each node in the product hierarchy diagram.

[0010] In some embodiments of this application, before obtaining the baseline node in the product hierarchy graph, performing product demand forecasting on the baseline node, and obtaining the baseline demand forecast value of the baseline node, the method further includes:

[0011] Set each node in the product hierarchy diagram as a target node in sequence;

[0012] Based on the product information of the target products associated with the target node, determine the initial demand forecast value and initial forecast error corresponding to the target node;

[0013] Based on the initial demand forecast and initial forecast error of each target node, a baseline level in the product hierarchy diagram is determined, and the nodes in the baseline level are set as baseline nodes.

[0014] In some embodiments of this application, determining the initial demand forecast value and initial forecast error corresponding to the target node based on the product information of the target product associated with the target node includes:

[0015] The product information of the target product associated with the target node is input into a preset prediction model. The sales data in the product information is fitted by the prediction model to obtain the initial demand prediction value of the target node for a historical time period.

[0016] The initial prediction error of the target node is calculated based on the initial demand forecast and the actual product demand for the historical time period in the product information.

[0017] In some embodiments of this application, determining the baseline level in the product hierarchy diagram based on the initial demand forecast value and initial forecast error of each target node, and setting the nodes in the baseline level as baseline nodes, includes:

[0018] Set each level in the product hierarchy diagram as the target level in sequence;

[0019] The initial demand prediction value and initial prediction error of each node corresponding to the target level are weighted and processed to obtain the hierarchical prediction error of the target level.

[0020] The target level with the smallest hierarchical error is taken as the reference level, and the node corresponding to the reference level is set as the reference node.

[0021] In some embodiments of this application, the step of processing the baseline demand forecast value of the baseline node according to the cascading relationship of the product hierarchy diagram to obtain the target demand forecast value of each node in the product hierarchy diagram includes:

[0022] Obtain the target level of the node to be predicted in the product hierarchy diagram;

[0023] If the target level is above the baseline level, the baseline demand prediction value is summarized according to the cascading relationship between the node to be predicted and the baseline node in the product hierarchy diagram to obtain the target demand prediction value of the node to be predicted.

[0024] If the target level is below the baseline level, the baseline demand prediction value is split according to the cascading relationship between the node to be predicted and the baseline node in the product hierarchy diagram to obtain the target demand prediction value of the node to be predicted.

[0025] In some embodiments of this application, obtaining the baseline node in the product hierarchy graph, performing product demand forecasting on the baseline node, and obtaining the baseline demand forecast value of the baseline node includes:

[0026] Extract the product information of the target product corresponding to the baseline node in the product hierarchy graph;

[0027] The product information of the target product corresponding to the benchmark node is input into the preset prediction model, and the data is fitted through the preset prediction model to obtain the benchmark demand prediction value of the benchmark node.

[0028] In some embodiments of this application, obtaining the baseline node in the product hierarchy graph, performing product demand forecasting on the baseline node, and obtaining the baseline demand forecast value of the baseline node includes:

[0029] Obtain a preset sales target and determine the target node corresponding to the preset sales target;

[0030] Based on the cascading relationship between the target node and the baseline node in the product hierarchy diagram, the preset sales targets are summarized or broken down to obtain the baseline demand forecast value of the baseline node.

[0031] In some embodiments of this application, after processing the baseline demand forecast value of the baseline node according to the cascading relationship of the product hierarchy diagram to obtain the target demand forecast value of the product corresponding to each node in the product hierarchy diagram, the method further includes:

[0032] Obtain the first baseline demand forecast value determined by the preset forecast model, and obtain the second baseline demand forecast value determined according to the preset sales target;

[0033] Calculate the ratio between the first baseline demand forecast and the second baseline demand forecast. If the ratio exceeds a preset threshold, output a model update prompt.

[0034] On the other hand, this application provides a product demand forecasting device, the product demand forecasting device comprising:

[0035] The acquisition module is used to acquire product information of the target product;

[0036] The segmentation module is used to segment each of the target products according to the product information to obtain a product hierarchy diagram;

[0037] The prediction module is used to obtain the baseline node in the product hierarchy diagram, perform product demand prediction on the baseline node, and obtain the baseline demand prediction value of the baseline node.

[0038] The processing module is used to process the baseline demand forecast value of the baseline node according to the cascading relationship of the product hierarchy diagram, and obtain the target demand forecast value of the product corresponding to each node in the product hierarchy diagram.

[0039] On the other hand, this application also provides a product demand forecasting device, which is a charging cabinet and / or a server, and the product demand forecasting device includes:

[0040] One or more processors;

[0041] Memory; and

[0042] One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the steps in the product demand forecasting method.

[0043] On the other hand, this application also provides a computer-readable storage medium having a computer program stored thereon, the computer program being loaded by a processor to perform the steps in the product demand forecasting method.

[0044] The product demand forecasting method in this application should include: obtaining product information of target products; dividing each target product according to the product information to obtain a product hierarchy diagram; obtaining a baseline node in the product hierarchy diagram, performing product demand forecasting on the baseline node to obtain a baseline demand forecast value for the baseline node; processing the baseline demand forecast value of the baseline node according to the cascading relationship of the product hierarchy diagram to obtain the target demand forecast value of the product corresponding to each node in the product hierarchy diagram. In this embodiment, the target products are first divided according to the product information of the target products to obtain a product hierarchy diagram. Each node in the product hierarchy diagram corresponds to different dimensions of target products after classification. The cascading relationship between target products at each node can be clearly and accurately determined through the product hierarchy diagram. Then, a baseline node is obtained, which is the node with the highest prediction accuracy. The server performs product demand forecasting for different nodes according to the baseline node in the product hierarchy diagram to obtain the target demand forecast value of the product corresponding to each node in the product hierarchy diagram, thereby achieving accurate prediction of product demand in different dimensions. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a schematic diagram of a scenario for the product demand forecasting method provided in the embodiments of this application;

[0047] Figure 2 This is a schematic flowchart of one embodiment of the product demand forecasting method in this application.

[0048] Figure 3 This is a schematic diagram of an embodiment of the product hierarchy relationship diagram in the product demand forecasting method provided in this application;

[0049] Figure 4 This is a schematic flowchart of an embodiment of the product demand forecasting method for determining a baseline node provided in this application.

[0050] Figure 5 This is a schematic flowchart of an embodiment of the product demand forecasting method provided in this application for determining the baseline demand forecast value;

[0051] Figure 6 This is a flowchart illustrating another embodiment of the product demand forecasting method provided in this application for determining the baseline demand forecast value;

[0052] Figure 7 This is a schematic flowchart of an embodiment of the product demand forecasting method provided in this application for updating the preset forecasting model;

[0053] Figure 8 This is a schematic diagram of an embodiment of the product demand forecasting device provided in this application.

[0054] Figure 9 This is a schematic diagram of an embodiment of the product demand forecasting device provided in this application. Detailed Implementation

[0055] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of the present invention.

[0056] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0057] In this application, the term "exemplary" is used to mean "serving as an example, illustration, or description." Any embodiment described as "exemplary" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed in this application.

[0058] This application provides a product demand forecasting method, apparatus, device, and computer-readable storage medium, which will be described in detail below.

[0059] The product demand forecasting method in this embodiment of the invention is applied to a product demand forecasting device, which is provided in a product demand forecasting equipment. The product demand forecasting equipment includes one or more processors, a memory, and one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the processor to implement the product demand forecasting method.

[0060] like Figure 1 As shown, Figure 1 This is a schematic diagram of a product demand forecasting method in an embodiment of this application. The product demand forecasting scenario in this embodiment includes a product demand forecasting device 100 (the product demand forecasting device 100 integrates a product demand forecasting apparatus). The product demand forecasting device 100 runs a computer-readable storage medium corresponding to the product demand forecasting to perform product demand forecasting.

[0061] Understandable Figure 1 The product demand forecasting device in the product demand forecasting scenario shown, or the device included in the product demand forecasting device, does not constitute a limitation on the embodiments of the present invention. That is, the number or type of device included in the product demand forecasting scenario, or the number or type of device included in each device, does not affect the overall implementation of the technical solution in the embodiments of the present invention, and can all be considered as equivalent substitutions or derivatives of the technical solutions claimed in the embodiments of the present invention.

[0062] In this embodiment of the invention, the product demand forecasting device 100 is mainly used for: acquiring product information of target products; dividing each target product according to the product information to obtain a product hierarchy diagram; acquiring a reference node in the product hierarchy diagram, performing product demand forecasting on the reference node to obtain a reference demand forecast value for the reference node; and processing the reference demand forecast value of the reference node according to the cascading relationship of the product hierarchy diagram to obtain the target demand forecast value of the product corresponding to each node in the product hierarchy diagram.

[0063] In this embodiment of the invention, the product demand forecasting device 100 can be an independent product demand forecasting device, or a network or cluster of multiple product demand forecasting devices. For example, the product demand forecasting device 100 described in this embodiment includes, but is not limited to, computers, network hosts, single network product demand forecasting devices, sets of multiple network product demand forecasting devices, or cloud product demand forecasting devices composed of multiple product demand forecasting devices. The cloud product demand forecasting device is composed of a large number of computer or network product demand forecasting devices based on cloud computing.

[0064] Those skilled in the art will understand that Figure 1 The application environment shown is merely one application scenario of the solution in this application and does not constitute a limitation on the application scenario of the solution in this application. Other application environments may include those that are more specific to this application. Figure 1 The more or fewer product demand forecasting devices shown, or the network connectivity of product demand forecasting devices, for example... Figure 1 Only one product demand forecasting device is shown in the diagram. It is understood that the product demand forecasting scenario may also include one or more other product demand forecasting devices, which are not limited here. The product demand forecasting device 100 may also include a memory for storing data, such as storing relevant data on product demand forecasting.

[0065] Furthermore, in the product demand forecasting scenario of this application, the product demand forecasting device 100 may be equipped with a display device, or the product demand forecasting device 100 may not have a display device but may communicate with an external display device 200. The display device 200 is used to output the results of the product demand forecasting method executed in the product demand forecasting device. The product demand forecasting device 100 can access a background database 300 (the background database may be located in the local storage of the product demand forecasting device, or it may be located in the cloud). The background database 300 stores information related to product demand forecasting, such as data related to product demand forecasting.

[0066] It should be noted that, Figure 1The schematic diagram of the product demand forecasting method shown is merely an example. The product demand forecasting scenarios described in the embodiments of the present invention are intended to more clearly illustrate the technical solutions of the embodiments of the present invention and do not constitute a limitation on the technical solutions provided in the embodiments of the present invention.

[0067] like Figure 2 As shown, Figure 2 This is a flowchart illustrating one embodiment of the product demand forecasting method in this application. The product demand forecasting method includes steps 201-204:

[0068] 201. Obtain product information for the target product.

[0069] In this embodiment, the product demand forecasting method is applied to a product demand forecasting device. The type of product demand forecasting device is not specifically limited. The product demand forecasting device can be a server or a terminal, such as a computer, tablet, or other electronic device. In this embodiment, a server is used as an example to illustrate the product demand forecasting device.

[0070] The server receives product demand forecasting requests. The triggering method for these requests is not specifically limited. Specifically, they can be triggered by the user, such as by the user entering a forecast command on their mobile phone, which then sends the request to the server. Alternatively, the server can trigger the requests automatically, for example, by pre-setting the automatic triggering condition to the beginning of each month. When the server detects the beginning of the month, it automatically triggers the product demand forecasting request.

[0071] After receiving a product demand forecasting request, the server obtains the product information of the target product. The type and quantity of the target product are not limited. For example, for a shoe store, the target product is shoes. Product information refers to product-related information. For an office enterprise, the target product is office supplies. Product information includes, but is not limited to, product name, product quantity, product type, product color, product use, product batch number, and product shelf life, etc.

[0072] To facilitate understanding, this embodiment uses shoes as the target product. The shoe manufacturer triggers a product demand forecasting request on the server. After receiving the product demand forecasting request, the server obtains information such as the type, color, style, and size of the shoes.

[0073] 202. Based on the product information, the target products are divided to obtain a product hierarchy diagram.

[0074] After obtaining the product information of the target product, the server determines the relationship between the target products according to the product information. The server classifies the target products according to the product relationship between the target products to form a product hierarchy diagram. The product hierarchy diagram in this embodiment is used to reflect the cascading relationship between products. The form of the product hierarchy diagram is not limited.

[0075] like Figure 3 As shown, Figure 3 This is a schematic diagram of an embodiment of the product hierarchy diagram in the product demand forecasting method provided in this application. In this embodiment, level 0 in the product hierarchy diagram represents the full category of the target product, level 1 represents the level classified according to the first classification dimension, level 2 represents the level that classifies the target products corresponding to each node in level 1 according to the second classification dimension, and level 3 represents the level that classifies the target products corresponding to each node in level 2 according to the third classification dimension.

[0076] For example, if the target product is shoes, level 0 in the product hierarchy diagram represents all types of shoes. Level 1 represents the classification by shoe type, with category 1 being athletic shoes, category 2 being leather shoes, category 3 being cloth shoes, and so on. Level 2 represents the classification of the target products corresponding to each node in level 1 by men and women, with sub-category 1-1 being men's athletic shoes, sub-category 1-2 being women's athletic shoes, sub-category 2-1 being men's leather shoes, sub-category 2-2 being women's leather shoes, sub-category 3-1 being men's cloth shoes, sub-category 3-2 being women's cloth shoes, and so on. Level 3 represents the classification of the target products corresponding to each node in level 2 by size, with sub-category 1-1-1 being men's athletic shoes size 39, sub-category 1-1-2 being men's athletic shoes size 40, sub-category 1-1-3 being men's athletic shoes size 41, and so on.

[0077] 203. Obtain the baseline node in the product hierarchy graph, perform product demand forecasting on the baseline node, and obtain the baseline demand forecast value of the baseline node.

[0078] In this embodiment, after the server obtains the product hierarchy relationship diagram, the server determines the benchmark node in the product hierarchy relationship diagram. The benchmark node is the node with the highest prediction accuracy. The method of determining the benchmark node is not limited. The benchmark node can be set by the user or determined by the server according to rules.

[0079] After the server obtains the baseline node in the product hierarchy graph, the server performs product demand forecasting for the target products corresponding to the baseline node. In this embodiment, the method of forecasting product demand for the baseline node is not limited, but specifically includes:

[0080] Implementation Method 1: A pre-set prediction model is installed on the server. The pre-set prediction model is an algorithm for product demand prediction obtained through machine learning training. The server inputs the product information of the target product corresponding to the baseline node into the pre-set prediction model. The pre-set prediction model processes the product information to obtain the baseline demand prediction value of the baseline node.

[0081] Method 2: The server pre-stores product demand data for historical time periods. The server fits the product demand data of each product to obtain the baseline demand forecast value of the baseline node.

[0082] In this embodiment, the server obtains the baseline node in the product hierarchy graph. The baseline node is the node with the highest product demand forecast accuracy. The server performs product demand forecasting on the baseline node to obtain the baseline demand forecast value. Based on the baseline demand forecast value of the baseline node, the server performs product demand forecasting on other nodes in the product hierarchy graph. Specifically:

[0083] 204. Based on the cascading relationship of the product hierarchy diagram, process the baseline demand forecast value of the baseline node to obtain the target demand forecast value of the product corresponding to each node in the product hierarchy diagram.

[0084] The server processes the baseline demand forecast values ​​of the baseline nodes based on the cascading relationships in the product hierarchy diagram to obtain the target demand forecast values ​​for the products corresponding to each node in the product hierarchy diagram; specifically, this includes:

[0085] (1) Obtain the target level of the node to be predicted in the product hierarchy diagram;

[0086] (2) If the target level is above the baseline level, the baseline demand prediction value is obtained by summarizing the baseline demand prediction value based on the cascading relationship between the node to be predicted and the baseline node in the product hierarchy diagram.

[0087] (3) If the target level is below the baseline level, the baseline demand prediction value is split according to the cascading relationship between the node to be predicted and the baseline node in the product hierarchy diagram to obtain the target demand prediction value of the node to be predicted.

[0088] That is, the server obtains the target level of the node to be predicted in the product hierarchy graph. The server determines the relationship between the target level and the baseline level, which is level K. The target level is above the baseline level. The server traverses levels 0 to (K-1) and summarizes y(K,j) upwards to each level above it according to the hierarchical relationship of the categories, obtaining all the summarized values ​​from y(0,) to y(K-1,j). For example, if K=2 (that is, the level with the smallest level prediction error is the product sub-category), summing all y(2,j) can give y(0,), summing y(2, sub-category 3-1) to y(2, sub-category 3-n) can give y(1, major category 3), and so on, to obtain the target demand prediction value of the node to be predicted.

[0089] The target level is below the base level. The server traverses levels (K+1) to the maximum level, splitting y(K,j) downwards according to the category hierarchy into each of the lower levels, obtaining all split values ​​from y(K+1,j) to y(maximum level,j). The splitting ratio can be based on the relative sales proportion of the lower levels over a historical period; for example, if K=2 (i.e., the level with the smallest prediction error is the product sub-category), splitting y(2, sub-category 3-3) yields y(3, SKU3-3-1) to y(K+1,j). (3,SKU3-3-n) The split ratio is based on the relative sales ratio of SKUs from SKU3-3-1 to SKU3-3-n over a historical period. For example, if the relative sales ratio of SKU3-3-1, SKU3-3-2, SKU3-3-3, and SKU3-3-n over the past six months is 5:2:2:1, then the demand forecast y(3,SKU3-3-11) for SKU3-3-1 after splitting is half of y(2,sub-category 3-3), thus obtaining the target demand forecast value for the node to be predicted.

[0090] In this embodiment, the server first divides the target products according to their product information to obtain a product hierarchy diagram. Each node in the product hierarchy diagram corresponds to a different dimension of the target products after classification. The product hierarchy diagram can clearly and accurately determine the cascading relationship between target products at each node. Then, a baseline node is obtained, which is the node with the highest prediction accuracy. The server performs product demand prediction for different nodes based on the baseline node in the product hierarchy diagram to obtain the target demand prediction value of the products corresponding to each node in the product hierarchy diagram, thereby achieving accurate prediction of product demand in different dimensions.

[0091] Reference Figure 4 , Figure 4 This is a schematic flowchart of an embodiment of the product demand forecasting method for determining a baseline node provided in this application.

[0092] In some embodiments of this application, before step 203, where the server obtains the baseline node in the product hierarchy graph and performs product demand forecasting on the baseline node to obtain the baseline demand forecast value, the server determines the baseline node in the product hierarchy graph, specifically including steps 301-303:

[0093] 301. Set each node in the product hierarchy diagram as a target node in sequence.

[0094] The server sequentially sets each node in the product hierarchy diagram as a target node, and determines the product demand forecast value for each node. Specifically:

[0095] 302. Based on the product information of the target product associated with the target node, determine the initial demand forecast value and initial forecast error corresponding to the target node.

[0096] The server has a preset prediction model. The server obtains product information from the target node and inputs this information into the preset prediction model. The server then uses this model to predict product demand over a historical time period, obtaining an initial demand forecast value. Finally, the server compares this initial demand forecast value with the actual product demand over the historical time period to obtain an initial prediction error. In other words, in this embodiment, if the server were to predict product demand for a future point in time, since there are no actual product demand values, the server could not obtain a prediction error. Therefore, the server uses a preset prediction model to predict product demand over a historical time period, specifically including:

[0097] (1) Input the product information of the target product associated with the target node into the preset prediction model, and fit the sales data in the product information through the prediction model to obtain the initial demand prediction value of the target node for the historical time period.

[0098] (2) Calculate the initial prediction error of the target node based on the initial demand forecast value and the actual product demand value of the historical time period in the product information.

[0099] That is, the server inputs the product information of the target product associated with the target node into the preset prediction model. The server fits the sales data in the product information to the prediction model to obtain the initial demand prediction value of the target node in the historical time period. The server calculates the initial demand prediction value and the actual product demand value in the historical time period in the product information to obtain the initial prediction error of the target node.

[0100] 303. Based on the initial demand forecast value and initial forecast error of each target node, determine the baseline level in the product hierarchy diagram, and set the nodes in the baseline level as baseline nodes.

[0101] The server processes the initial demand forecast and initial forecast error of the target node corresponding to each product level in the product hierarchy diagram to determine the baseline level in the product hierarchy diagram. Specifically, the server weights the initial forecast errors corresponding to the target nodes of each level to obtain the level-specific forecast error. The server sets the level with the smallest level-specific forecast error as the baseline level and sets the nodes in the baseline level as the baseline nodes. This includes:

[0102] (1) Set each level in the product hierarchy diagram as the target level in sequence;

[0103] (2) The initial demand prediction value and initial prediction error of each node corresponding to the target level are weighted and processed to obtain the level prediction error of the target level.

[0104] (3) Take the target level with the smallest hierarchical error as the reference level, and set the node corresponding to the reference level as the reference node.

[0105] For example, the server stores a preset prediction model. The type of preset prediction model is not limited; for example, the preset prediction model could be the time series model Arima or the machine learning model LightGBM. The server iterates through each level (0 to 3). For all products in level 0, the server uses the time series model Arima to fit the historical sales data of all products. For level i, the server uses the machine learning model LightGBM to fit the historical sales data of each category / SKU j. The server extracts features such as the sales volume of the same period in the past year, the sales volume of the past three months, and the fluctuation of the past three months for each category to make predictions. Based on the prediction error e(i,j) and the demand prediction value y(i,j) of each category / SKU j in each level i obtained during the fitting process, the server calculates the level prediction error for each level based on the error e(i,j) of all categories / SKUs in each level.

[0106]

[0107] The server compares the prediction errors of each level in the product hierarchy graph. The server selects the level with the smallest prediction error as the baseline level and sets the nodes in that baseline level as baseline nodes. In this embodiment, the server determines the initial prediction error of each node based on the actual and predicted product demand values ​​for the target node over a historical period in the product hierarchy graph. Then, the server weights the initial prediction errors of each level to obtain the hierarchical prediction error. Based on this hierarchical prediction error, the server determines the baseline level and baseline node, which makes product demand prediction more accurate.

[0108] Reference Figure 5 , Figure 5 This is a schematic flowchart of an embodiment of the product demand forecasting method provided in this application for determining the baseline demand forecast value.

[0109] In some embodiments of this application, the baseline demand forecast value of the baseline node in the product hierarchy diagram is determined, specifically including steps 401-402:

[0110] 401. Extract the product information of the target product corresponding to the base node in the product hierarchy diagram.

[0111] The server extracts product information corresponding to the target product from the baseline node in the product hierarchy graph. The product information includes product name, product number, product production date, and product historical sales records, etc.

[0112] 402. Input the product information of the target product corresponding to the benchmark node into the preset prediction model, and perform data fitting through the preset prediction model to obtain the benchmark demand prediction value of the benchmark node.

[0113] The server inputs the product information of the target product corresponding to the baseline node into a preset prediction model. The server then fits the data using the preset prediction model to obtain the baseline demand prediction value for the baseline node. In this embodiment, the server uses the preset prediction model to predict product demand for the baseline node, obtaining the baseline demand prediction value, thereby achieving accurate product demand prediction based on the baseline demand prediction value.

[0114] Reference Figure 6 , Figure 6 This is a schematic flowchart of another embodiment of the product demand forecasting method provided in this application for determining the baseline demand forecast value.

[0115] In some embodiments of this application, determining the baseline demand forecast value of the baseline node in the product hierarchy diagram further includes steps 501-502:

[0116] 501. Obtain the preset sales target and determine the target node corresponding to the preset sales target.

[0117] The server obtains the preset sales target, which refers to the product sales target stored in the server beforehand. The server determines the target node corresponding to the preset sales target. That is, the preset sales target can be the product demand forecast value corresponding to all product categories in the product hierarchy diagram; the preset sales target can also be the product demand forecast value of a specified node in the product hierarchy diagram.

[0118] 502. Based on the cascading relationship between the target node and the benchmark node in the product hierarchy diagram, the preset sales target is summarized or split to obtain the benchmark demand forecast value of the benchmark node.

[0119] The server summarizes or breaks down preset sales targets based on the cascading relationship between target nodes and baseline nodes in the product hierarchy diagram, and obtains the baseline demand forecast value for the baseline node. For example, if the server determines that the target node is a level 0 node in the product hierarchy diagram and the baseline node is a level 1 node in the product hierarchy diagram, the server breaks down the product sales targets to obtain the baseline demand forecast value corresponding to each baseline node.

[0120] In this embodiment, users can manually customize product sales targets. The server breaks down or summarizes the product sales targets based on the relationship between the target nodes and the baseline nodes corresponding to the product sales targets, and obtains the baseline demand forecast value corresponding to the baseline nodes. Furthermore, the server determines the target demand forecast value corresponding to each node in the product hierarchy diagram based on the baseline demand forecast value corresponding to the baseline nodes, thus realizing accurate prediction of product demand in different dimensions.

[0121] Reference Figure 7 , Figure 7 This is a schematic diagram of an embodiment of the product demand forecasting method provided in this application for updating a preset forecasting model.

[0122] In some embodiments of this application, the product demand forecasting method specifically includes steps 501-503:

[0123] 601. Obtain the first baseline demand forecast value determined by the preset forecast model, and obtain the second baseline demand forecast value determined according to the preset sales target.

[0124] The server obtains the first baseline demand forecast value determined by the preset forecast model, and the server obtains the second baseline demand forecast value determined according to the preset sales target.

[0125] 602. Calculate the ratio between the first baseline demand forecast and the second baseline demand forecast. If the ratio exceeds a preset threshold, output a model update prompt.

[0126] The server calculates the ratio between the first baseline demand forecast and the second baseline demand forecast. The server compares the ratio with a preset threshold, which can be flexibly set according to specific scenarios. For example, the preset threshold can be set to 80%. If the ratio does not exceed the preset threshold, no action is taken; if the ratio exceeds the preset threshold, a model update prompt is output.

[0127] In this embodiment, the difference between the product demand forecast value obtained by the preset forecast model and the product demand forecast value set manually is determined to determine whether to update the preset forecast model, so that the preset forecast model can be updated in a timely manner.

[0128] In this embodiment, the demand forecast value y is an objective forecast result without human intervention, and the sales plan value p is a plan result that aligns with management objectives and combines the forecast result. By comparing the errors of y and p, if y is closer to the actual sales situation, the way management objectives are set can be further optimized. If p is better, the forecast model can be optimized to provide better guidance to planners. Thus, a closed loop of demand planning management can be formed.

[0129] To better implement the product demand forecasting method in the embodiments of this application, based on the product demand forecasting method, the embodiments of this application also provide a product demand forecasting device, such as... Figure 8 As shown, Figure 8 This is a schematic diagram of an embodiment of the product demand forecasting device provided in this application; the product demand forecasting device includes:

[0130] Module 701 is used to acquire product information of the target product;

[0131] The segmentation module 702 is used to segment each of the target products according to the product information to obtain a product hierarchy diagram;

[0132] The prediction module 703 is used to obtain the baseline node in the product hierarchy diagram, perform product demand prediction on the baseline node, and obtain the baseline demand prediction value of the baseline node.

[0133] The processing module 704 is used to process the baseline demand forecast value of the baseline node according to the cascading relationship of the product hierarchy diagram, and obtain the target demand forecast value of the product corresponding to each node in the product hierarchy diagram.

[0134] In some embodiments of this application, the product demand forecasting device includes:

[0135] Set each node in the product hierarchy diagram as a target node in sequence;

[0136] Based on the product information of the target products associated with the target node, determine the initial demand forecast value and initial forecast error corresponding to the target node;

[0137] Based on the initial demand forecast and initial forecast error of each target node, a baseline level in the product hierarchy diagram is determined, and the nodes in the baseline level are set as baseline nodes.

[0138] In some embodiments of this application, the product demand forecasting device performs the step of determining the initial demand forecast value and initial forecast error corresponding to the target node based on the product information of the target product associated with the target node, including:

[0139] The product information of the target product associated with the target node is input into a preset prediction model. The sales data in the product information is fitted by the prediction model to obtain the initial demand prediction value of the target node for a historical time period.

[0140] The initial prediction error of the target node is calculated based on the initial demand forecast and the actual product demand for the historical time period in the product information.

[0141] In some embodiments of this application, the product demand forecasting device performs the step of determining a baseline level in the product hierarchy diagram based on the initial demand forecast values ​​and initial forecast errors of each target node, and setting the nodes in the baseline level as baseline nodes, including:

[0142] Set each level in the product hierarchy diagram as the target level in sequence;

[0143] The initial demand prediction value and initial prediction error of each node corresponding to the target level are weighted and processed to obtain the hierarchical prediction error of the target level.

[0144] The target level with the smallest hierarchical error is taken as the reference level, and the node corresponding to the reference level is set as the reference node.

[0145] In some embodiments of this application, the processing module 704 in the product demand forecasting device includes:

[0146] Obtain the target level of the node to be predicted in the product hierarchy diagram;

[0147] If the target level is above the baseline level, the baseline demand prediction value is summarized according to the cascading relationship between the node to be predicted and the baseline node in the product hierarchy diagram to obtain the target demand prediction value of the node to be predicted.

[0148] If the target level is below the baseline level, the baseline demand prediction value is split according to the cascading relationship between the node to be predicted and the baseline node in the product hierarchy diagram to obtain the target demand prediction value of the node to be predicted.

[0149] In some embodiments of this application, the prediction module 703 in the product demand prediction device includes:

[0150] Extract the product information of the target product corresponding to the baseline node in the product hierarchy graph;

[0151] The product information of the target product corresponding to the benchmark node is input into the preset prediction model, and the data is fitted through the preset prediction model to obtain the benchmark demand prediction value of the benchmark node.

[0152] In some embodiments of this application, the product demand forecasting device forecasting module 703 includes:

[0153] Obtain a preset sales target and determine the target node corresponding to the preset sales target;

[0154] Based on the cascading relationship between the target node and the baseline node in the product hierarchy diagram, the preset sales targets are summarized or broken down to obtain the baseline demand forecast value of the baseline node.

[0155] In some embodiments of this application, the product demand forecasting device includes:

[0156] Obtain the first baseline demand forecast value determined by the preset forecast model, and obtain the second baseline demand forecast value determined according to the preset sales target;

[0157] Calculate the ratio between the first baseline demand forecast and the second baseline demand forecast. If the ratio exceeds a preset threshold, output a model update prompt.

[0158] In this embodiment, the product demand forecasting device acquires product information of the target products; divides each target product according to the product information to obtain a product hierarchy diagram; acquires a baseline node in the product hierarchy diagram, performs product demand forecasting on the baseline node to obtain a baseline demand forecast value for the baseline node; processes the baseline demand forecast value of the baseline node according to the cascading relationship of the product hierarchy diagram to obtain the target demand forecast value of the product corresponding to each node in the product hierarchy diagram. In this embodiment, the target products are first divided according to the product information of the target products to obtain a product hierarchy diagram. Each node in the product hierarchy diagram corresponds to different dimensions of target products after classification. The cascading relationship between target products at each node can be clearly and accurately determined through the product hierarchy diagram. Then, a baseline node is acquired. The baseline node is the node with the highest prediction accuracy. The server performs product demand forecasting on different nodes according to the baseline node in the product hierarchy diagram to obtain the target demand forecast value of the product corresponding to each node in the product hierarchy diagram, thereby achieving accurate prediction of product demand in different dimensions.

[0159] This invention also provides a product demand forecasting device, such as... Figure 9 As shown, Figure 9 It shows a schematic diagram of the product demand forecasting device involved in the embodiment of the present invention.

[0160] The product demand forecasting device integrates any of the product demand forecasting apparatuses provided in the embodiments of the present invention. The product demand forecasting device is a server, and the product demand forecasting device includes:

[0161] One or more processors;

[0162] Memory; and

[0163] One or more applications, wherein the one or more applications are stored in the memory and configured by the processor to perform the steps of the product demand forecasting method described in any of the embodiments of the above-described product demand forecasting method.

[0164] Specifically, a product demand forecasting device may include components such as a processor 801 with one or more processing cores, a memory 802 with one or more computer-readable storage media, a power supply 803, and an input unit 804. Those skilled in the art will understand that... Figure 9 The product demand forecasting device structure shown does not constitute a limitation on the product demand forecasting device. It may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein:

[0165] The processor 801 is the control center of the product demand forecasting device. It connects various parts of the device via interfaces and lines, and executes software programs and / or modules stored in the memory 802, as well as calling data stored in the memory 802, to perform various functions and process data, thereby providing overall monitoring of the product demand forecasting device. Optionally, the processor 801 may include one or more processing cores; preferably, it may integrate an application processor and a modem processor, where the application processor primarily handles the operating system, user interface, and applications, while the modem processor primarily handles wireless communication. It is understood that the modem processor may not be integrated into the processor 801.

[0166] The memory 802 can be used to store software programs and modules. The processor 801 executes various functional applications and data processing by running the software programs and modules stored in the memory 802. The memory 802 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created based on product demand and predictions of device usage. In addition, the memory 802 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 802 may also include a memory controller to provide the processor 801 with access to the memory 802.

[0167] The product demand forecasting device also includes a power supply 803 that supplies power to various components. Preferably, the power supply 803 can be logically connected to the processor 801 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 803 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0168] The product demand forecasting device may also include an input unit 804, which can be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0169] Although not shown, the product demand forecasting device may also include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 801 in the product demand forecasting device loads the executable files corresponding to the processes of one or more application programs into the memory 802 according to the following instructions, and the processor 801 runs the application programs stored in the memory 802 to realize various functions, as follows:

[0170] Obtain product information for the target product;

[0171] Based on the product information, each target product is divided into categories to obtain a product hierarchy diagram;

[0172] Obtain the baseline node in the product hierarchy diagram, perform product demand forecasting on the baseline node, and obtain the baseline demand forecast value of the baseline node.

[0173] Based on the cascading relationships in the product hierarchy diagram, the baseline demand forecast value of the baseline node is processed to obtain the target demand forecast value of the product corresponding to each node in the product hierarchy diagram.

[0174] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0175] Therefore, embodiments of the present invention provide a computer-readable storage medium, which may include: read-only memory (ROM), random access memory (RAM), a disk, or an optical disk, etc. A computer program is stored thereon, which is loaded by a processor to execute any of the product demand forecasting methods provided in the embodiments of the present invention. For example, the computer program loaded by the processor can perform the following steps:

[0176] Obtain product information for the target product;

[0177] Based on the product information, each target product is divided into categories to obtain a product hierarchy diagram;

[0178] Obtain the baseline node in the product hierarchy diagram, perform product demand forecasting on the baseline node, and obtain the baseline demand forecast value of the baseline node.

[0179] Based on the cascading relationships in the product hierarchy diagram, the baseline demand forecast value of the baseline node is processed to obtain the target demand forecast value of the product corresponding to each node in the product hierarchy diagram.

[0180] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the detailed descriptions of other embodiments above, which will not be repeated here.

[0181] In practice, each of the above units or structures can be implemented as an independent entity or can be arbitrarily combined to be implemented as the same or several entities. For the specific implementation of each of the above units or structures, please refer to the previous method embodiments, which will not be repeated here.

[0182] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0183] The above provides a detailed description of a product demand forecasting method provided by the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A product demand forecasting method, characterized in that, The product demand forecasting method includes: Obtain product information for the target product; Based on the product information, each target product is divided into categories to obtain a product hierarchy diagram; Set each node in the product hierarchy diagram as a target node in sequence; Based on the product information of the target products associated with the target node, determine the initial demand forecast value and initial forecast error corresponding to the target node; Based on the initial demand forecast and initial forecast error of each target node, the baseline level in the product hierarchy diagram is determined, and the nodes in the baseline level are set as baseline nodes. Obtain the baseline node in the product hierarchy diagram, perform product demand forecasting on the baseline node, and obtain the baseline demand forecast value of the baseline node. Based on the cascading relationship of the product hierarchy diagram, the baseline demand forecast value of the baseline node is processed to obtain the target demand forecast value of the product corresponding to each node in the product hierarchy diagram. The initial demand forecast and initial forecast error are used to determine the hierarchical forecast error of each level in the product hierarchy diagram. The benchmark level is the level with the smallest hierarchical forecast error, the benchmark node is a node in the benchmark level, and the benchmark node is the node with the highest forecast accuracy.

2. The product demand forecasting method according to claim 1, characterized by, The step of determining the initial demand forecast value and initial forecast error corresponding to the target node based on the product information of the target product associated with the target node includes: The product information of the target product associated with the target node is input into a preset prediction model. The sales data in the product information is fitted by the prediction model to obtain the initial demand prediction value of the target node for a historical time period. The initial prediction error of the target node is calculated based on the initial demand forecast and the actual product demand for the historical time period in the product information.

3. The product demand forecasting method of claim 1, wherein, The step of determining the baseline level in the product hierarchy diagram based on the initial demand forecast value and initial forecast error of each target node, and setting the nodes in the baseline level as baseline nodes, includes: Set each level in the product hierarchy diagram as the target level in sequence; The initial demand prediction value and initial prediction error of each node corresponding to the target level are weighted and processed to obtain the hierarchical prediction error of the target level. The target level with the smallest hierarchical error is taken as the reference level, and the node corresponding to the reference level is set as the reference node.

4. The product demand forecasting method according to any one of claims 1 to 3, characterized in that, The step of processing the baseline demand forecast value of the baseline node according to the cascading relationship of the product hierarchy diagram to obtain the target demand forecast value of each node in the product hierarchy diagram includes: Obtain the target level of the node to be predicted in the product hierarchy diagram; If the target level is above the baseline level, the baseline demand prediction value is summarized according to the cascading relationship between the node to be predicted and the baseline node in the product hierarchy diagram to obtain the target demand prediction value of the node to be predicted. If the target level is below the baseline level, the baseline demand prediction value is split according to the cascading relationship between the node to be predicted and the baseline node in the product hierarchy diagram to obtain the target demand prediction value of the node to be predicted.

5. The product demand forecasting method according to claim 1, characterized in that, The step of obtaining the baseline node in the product hierarchy diagram, performing product demand forecasting on the baseline node, and obtaining the baseline demand forecast value of the baseline node includes: Extract the product information of the target product corresponding to the baseline node in the product hierarchy graph; The product information of the target product corresponding to the benchmark node is input into the preset prediction model, and the data is fitted through the preset prediction model to obtain the benchmark demand prediction value of the benchmark node.

6. The product demand forecasting method of claim 1, wherein, The step of obtaining the baseline node in the product hierarchy diagram, performing product demand forecasting on the baseline node, and obtaining the baseline demand forecast value of the baseline node includes: Obtain a preset sales target and determine the target node corresponding to the preset sales target; Based on the cascading relationship between the target node and the baseline node in the product hierarchy diagram, the preset sales targets are summarized or broken down to obtain the baseline demand forecast value of the baseline node.

7. The product demand forecasting method according to claim 5 or 6, characterized by, After processing the baseline demand forecast value of the baseline node according to the cascading relationship of the product hierarchy diagram to obtain the target demand forecast value of the product corresponding to each node in the product hierarchy diagram, the method further includes: Obtain the first baseline demand forecast value determined by the preset forecast model, and obtain the second baseline demand forecast value determined according to the preset sales target; Calculate the ratio between the first baseline demand forecast and the second baseline demand forecast. If the ratio exceeds a preset threshold, output a model update prompt.

8. A product demand forecasting apparatus characterized by comprising: The product demand forecasting device includes: The acquisition module is used to acquire product information of the target product; The segmentation module is used to segment each of the target products according to the product information to obtain a product hierarchy diagram; The prediction module is used to obtain the baseline node in the product hierarchy diagram, perform product demand prediction on the baseline node, and obtain the baseline demand prediction value of the baseline node. The processing module is used to process the baseline demand forecast value of the baseline node according to the cascading relationship of the product hierarchy diagram, and obtain the target demand forecast value of the product corresponding to each node in the product hierarchy diagram. The product demand forecasting device is also used to set each node in the product hierarchy diagram as a target node in sequence. Based on the product information of the target products associated with the target node, determine the initial demand forecast value and initial forecast error corresponding to the target node; Based on the initial demand forecast and initial forecast error of each target node, the baseline level in the product hierarchy diagram is determined, and the nodes in the baseline level are set as baseline nodes. The initial demand forecast and initial forecast error are used to determine the hierarchical forecast error of each level in the product hierarchy diagram. The benchmark level is the level with the smallest hierarchical forecast error, the benchmark node is a node in the benchmark level, and the benchmark node is the node with the highest forecast accuracy.

9. A product demand forecasting device characterized by comprising: The product demand forecasting equipment includes: One or more processors; Memory; and One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the steps in the product demand forecasting method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores a computer program, which is loaded by a processor to perform the steps in the product demand forecasting method according to any one of claims 1 to 7.

Images