Commodity sales volume expectation and pre-business planning method and system

Abstract

The invention relates to the technical field of business planning, and discloses a commodity sales volume expectation and pre-business planning method and system, and the method comprises the steps: obtaining historical orders in an analysis period, extracting a feature tag of each historical order through a regular matching algorithm, and inputting the feature tag into a tag statistical form; obtaining a historical sales directory of each adjacent merchant, and extracting historical discount data and historical sales volume data of each competitive commodity from the historical sales directory to generate a plurality of corresponding competitive product historical discount labels and historical discount sales volumes; on the basis of the time nodes, time stamps are associated with the various feature tags, the competing product historical discount tags and the historical discount sales volume, and the time stamps are input into a preset correlation analysis model to serve as training data to train the correlation analysis model; inputting competing product historical discount labels of the adjacent commercial tenants into the correlation analysis model to generate discount sales correlation expressions of similar commodities of the commercial tenants; the method and the device have the effect of improving the commodity sales volume expectation and stock prediction accuracy.

Description

Technical Field

[0001] The present application relates to the technical field of business planning, and more particularly to a method and system for commodity sales forecasting and pre-business planning. Background Art

[0002] Many businesses need to prepare goods before opening for business. However, the storage of goods requires a certain cost. Too little stock may lead to insufficient supply and reduce operating income. Too much stock may lead to excessive storage costs or waste of goods due to unsalable goods. Especially for restaurants, some fresh ingredients have a short shelf life. Once unsalable, it will cause great economic losses. At present, many businesses mainly prepare goods before opening for business based on the sales volume of goods on similar dates under previous market conditions to estimate future sales volume of goods, so as to prepare goods. However, there are many factors that affect the sales volume of goods, such as the nature of holidays / working days, weather conditions, promotions, the number of surrounding competing merchants, promotions of surrounding competing merchants, etc. If only manual analysis is relied upon, it is easy to miss the analysis of influencing factors and misjudge the degree of influence of influencing factors. Therefore, the above-mentioned related technologies have the problem of poor accuracy in product sales expectations and stocking predictions. Summary of the invention

[0003] In order to improve the accuracy of commodity sales forecast and inventory prediction, the present application provides a commodity sales forecast and pre-business planning method and system.

[0004] The first invention objective of the present application is achieved by adopting the following technical solution: A method for commodity sales forecasting and pre-business planning, comprising: Obtain historical orders within the preset analysis period, extract the feature labels of each historical order through the pre-trained regular matching algorithm and input them into the label statistics form; Obtaining historical sales catalogs of each neighboring merchant, extracting historical discount data and historical sales data of each competing product from each of the historical sales catalogs, so as to generate corresponding historical discount tags and historical discount sales of several competing products; Based on the time nodes of historical orders, various feature tags, competitor product historical discount tags, and historical discount sales are associated with corresponding time tags, and input into a preset correlation analysis model. The correlation analysis model is trained based on several feature tags and competitor product discount tags as training data. Input the historical discount labels of competing products of neighboring merchants into the correlation analysis model to generate the discount sales correlation formula of similar products of this merchant; The feature labels include product labels, date labels, time period labels, and discount labels; the competing products refer to products that are in competition with the products sold by this merchant, and the similar products refer to products sold by this merchant that are in competition with the corresponding competing products.

[0005] By adopting the above technical solution, historical orders within the analysis period are obtained, key information of each historical order is extracted from each historical order through a regular matching algorithm, characteristic tags of key information such as product tags, date tags, time tags, discount tags, etc. are obtained, and each characteristic tag is input into a tag statistics form to collect key information of each historical order; historical sales catalogs of each neighboring merchant are obtained, and historical discount data and historical sales data of each competing product are extracted from each historical sales catalog, so as to generate corresponding historical discount tags of several competing products according to the historical discount data, and calculate several historical discount sales according to the corresponding relationship between the historical discount data and the historical sales data; based on the time nodes of historical orders, The acquired feature labels, competitor product historical discount labels, and historical discount sales are associated with corresponding time tags and input into the correlation analysis model, so as to train the correlation analysis model based on the feature labels, competitor product historical discount labels, and historical discount sales as training data, so as to optimize the correlation analysis model; after the correlation analysis model is trained, the competitor product historical discount labels of each neighboring merchant are input into the correlation analysis model to predict the expected discounts of the competing products of each neighboring merchant in the next sales cycle, and further analyze the discount-sales relationship of similar products of the merchant, so as to guide the merchant to formulate discounts for similar products, and prepare stocks according to the expected sales, thereby improving the accuracy of product sales expectations and preparation forecasts.

[0006] In a preferred example of the present application, before extracting the historical discount data and historical sales data of each competing product from each of the historical sales catalogs to generate a corresponding number of competing product historical discount tags and historical discount sales, the method includes: Obtain the historical sales catalog of each neighboring merchant, extract the historical discount data and historical sales data of each neighboring commodity from the historical sales catalog, and draw a number of first-class discount sales curves; Obtain the discount tags of each historical order within the analysis period to collect historical discount data and historical sales data of each store's products, and draw several second-category discount sales curves; Input a number of first-class discount sales curves and second-class discount sales curves into the correlation analysis model to analyze the products that are in a competitive relationship with each other based on the discount and sales relationship between various types of in-store products and adjacent products, and mark similar products and competing products; The neighboring commodities refer to commodities sold by neighboring merchants; the in-store commodities refer to commodities sold by the in-store merchant.

[0007] By adopting the above technical solution, the historical sales catalog of each neighboring merchant is obtained, so as to extract the historical discount data and historical sales data of each neighboring commodity in the neighboring merchant from the historical sales catalog, and draw a number of first-class discount sales curves according to the relationship between discount and sales; the discount labels of each historical order of the merchant within the analysis period are obtained, and the discount data and sales data corresponding to each discount label are counted, and then a number of second-class discount sales curves are drawn according to the relationship between discount and sales; the several first-class discount sales curves and the second-class discount sales curves are input into the correlation analysis model, so as to analyze the commodities with mutual competition relationship according to the discount and sales relationship between various types of the store's commodities and neighboring commodities, define the neighboring commodities with mutual competition relationship as competing commodities, and define the store's commodities with mutual competition relationship as similar commodities.

[0008] In a preferred example of the present application: the relevance analysis model has a built-in neural network algorithm; Based on the time nodes of historical orders, various feature tags, competitor product historical discount tags, and historical discount sales are associated with corresponding time tags, and input into a preset correlation analysis model. The correlation analysis model is trained based on several feature tags and competitor product discount tags as training data, including: Based on the time nodes in historical orders and the acquisition time nodes of historical sales catalogs, various feature tags, historical discount tags of competing products, and historical discount sales are associated with corresponding time tags; Various feature labels associated with time tags, historical discount labels of competing products, and historical discounted sales are input into the correlation analysis model as training data. The relationship between discounts and sales of similar products, the relationship between discounts and sales of several competing products, and the relationship between sales of similar products and competing products are used as analysis targets to conduct targeted training on the neural network algorithm.

[0009] By adopting the above technical solution, the correlation analysis model is built with a neural network algorithm so that the correlation analysis model has the characteristics of analyzing complex data. Based on the time nodes in the historical orders and the time nodes for obtaining the historical sales catalog, various feature labels, historical discount labels of competing products and historical discounted sales are associated with corresponding time tags, so that the subsequent data analysis can be combined with the time dimension for correlation analysis; various feature labels, historical discount labels of competing products and historical discounted sales associated with time tags are used as training data, and training goals are set to analyze the relationship between discounts and sales of similar products, analyze the relationship between discounts and sales of several competing products, and analyze the relationship between sales of similar products and competing products. Targeted training is performed on the neural network algorithm to improve the accuracy of the correlation analysis model in analyzing the sales competition relationship between different products and the relationship between product discounts and sales.

[0010] In a preferred example of the present application, the historical discount tags of competing products of each neighboring merchant are input into the correlation analysis model to generate a discount sales correlation formula of similar products of the merchant, including: Input the historical discount labels of competing products of each neighboring merchant into the correlation analysis model, predict a number of predicted discount labels of competing products of each neighboring merchant in the next sales period, and determine a number of predicted sales of competing products based on the predicted discount labels of each competing product and the corresponding type of discount sales curve; Analyze the sales relationship between similar products and competing products based on the predicted sales of several competing products to evaluate the predicted sales of similar products; Based on the predicted sales volume of the same category as the sales volume base, the discount sales volume correlation formula of the same category of products is generated in combination with the corresponding second category discount sales volume curve.

[0011] By adopting the above technical solution, the historical discount labels of competing products of each neighboring merchant are input into the correlation analysis model, so as to analyze the discount rules of neighboring merchants according to the historical discount labels of neighboring merchants, and then predict the discount settings of neighboring merchants for several competing products, so as to determine several predicted discount labels of competing products, and determine the predicted sales volume of competing products of several competing products according to the predicted discount labels of each competing product and the corresponding first-class discount sales curve; based on the predicted sales volume of several competing products, the sales relationship between similar products and competing products is analyzed in the correlation analysis model to evaluate the similar predicted sales volume corresponding to the similar products sold by this merchant; taking the similar predicted sales volume as the sales volume base, combined with the corresponding second-class discount sales curve, a discount sales volume correlation formula is generated, so that this merchant can formulate discounts and prepare goods according to the expected sales volume.

[0012] In a preferred example, the present application also includes: The average price of each competing product and similar product in the previous analysis period is calculated to determine the benchmark price of each competing product and similar product. The discounts are calculated based on the sales price and the benchmark price.

[0013] By adopting the above technical solution, since each merchant may offer discount promotions on goods at the same time, and some merchants may even sell at discounted prices for a long time, in order to eliminate the impact of long-term discounted sales on the actual price of goods, the average price of the goods in the previous analysis period is used as the benchmark price, and the discount is calculated based on the actual sales price and the benchmark price, thereby improving the authenticity and reliability of the discount data.

[0014] The second invention objective of this application is achieved by the following technical solution: A commodity sales forecasting and pre-opening planning system, applied to any of the above commodity sales forecasting and pre-opening planning methods, comprising: The feature label extraction module is used to obtain historical orders within a preset analysis period, extract the feature labels of each historical order through a pre-trained regular matching algorithm, and input them into the label statistics form; A neighboring product analysis module is used to obtain the historical sales catalog of each neighboring merchant, extract the historical discount data and historical sales data of each competing product from each of the historical sales catalogs, and generate a number of corresponding historical discount tags and historical discount sales of competing products; The analysis model training module is used to associate the corresponding time tags with various feature tags, competitor product historical discount tags, and historical discount sales based on the time nodes of historical orders, and input them into the preset correlation analysis model, and train the correlation analysis model based on several feature tags and competitor product discount tags as training data; The discount sales volume correlation analysis module is used to input the historical discount tags of competing products of neighboring merchants into the correlation analysis model to generate the discount sales volume correlation formula of similar products of this merchant; The feature labels include product labels, date labels, time period labels, and discount labels; the competing products refer to products that are in competition with the products sold by this merchant, and the similar products refer to products sold by this merchant that are in competition with the corresponding competing products.

[0015] In a preferred example of the present application: the adjacent product analysis module includes: A first-class discount sales analysis submodule is used to obtain the historical sales catalog of each neighboring merchant, extract the historical discount data and historical sales data of each neighboring commodity from the historical sales catalog, and draw a number of first-class discount sales curves; The second-category discount sales analysis submodule is used to obtain the discount tags of each historical order within the analysis period, so as to count the historical discount data and historical sales data of each store's products and draw several second-category discount sales curves; The commodity competition analysis submodule is used to input a number of first-class discount sales curves and second-class discount sales curves into the correlation analysis model, so as to analyze the commodities that are in a competitive relationship with each other according to the discount and sales relationship between various types of in-store commodities and neighboring commodities, and mark similar commodities and competing commodities; The neighboring commodities refer to commodities sold by neighboring merchants; the in-store commodities refer to commodities sold by the in-store merchant.

[0016] In a preferred example of the present application: the analysis model training module includes: The tag time tagging submodule includes the time tags corresponding to various feature tags, competitor product historical discount tags, and historical discount sales based on the time nodes in historical orders and the acquisition time nodes of historical sales catalogs; The algorithm targets the training submodule, including inputting various feature labels associated with time tags, historical discount labels of competing products, and historical discounted sales as training data into the correlation analysis model, taking the relationship between discounts and sales of similar products, the relationship between discounts and sales of several competing products, and the relationship between sales of similar products and competing products as analysis targets, and conducting targeted training on the neural network algorithm; The correlation analysis model has a built-in neural network algorithm.

[0017] The third invention objective of this application is achieved by the following technical solution: A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned method for predicting commodity sales volume and pre-business planning are implemented.

[0018] The fourth invention objective of this application is achieved by the following technical solution: A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned commodity sales forecasting and pre-business planning method.

[0019] In summary, the present application includes at least one of the following beneficial technical effects: 1. Obtain historical orders within the analysis period, extract key information from each historical order through a regular matching algorithm, obtain feature tags of key information such as product tags, date tags, time tags, discount tags, etc., and input each feature tag into a tag statistics form to collect key information of each historical order; obtain historical sales catalogs of each neighboring merchant, extract historical discount data and historical sales data of each competing product from each historical sales catalog, and generate corresponding historical discount tags of several competing products according to the historical discount data, and calculate several historical discount sales according to the corresponding relationship between historical discount data and historical sales data; based on the time nodes of historical orders, The feature labels, competitor historical discount labels, and historical discount sales are associated with corresponding time tags and input into the correlation analysis model, so as to train the correlation analysis model based on the feature labels, competitor historical discount labels, and historical discount sales as training data, so as to optimize the correlation analysis model; after the correlation analysis model is trained, the competitor historical discount labels of each neighboring merchant are input into the correlation analysis model to predict the expected discounts of the competing products of each neighboring merchant in the next sales cycle, and further analyze the discount-sales relationship of similar products of the merchant, so as to guide the merchant to formulate discounts for similar products and prepare stocks according to the expected sales, thereby improving the accuracy of product sales expectations and preparation forecasts.

[0020] 2. Obtain the historical sales catalog of each neighboring merchant to extract the historical discount data and historical sales data of each neighboring commodity in the neighboring merchant from the historical sales catalog, and draw a number of first-class discount sales curves according to the relationship between discount and sales; obtain the discount tags of each historical order of the merchant during the analysis period, and count the discount data and sales data corresponding to each discount tag, and then draw a number of second-class discount sales curves according to the relationship between discount and sales; input a number of first-class discount sales curves and second-class discount sales curves into the correlation analysis model, so as to analyze the commodities with mutual competition relationship according to the discount and sales relationship between various types of own store commodities and neighboring commodities, define the neighboring commodities with mutual competition relationship as competing commodities, and define the own store commodities with mutual competition relationship as similar commodities.

[0021] 3. The correlation analysis model has a built-in neural network algorithm so that the correlation analysis model has the characteristics of analyzing complex data. Based on the time nodes in the historical orders and the time nodes for obtaining the historical sales catalog, various feature labels, historical discount labels of competing products and historical discounted sales are associated with corresponding time tags, so that the subsequent data analysis can be combined with the time dimension for correlation analysis; various feature labels, historical discount labels of competing products and historical discounted sales associated with time tags are used as training data, and training goals are set to analyze the relationship between discounts and sales of similar products, analyze the relationship between discounts and sales of several competing products, and analyze the relationship between sales of similar products and competing products. Targeted training is performed on the neural network algorithm to improve the accuracy of the correlation analysis model's analysis of the sales competition relationship between different products and the relationship between product discounts and sales. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a flow chart of the method for predicting commodity sales volume and pre-business planning in Example 1 of the present application.

[0023] Figure 2 It is a principle block diagram of the commodity sales forecasting and pre-business planning system in the second embodiment of the present application.

[0024] Figure 3 It is a schematic diagram of the equipment in Example 3 of the present application. DETAILED DESCRIPTION

[0025] The following is combined with Figures 1 to 3 This application is described in further detail.

[0026] Embodiment 1

[0027] Reference Figure 1 The present application discloses a method for predicting commodity sales and pre-business planning, which specifically includes the following steps: S10: Obtain historical orders within a preset analysis period, extract feature labels of each historical order through a pre-trained regular matching algorithm, and input them into a label statistics form.

[0028] In this embodiment, the analysis period refers to a preset period of historical orders required for collecting training data. Preferably, the analysis period can be set to 12 months or 13 months according to actual needs; the feature tags include product tags, date tags, time period tags, and discount tags.

[0029] Specifically, historical orders within the most recent analysis period are obtained from the order database, key information of each historical order is extracted from each historical order through a regular matching algorithm, feature tags of key information such as product tags, date tags, time period tags, discount tags, etc. are obtained, and each feature tag is input into a tag statistics form to collect key information of each historical order.

[0030] Furthermore, the average price of each competing product and similar product in the previous analysis period is counted to determine the benchmark price of each competing product and similar product, and the discounts are calculated based on the sales price and the benchmark price.

[0031] Since various merchants may offer discount promotions for goods at the same time, and some merchants may even sell goods at discounted prices for a long time, in order to eliminate the impact of long-term discounted sales on the actual price of goods, the average price of the goods in the previous analysis period is used as the benchmark price, and the discount is calculated based on the actual sales price and the benchmark price, thereby improving the authenticity and reliability of the discount data.

[0032] S20: Obtain the historical sales catalog of each neighboring merchant, and extract the historical discount data and historical sales data of each competing product from each of the historical sales catalogs to generate a corresponding number of competing product historical discount tags and historical discount sales.

[0033] In this embodiment, the sales catalog refers to a form containing price, sales volume, and discount data obtained based on the online ordering platform interface of neighboring merchants; competing products refer to products that are in competition with products sold by this merchant, and similar products refer to products sold by this merchant that are in competition with corresponding competing products; among them, since the discount of products may change, historical discount sales refer to sales counted during a specific discount sales period, and each historical discount sales volume corresponds to a competitor historical discount label.

[0034] Specifically, the historical sales catalogs of each neighboring merchant are obtained, and the historical discount data and historical sales data of each competing product are extracted from each historical sales catalog, so as to generate several corresponding historical discount labels of competing products according to the historical discount data, and calculate several historical discount sales according to the corresponding relationship between the historical discount data and the historical sales data.

[0035] Before step S20, the following steps are included: S21: Obtain the historical sales catalog of each neighboring merchant, extract the historical discount data and historical sales data of each neighboring commodity from the historical sales catalog, and draw a number of first-class discount sales curves.

[0036] In this embodiment, the neighboring commodities refer to commodities sold by neighboring merchants.

[0037] Specifically, the historical sales catalog of each neighboring merchant is obtained to extract the historical discount data and historical sales data of each neighboring commodity in the neighboring merchant from the historical sales catalog, and a plurality of first-class discount sales curves are drawn according to the relationship between the discount and the sales.

[0038] S22: Obtain the discount tags of each historical order within the analysis period to collect historical discount data and historical sales data of each store product to draw a number of second-category discount sales curves.

[0039] Specifically, the discount labels of each historical order of the merchant within the analysis period are obtained, and the discount data and sales data corresponding to each discount label are counted, and then several second-class discount sales curves are drawn according to the relationship between the discount and the sales volume.

[0040] S23: Input a number of first-class discount sales curves and second-class discount sales curves into the correlation analysis model to analyze the products that are in a competitive relationship with each other based on the relationship between the discounts and sales of various types of in-store products and adjacent products, and mark similar products and competing products.

[0041] Specifically, several first-class discount sales curves and second-class discount sales curves are input into the correlation analysis model, so as to analyze the competitive products according to the discount and sales relationship between various types of in-store products and adjacent products, define the competitive adjacent products as competing products, and define the in-store products that are in a competitive relationship as similar products.

[0042] S30: Based on the time nodes of historical orders, various feature tags, competitor product historical discount tags, and historical discount sales are associated with corresponding time tags, and input into a preset correlation analysis model. The correlation analysis model is trained based on several feature tags and competitor product discount tags as training data.

[0043] In this embodiment, the time node of the historical order may specifically be the order time, or may be set to the meal delivery time, payment time, etc. according to actual needs; the correlation analysis model has a built-in neural network algorithm.

[0044] Specifically, based on the time nodes of historical orders, the corresponding time tags are associated with each feature label, competitor's historical discount label, and historical discount sales volume, and are input into the correlation analysis model. The correlation analysis model is trained based on the feature labels, competitor's historical discount labels, and historical discount sales volume as training data, so as to optimize the correlation analysis model.

[0045] Wherein, in step S30, it includes: S31: Based on the time nodes in the historical orders and the time nodes for obtaining the historical sales catalog, various feature tags, competitor product historical discount tags, and historical discount sales are associated with corresponding time tags.

[0046] Specifically, the correlation analysis model has a built-in neural network algorithm so that the correlation analysis model has the characteristics of analyzing complex data. Based on the time nodes in historical orders and the time nodes for obtaining historical sales catalogs, various feature labels, historical discount labels of competing products and historical discounted sales are associated with corresponding time tags, so that subsequent data analysis can combine the time dimension for correlation analysis.

[0047] S32: Various feature labels associated with time tags, historical discount labels of competing products, and historical discounted sales are input into the correlation analysis model as training data, and the relationship between discounts and sales of similar products, the relationship between discounts and sales of several competing products, and the relationship between sales of similar products and competing products are used as analysis targets to conduct targeted training on the neural network algorithm.

[0048] Specifically, various feature labels associated with time tags, historical discount labels and historical discounted sales of competing products are used as training data, and training goals are set to analyze the relationship between discounts and sales of similar products, analyze the relationship between discounts and sales of several competing products, and analyze the relationship between sales of similar products and competing products. Targeted training is performed on the neural network algorithm to improve the accuracy of the correlation analysis model in analyzing the sales competition relationship between different products, and the relationship between product discounts and sales.

[0049] S40: Input the historical discount labels of competing products of each neighboring merchant into the correlation analysis model to generate a discount sales correlation formula for similar products of the merchant.

[0050] Specifically, after the correlation analysis model is trained, the historical discount labels of the competing products of each neighboring merchant are input into the correlation analysis model to predict the expected discounts of the competing products of each neighboring merchant in the next sales cycle, and further analyze the discount-sales relationship of similar products of this merchant, so as to guide this merchant to formulate discounts for similar products and prepare stocks according to the expected sales, thereby improving the accuracy of product sales expectations and stocking forecasts.

[0051] Wherein, in step S40, it includes: S41: Input the historical discount labels of competing products of each neighboring merchant into the correlation analysis model, predict a number of predicted discount labels of competing products of each neighboring merchant in the next sales period, and determine a number of predicted sales of competing products based on each predicted discount label of competing products and a corresponding type of discount sales curve.

[0052] Specifically, the historical discount labels of competing products of each neighboring merchant are input into the correlation analysis model to analyze the discount rules of neighboring merchants based on their historical discount labels, and then predict the discount settings of neighboring merchants for several competing commodities to determine several competing product predicted discount labels, and determine the competing product predicted sales of several competing commodities based on each competing product predicted discount label and a corresponding type of discount sales curve.

[0053] S42: Analyze the sales relationship between similar products and competing products based on the predicted sales of several competing products to evaluate the predicted sales of similar products.

[0054] Specifically, based on the predicted sales of several competing products, the sales relationship between similar products and competing products is analyzed in the correlation analysis model to evaluate the similar predicted sales corresponding to the similar products sold by the merchant.

[0055] S43: Based on the predicted sales volume of the same category as the sales volume base, the discount sales volume correlation formula of the same category of products is generated in combination with the corresponding second category discount sales volume curve.

[0056] Specifically, the predicted sales volume of the same type is used as the sales volume base, and combined with the corresponding two-category discount sales volume curve, a discount sales volume correlation formula is generated so that the merchant can formulate discounts and prepare stocks according to the expected sales volume.

[0057] It should be understood that the serial numbers of the steps in the above embodiments do not imply a sequence of execution. The execution sequence of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0058] Embodiment 2

[0059] A commodity sales volume forecasting and pre-opening planning system, which corresponds to the commodity sales volume forecasting and pre-opening planning method in the above-mentioned embodiment.

[0060] like Figure 2 As shown in the figure, the commodity sales forecast and pre-opening planning system includes a feature label extraction module, a neighboring commodity analysis module, an analysis model training module, and a discount sales correlation analysis module. The detailed description of each functional module is as follows: The feature label extraction module is used to obtain historical orders within a preset analysis period, extract the feature labels of each historical order through a pre-trained regular matching algorithm, and input them into the label statistics form; A neighboring product analysis module is used to obtain the historical sales catalog of each neighboring merchant, extract the historical discount data and historical sales data of each competing product from each of the historical sales catalogs, and generate a number of corresponding historical discount tags and historical discount sales of competing products; The analysis model training module is used to associate the corresponding time tags with various feature tags, competitor product historical discount tags, and historical discount sales based on the time nodes of historical orders, and input them into the preset correlation analysis model, and train the correlation analysis model based on several feature tags and competitor product discount tags as training data; The discount sales volume correlation analysis module is used to input the historical discount tags of competing products of neighboring merchants into the correlation analysis model to generate the discount sales volume correlation formula of similar products of this merchant.

[0061] Among them, the adjacent product analysis module also includes: A first-class discount sales analysis submodule is used to obtain the historical sales catalog of each neighboring merchant, extract the historical discount data and historical sales data of each neighboring commodity from the historical sales catalog, and draw a number of first-class discount sales curves; The second-category discount sales analysis submodule is used to obtain the discount tags of each historical order within the analysis period, so as to count the historical discount data and historical sales data of each store's products and draw several second-category discount sales curves; The product competition analysis submodule is used to input a number of first-class discount sales curves and second-class discount sales curves into the correlation analysis model, so as to analyze the products that are in competition with each other according to the discount and sales relationship between various types of in-store products and adjacent products, and mark similar products and competing products.

[0062] The analysis model training module also includes: The tag time tagging submodule includes the time tags corresponding to various feature tags, competitor product historical discount tags, and historical discount sales based on the time nodes in historical orders and the acquisition time nodes of historical sales catalogs; The algorithm targets the training sub-module, including inputting various feature labels associated with time tags, historical discount labels and historical discounted sales of competing products as training data into the correlation analysis model, and taking the relationship between discounts and sales of similar products, the relationship between discounts and sales of several competing products, and the relationship between sales of similar products and competing products as analysis targets to conduct targeted training on the neural network algorithm.

[0063] Among them, the discount sales volume correlation analysis module also includes: The competitor product predicted sales volume calculation submodule is used to input the competitor product historical discount labels of each neighboring merchant into the correlation analysis model, predict a number of competitor product predicted discount labels of each neighboring merchant in the next sales period, and determine a number of competitor product predicted sales volumes based on each competitor product predicted discount label and a corresponding type of discount sales volume curve; The similar forecast sales evaluation submodule is used to analyze the sales relationship between similar products and competing products based on the forecast sales of several competing products, so as to evaluate the similar forecast sales; The discount sales volume correlation formula generation submodule is used to generate the discount sales volume correlation formula of similar products based on the similar predicted sales volume as the sales volume base and in combination with the corresponding second-category discount sales volume curve.

[0064] For the specific limitations of the commodity sales forecasting and pre-opening planning system, please refer to the limitations of the commodity sales forecasting and pre-opening planning method in the above text, which will not be repeated here; each module in the above-mentioned commodity sales forecasting and pre-opening planning system can be implemented in whole or in part through software, hardware and a combination thereof; each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of each of the above modules.

[0065] Embodiment 3

[0066] A computer device, which may be a server, may have an internal structure as shown in Figure 3 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store analysis cycles, historical orders, regular matching algorithms, feature labels, label statistics forms, historical sales catalogs, historical discount data, historical sales data, historical discount labels of competing products, historical discount sales, several time nodes, time tags, correlation analysis models, discount sales correlation formulas and other data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it realizes the commodity sales forecast and pre-business planning method.

[0067] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program: S10: Obtain historical orders within a preset analysis period, extract feature labels of each historical order through a pre-trained regular matching algorithm, and input the feature labels into a label statistics form; S20: Obtain historical sales catalogs of each neighboring merchant, and extract historical discount data and historical sales data of each competing product from each of the historical sales catalogs to generate corresponding historical discount tags and historical discount sales of several competing products; S30: Based on the time nodes of historical orders, various feature tags, competitor product historical discount tags, and historical discount sales are associated with corresponding time tags, and input into a preset correlation analysis model, and the correlation analysis model is trained based on a number of feature tags and competitor product discount tags as training data; S40: Input the historical discount labels of competing products of each neighboring merchant into the correlation analysis model to generate a discount sales correlation formula for similar products of the merchant.

[0068] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: S10: Obtain historical orders within a preset analysis period, extract feature labels of each historical order through a pre-trained regular matching algorithm, and input the feature labels into a label statistics form; S20: Obtain historical sales catalogs of each neighboring merchant, and extract historical discount data and historical sales data of each competing product from each of the historical sales catalogs to generate corresponding historical discount tags and historical discount sales of several competing products; S30: Based on the time nodes of historical orders, various feature tags, competitor product historical discount tags, and historical discount sales are associated with corresponding time tags, and input into a preset correlation analysis model, and the correlation analysis model is trained based on a number of feature tags and competitor product discount tags as training data; S40: Input the historical discount labels of competing products of each neighboring merchant into the correlation analysis model to generate a discount sales correlation formula for similar products of the merchant.

[0069] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink), DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0070] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0071] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, it should be understood by those skilled in the art that the technical solutions described in the aforementioned embodiments may still be modified, or some of the features thereof may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for commodity sales forecasting and pre-business planning, characterized in that: include: Obtain historical orders within the preset analysis period, extract the feature labels of each historical order through the pre-trained regular matching algorithm and input them into the label statistics form; Obtaining historical sales catalogs of each neighboring merchant, extracting historical discount data and historical sales data of each competing product from each of the historical sales catalogs, so as to generate corresponding historical discount tags and historical discount sales of several competing products; Based on the time nodes of historical orders, various feature tags, competitor product historical discount tags, and historical discount sales are associated with corresponding time tags, and input into a preset correlation analysis model. The correlation analysis model is trained based on several feature tags and competitor product discount tags as training data. Input the historical discount labels of competing products of neighboring merchants into the correlation analysis model to generate the discount sales correlation formula of similar products of this merchant; The feature labels include product labels, date labels, time period labels, and discount labels; the competing products refer to products that are in competition with the products sold by this merchant, and the similar products refer to products sold by this merchant that are in competition with the corresponding competing products.

2. A method for predicting commodity sales and pre-business planning according to claim 1, characterized in that: Before extracting the historical discount data and historical sales data of each competing product from each of the historical sales catalogs to generate a corresponding number of competing product historical discount tags and historical discount sales, the method includes: Obtain the historical sales catalog of each neighboring merchant, extract the historical discount data and historical sales data of each neighboring commodity from the historical sales catalog, and draw a number of first-class discount sales curves; Obtain the discount tags of each historical order within the analysis period to collect historical discount data and historical sales data of each store's products, and draw several second-category discount sales curves; Input a number of first-class discount sales curves and second-class discount sales curves into the correlation analysis model to analyze the competitive products according to the discount and sales relationship between various types of in-store products and adjacent products, and mark similar products and competitive products; The neighboring commodities refer to commodities sold by neighboring merchants; the in-store commodities refer to commodities sold by the in-store merchant.

3. A method for predicting commodity sales and pre-business planning according to claim 1, characterized in that: The relevance analysis model has a built-in neural network algorithm; Based on the time nodes of historical orders, various feature tags, competitor product historical discount tags, and historical discount sales are associated with corresponding time tags, and input into a preset correlation analysis model. The correlation analysis model is trained based on several feature tags and competitor product discount tags as training data, including: Based on the time nodes in historical orders and the acquisition time nodes of historical sales catalogs, various feature tags, historical discount tags of competing products, and historical discount sales are associated with corresponding time tags; Various feature labels associated with time tags, historical discount labels of competing products, and historical discounted sales are input into the correlation analysis model as training data. The relationship between discounts and sales of similar products, the relationship between discounts and sales of several competing products, and the relationship between sales of similar products and competing products are used as analysis targets to conduct targeted training on the neural network algorithm.

4. A method for predicting commodity sales and pre-business planning according to claim 1, characterized in that: The historical discount labels of competing products of neighboring merchants are input into the correlation analysis model to generate a correlation formula for the discount sales of similar products of the merchant, including: Input the historical discount labels of competing products of each neighboring merchant into the correlation analysis model, predict a number of predicted discount labels of competing products of each neighboring merchant in the next sales period, and determine a number of predicted sales of competing products based on the predicted discount labels of each competing product and the corresponding type of discount sales curve; Analyze the sales relationship between similar products and competing products based on the predicted sales of several competing products to evaluate the predicted sales of similar products; Based on the predicted sales volume of the same category as the sales volume base, the discount sales volume correlation formula of the same category of products is generated in combination with the corresponding second category discount sales volume curve.

5. A method for predicting commodity sales and pre-business planning according to claim 4, characterized in that: Also includes: The average price of each competing product and similar product in the previous analysis period is calculated to determine the benchmark price of each competing product and similar product. The discounts are calculated based on the sales price and the benchmark price.

6. A commodity sales forecasting and pre-business planning system, characterized in that: The method for predicting sales volume and pre-business planning of goods as claimed in any one of claims 1 to 5 comprises: The feature label extraction module is used to obtain historical orders within a preset analysis period, extract the feature labels of each historical order through a pre-trained regular matching algorithm, and input them into the label statistics form; A neighboring product analysis module is used to obtain the historical sales catalog of each neighboring merchant, extract the historical discount data and historical sales data of each competing product from each of the historical sales catalogs, and generate a number of corresponding historical discount tags and historical discount sales of competing products; The analysis model training module is used to associate the corresponding time tags with various feature tags, competitor product historical discount tags, and historical discount sales based on the time nodes of historical orders, and input them into the preset correlation analysis model, and train the correlation analysis model based on several feature tags and competitor product discount tags as training data; The discount sales volume correlation analysis module is used to input the historical discount tags of competing products of neighboring merchants into the correlation analysis model to generate the discount sales volume correlation formula of similar products of this merchant; The feature labels include product labels, date labels, time period labels, and discount labels; the competing products refer to products that are in competition with the products sold by this merchant, and the similar products refer to products sold by this merchant that are in competition with the corresponding competing products.

7. A commodity sales forecasting and pre-business planning system according to claim 6, characterized in that: The adjacent commodity analysis module includes: A first-class discount sales analysis submodule is used to obtain the historical sales catalog of each neighboring merchant, extract the historical discount data and historical sales data of each neighboring commodity from the historical sales catalog, and draw a number of first-class discount sales curves; The second-category discount sales analysis submodule is used to obtain the discount tags of each historical order within the analysis period, so as to count the historical discount data and historical sales data of each store's products and draw several second-category discount sales curves; The commodity competition analysis submodule is used to input a number of first-class discount sales curves and second-class discount sales curves into the correlation analysis model, so as to analyze the commodities that are in a competitive relationship with each other according to the discount and sales relationship between various types of in-store commodities and neighboring commodities, and mark similar commodities and competing commodities; The neighboring commodities refer to commodities sold by neighboring merchants; the in-store commodities refer to commodities sold by the in-store merchant.

8. A commodity sales forecasting and pre-business planning system according to claim 6, characterized in that: The analysis model training module includes: The tag time tagging submodule includes the time tags corresponding to various feature tags, competitor product historical discount tags, and historical discount sales based on the time nodes in historical orders and the acquisition time nodes of historical sales catalogs; The algorithm targets the training submodule, including inputting various feature labels associated with time tags, historical discount labels of competing products, and historical discounted sales as training data into the correlation analysis model, taking the relationship between discounts and sales of similar products, the relationship between discounts and sales of several competing products, and the relationship between sales of similar products and competing products as analysis targets, and conducting targeted training on the neural network algorithm; The correlation analysis model has a built-in neural network algorithm.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method for predicting commodity sales and pre-business planning as described in any one of claims 1 to 5 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for predicting sales volume and pre-business planning of goods as claimed in any one of claims 1 to 5 are implemented.

Images