Product sales volume prediction method and device, equipment and medium

By obtaining sales information of target products and competitors, establishing prediction models and promotional diversion impact models, and adjusting sales forecasts, the problem of inaccurate prediction in the existing technology is solved, and the accuracy of sales forecasts and the efficiency of marketing plans is improved.

CN119006032APending Publication Date: 2024-11-22SHENGDOUSHI SHANGHAI SCI & TECH DEV CO LTD
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202310554479.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-05-16
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The prior art artificial analysis of historical data in product sales forecasts is inefficient and fails to accurately estimate the impact of competitor promotion activities on target product sales, resulting in inaccurate predictions.

Method used

By obtaining sales characteristic information of the target product and promotional sales increase information of competitors, establish sales forecast models and promotional diversion impact forecast models, and adjust predictive sales to reflect the impact of competitors' promotion activities.

Benefits of technology

It improves the accuracy and interpretability of sales forecasts, can more accurately estimate the impact of competitor promotions on target product sales, and improves the efficiency of marketing plans.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119006032A_ABST
    Figure CN119006032A_ABST
Patent Text Reader

Abstract

The invention provides a product sales prediction method and device, equipment and a medium, and relates to the technical field of computers, in particular to the technical field of data processing. According to the implementation scheme, sales volume feature information of a target product is acquired, wherein the sales volume feature information comprises corresponding historical daily sales volume information of the target product when sales promotion is not performed; determining a predicted sales volume of the target product in the target time period based on the sales volume feature information of the target product; promotion sales volume increase information of a competitive product corresponding to the first target product is acquired, the promotion sales volume increase information of the competitive product can influence the sales volume of the target product, and the promotion sales volume increase information is determined based on historical promotion sales volume information corresponding to promotion of the competitive product and historical daily sales volume information corresponding to non-promotion of the competitive product; and adjusting the predicted sales volume of the target product in the target time period at least based on the promotion sales volume increase information of the competitive product corresponding to the target product.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the field of computers, in particular to the field of data processing technology, and specifically to a method, device, electronic device, computer-readable storage medium, and computer program product for predicting sales of a product. Background Art

[0002] When formulating marketing and production plans, the accuracy of product sales forecasts can directly affect the product's production costs, sales costs and marketing effectiveness.

[0003] The methods described in this section are not necessarily methods that have been previously conceived or employed. Unless otherwise indicated, it should not be assumed that any method described in this section is considered to be prior art simply because it is included in this section. Similarly, unless otherwise indicated, the issues mentioned in this section should not be considered to have been recognized in any prior art. Summary of the invention

[0004] The present disclosure provides a product sales forecasting method, device, electronic device, computer-readable storage medium, and computer program product.

[0005] According to one aspect of the present disclosure, a sales forecasting method for a product is provided, comprising: obtaining sales characteristic information of a target product, the sales characteristic information comprising historical daily sales information corresponding to a target product without promotion; determining predicted sales of the target product within a target time period based on the sales characteristic information of the target product; obtaining promotional sales increase information of a competing product corresponding to the target product, the promotional sales increase information of the competing product being able to affect the sales of the target product, the promotional sales increase information being determined based on historical promotional sales information corresponding to a promotion and historical daily sales information corresponding to a promotion without promotion of the competing product; and adjusting predicted sales of the target product within the target time period based at least on the promotional sales increase information of the competing product corresponding to the target product.

[0006] According to another aspect of the present disclosure, a method for predicting raw material shipments of a product is provided, comprising: determining the predicted sales volume of a target product within a target time period using the sales volume prediction method as described above; determining the raw material demand for the target product within the target time period based on the predicted sales volume of the target product within the target time period; and determining the raw material shipment volume of the target product within the target time period based on the raw material demand for the target product within the target time period.

[0007] According to another aspect of the present disclosure, a sales prediction device for a product is provided, comprising: a first acquisition unit, configured to acquire sales characteristic information of a target product, the sales characteristic information comprising historical daily sales information corresponding to a target product when there is no promotion; a first determination unit, configured to determine the predicted sales of the target product within a target time period based on the sales characteristic information of the target product; a second acquisition unit, configured to acquire promotional sales increase information of a competing product corresponding to the target product, the promotional sales increase information of the competing product being able to affect the sales of the target product, the promotional sales increase information being determined based on historical promotional sales information corresponding to a promotion and historical daily sales information corresponding to a promotion without a promotion of the competing product; and an adjustment unit, configured to adjust the predicted sales of the target product within the target time period based at least on the promotional sales increase information of the competing product corresponding to the target product.

[0008] According to another aspect of the present disclosure, a product shipment prediction device is provided, comprising: the sales volume prediction device for the product as described above, configured to determine the predicted sales volume of a target product within a target time period; a second determination unit, configured to determine the raw material demand of the target product within the target time period based on the predicted sales volume of the target product within the target time period; and a third determination unit, configured to determine the raw material shipment volume of the target product within the target time period based on the raw material demand of the target product within the target time period.

[0009] According to another aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute any of the above methods.

[0010] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute any of the above methods.

[0011] According to another aspect of the present disclosure, a computer program product is provided, comprising a computer program, wherein the computer program implements any of the above methods when executed by a processor.

[0012] According to one or more embodiments of the present disclosure, the accuracy of product sales forecasting can be improved.

[0013] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The accompanying drawings illustrate embodiments in some examples and constitute a part of the specification, and together with the text description of the specification, are used to explain exemplary implementations of the embodiments. The embodiments shown are for illustrative purposes only and do not limit the scope of the claims. In all drawings, the same reference numerals refer to similar but not necessarily identical elements.

[0015] Figure 1 A flowchart showing a method for predicting sales volume of a product according to an exemplary embodiment of the present disclosure is shown;

[0016] Figure 2 A schematic diagram showing a sales volume prediction model and a promotion diversion impact prediction model according to an exemplary embodiment of the present disclosure is shown;

[0017] Figure 3 A flowchart showing a method for training a sales volume prediction model and a promotion diversion impact prediction model according to an exemplary embodiment of the present disclosure is shown;

[0018] Figure 4 A flowchart showing a method for predicting sales volume of a product according to an exemplary embodiment of the present disclosure is shown;

[0019] Figure 5 A flow chart showing a method for predicting raw material shipment volume of a product according to an exemplary embodiment of the present disclosure is shown;

[0020] Figure 6 A flowchart of a method for training a loss prediction model according to an exemplary embodiment of the present disclosure is shown;

[0021] Fig. 7A and Figure 7B A schematic diagram showing a historical daily sales curve of a product according to an exemplary embodiment of the present disclosure

[0022] Figure 8 A structural block diagram of a sales volume prediction device for a product according to an exemplary embodiment of the present disclosure is shown;

[0023] Fig. 9 A structural block diagram of a device for predicting the raw material shipment volume of a product according to an exemplary embodiment of the present disclosure is shown;

[0024] Fig.10 A block diagram of an electronic device according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0025] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope of the present disclosure. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0026] In the present disclosure, unless otherwise specified, the use of the terms "first", "second", etc. to describe various elements is not intended to limit the positional relationship, timing relationship, or importance relationship of these elements, and such terms are only used to distinguish one element from another element. In some examples, the first element and the second element may refer to the same instance of the element, and in some cases, based on the description of the context, they may also refer to different instances.

[0027] The terms used in the description of various examples in this disclosure are only for the purpose of describing specific examples and are not intended to be limiting. Unless the context clearly indicates otherwise, if the number of elements is not specifically limited, the element can be one or more. In addition, the term "and / or" used in this disclosure covers any one of the listed items and all possible combinations.

[0028] In actual application scenarios, it is necessary to predict product sales and formulate product production and marketing plans based on the predicted sales, so as to more efficiently produce and sell corresponding products, such as meals. In related technologies, historical sales data is usually analyzed manually, and product sales are predicted based on the results of data analysis. This method has high labor costs and low efficiency, and fails to accurately estimate the impact of various promotional activities on the sales of target products.

[0029] Based on this, the present disclosure provides a product sales forecasting method, which first determines the predicted sales of the product based on the sales feature information of the product itself, and then adjusts the predicted sales based on the promotional sales information of competing products, so that the promotional sales information of competing products can be explicitly introduced in the sales forecasting process to accurately estimate the impact of competing product promotion activities on the sales of target products, thereby improving the interpretability and accuracy of sales forecasts.

[0030] The embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0031] Figure 1 FIG. 1 is a flowchart of a method 100 for predicting sales volume of a product according to an exemplary embodiment of the present disclosure. Figure 1 As shown, the method 100 includes:

[0032] Step S101, obtaining sales characteristic information of a target product, where the sales characteristic information includes historical daily sales information corresponding to the target product when there is no promotion;

[0033] Step S102: determining the predicted sales volume of the target product within the target time period based on the sales volume characteristic information of the target product;

[0034] Step S103, obtaining promotional sales increase information of a competing product corresponding to the target product, wherein the promotional sales increase information of the competing product can affect the sales of the target product, and the promotional sales increase information is determined based on the historical promotional sales information corresponding to the competitive product when there is a promotion and the historical daily sales information corresponding to the competitive product when there is no promotion; and

[0035] Step S104: Adjust the predicted sales volume of the target product within the target time period based at least on the promotional sales volume increase information of the competing product corresponding to the target product.

[0036] The inventors note that in actual application scenarios, the sales volume of the target product will also be affected by the promotional activities of its competitors. In one example, when method 100 is applied to the catering business, the sales volume of the same category of food or beverages will inevitably affect each other. Therefore, by using the sales volume forecasting method of the product in the embodiment of the present disclosure, the promotional sales volume information of the competitors can be explicitly introduced into the sales volume forecasting process to accurately estimate the impact of the promotional activities of the competitors on the sales volume of the target product, thereby improving the interpretability and accuracy of the sales volume forecast.

[0037] In some examples, the target time period may be set according to actual needs. For example, when planning to promote a target product, the promotion period may be set as the target time period, so that the predicted sales volume of the target product can be more accurately determined for the product promotion period, so as to formulate a more accurate and efficient marketing plan and improve the promotion effect.

[0038] In some examples, the sales feature information may also include historical sales information of the target product. By using the historical sales information of the target product to predict the predicted sales of the product in a future target time period, the accuracy of the prediction can be improved.

[0039] In actual application scenarios, the target product may be an existing product with a historical sales record, or a new product without a historical sales record. For existing products that have been sold and new products that have not been sold, different methods can be used to determine the sales feature information of the target product, so as to more accurately determine its predicted sales.

[0040] In some examples, the competing products may be determined based on preset rules, for example, multiple products may be classified, and other products belonging to the same category as the target product may be considered as competing products of the target product. In one example, when the method 100 is applied to a catering business, the multiple products included in the catering business may be classified, for example, into staple food, snacks, desserts, beverages, etc., so that the competing products of each product can be determined based on the classification results.

[0041] According to some embodiments, determining the predicted sales volume of the target product within the target time period based on the sales volume characteristic information of the target product in step S102 includes: inputting the sales volume characteristic information of the target product into a sales volume prediction model to obtain the predicted sales volume of the target product output by the sales volume prediction model. Thus, the sales volume prediction model can be used to simply and accurately determine the predicted sales volume of the target product.

[0042] In some examples, the initial sales prediction model may be established based on various types of neural network structures, such as feedforward neural networks, feedback neural networks, recurrent neural networks, etc. Furthermore, the established initial sales prediction model may be trained using labeled sample data to obtain a sales prediction model. The training process of the sales prediction model may include, for example: obtaining sales feature information of sample products and the actual sales of sample products; inputting the sales feature information of sample products into the initial sales prediction model to obtain the predicted sales of sample products output by the initial sales prediction model; calculating the loss value based on the actual sales and predicted sales; and adjusting the parameters of the initial sales prediction model based on the loss value. By repeatedly performing the above-mentioned parameter adjustment steps using the training data set, a sales prediction model with optimized performance can be obtained to more accurately obtain the predicted sales of the target product.

[0043] According to some embodiments, in step S104, adjusting the predicted sales volume of the target product in the target time period based at least on the promotion sales volume increase information of the competing product corresponding to the target product includes: determining the promotion diversion influence coefficient based on the sales volume characteristic information of the target product and the promotion sales volume increase information of the competing product corresponding to the target product, the promotion diversion influence coefficient being used to characterize the degree of influence of the promotion of the competing product on the sales volume of the target product; and adjusting the predicted sales volume of the target product in the target time period based on the promotion diversion influence coefficient. Thus, the promotion diversion influence coefficient can be determined in combination with the promotion sales volume increase information of the competing product corresponding to the target product, and the promotion diversion influence coefficient is used to characterize the degree of influence of the promotion activities of the competing product on the sales volume of the target product, so as to adjust the predicted sales volume of the target product based on this.

[0044] In some examples, a preset adjustment formula may be used to adjust the predicted sales volume of the target product based on the promotion diversion impact coefficient.

[0045] For example, the following formula may be used to adjust the predicted sales volume of the target product within the target time period:

[0046]

[0047] In the formula, y0 is the initial predicted sales volume of the target product determined based on the sales volume characteristic information of the target product using the method described above, c is the promotion diversion impact coefficient of the target product's competitor on the target product's sales volume, and x is the promotion sales increase information of the target product's competitor. Based on this, the predicted sales volume of the target product can be adjusted more simply and accurately.

[0048] According to some embodiments, the method further includes: determining a product sales feature vector corresponding to the sales feature information, wherein, based on the sales feature information of the target product and the promotion sales increase information of the competing product corresponding to the target product, determining the promotion diversion impact coefficient includes: inputting the product sales feature vector and the promotion sales increase information of the competing product corresponding to the target product into the promotion diversion impact prediction model to obtain the promotion diversion impact coefficient output by the promotion diversion impact prediction model. Thus, the product sales feature vector output by the hidden layer of the sales prediction model can be used to efficiently and accurately indicate the sales feature information of the target product, so that the promotion diversion impact prediction model can obtain a more accurate promotion diversion impact coefficient.

[0049] In some embodiments, when product sales feature information is obtained using a sales prediction model, the sales prediction model may include a hidden layer, so that a product sales feature vector corresponding to the sales feature information output by the hidden layer can be obtained.

[0050] In actual application scenarios, there may be multiple competing products corresponding to the target product, or there may be multiple promotional activities for one competing product corresponding to the target product. Based on this, according to some embodiments, the promotional sales increase information of the competing product corresponding to the target product includes multiple sub-competing product promotional sales increase information corresponding to multiple competing product promotional activities, and the promotional diversion impact prediction model includes a feature determination module and an influence coefficient calculation module, and wherein the product sales feature vector and the promotional sales increase information of the competing product corresponding to the target product are input into the promotional diversion impact prediction model to obtain the promotional diversion impact coefficient output by the promotional diversion impact prediction model, including: inputting the product sales feature vector and the multiple sub-competing product promotional sales increase information into the feature determination module to obtain multiple sub-competing product activity impact features corresponding to multiple competing product promotional activities output by the feature determination module, each sub-competing product activity impact feature indicating the degree to which the corresponding competing product promotional activity affects the sales of the target product; and inputting the predicted sales of the target product and the multiple sub-competing product activity impact features into the influence coefficient calculation module to obtain the promotional diversion impact coefficient output by the influence coefficient calculation module. Therefore, for multiple competitive product promotion activities, the impact characteristics of multiple sub-competitive product activities can be used to indicate the impact of each competitive product promotion activity on the sales volume of the target product, thereby improving the accuracy and explainability of sales volume forecasting.

[0051] In one example, when there are m different types of competitive product promotion activities, the sub-competitive product promotion sales increase information CE can be used m (m≥0) to indicate the sales increase information corresponding to each competitive product promotion activity. In this case, the sales increase information of m sub-competitive product promotions can be concatenated with the product sales feature vector to obtain m sub-competitive product activity impact features {E1, E2, …, E m}, so that feature E can be affected based on sub-competitive product activities m The promotion diversion impact coefficient is determined by combining the predicted sales volume y0 of the target product. In some embodiments, the feature determination module may include a multilayer perceptron and a feature extraction network, so that the product sales feature vector may be input into the multilayer perceptron, and the result output by the multilayer perceptron is then input into the feature extraction network to obtain further optimized feature information for the target product, and then combined with the promotion sales increase information of multiple sub-competitive products to obtain multiple sub-competitive product activity impact features.

[0052] In one example, the promotion diversion impact coefficient c may be determined using the following formula:

[0053]

[0054] In actual application scenarios, the sales volume of the target product will also be affected by its own promotional activities. Based on this, according to some embodiments, method 100 also includes: obtaining promotional sales increase information of the target product, wherein in step S104, adjusting the predicted sales volume of the target product in the target time period based at least on the promotional sales increase information of the competing product corresponding to the target product includes: adjusting the predicted sales volume of the target product in the target time period based on the promotional sales increase information of the target product and the promotional sales increase information of the competing product corresponding to the product. Thus, the impact of promotions on product sales can be fully considered, and the predicted sales volume of the product can be more accurately determined. Thus, the predicted sales volume of the target product in the promotion time period can be more accurately determined in combination with the promotional sales increase information of the target product.

[0055] According to some embodiments, obtaining the promotion sales increase information of the target product includes: obtaining the historical promotion sales information of the target product when there is a promotion; and determining the promotion sales increase information of the target product based on the historical daily sales information and the historical promotion sales information of the target product. By using the historical daily sales information and the historical promotion sales information of the target product for prediction, the predicted sales of the target product can be determined more simply and accurately, and the raw material shipment volume can be predicted more accurately.

[0056] In some examples, after determining the corresponding promotional sales increase information based on historical daily sales information and historical promotional sales information, the sum of the initial predicted sales and the promotional sales increase information is calculated, and the calculation result is used as the predicted sales of the product in the target time period.

[0057] It should be understood that, generally speaking, the impact of promotion on product sales is directly related to the strength of the promotion. Based on this, in some examples, other information can be further combined to determine the promotion sales increase information of the target product, for example, information that can characterize the promotion strength of the target product, such as discount information, gift value information, etc., can be included, so that the promotion sales increase information of the target product can be determined more accurately, so as to more accurately determine the predicted sales of the target product.

[0058] According to some embodiments, adjusting the predicted sales volume of the target product within the target time period based on the promotional sales volume increase information of the target product and the promotional sales volume increase information of the competitor corresponding to the target product includes: determining the self-promotional impact amount and the promotional diversion impact coefficient based on the sales volume characteristic information of the target product, the promotional sales volume increase information of the target product and the promotional sales volume increase information of the competitor corresponding to the target product; and adjusting the predicted sales volume of the target product within the target time period based on the self-promotional impact amount and the promotional diversion impact coefficient. Thus, the sales volume characteristic information of the target product and the promotional sales volume increase information corresponding to each of the target product and the competitor can be combined to more accurately determine the impact of the promotional activities of the target product itself and the promotional activities of the competitor on the sales volume of the target product, so as to obtain a more accurate predicted sales volume of the target product.

[0059] In one example, based on obtaining the product sales feature vector of the target product output by the hidden layer of the sales prediction model, the product sales feature vector, the target product's promotional sales increase information, and the promotional sales increase information of the competing product corresponding to the target product are input into the promotion diversion impact prediction model to obtain the promotional impact amount and promotional diversion impact coefficient output by the promotion diversion impact prediction model to improve the prediction efficiency and accuracy.

[0060] In actual application scenarios, there may be multiple promotional activities for the target product. Based on this, in some examples, the promotional sales increase information of the target product includes multiple sub-target product promotional sales increase information corresponding to multiple target product promotion activities, and wherein, the product sales feature vector, multiple sub-competitive product promotional sales increase information and multiple sub-target product promotional sales increase information may be input into the feature determination module using the promotional diversion impact prediction model described above, so as to obtain multiple sub-competitive product activity impact features corresponding to multiple competitive product promotion activities and multiple sub-target product activity impact features corresponding to multiple target product promotion activities output by the feature determination module, and each sub-target product activity impact feature is used to indicate the degree to which the corresponding self-promotional activity affects the sales of the target product, and then the predicted sales of the target product, multiple sub-competitive product activity impact features and multiple sub-target product activity impact features are input into the influence coefficient calculation module to obtain the promotional diversion impact coefficient and self-promotion impact amount output by the influence coefficient calculation module. Thus, for multiple target product promotion activities, multiple sub-target product activity impact features can be used to indicate the impact of each self-promotion activity of the target product on its own sales, thereby improving the accuracy and interpretability of sales prediction.

[0061] In one example, when there are n different target product activities, the sub-target product promotion sales increase information SE can be used n(n≥0) to indicate the sales increase information corresponding to the promotion activities of each target product. In this case, the sales increase information of the n sub-target products can be concatenated with the product sales feature vector to obtain n sub-target product activity image features corresponding to their own activities {E m+1 ,E m+2 ,…,E m+n}, so that it can be based on the sub-target product activity image feature E m+n And the predicted sales volume y0 of the target product is used to determine the impact of its own promotion.

[0062] In one example, the following formula may be used to determine the self-promotion influence x:

[0063]

[0064] In some examples, the following formula may be used to adjust the predicted sales of a product within a target time period:

[0065]

[0066] Wherein, y0 is the predicted sales volume of the product preliminarily determined based on the sales volume characteristic information of the target product using the method described above, c is the promotion diversion influence coefficient indicating the influence of the target product's competitor products on the sales volume of the target product, and x is the product's own promotion sales increase information.

[0067] Figure 2 A schematic diagram of a sales volume prediction model and a promotion diversion impact prediction model according to an exemplary embodiment of the present disclosure is shown.

[0068] like Figure 2 As shown, the sales prediction model is used to output the predicted sales of the target product based on the sales feature information of the target product. And the sales prediction model includes a hidden layer, and the product sales feature vector corresponding to the sales feature information output by the hidden layer can be obtained in the sales prediction process, so that the promotion diversion impact prediction model can determine the impact of the promotion activities of at least one of the target product and the corresponding competitor of the target product on the sales of the target product based on the product sales feature vector. In this example, the promotion diversion impact prediction model includes a feature determination module and an impact coefficient calculation module. The feature determination module can output the competitive product activity impact feature and the self-activity impact feature based on the product sales feature vector, the competitive product promotion sales increase information and the target product promotion sales increase information. The impact coefficient calculation module can calculate the self-promotion impact amount and the promotion diversion impact coefficient based on the competitive product activity impact feature and the self-activity impact feature, so that the initial predicted sales can be adjusted based on this to obtain a more accurate predicted sales of the target product.

[0069] In this example, the predicted sales adjustment module can obtain the adjusted predicted sales of the target product based on the predicted sales y0 of the target product output by the sales prediction model and the promotion diversion impact prediction model, the promotion diversion impact coefficient c and the target product's own promotion influence x. For example, the adjusted predicted sales y can be calculated using the preset formula y=f(y0,x,c).

[0070] According to some embodiments, sample data may be used to jointly perform supervised training on a sales volume prediction model and a promotion diversion impact prediction model. Figure 3 FIG. 3 is a flowchart of a method 300 for training a sales volume prediction model and a promotion diversion impact prediction model according to an exemplary embodiment of the present disclosure. Figure 3 As shown, the training method 300 includes:

[0071] Step S301, obtaining sales characteristic information of the sample product, promotional sales increase information of the competing products corresponding to the sample product, and the actual sales volume of the sample product;

[0072] Step S302: input the sales feature information of the sample product into the initial sales prediction model to obtain the predicted sales and product sales feature vector of the sample product in the target time period output by the initial sales prediction model;

[0073] Step S303: input the product sales feature vector of the sample product and the promotion sales increase information of the competing product corresponding to the sample product into the promotion diversion impact prediction model to obtain the promotion diversion impact coefficient output by the promotion diversion impact prediction model;

[0074] Step S304: adjusting the predicted sales volume of the sample product within the target time period based on the promotion diversion impact coefficient;

[0075] Step S305, calculating the loss value based on the actual sales volume and the adjusted predicted sales volume; and

[0076] Step S306: Based on the loss value, adjust the parameters of the sales forecasting model and the promotion diversion impact forecasting model.

[0077] Therefore, by using labeled data for supervised training, the prediction accuracy of the model can be improved to obtain more accurate predicted sales of the target product. In some examples, the sample data can also include the promotional sales increase information of the sample product itself, so that the promotion diversion impact prediction model can obtain more accurate self-promotion impact and improve the product sales prediction accuracy.

[0078] In some examples, the loss value in step S305 may be an absolute mean error calculated based on the actual sales volume and the adjusted predicted sales volume, but is not limited thereto. In the actual training process, a corresponding loss value calculation method may be set according to the needs, for example, the mean square error may be calculated, which is not limited thereto.

[0079] Figure 4 FIG. 4 is a flowchart of a method 400 for predicting sales volume of a product according to an exemplary embodiment of the present disclosure. Figure 4 As shown, the method 400 includes:

[0080] Step S401, obtaining sales characteristic information of a target product, where the sales characteristic information includes historical daily sales information corresponding to the target product when there is no promotion;

[0081] Step S402: inputting the sales feature information of the target product into the sales prediction model to obtain the predicted sales of the target product output by the sales prediction model;

[0082] Step S403, obtaining a product sales feature vector corresponding to the sales feature information output by the hidden layer of the sales prediction model;

[0083] Step S404: Acquire promotional sales increase information of the target product and promotional sales increase information of the competing products corresponding to the target product;

[0084] Step S405: input the product sales feature vector, the promotion sales increase information of the target product, and the promotion sales increase information of the competing products corresponding to the target product into the promotion diversion impact prediction model to obtain the own promotion impact amount and the promotion diversion impact coefficient; and

[0085] Step S406: Adjust the predicted sales volume of the target product within the target time period based on the promotion influence and the promotion diversion influence coefficient.

[0086] Therefore, the sales characteristic information of the target product and the promotional sales increase information corresponding to the target product and the competing products can be combined to more accurately determine the impact of the target product's own promotional activities and the competing products' promotional activities on the sales of the target product, so as to obtain a more accurate predicted sales volume of the target product.

[0087] In actual application scenarios, new products that have not been sold do not have corresponding historical sales information. However, among the existing products that have been sold, there are usually similar products that are relatively similar to the new product. Based on the historical sales information of similar products, the predicted sales of the new product can also be predicted. Based on this, in some embodiments, the target product is a product that has not been sold, wherein obtaining the sales characteristic information of the target product includes: determining similar products that have been sold corresponding to the target product; and obtaining the historical daily sales information of similar products. In this way, the predicted sales of the new product can be predicted based on the historical sales information of similar existing products that are relatively similar to the new product.

[0088] In some examples, similar products can be determined based on preset rules, for example, multiple products can be classified, and then other products that belong to the same category as the target product and have similar raw materials can be used as similar products of the target product. In one example, when method 100 is applied to a catering business, the multiple products included in the catering business can be classified, for example, into staple food, snack food, dessert, beverage, etc. When the target product belongs to the staple food category, similar products whose raw materials are relatively close to the target product can be determined from the corresponding existing products of the staple food category.

[0089] In some examples, competitors and similar products for the same target product may be the same or different. As described above, when method 100 is applied to the catering business, a relatively coarse-grained classification method of catering products may be configured, such as including staple food, snacks, desserts, beverages, etc., and then other products belonging to the same category as the target product are used as competitors of the target product to more comprehensively analyze the impact of other products on the sales of the target product. At the same time, a more fine-grained classification method of catering products may be configured, further introducing classification dimensions such as raw material category and price level, and determining similar products that are relatively close to the target product in each classification dimension based on a more fine-grained classification method, so as to use the sales information of similar products to more accurately indicate the possible sales information of the target product.

[0090] Further, it should be understood that, generally speaking, the target product cannot completely replace the corresponding similar products. Based on this, in some embodiments, based on the sales characteristic information of the target product, determining the predicted sales of the target product in the target time period includes: determining the substitution coefficient between the target product and similar products; and determining the predicted sales of the target product in the target time period based on the substitution coefficient between the target product and similar products and the historical daily sales information of similar products. In this way, the differences between the target product and similar products can be fully considered, and the predicted sales of the target product can be determined more accurately.

[0091] In some examples, determining the predicted sales volume of a product within a target time period based on a substitution coefficient between the target product and similar products, and historical daily sales information and / or historical promotional sales information of similar products includes: determining the initial predicted sales volume of the target product within the target time period based on historical daily sales information and / or historical promotional sales information of similar products; and determining the predicted sales volume of the target product based on the substitution coefficient between the target product and similar products and the initial predicted sales volume of the target product. For example, the product of the substitution coefficient between the target product and similar products and the initial predicted sales volume of the target product may be calculated and determined as the predicted sales volume of the target product.

[0092] In some embodiments, the target product includes existing products and new products. For example, when method 100 is applied to a catering business, the product may be a new set meal consisting of a plurality of meals and drinks, and the new set meal is obtained by replacing the existing meals in the old set meal with the new meals. In this case, the predicted sales volume of the new set meal may be determined in the following manner: based on the historical daily sales volume information and / or historical promotional sales volume information of the replaced old set meal, the initial predicted sales volume of the new set meal in the target time period is determined; and based on the substitution coefficient between the new meal meal and the replaced existing meal meal and the initial predicted sales volume of the new set meal, the predicted sales volume of the new set meal is determined.

[0093] According to some embodiments, determining the substitution coefficient between the target product and similar products includes: determining at least one competing product corresponding to the similar product; and determining the substitution coefficient between the target product and similar products based on historical sales information of the similar product and historical sales information of at least one competing product corresponding to the similar product. Thus, the accuracy of the determined substitution coefficient between the target product and similar products can be improved, thereby improving the accuracy of the predicted sales.

[0094] As described above, similar products and at least one competing product corresponding to the similar products can be determined based on preset rules. In one example, when method 100 is applied to a catering business, the various products included in the catering business can be classified, for example, into staple food, snack food, dessert, beverage, etc. When the target product belongs to the staple food category, similar products whose raw materials are relatively close to the target product can be determined from the existing products corresponding to the staple food category, and then other existing products corresponding to the staple food category can be determined as competing products of the similar product.

[0095] In some examples, the substitution coefficient can also be determined more accurately based on other information. For example, when the method 100 is applied to the catering business, the historical number of consumers of similar products can be determined based on the historical sales information of similar products, and then the substitution coefficient between the target product and the similar product can be determined based on the ratio of the historical number of consumers to the total number of historical consumers of the corresponding restaurant. Furthermore, the substitution coefficient can also be further corrected by using the flow of people at the location of the corresponding restaurant, the grade information of the corresponding restaurant, etc., to improve its accuracy.

[0096] According to another aspect of the present disclosure, a method for predicting raw material shipment volume of a product is also provided. Figure 5 FIG. 5 is a flowchart of a method 500 for predicting the raw material shipment volume of a product according to an exemplary embodiment of the present disclosure. Figure 5 As shown, the method 500 includes:

[0097] Step S501, using method 100 to determine the predicted sales volume of a target product within a target time period;

[0098] Step S502: determining the raw material demand of the target product within the target time period based on the predicted sales volume of the target product within the target time period; and

[0099] Step S503: Determine the raw material shipment volume of the target product within the target time period based on the raw material demand volume of the target product within the target time period.

[0100] In some examples, the raw material shipment volume indicates the amount of raw materials that need to be provided to the product manufacturer and seller, that is, this indicator is applied to the inventory management of raw materials. The raw material demand volume indicates the amount of raw materials actually consumed by the product manufacturer and seller in the process of making the product. It can be understood that the raw material shipment volume, on the basis of covering the raw material demand volume, also needs to fully consider the loss of raw materials during transportation and warehousing. By more accurately determining the raw material shipment volume, it is possible to reduce waste and reduce costs while meeting production needs.

[0101] In actual application scenarios, it is necessary to determine the raw material shipment volume of a product based on the predicted sales volume of the product, and based on this, provide the raw materials of the product to the product manufacturer and seller so that they can make and sell the corresponding products, such as meals, etc. In related technologies, data analysis is usually performed manually, and then the raw material shipment volume of the product is determined based on the results of the data analysis. This method has high labor costs and low efficiency.

[0102] Based on this, the present disclosure provides a product shipment prediction method, which can determine the raw material demand based on the predicted sales volume of the target product obtained by using method 100, and then determine the corresponding raw material shipment volume to improve the prediction efficiency and accuracy of the raw material shipment volume.

[0103] In some examples, the target product may include various products in multiple application scenarios. For example, when the method 100 is applied to the catering business, the products may include different types of meals or drinks, and the raw materials may include various types of ingredients corresponding to the various meals or drinks. For another example, when the method 100 is applied to the clothing manufacturing industry, the products may include various types of clothing, and the raw materials may include various types of fabrics corresponding to the various clothing. It should be understood that the method 100 may also be applied to other scenarios, without limitation.

[0104] Therefore, the predicted sales volume of the product can be determined based on the sales volume characteristic information of the product. It should be understood that, generally speaking, the raw material ingredient list of the product is fixed. When the predicted sales volume of the product is determined, the raw material demand of the target product can be obtained based on the corresponding raw material ingredient list. When the raw material demand of the product is determined, the influence of raw material loss and the like can be further considered to determine the corresponding raw material shipment volume, thereby improving the prediction efficiency and accuracy of the raw material shipment volume.

[0105] In some examples, there is a certain linear mapping relationship between the predicted sales volume of the target product and the raw material demand of the target product. Based on this, in step S503, the raw material demand of the target product can be obtained based on the corresponding certain linear mapping relationship.

[0106] According to some embodiments, determining the raw material shipment quantity of the target product within the target time period based on the raw material demand quantity of the target product within the target time period in step S503 includes: determining the raw material loss information of the target product within the target time period; and determining the raw material shipment quantity of the target product within the target time period based on the raw material demand quantity and the raw material loss information of the target product within the target time period. Thus, the loss of raw materials during the process of raw material transportation, product production, etc. can be fully considered, and the raw material shipment quantity of the target product can be determined more accurately based on this.

[0107] In some examples, the raw material loss information may be in the form of a raw material loss rate, and the raw material shipment volume of the target product may be determined using the following formula:

[0108] b=a×r

[0109] In the formula, b is the raw material shipment volume of the target product, a is the raw material demand of the target product, and r is the raw material loss rate of the target product. Based on this, the raw material shipment volume of the target product can be accurately and efficiently determined.

[0110] Furthermore, the neural network model may be used to predict the raw material loss information of the product within the target time period. In some embodiments, determining the raw material loss information of the target product within the target time period includes: inputting the raw material demand and raw material shipment volume of the target product within a preset time period before the target time period into the loss prediction model to obtain the raw material loss information of the target product within the target time period output by the loss prediction model. Thus, the neural network model can be used to efficiently and accurately predict the raw material loss information of the target product.

[0111] According to some embodiments, the loss prediction model may be trained in a supervised manner using sample data. Figure 6 FIG. 6 is a flowchart of a method 600 for training a loss prediction model according to an exemplary embodiment of the present disclosure. Figure 6 As shown, the method 600 includes:

[0112] Step S601, obtaining the raw material demand and raw material shipment of the target product within a preset period of time before the sample time period and the actual raw material loss information of the target product within the sample time period;

[0113] Step S602: input the raw material demand and raw material shipment volume of the target product within a preset period of time before the sample time period into the loss prediction model to obtain the predicted raw material loss information of the target product within the sample time period output by the loss prediction model;

[0114] Step S603, calculating the loss value based on the actual raw material loss information and the predicted raw material loss information; and

[0115] Step S604: Adjust the parameters of the loss prediction model based on the loss value.

[0116] Therefore, by using sample data for supervised training, the performance of the loss prediction model can be improved, thereby improving the accuracy of the prediction of raw material loss information of the target product.

[0117] The exemplary embodiments of the present disclosure will be further described below in conjunction with examples.

[0118] In one example, the raw material shipment volume of a target product may be predicted using the following process:

[0119] Step S11: Obtain historical daily sales information of the target product when there is no promotion.

[0120] Step S12: Determine the historical daily sales curve of the target product based on the historical daily sales information of the target product.

[0121] Step S13: predicting the predicted daily sales volume of the target product in the target time period without promotion based on the historical daily sales volume curve.

[0122] When the predicted daily sales volume of the target product is determined, the impact of the target product's own promotional activities on its sales volume can be further considered, and then the following steps S14-S15 are performed:

[0123] Step S14: Determine corresponding promotional sales increase information based on historical daily sales information and historical promotional sales information.

[0124] Step S15: Calculate the sum of the predicted daily sales volume and the promotional sales volume increase information to obtain the predicted sales volume of the target product within the target time period.

[0125] Fig. 7A and Figure 7B A schematic diagram showing a historical daily sales curve of a product according to an exemplary embodiment of the present disclosure. Fig. 7A and Figure 7B As shown, based on the historical daily sales information of product A and product B, the historical daily sales curves corresponding to product A and product B can be determined, and then the predicted daily sales corresponding to product A and product B can be predicted based on the historical daily sales curves.

[0126] Continue to see Fig. 7A and Figure 7B As shown, in some examples, the promotional sales increase information corresponding to product A and product B can also be represented in the form of a promotional sales increase curve, and then the predicted sales corresponding to product A and product B can be determined based on the historical sales curve and the promotional sales increase curve.

[0127] In one example, the target product may be a new product that has not been sold. In this case, the target product does not have corresponding historical sales information. However, there are usually similar products that are similar to the new product among the existing products that have been sold. Therefore, the method described above can also be used to predict the sales of the new target product based on the historical sales information of similar products.

[0128] After the predicted sales volume of the target product is determined, the impact of the promotion activities of the target product's competitors on its sales volume can be further considered, and then the following steps S16 and S17 are performed:

[0129] Step S16: Obtain promotional sales increase information of competing products of the target product.

[0130] Step S17: Based on the promotional sales increase information of the competing products, adjust the predicted sales of the target product within the target time period.

[0131] In one example, the following formula may be used to adjust the predicted sales volume of the product within the target time period:

[0132]

[0133] In the formula, y0 is the predicted sales volume determined based on the historical daily sales volume information and historical promotional sales volume information of the target product, c is the impact coefficient of the target product's competitor's impact on the target product's sales volume, and x is the promotional sales increase information of the target product's competitor. Exemplarily, the impact coefficient can be set manually based on empirical data. Based on this, the predicted sales volume of the target product can be adjusted more accurately.

[0134] After obtaining the predicted sales volume of the target product, the following steps S18 to S20 can be used to predict the raw material shipment volume of the target product within the target time period:

[0135] Step S18: Determine the raw material demand a of the target product within the target time period based on the predicted sales volume of the target product within the target time period.

[0136] Step S19: input the raw material demand a of the target product within the target time period into the loss prediction model to obtain the raw material loss rate r of the target product within the target time period output by the loss prediction model.

[0137] Step S20: Based on the raw material demand a and the raw material loss rate r of the target product within the target time period, the raw material delivery volume b of the product is determined using the following formula:

[0138] b=a×r

[0139] In this way, the raw material shipment volume of the target product within the target time period can be determined accurately and efficiently, improving the accuracy of inventory management.

[0140] In one example, the raw material shipment volume of a target product may be predicted using the following process:

[0141] Step S31, obtaining historical daily sales information of the target product;

[0142] Step S32: input the historical daily sales information of the target product into the sales prediction model to obtain the predicted sales volume y0 of the target product when there is no promotion in the target time period output by the sales prediction model, and at the same time obtain the product sales feature vector output by the hidden layer included in the sales prediction model;

[0143] Step S33: Obtain information about at least one target product promotion activity of the target product itself within the target time period and information about at least one competitive product promotion activity of the target product within the target time period to determine at least one sub-target product promotion sales increase information SE n (n≥0) and at least one sub-competitive product promotion sales increase information CE m(m≥0).

[0144] Step S34: concatenate the sales increase information of the m sub-competitive products with the product sales feature vector to obtain the m sub-competitive product activity impact features {E1, E2, ..., E m}, and concatenate the n sub-target product promotion sales increase information with the product sales feature vector respectively to obtain n sub-target product activity impact features corresponding to their own activities {E m+1 ,E m+2 ,…,E m+n}.

[0145] Step S35: based on the competitive product activity impact features {E1, E2, ..., E m} and target product activity impact characteristics {E m+1 ,E m+2 ,…,E m+n Obtain an explainable promotional activity impact, which may include a self-promotional impact x that can indicate the incremental sales volume generated by the promotional activities of the target product itself and a promotional diversion impact coefficient c that can indicate the degree of influence of the promotional activities of the target product's competitors on the sales volume of the target product.

[0146] Step S36, based on the obtained promotion diversion impact coefficient c and the self-promotion impact x, the predicted sales volume y0 of the target product is adjusted to obtain a more accurate predicted sales volume of the target product, and the influence of multiple competing product promotion activities and multiple target product promotion activities on sales volume can be explicitly introduced in the sales forecasting process, thereby improving the accuracy and interpretability of the sales forecast.

[0147] After obtaining the predicted sales volume of the target product, the following steps S37 to S39 can be used to predict the raw material shipment volume of the target product within the target time period:

[0148] Step S37: Determine the raw material demand a of the target product within the target time period based on the predicted sales volume of the target product within the target time period.

[0149] Step S38: input the raw material demand a of the target product within the target time period into the loss prediction model to obtain the raw material loss rate t of the target product within the target time period output by the loss prediction model.

[0150] Step S39: Based on the raw material demand a and the raw material loss rate r of the target product within the target time period, the raw material shipment volume b of the product is determined using the formula b=a×r.

[0151] In this way, the raw material shipment volume of the target product within the target time period can be determined accurately and efficiently. At the same time, by utilizing the above method, a quantitative relationship can be established between the daily sales volume of the target product, the promotional activity data of the target product itself, the promotional activity data of the target product's competitors, and the raw material shipment volume of the target product, which helps to more accurately market and manage the target product's inventory and improve the accuracy of inventory management.

[0152] According to another aspect of the present disclosure, a device for predicting sales volume of a product is also provided. Figure 8 FIG. 8 shows a structural block diagram of a sales volume prediction device 800 for a product according to an exemplary embodiment of the present disclosure. Figure 8 As shown, the device 800 includes:

[0153] The first acquisition unit 801 is configured to acquire sales characteristic information of a target product, where the sales characteristic information includes historical daily sales information corresponding to the target product when there is no promotion;

[0154] The first determining unit 802 is configured to determine the predicted sales volume of the target product within the target time period based on the sales volume characteristic information of the target product;

[0155] The second acquisition unit 803 is configured to acquire promotion sales increase information of a competing product corresponding to the target product, where the promotion sales increase information of the competing product can affect the sales of the target product, and the promotion sales increase information is determined based on the historical promotion sales information corresponding to the competitive product when there is a promotion and the historical daily sales information corresponding to the competitive product when there is no promotion; and

[0156] The adjustment unit 804 is configured to adjust the predicted sales volume of the target product within the target time period based on at least the promotion sales volume increase information of the competing product corresponding to the target product.

[0157] The operations of units 801 to 804 of the product sales prediction device 800 are similar to the operations of steps S101 to S104 described above, and are not described in detail here.

[0158] According to another aspect of the present disclosure, a device for predicting raw material shipment volume of a product is also provided. Fig. 9 FIG. 9 is a block diagram showing a structure of a raw material shipment volume prediction device 900 for a product according to an exemplary embodiment of the present disclosure. Fig. 9 As shown, the device 900 includes:

[0159] The product sales prediction device 800 as above is configured to determine the predicted sales of a target product within a target time period;

[0160] The second determining unit 901 is configured to determine the raw material demand of the target product within the target time period based on the predicted sales volume of the target product within the target time period; and

[0161] The third determining unit 902 is configured to determine the raw material shipment quantity of the target product within the target time period based on the raw material demand quantity of the target product within the target time period.

[0162] The operations of the device 800 and units 901 - 902 of the device 900 for predicting the raw material shipment volume of a product are similar to the operations of steps S501 - S503 described above, and are not described in detail here.

[0163] According to another aspect of the present disclosure, an electronic device is also provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned product sales forecasting method and / or raw material shipment forecasting method.

[0164] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is also provided, wherein the computer instructions are used to enable a computer to execute the above-mentioned product sales forecasting method and / or raw material shipment forecasting method.

[0165] According to another aspect of the present disclosure, a computer program product is also provided, including a computer program, wherein when the computer program is executed by a processor, the above-mentioned product sales forecasting method and / or raw material shipment forecasting method is implemented.

[0166] See also Fig.10 , a block diagram of an electronic device 1000 that can be used as the present disclosure will now be described, which is an example of a hardware device that can be applied to various aspects of the present disclosure. The electronic device can be a computer device of different types, such as a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0167] Fig.10 1 shows a block diagram of an electronic device according to an embodiment of the present disclosure. Fig.10 As shown, the electronic device 1000 may include at least one processor 1001 , a working memory 1002 , an I / O device 1004 , a display device 1005 , a storage device 1006 , and a communication interface 1007 that can communicate with each other via a system bus 1003 .

[0168] The processor 1001 may be a single processing unit or multiple processing units, all of which may include a single or multiple computing units or multiple cores. The processor 1001 may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuits, and / or any device that manipulates signals based on operating instructions. The processor 1001 may be configured to obtain and execute computer-readable instructions stored in a working memory 1002, a storage device 1006, or other computer-readable media, such as program codes of an operating system 1002a, program codes of an application program 1002b, and the like.

[0169] Working memory 1002 and storage device 1006 are examples of computer-readable storage media for storing instructions, which are executed by processor 1001 to implement the various functions described above. Working memory 1002 may include both volatile memory and non-volatile memory (e.g., RAM, ROM, etc.). In addition, storage device 1006 may include a hard disk drive, a solid-state drive, a removable medium, including external and removable drives, a memory card, a flash memory, a floppy disk, an optical disk (e.g., CD, DVD), a storage array, a network attached storage, a storage area network, etc. Working memory 1002 and storage device 1006 may all be collectively referred to herein as memory or computer-readable storage media, and may be a non-transitory medium capable of storing computer-readable, processor-executable program instructions as computer program code, which may be executed by processor 1001 as a specific machine configured to implement the operations and functions described in the examples herein.

[0170] The I / O device 1004 may include an input device and / or an output device. The input device may be any type of device capable of inputting information to the electronic device 1000, and may include but is not limited to a mouse, a keyboard, a touch screen, a track pad, a track ball, a joystick, a microphone, and / or a remote controller. The output device may be any type of device capable of presenting information, and may include but is not limited to a video / audio output terminal, a vibrator, and / or a printer.

[0171] The communication interface 1007 allows the electronic device 1000 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks, and may include but is not limited to a modem, a network card, an infrared communication device, a wireless communication transceiver and / or a chipset, such as a Bluetooth device, an 802.11 device, a WiFi device, a WiMax device, a cellular communication device and / or the like.

[0172] The application 1002b in the working register 1002 can be loaded to execute the various methods and processes described above, for example Figure 1In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 1000 via the storage device 1006 and / or the communication interface 1007. When the computer program is loaded and executed by the processor 1001, one or more steps of the product sales volume forecasting method and / or the raw material shipment volume forecasting method described above may be executed.

[0173] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0174] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0175] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0176] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0177] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0178] A computing system may include clients and servers. Clients and servers are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship to each other.

[0179] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0180] Although the embodiments or examples of the present disclosure have been described with reference to the accompanying drawings, it should be understood that the above-mentioned methods, systems and devices are merely exemplary embodiments or examples, and the scope of the present invention is not limited by these embodiments or examples, but only by the claims after authorization and their equivalent scope. Various elements in the embodiments or examples can be omitted or replaced by their equivalent elements. In addition, each step can be performed in an order different from that described in the present disclosure. Further, the various elements in the embodiments or examples can be combined in various ways. It is important that with the evolution of technology, many elements described herein can be replaced by equivalent elements that appear after the present disclosure.

Claims

1. A method for predicting sales volume of a product, comprising: Acquire sales characteristic information of a target product, wherein the sales characteristic information includes historical daily sales information of the target product corresponding to a non-promotion period; Determining the predicted sales volume of the target product within a target time period based on the sales volume characteristic information of the target product; Acquire promotion sales increase information of a competing product corresponding to the target product, wherein the promotion sales increase information is determined based on historical promotion sales information corresponding to a promotion and historical daily sales information corresponding to a promotion of the competing product; and The predicted sales volume of the target product within the target time period is adjusted based at least on the promotional sales volume increase information of the competing product corresponding to the target product.

2. The method of claim 1, wherein: The adjusting the predicted sales volume of the target product within the target time period based at least on the promotion sales volume increase information of the competing product corresponding to the target product includes: Determine a promotion diversion impact coefficient based on the sales characteristic information of the target product and the promotion sales increase information of the competing product corresponding to the target product, wherein the promotion diversion impact coefficient is used to characterize the influence of the promotion of the competing product on the sales of the target product; and Based on the promotion diversion impact coefficient, the predicted sales volume of the target product within the target time period is adjusted.

3. The method of claim 2, further comprising: Determine the product sales feature vector corresponding to the sales feature information, Wherein, determining the promotion diversion impact coefficient based on the sales characteristic information of the target product and the promotion sales increase information of the competing product corresponding to the target product includes: The product sales feature vector and the promotion sales increase information of the competing product corresponding to the target product are input into the promotion diversion impact prediction model to obtain the promotion diversion impact coefficient output by the promotion diversion impact prediction model.

4. The method of claim 3, wherein: The promotion sales increase information of the competing product corresponding to the target product includes promotion sales increase information of multiple sub-competing products corresponding to multiple competing product promotion activities, and the promotion diversion impact prediction model includes a feature determination module and an impact coefficient calculation module. And wherein, the step of inputting the product sales feature vector and the promotion sales increase information of the competing product corresponding to the target product into the promotion diversion impact prediction model to obtain the promotion diversion impact coefficient output by the promotion diversion impact prediction model includes: Inputting the product sales feature vector and the multiple sub-competitive product promotion sales increase information into the feature determination module to obtain multiple sub-competitive product activity impact features corresponding to the multiple competitive product promotion activities, each of which indicates the degree to which the corresponding competitive product promotion activity affects the sales of the target product; and The predicted sales volume of the target product and the activity impact characteristics of the multiple sub-competing products are input into the impact coefficient calculation module to obtain the promotion diversion impact coefficient output by the impact coefficient calculation module.

5. The method of claim 3, wherein: The determining, based on the sales characteristic information of the target product, the predicted sales volume of the target product within the target time period includes: The sales feature information of the target product is input into the sales prediction model to obtain the predicted sales volume of the target product output by the sales prediction model.

6. The method of claim 5, wherein: The training process of the sales volume prediction model and the promotion diversion impact prediction model includes: Obtaining sales characteristic information of sample products, promotional sales increase information of competing products corresponding to the sample products, and actual sales of the sample products; Inputting the sales feature information of the sample product into the initial sales prediction model to obtain the predicted sales and product sales feature vector of the sample product within the target time period output by the initial sales prediction model; Inputting the product sales feature vector of the sample product and the promotion sales increase information of the competing product corresponding to the sample product into the promotion diversion impact prediction model to obtain the promotion diversion impact coefficient output by the promotion diversion impact prediction model; Based on the promotion diversion impact coefficient, adjusting the predicted sales volume of the sample product within the target time period; Calculating a loss value based on the actual sales volume and the adjusted predicted sales volume; and Based on the loss value, the parameters of the sales volume prediction model and the promotion diversion impact prediction model are adjusted.

7. The method according to any one of claims 1 to 6, further comprising: Obtaining promotional sales increase information of the target product, The adjusting the predicted sales volume of the target product within the target time period based at least on the promotion sales volume increase information of the competing product corresponding to the target product includes: Based on the promotional sales increase information of the target product and the promotional sales increase information of the competing product corresponding to the product, the predicted sales of the target product in the target time period is adjusted.

8. The method of claim 7, wherein: The obtaining of the promotion sales increase information of the target product includes: Obtaining historical promotional sales information of the target product when there is a promotion; and Based on the historical daily sales information and historical promotional sales information of the target product, promotional sales increase information of the target product is determined.

9. The method of claim 7, wherein: The adjusting the predicted sales volume of the target product within the target time period based on the promotion sales volume increase information of the target product and the promotion sales volume increase information of the competing product corresponding to the target product includes: Determine the own promotional impact amount and the promotional diversion impact coefficient based on the sales characteristic information of the target product, the promotional sales increase information of the target product, and the promotional sales increase information of the competing products corresponding to the target product; and Based on the self-promotion influence and the promotion diversion influence coefficient, the predicted sales volume of the target product in the target time period is adjusted.

10. The method according to any one of claims 1 to 6, wherein the target product is an unsold product, wherein: The step of obtaining the sales characteristic information of the target product includes: Determine similar products that have been sold corresponding to the target product; Obtaining historical daily sales information of similar products; and Based on the historical daily sales information of the similar products, the sales feature information of the target product is determined.

11. The method of claim 10, wherein: The determining, based on the sales characteristic information of the target product, the predicted sales volume of the target product within the target time period includes: Determining a substitution coefficient between the target product and the similar product; and Based on the substitution coefficient between the target product and the similar products, and the historical daily sales information of the similar products, the predicted sales volume of the target product within the target time period is determined.

12. The method of claim 11, wherein: Determining the substitution coefficient between the target product and the similar product includes: determining at least one competing product corresponding to the similar product; Based on the historical sales information of the similar product and the historical sales information of at least one competing product corresponding to the similar product, a substitution coefficient between the target product and the similar product is determined.

13. A method for predicting raw material shipment volume of a product, comprising: Determine the predicted sales volume of the target product within the target time period using the method described in any one of claims 1 to 12; Determining the raw material demand of the target product within the target time period based on the predicted sales volume of the target product within the target time period; as well as Based on the raw material demand of the target product within the target time period, the raw material shipment volume of the target product within the target time period is determined.

14. The method of claim 13, wherein: Determining the raw material shipment volume of the target product within the target time period based on the raw material demand volume of the target product within the target time period includes: Determining the raw material loss information of the target product within the target time period; and The raw material shipment volume of the target product within the target time period is determined based on the raw material demand and raw material loss information of the target product within the target time period.

15. The method of claim 14, wherein: Determining the raw material loss information of the target product within the target time period includes: The raw material demand and raw material shipment volume of the target product within a preset period of time before the target time period are input into the loss prediction model to obtain the raw material loss information of the target product within the target time period output by the loss prediction model.

16. The method of claim 15, wherein: The loss prediction model is trained in the following way: Obtaining the raw material demand and raw material shipment of the target product within a preset period of time before the sample time period and the actual raw material loss information of the target product within the sample time period; Inputting the raw material demand and raw material shipment of the target product within a preset period of time before the sample time period into the loss prediction model to obtain the predicted raw material loss information of the target product within the sample time period output by the loss prediction model; Calculating a loss value based on the actual raw material loss information and the predicted raw material loss information; and Based on the loss value, parameters of the loss prediction model are adjusted.

17. A sales volume forecasting device for a product, comprising: A first acquisition unit is configured to acquire sales characteristic information of a target product, wherein the sales characteristic information includes historical daily sales information corresponding to the target product when there is no promotion; A first determining unit is configured to determine the predicted sales volume of the target product within a target time period based on the sales volume characteristic information of the target product; a second acquisition unit configured to acquire promotion sales increase information of a competing product corresponding to the target product, wherein the promotion sales increase information is determined based on historical promotion sales information corresponding to a promotion and historical daily sales information corresponding to a promotion of the competing product; and The adjustment unit is configured to adjust the predicted sales volume of the target product within the target time period based at least on the promotional sales volume increase information of the competing product corresponding to the target product.

18. A device for predicting the shipment volume of raw materials for a product, comprising: The apparatus of claim 17, configured to determine predicted sales of a target product within a target time period; A second determining unit is configured to determine the raw material demand of the target product within the target time period based on the predicted sales volume of the target product within the target time period; as well as The third determining unit is configured to determine the raw material shipment quantity of the target product within the target time period based on the raw material demand quantity of the target product within the target time period.

19. An electronic device comprising: at least one processor; as well as a memory communicatively coupled to the at least one processor; in The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 16.

20. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-16.

21. A computer program product comprising a computer program, wherein: When the computer program is executed by a processor, the method according to any one of claims 1 to 16 is implemented.

Citation Information

Cited By

  • Inventory prediction method and device

    CN120046794A

  • Inventory prediction method and apparatus

    WO2026166138A1