Purchaser collaborative management system and method based on sales plan prediction

Abstract

The embodiment of the invention provides a sales plan prediction-based purchaser collaborative management system and method, and belongs to the field of data management. The system comprises a first acquisition module for acquiring first associated data and second associated data; the first analysis module performs data association analysis on the first association data to obtain a first association feature; a second analysis module performs data association analysis on the second association data to obtain a second association feature; the model establishment module establishes a supply prediction model according to the first association data in combination with the first association features and establishes a purchase prediction model according to the second association data in combination with the second association features; the second acquisition module acquires third associated data and fourth associated data; the data prediction module predicts the third associated data according to the supply prediction model to obtain a first prediction result, and predicts the fourth associated data according to the purchase prediction model to obtain a second prediction result; the result determination module determines a target management result according to the first prediction result and the second prediction result.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to a buyer collaborative management system and method based on sales plan forecasting. Background Art

[0002] In modern supply chain management, collaborative management between buyers and suppliers is a key link to ensure the efficient operation of the supply chain. Traditional supply chain management usually relies on manual predictions based on historical data and experience to formulate procurement and supply plans. Although this method can meet the needs to a certain extent, it is often not flexible and accurate enough when facing market fluctuations, seasonal changes and other uncertain factors. Therefore, there is an urgent need for a buyer collaborative management system based on sales plan forecasts, so as to achieve efficient collaborative management between supply and procurement, and improve the accuracy and timeliness of forecasts. Summary of the invention

[0003] The main purpose of the embodiments of the present invention is to provide a buyer collaborative management system and method based on sales plan forecasting, aiming to solve the problem that traditional supply chain management usually relies on manual prediction of historical data and experience to formulate procurement and supply plans, which is often not flexible and accurate when facing market fluctuations, seasonal changes and other uncertain factors.

[0004] In a first aspect, an embodiment of the present invention provides a buyer collaborative management system based on sales plan forecasting, including:

[0005] A first acquisition module is used to obtain first associated data corresponding to a first sales plan performed by a supply object on a purchase object and to obtain second associated data corresponding to a second sales plan performed by the purchase object on the supply object;

[0006] A first analysis module, configured to perform data association analysis on the first association data to obtain a first association feature corresponding to the first association data;

[0007] A second analysis module, configured to perform data association analysis on the second association data to obtain a second association feature corresponding to the second association data;

[0008] a model building module, configured to build a supply forecast model based on the first correlation data in combination with the first correlation feature, and to build a purchase forecast model based on the second correlation data in combination with the second correlation feature;

[0009] A second acquisition module is used to obtain third associated data corresponding to a third sales plan performed by the supply object on the purchase object and to obtain fourth associated data corresponding to a fourth sales plan performed by the purchase object on the supply object;

[0010] a data prediction module, configured to perform supply prediction on the third related data according to the supply prediction model to obtain a first prediction result, and perform purchase prediction on the fourth related data according to the purchase prediction model to obtain a second prediction result;

[0011] A result determination module is used to determine a target management result based on the first prediction result and the second prediction result, wherein the target management result includes a first management result corresponding to the third sales plan performed by the supplier to the purchasing object and a second management result corresponding to the fourth sales plan performed by the purchasing object to the supplier.

[0012] In a second aspect, an embodiment of the present invention provides a method for collaborative management of buyers based on sales plan forecasting, comprising:

[0013] Obtaining first associated data corresponding to a first sales plan performed by a supply object on a purchase object and obtaining second associated data corresponding to a second sales plan performed by the purchase object on the supply object;

[0014] Performing data association analysis on the first associated data to obtain a first associated feature corresponding to the first associated data;

[0015] Performing data association analysis on the second association data to obtain a second association feature corresponding to the second association data;

[0016] Establishing a supply forecasting model based on the first associated data combined with the first associated feature, and establishing a procurement forecasting model based on the second associated data combined with the second associated feature;

[0017] Obtaining third associated data corresponding to a third sales plan performed by the supplier on the purchaser and obtaining fourth associated data corresponding to a fourth sales plan performed by the purchaser on the supplier;

[0018] Performing supply forecasting on the third related data according to the supply forecasting model to obtain a first forecasting result, and performing procurement forecasting on the fourth related data according to the procurement forecasting model to obtain a second forecasting result;

[0019] The target management result is determined based on the first prediction result and the second prediction result, and the target management result includes the first management result corresponding to the third sales plan performed by the supplier to the purchasing object and the second management result corresponding to the fourth sales plan performed by the purchasing object to the supplier.

[0020] The embodiment of the present invention provides a buyer collaborative management system and method based on sales plan forecasting, the system comprising: a first acquisition module, used to obtain first associated data corresponding to a first sales plan carried out by a supply object on a purchase object and to obtain second associated data corresponding to a second sales plan carried out by a purchase object on the supply object; a first analysis module, used to perform data association analysis on the first associated data to obtain first associated features corresponding to the first associated data; a second analysis module, used to perform data association analysis on the second associated data to obtain second associated features corresponding to the second associated data; a model building module, used to establish a supply forecasting model based on the first associated data combined with the first associated features, and to analyze the supply forecasting model based on the second associated data combined with the second associated features. Establish a purchase forecasting model; a second acquisition module, used to obtain the third associated data corresponding to the third sales plan of the supply object to the purchase object and obtain the fourth associated data corresponding to the fourth sales plan of the purchase object to the supply object; a data forecasting module, used to obtain a first forecast result by performing supply forecasting on the third associated data according to the supply forecasting model, and obtain a second forecast result by performing purchase forecasting on the fourth associated data according to the purchase forecasting model; a result determination module, used to determine the target management result according to the first forecasting result and the second forecasting result, and the target management result includes the first management result corresponding to the third sales plan of the supply object to the purchase object and the second management result corresponding to the fourth sales plan of the purchase object to the supply object. The system collects and organizes the first associated data and the second associated data of the supply object and the purchase object under different sales and purchase plans through the first acquisition module. These data provide a solid foundation for subsequent analysis and model establishment, making decision-making more data-driven, reducing the influence of human errors and subjective judgments, and then through the first analysis module and the second analysis module, the first associated data and the second associated data are deeply analyzed for data association, and key associated features are extracted from them. These features provide an important reference for the optimization of the supply chain. Therefore, the model building module establishes a supply forecast model and a procurement forecast model according to the first associated data and the second associated data and their corresponding first associated features and second associated features. These models can accurately predict future supply and procurement needs, help enterprises prepare in advance, and avoid inventory backlogs or insufficient supply. Finally, the data forecast module uses the established supply forecast model and procurement forecast model to perform real-time forecasts and generate first forecast results and second forecast results. This real-time response capability enables the supply object and the procurement object to quickly adjust production and supply strategies to better meet market demand. Then, the result determination module comprehensively evaluates each link of the supply chain according to the first forecast result and the second forecast result, and finally determines the target management result. This comprehensive evaluation and decision-making process improves the accuracy and scientificity of decision-making and avoids the deviation that may be caused by a single forecast.Through accurate supply and procurement forecasts, inventory and production can be managed more effectively, inventory costs and operational risks can be reduced. At the same time, losses caused by demand fluctuations can be avoided, thereby optimizing supply chain management, improving decision-making accuracy and response speed, reducing costs and risks, and ultimately achieving higher customer satisfaction and market competitiveness. It also solves the problem that traditional supply chain management usually relies on manual procurement and supply planning based on historical data and experience forecasts, which is often not flexible and accurate enough when facing market fluctuations, seasonal changes and other uncertain factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0022] Figure 1 A schematic diagram of the module structure of a buyer collaborative management system based on sales plan forecasting provided by an embodiment of the present invention;

[0023] Figure 2 A flowchart of a method for collaborative management of purchasers based on sales plan forecasting provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0026] It should be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0027] The embodiment of the present invention provides a buyer collaborative management system and method based on sales plan forecasting. The buyer collaborative management system based on sales plan forecasting can be applied to a terminal device, which can be an electronic device such as a tablet computer, a laptop computer, a desktop computer, a personal digital assistant, and a wearable device. The terminal device can be a server or a server cluster.

[0028] Some embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0029] Please refer to Figure 1 , Figure 1 A schematic diagram of the module structure of a buyer collaborative management system 200 based on sales plan forecasting provided in an embodiment of the present invention. The buyer collaborative management system 200 based on sales plan forecasting includes a first acquisition module 201, a first analysis module 202, a second analysis module 203, a model building module 204, a second acquisition module 205, a data prediction module 206, and a result determination module 207, wherein the first acquisition module 201 is used to obtain the first associated data corresponding to the first sales plan of the supply object to the purchase object and to obtain the second associated data corresponding to the second sales plan of the purchase object to the supply object; the first analysis module 202 is used to perform data association analysis on the first associated data to obtain the first associated feature corresponding to the first associated data; the second analysis module 203 is used to perform data association analysis on the second associated data to obtain the second associated feature corresponding to the second associated data; the model building module 204 is used to establish a supply forecasting model according to the first associated data combined with the first associated feature. type, and establish a procurement forecast model according to the second associated data combined with the second associated feature; a second acquisition module 205, used to obtain the third associated data corresponding to the third sales plan performed by the supply object on the procurement object and to obtain the fourth associated data corresponding to the fourth sales plan performed by the procurement object on the supply object; a data prediction module 206, used to perform supply forecasting on the third associated data according to the supply forecasting model to obtain a first forecasting result, and to perform procurement forecasting on the fourth associated data according to the procurement forecasting model to obtain a second forecasting result; a result determination module 207, used to determine a target management result according to the first forecasting result and the second forecasting result, the target management result including the first management result corresponding to the third sales plan performed by the supply object on the procurement object and the second management result corresponding to the fourth sales plan performed by the procurement object on the supply object.

[0030] For example, in a commodity supply chain, a supply object refers to a unit or individual that provides commodities or services, and a purchase object refers to a unit or individual that purchases commodities or services. The supply object and the purchase object together constitute a complete commodity supply chain. When a supply object provides commodities to a purchase object, in order to ensure that the needs of the purchase object can be met in a timely manner and to alleviate the problem of its own storage or backlog of goods, the supply object will formulate a first sales plan for the preparation of goods for the purchase object based on historical experience or historical data. The first sales plan is the reference information in the historical records that the supply object relies on when providing commodities to the purchase object. The reference information includes but is not limited to user needs of market research, historical sales data, market trends, etc.

[0031] Exemplarily, the first sales plan is a plan formulated from the perspective of the supplier's preparation of goods for the purchasing object in the historical records, including reference information that the supplier relies on when providing goods to the purchasing object. Then, the first acquisition module 201 obtains the first related data required by the supplier to make supply forecasts for the purchasing object in the historical records from the first sales plan.

[0032] Exemplarily, when a purchasing object makes a commodity reservation for a supply object, in order to ensure timeliness of obtaining the commodity, the purchasing object needs to formulate a detailed second sales plan when purchasing from the supply object. The second sales plan is the demand information in the historical records that the purchasing object relies on when purchasing from the supply object. The demand information includes but is not limited to the sales status of the goods, the market response of the goods, historical sales data, etc.

[0033] Exemplarily, the second sales plan is a plan formulated from the perspective of the purchasing object's purchase of goods from the supply object in the historical records, including the demand information that the purchasing object relies on when purchasing goods from the supply object, and then the first acquisition module 201 obtains the second related data required by the purchasing object in the historical records when making purchase forecasts for the supply object from the second sales plan.

[0034] Exemplarily, the first analysis module 202 determines an association analysis method, such as correlation analysis, regression analysis, principal component analysis, cluster analysis, etc., and then performs association analysis on different parameters in the first association data according to the determined association analysis method. For example, a linear relationship between any two parameters in the first association data is identified through correlation analysis, or a causal relationship between any two parameters in the first association data is identified through regression analysis, thereby obtaining a first association feature corresponding to the first association data. The first association feature is used to characterize the association relationship between any two parameters in the first association data.

[0035] Exemplarily, the second analysis module 203 determines an association analysis method, such as correlation analysis, regression analysis, principal component analysis, cluster analysis, etc., and then performs association analysis on different parameters in the second association data according to the determined association analysis method. For example, a linear relationship between any two parameters in the second association data is identified through correlation analysis, or a causal relationship between any two parameters in the second association data is identified through regression analysis, thereby obtaining a second association feature corresponding to the second association data. The second association feature is used to characterize the association relationship between any two parameters in the second association data.

[0036] Exemplarily, the model building module 204 uses the first correlation feature to perform data association on the first correlation data according to the parameter fusion layer of the supply prediction model to obtain the first fused data, and then performs regression analysis on the first fused data according to the data regression layer of the supply prediction model to obtain the supply prediction result, and then performs error calculation on the supply prediction result and the actual supply result according to the data loss layer of the supply prediction model to obtain the error value, and then performs model training on the supply prediction model according to the error value until the supply prediction model is obtained. The model building module 204 uses the second correlation feature to perform data association on the second correlation data according to the parameter fusion layer of the purchase prediction model to obtain the second fused data, and then performs regression analysis on the second fused data according to the data regression layer of the purchase prediction model to obtain the purchase prediction result, and then performs error calculation on the purchase prediction result and the actual purchase result according to the data loss layer of the purchase prediction model to obtain the error value, and then performs model training on the purchase prediction model according to the error value until the purchase prediction model is obtained.

[0037] Exemplarily, the second acquisition module 205 obtains the third associated data corresponding to the third sales plan of the supply object to the purchase object and obtains the fourth associated data corresponding to the fourth sales plan of the purchase object to the supply object. The third sales plan is a plan formulated from the perspective of the supply object preparing goods for the purchase object at the current time, including reference information that the supply object relies on when providing goods to the purchase object. The fourth sales plan is a plan formulated from the perspective of the purchase object purchasing goods from the supply object at the current time, including demand information that the purchase object relies on when purchasing goods from the supply object.

[0038] Exemplarily, the data prediction module 206 performs supply prediction on the third related data according to the supply prediction model to obtain a first prediction result of the goods that need to be prepared from the perspective of the supply object, and performs procurement prediction on the fourth related data according to the procurement prediction model to obtain a second prediction result of the goods procurement from the perspective of the procurement object.

[0039] Exemplarily, the result determination module 207 fuses the first prediction result predicted from the perspective of the supply object and the second prediction result predicted from the perspective of the purchasing object to obtain a target fusion result, thereby preparing the goods to be prepared according to the target fusion result and then informing the supply object of the corresponding preparation quantity and informing the purchasing object of the purchase quantity that can be purchased, and then determining the first management result corresponding to the third sales plan of the supply object for the purchasing object according to the corresponding preparation quantity of the supply object and determining the second management result corresponding to the fourth sales plan of the purchasing object for the supply object according to the purchase quantity that the purchasing object can purchase, thereby obtaining the target management result based on the first management result and the second management result.

[0040] For example, the difference between the first prediction result and the second prediction result is calculated. When the difference is within the preset range, it means that the gap between the first prediction result and the second prediction result is small, and the prediction reliability is high. Then the supplier can be informed to prepare goods according to the maximum value of the first prediction result and the second prediction result, and the purchaser can be informed to purchase goods according to the second prediction result. When the difference is not within the preset range, it means that the gap between the first prediction result and the second prediction result is large, and the prediction reliability is low. Then the target management result is that the third sales plan and the fourth sales plan need to be readjusted until the purchase quantity of the purchaser and the corresponding preparation quantity of the supply are obtained.

[0041] In some embodiments, the first analysis module includes: a first parameter screening module, used to obtain the first target parameter from the first associated data, and to obtain the second target parameter by eliminating any first parameter to be screened from the first target parameter; a first data determination module, used to obtain the fifth associated data corresponding to the second target parameter from the first associated data; a first reliable calculation module, used to calculate the first reliable value corresponding to the first parameter to be screened in the supply forecasting model according to the first associated data and the fifth associated data; a first data screening module, used to screen the first target parameter according to the first reliable value to obtain the third target parameter, and obtain the third data corresponding to the third target parameter from the first associated data; a first relationship calculation module, used to perform data association analysis on any two parameters of the third target parameter according to the third data to obtain the first association feature.

[0042] Exemplarily, the first analysis module includes a first parameter screening module, a first data determination module, a first reliable calculation module, a first data screening module, and a first relationship calculation module.

[0043] Exemplarily, the first parameter screening module extracts the first target parameter from the first associated data. The first target parameter is all parameter variables directly related to the supply forecast. Any first parameter to be screened is removed from the first target parameter to obtain the second target parameter. That is, after any one parameter is removed from the first target parameter, the remaining parameter is determined as the second target parameter.

[0044] Exemplarily, the first data determination module extracts the fifth associated data corresponding to the second target parameter from the first associated data, that is, determines the remaining data after removing the data corresponding to the first parameter to be screened from the first associated data as the fifth associated data corresponding to the second target parameter.

[0045] Exemplarily, the first reliable calculation module uses a selected prediction model (such as linear regression, decision tree, random forest, etc.), takes the first associated data and the fifth associated data as input, trains the supply prediction model to obtain the first prediction model corresponding to the first associated data and the second prediction model corresponding to the fifth associated data, and then evaluates the first reliability of the first parameter to be screened by observing the change in the prediction error of the first prediction model and the second prediction model under different parameter combinations. If the model error increases significantly after removing the first parameter to be screened, it means that the first parameter to be screened has a higher reliability.

[0046] Exemplarily, the first data screening module determines a preset threshold value, and compares the preset threshold value with the first reliability value. When the first reliability value is less than the preset threshold value, it indicates that the first parameter to be screened is relatively stable in the supply forecast model. When the first reliability value is greater than or equal to the preset threshold value, it indicates that the first parameter to be screened is relatively random in the supply forecast model. Thus, the first parameter to be screened corresponding to the first reliability value being greater than or equal to the preset threshold value is removed from the first target parameter to obtain the third target parameter. Then, the third data corresponding to the third target parameter is extracted from the first associated data.

[0047] Exemplarily, the first relationship calculation module performs data association analysis on any two parameters in the third target parameters by correlation analysis, regression analysis, cluster analysis or the like to obtain a first association feature, so as to discover a potential relationship between any two parameters in the third target parameters.

[0048] In some embodiments, the first reliability calculation module includes: a first segmentation module, which is used to determine a preset distance and a first target window, and perform data segmentation on the fifth associated data according to the first target window to obtain first window data; a first analysis module, which is used to calculate the first data distance between any two of the second target parameters in the first window data, and determine the first data association degree corresponding to the first window data according to the first data distance and the preset distance; a first mean calculation module, which is used to perform mean calculation based on the first data association degrees of all the first window data to obtain the first mean association degree corresponding to the fifth associated data; a second segmentation module, which is used to perform data segmentation on the first associated data according to the first target window to obtain second window data; a second analysis module, which is used to calculate the second data distance between any two of the first target parameters in the second window data, and determine the second data association degree corresponding to the second window data according to the second data distance and the preset distance; a second mean calculation module, which is used to perform mean calculation based on the second data association degrees of all the second window data to obtain the second mean association degree corresponding to the first associated data; a numerical calculation module, which is used to determine the first reliability value corresponding to the first parameter to be screened in the supply forecast model according to the first mean association degree and the second mean association degree; wherein the first reliability value is obtained according to the following formula:

[0049]

[0050] in, represents the first reliable value corresponding to the kth first parameter to be screened; ln represents taking the natural logarithm, N represents the window size corresponding to the first target window, represents the first data association degree corresponding to the i-th first window data, represents the second data association degree corresponding to the j-th second window data, represents the first mean correlation, represents the second mean correlation.

[0051] Exemplarily, the first segmentation module determines a preset distance and a first target window. The preset distance is a standard for measuring the similarity between data, and the first target window is a time or sample window size for segmenting data. According to the first target window, the fifth associated data is segmented by a sliding window to obtain first window data.

[0052] Exemplarily, the first analysis module calculates the first data distance between any two second target parameters in the first window data. The first data distance can be Euclidean distance, Manhattan distance, cosine similarity, etc., so that the calculated first data distance is compared with the preset distance. When the first data distance is less than the preset distance, the first data association degree is increased by 1, and when the first data distance is greater than or equal to the preset distance, the first data association degree remains unchanged. Thus, after obtaining the first data distance between any two second target parameters in the first window data, the corresponding first data association degree is obtained.

[0053] Exemplarily, the first mean value calculation module performs mean value calculation according to the first data association degree of all the first window data to obtain the first mean association degree corresponding to the fifth association data. The mean association degree reflects the overall association status of the fifth association data.

[0054] Exemplarily, the second segmentation module performs data segmentation on the first associated data by using a sliding window method according to the first target window to obtain second window data.

[0055] Exemplarily, the second analysis module calculates the second data distance between any two first target parameters in the second window data, and the calculation method of the data distance is the same as that of the first analysis module. According to the calculated second data distance and the preset distance, the second data association degree corresponding to the second window data is determined. The determination method of the second data association degree is the same as that of the first analysis module.

[0056] Exemplarily, the second mean value calculation module performs mean value calculation according to the second data association degree of all the second window data to obtain the second mean association degree corresponding to the first association data. The second mean association degree reflects the overall association status of the first association data.

[0057] Exemplarily, the numerical calculation module determines the first reliability value corresponding to the first parameter to be screened in the supply forecast model by fusing the first mean correlation and the second mean correlation according to the following formula:

[0058]

[0059] in, represents the first reliable value corresponding to the kth first parameter to be screened; ln represents taking the natural logarithm, N represents the window size corresponding to the first target window, represents the first data association degree corresponding to the i-th first window data, represents the second data association degree corresponding to the j-th second window data, represents the first mean correlation, represents the second mean correlation.

[0060] Exemplarily, there is a difference between the first parameter to be screened in the first associated data and the second associated data, so that when the first reliable value calculated by the formula is larger, this indicates that the first parameter to be screened has a larger change range in the supply forecasting model that affects the forecast result, and the randomness is higher. In other words, the performance of the first parameter to be screened in different situations is inconsistent, which may lead to an increase in the uncertainty of the forecast result. Therefore, the larger the first reliable value, the higher the randomness of the parameter in the data set, and it may not be suitable as a key variable of the forecasting model. On the contrary, when the first reliable value calculated by the formula is smaller, this indicates that the first parameter to be screened has a smaller change range in the supply forecasting model that affects the forecast result, and the stability is higher. This means that the performance of the first parameter to be screened in different situations is relatively consistent, which helps to improve the accuracy and reliability of the forecast results. Therefore, the smaller the first reliable value, the higher the stability of the parameter in the data set, and the more suitable it is as a key variable of the forecasting model, and then the stability and randomness of the first parameter to be screened can be effectively evaluated, so as to select the most suitable parameter for the supply forecasting model and improve the forecasting accuracy and reliability of the model.

[0061] In some embodiments, the first relationship calculation module includes: a third segmentation module, used to determine the second target window, and perform data segmentation on the third data according to the second target window to obtain third window data; a data disassembly module, used to obtain first analysis data corresponding to the first parameter and second analysis data corresponding to the second parameter from the third window data; a variance calculation module, used to calculate the first variance corresponding to the first analysis data and the second variance corresponding to the second analysis data; an association calculation module, used to calculate the covariance data corresponding to the first analysis data and the second analysis data, and determine the window association between the first parameter and the second parameter under the third window data according to the covariance data combined with the first variance and the second variance; a fusion calculation module, used to fuse the window association of all the third window data under the second target window to obtain the target association value between the first parameter and the second parameter; a relationship determination module, used to determine the first association feature between the first parameter and the second parameter according to the target association value.

[0062] Exemplarily, the third segmentation module determines a second target window. The second target window is used to segment the time or sample window size of the data. The third data is segmented using a sliding window method according to the second target window to obtain third window data.

[0063] Exemplarily, the data disassembly module extracts first analysis data corresponding to the first parameter and second analysis data corresponding to the second parameter from the third window data.

[0064] Exemplarily, the variance calculation module calculates a first variance corresponding to the first analysis data. The first variance reflects the degree of variation of the first parameter in the third window data. The second variance corresponding to the second analysis data is calculated. The second variance reflects the degree of variation of the second parameter in the third window data.

[0065] Exemplarily, the correlation calculation module calculates the covariance data corresponding to the first analysis data and the second analysis data. Covariance is used to measure the linear relationship between the first parameter and the second parameter. Thus, according to the calculated covariance data, combined with the first variance and the second variance, the window correlation between the first parameter and the second parameter under the third window data is determined. The window correlation reflects the correlation between the first parameter and the second parameter in a specific window. The product data is obtained by calculating the product between the first variance and the second variance, and then the window correlation between the first parameter and the second parameter under the third window data is obtained by dividing the covariance data by the product data.

[0066] Exemplarily, the fusion calculation module calculates an average value of the window correlation of all third window data under the second target window to obtain a target correlation value between the first parameter and the second parameter.

[0067] Exemplarily, the relationship determination module determines a reference threshold, and when the target association value is greater than or equal to the reference threshold, the first association feature between the first parameter and the second parameter is strongly correlated or significantly correlated, and when the target association value is less than the reference threshold, the first association feature between the first parameter and the second parameter is weakly correlated or uncorrelated.

[0068] In some embodiments, the second analysis module includes: a second parameter screening module, used to obtain the fourth target parameter from the second associated data, and to obtain the fifth target parameter by eliminating any second parameter to be screened from the fourth target parameter; a second data determination module, used to obtain the sixth associated data corresponding to the fifth target parameter from the second associated data; a second reliable calculation module, used to calculate the second reliable value corresponding to the second parameter to be screened in the procurement forecast model according to the second associated data and the sixth associated data; a second data screening module, used to screen the fourth target parameter according to the second reliable value to obtain the sixth target parameter, and obtain the fourth data corresponding to the sixth target parameter from the second associated data; a second relationship calculation module, used to perform data association analysis on any two parameters of the sixth target parameter according to the fourth data to obtain the second association feature.

[0069] Exemplarily, the second parameter screening module extracts the fourth target parameter from the second associated data. The fourth target parameter is all parameter variables directly related to the purchase forecast. Any second parameter to be screened is removed from the fourth target parameter to obtain the fifth target parameter. That is, after any one parameter is removed from the fourth target parameter, the remaining parameter is determined as the fifth target parameter.

[0070] Exemplarily, the second data determination module extracts the sixth associated data corresponding to the fifth target parameter from the second associated data, that is, determines the remaining data after removing the data corresponding to the second parameter to be screened from the second associated data as the sixth associated data corresponding to the fifth target parameter.

[0071] Exemplarily, the second reliable calculation module determines the preset distance and the third target window. The preset distance is a standard for measuring the similarity between data, and the third target window is the time or sample window size for segmenting data. According to the third target window, the sixth associated data is segmented by a sliding window to obtain the third window data, and then the third data distance between any two parameters in the third window data is calculated. The third data distance can be Euclidean distance, Manhattan distance, cosine similarity, etc., so that the calculated third data distance is compared with the preset distance. When the third data distance is less than the preset distance, the third data association degree is added by 1, and when the third data distance is greater than or equal to the preset distance, the third data association degree remains unchanged. Thus, after obtaining the third data distance between any two parameters in the third window data, the corresponding third data association degree is obtained, and then the third data association degree of all the third window data is averaged and calculated to obtain the third mean association degree corresponding to the sixth associated data. This mean association degree reflects the overall association of the sixth associated data. Then, the second associated data is segmented by a sliding window according to the third target window to obtain the fourth window data, thereby calculating the fourth data distance between any two parameters in the fourth window data. The calculation method of the data distance is the same as that of the third data distance. According to the calculated fourth data distance and the preset distance, the fourth data association degree corresponding to the fourth window data is determined, and then the mean value of the fourth data association degree of all the fourth window data is calculated to obtain the fourth mean association degree corresponding to the second associated data. Finally, the third mean association degree and the fourth mean association degree are integrated to determine the second reliable value corresponding to the second parameter to be screened in the procurement forecast model.

[0072] Exemplarily, the second data screening module determines a preset threshold value, and compares the preset threshold value with the second reliability value. When the second reliability value is less than the preset threshold value, it indicates that the second parameter to be screened is relatively stable in the purchase forecast model. When the second reliability value is greater than or equal to the preset threshold value, it indicates that the second parameter to be screened is relatively random in the purchase forecast model. Thus, the second parameter to be screened corresponding to the second reliability value being greater than or equal to the preset threshold value is removed from the fourth target parameter to obtain the sixth target parameter. Then, the fourth data corresponding to the sixth target parameter is extracted from the second associated data.

[0073] Exemplarily, the second relationship calculation module performs data association analysis on any two parameters in the sixth target parameters by correlation analysis, regression analysis, cluster analysis, etc. to obtain a second association feature to discover a potential relationship between any two parameters in the sixth target parameters.

[0074] In some embodiments, the model building module includes: a first parameter determination layer, which is used to use the first parameter determination layer of the supply prediction model to determine the first random parameter corresponding to the first associated data according to the first reliable value, and determine the first associated parameter corresponding to the first associated data according to the first associated feature; a first regression prediction layer, which is used to use the first fusion regression layer of the supply prediction model to perform fusion regression on the first associated data according to the first random parameter and the first associated parameter to obtain a third prediction result; a first loss calculation layer, which is used to use the first loss layer of the supply prediction model to obtain the actual supply result corresponding to the first associated data, and determine the first loss value according to the third prediction result and the actual supply result; a first model training layer, which is used to perform model training on the supply prediction model according to the first loss value to obtain the supply prediction model.

[0075] Exemplarily, the first parameter determination layer determines the first random parameter corresponding to the first association data according to the first reliable value. The first random parameters are those parameters that show high randomness in the supply forecast model, and the first association parameter corresponding to the first association data is determined according to the first association feature. The first association parameter is a parameter that has a significant association in the supply forecast model.

[0076] Exemplarily, the first regression prediction layer uses the first fusion regression layer of the supply prediction model to perform fusion regression on the first associated data according to the first random parameter and the first associated parameter to obtain a third prediction result. The fusion regression layer improves the prediction accuracy by combining different types of parameters and comprehensively considering their contributions.

[0077] Exemplarily, the first loss calculation layer uses the first loss layer of the supply prediction model to obtain the actual supply result corresponding to the first associated data, and determines the first loss value according to the third prediction result and the actual supply result. The loss value is usually calculated using methods such as mean square error (MSE) and mean absolute error (MAE) to reflect the difference between the predicted result and the actual result.

[0078] Exemplarily, the first model training layer performs model training on the supply prediction model according to the first loss value. The model training process includes the following sub-steps: using an optimization algorithm (such as gradient descent, Adam, etc.) to adjust the parameters of the model so as to minimize the loss value. During the training process, a validation set is used to evaluate the model performance to avoid overfitting; and when the loss value of the model reaches a preset threshold or the model performance no longer improves, the training is stopped. Thus, a trained supply prediction model is obtained, which can better predict the supply results in the first associated data.

[0079] In some embodiments, the model building module includes: a second parameter determination layer, which is used to use the second parameter determination layer of the procurement prediction model to determine the second random parameter corresponding to the second associated data according to the second reliability value, and determine the second associated parameter corresponding to the second associated data according to the second associated feature; a second regression prediction layer, which is used to use the second fusion regression layer of the procurement prediction model to perform fusion regression on the second associated data according to the second random parameter and the second associated parameter to obtain a fourth prediction result; a second loss calculation layer, which is used to use the second loss layer of the procurement prediction model to obtain the actual procurement result corresponding to the second associated data, and determine the second loss value according to the fourth prediction result and the actual procurement result; a second model training layer, which is used to perform model training on the procurement prediction model according to the second loss value to obtain the procurement prediction model.

[0080] Exemplarily, the second parameter determination layer determines the corresponding second random parameters in the second association data according to the second reliable value. The second random parameters are those parameters that show high randomness in the purchase prediction model, and the corresponding second association parameters in the second association data are determined according to the second association feature. The second association parameters are parameters with significant association in the purchase prediction model.

[0081] Exemplarily, the second regression prediction layer uses the second fusion regression layer of the procurement prediction model to perform fusion regression on the second associated data according to the second random parameter and the second associated parameter to obtain a fourth prediction result. The fusion regression layer improves the prediction accuracy by combining different types of parameters and comprehensively considering their contributions.

[0082] Exemplarily, the second loss calculation layer uses the second loss layer of the purchase prediction model to obtain the actual purchase result corresponding to the second associated data. And the second loss value is determined according to the fourth prediction result and the actual purchase result. The loss value is usually calculated using methods such as mean square error (MSE) and mean absolute error (MAE) to reflect the difference between the prediction result and the actual result.

[0083] Exemplarily, the second model training layer performs model training on the purchase prediction model according to the second loss value. The model training process includes the following sub-steps: using an optimization algorithm (such as gradient descent, Adam, etc.) to adjust the parameters of the model so as to minimize the loss value. During the training process, the validation set is used to evaluate the model performance to avoid overfitting, and when the loss value of the model reaches a preset threshold or the model performance is no longer improved, the training is stopped, thereby obtaining a trained purchase prediction model that can better predict the purchase results in the second associated data.

[0084] In some embodiments, the result determination module includes: a data comparison module, used to compare the first prediction result and the second prediction result to obtain a target comparison result; a first adjustment module, used to adjust the first prediction result according to the target comparison result to obtain the first target result corresponding to the supply object when the supply object supplies the purchasing object, and determine the first management result corresponding to the third sales plan performed by the supply object on the purchasing object according to the first target result; a second adjustment module, used to adjust the second prediction result according to the target comparison result to obtain the second target result corresponding to the purchase object when the purchasing object purchases the target object, and determine the second management result corresponding to the fourth sales plan performed by the purchasing object on the supply object according to the second target result; a result generation module, used to determine the target management result according to the first management result and the second management result.

[0085] Exemplarily, the data comparison module compares the first prediction result with the second prediction result to obtain a target comparison result. The comparison may be performed by calculating the difference between the two, such as an absolute difference, a relative difference, etc.

[0086] Exemplarily, the first adjustment module adjusts the first prediction result according to the target comparison result to obtain the first target result corresponding to when the supply object supplies the purchase object. The adjustment method can be based on the weighted average of the difference, the compensation mechanism or the correction of the reference historical data.

[0087] Exemplarily, after obtaining the first target result, the supply object is required to prepare goods according to the first target result. For example, when the current prepared quantity is insufficient, the first management result is to increase the purchase quantity; when the current prepared quantity is large, the first management result is to clear part of the inventory, and so on.

[0088] Exemplarily, the second adjustment module adjusts the second prediction result according to the target comparison result to obtain the second target result corresponding to when the purchase object purchases the target object. The adjustment method can be based on a weighted average of differences, a compensation mechanism, or a correction of reference historical data.

[0089] Exemplarily, after obtaining the second target result, the purchasing object is made to purchase goods according to the second target result. For example, when the current purchase quantity is insufficient to match, the second management result is to increase the purchase volume; when the current purchase quantity is large, the second management result is to renegotiate with the supplier, and so on.

[0090] Exemplarily, the result generation module combines the first management result and the second management result to obtain the final target management result, that is, the target management result includes both the first management result of the supplier preparing data for the purchasing object, and the second management result corresponding to the purchasing object purchasing data from the supplier.

[0091] In some embodiments, the system also includes: a real data acquisition module, used to obtain the real supply quantity of the supply object to the purchasing object under the third related data and the real purchase quantity of the purchasing object to the supply object under the fourth related data; a first optimization module, used to optimize the supply forecast model according to the third related data and the real supply quantity to obtain the optimized supply forecast model; a second optimization module, used to optimize the purchase forecast model according to the fourth related data and the real purchase quantity to obtain the optimized purchase forecast model.

[0092] Exemplarily, the real data acquisition module obtains the real supply quantity of the supply object to the purchase object under the third associated data, and obtains the real purchase quantity of the purchase object to the supply object under the fourth associated data.

[0093] Exemplarily, the first optimization module performs incremental training on the supply forecasting model according to the third associated data and the actual supply quantity to obtain an optimized supply forecasting model, thereby improving the accuracy of the forecast.

[0094] Exemplarily, the second optimization module performs incremental training on the purchase prediction model according to the fourth associated data and the actual purchase quantity to obtain an optimized purchase prediction model, thereby improving the accuracy of the prediction.

[0095] The embodiment of the present invention provides a buyer collaborative management system and method based on sales plan forecasting, the system comprising: a first acquisition module, used to obtain first associated data corresponding to a first sales plan carried out by a supply object on a purchase object and to obtain second associated data corresponding to a second sales plan carried out by a purchase object on the supply object; a first analysis module, used to perform data association analysis on the first associated data to obtain first associated features corresponding to the first associated data; a second analysis module, used to perform data association analysis on the second associated data to obtain second associated features corresponding to the second associated data; a model building module, used to establish a supply forecasting model based on the first associated data combined with the first associated features, and to analyze the supply forecasting model based on the second associated data combined with the second associated features. Establish a purchase forecasting model; a second acquisition module, used to obtain the third associated data corresponding to the third sales plan of the supply object to the purchase object and obtain the fourth associated data corresponding to the fourth sales plan of the purchase object to the supply object; a data forecasting module, used to obtain a first forecast result by performing supply forecasting on the third associated data according to the supply forecasting model, and obtain a second forecast result by performing purchase forecasting on the fourth associated data according to the purchase forecasting model; a result determination module, used to determine the target management result according to the first forecasting result and the second forecasting result, and the target management result includes the first management result corresponding to the third sales plan of the supply object to the purchase object and the second management result corresponding to the fourth sales plan of the purchase object to the supply object. The system collects and organizes the first associated data and the second associated data of the supply object and the purchase object under different sales and purchase plans through the first acquisition module. These data provide a solid foundation for subsequent analysis and model establishment, making decision-making more data-driven, reducing the influence of human errors and subjective judgments, and then through the first analysis module and the second analysis module, the first associated data and the second associated data are deeply analyzed for data association, and key associated features are extracted from them. These features provide an important reference for the optimization of the supply chain. Therefore, the model building module establishes a supply forecast model and a procurement forecast model according to the first associated data and the second associated data and their corresponding first associated features and second associated features. These models can accurately predict future supply and procurement needs, help enterprises prepare in advance, and avoid inventory backlogs or insufficient supply. Finally, the data forecast module uses the established supply forecast model and procurement forecast model to perform real-time forecasts and generate first forecast results and second forecast results. This real-time response capability enables the supply object and the procurement object to quickly adjust production and supply strategies to better meet market demand. Then, the result determination module comprehensively evaluates each link of the supply chain according to the first forecast result and the second forecast result, and finally determines the target management result. This comprehensive evaluation and decision-making process improves the accuracy and scientificity of decision-making and avoids the deviation that may be caused by a single forecast.Through accurate supply and procurement forecasts, inventory and production can be managed more effectively, inventory costs and operational risks can be reduced. At the same time, losses caused by demand fluctuations can be avoided, thereby optimizing supply chain management, improving decision-making accuracy and response speed, reducing costs and risks, and ultimately achieving higher customer satisfaction and market competitiveness. It also solves the problem that traditional supply chain management usually relies on manual procurement and supply planning based on historical data and experience forecasts, which is often not flexible and accurate enough when facing market fluctuations, seasonal changes and other uncertain factors.

[0096] like Figure 2 As shown, the buyer collaborative management method based on sales plan forecasting includes steps S101 to S107.

[0097] Step S101: obtaining first associated data corresponding to a first sales plan performed by a supplier on a purchaser and obtaining second associated data corresponding to a second sales plan performed by the purchaser on the supplier.

[0098] Step S102: performing data association analysis on the first association data to obtain a first association feature corresponding to the first association data.

[0099] Step S103: performing data association analysis on the second association data to obtain a second association feature corresponding to the second association data.

[0100] Step S104: establishing a supply forecasting model based on the first associated data in combination with the first associated features, and establishing a procurement forecasting model based on the second associated data in combination with the second associated features.

[0101] Step S105: obtaining third associated data corresponding to the third sales plan performed by the supplier on the purchaser and obtaining fourth associated data corresponding to the fourth sales plan performed by the purchaser on the supplier.

[0102] Step S106: performing supply forecasting on the third related data according to the supply forecasting model to obtain a first forecasting result, and performing procurement forecasting on the fourth related data according to the procurement forecasting model to obtain a second forecasting result.

[0103] Step S107, determine the target management result based on the first prediction result and the second prediction result, and the target management result includes the first management result corresponding to the third sales plan performed by the supplier to the purchasing object and the second management result corresponding to the fourth sales plan performed by the purchasing object to the supplier.

[0104] It should be noted that technical personnel in the relevant field can clearly understand that, for the convenience and conciseness of description, the specific working process of the buyer collaborative management method based on sales plan forecasting described above can refer to the corresponding process in the aforementioned buyer collaborative management system based on sales plan forecasting embodiment, and will not be repeated here.

[0105] An embodiment of the present invention also provides a storage medium for computer-readable storage, wherein the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement any step of a buyer collaborative management system based on sales plan forecasting as provided in the description of the embodiment of the present invention.

[0106] The storage medium may be an internal storage unit of the terminal device in the aforementioned embodiment, such as a hard disk or memory of the terminal device. The storage medium may also be an external storage device of the terminal device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the terminal device.

[0107] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware embodiment, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or transient medium). As is well known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0108] It should be understood that the term "and / or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations. It should be noted that, in this article, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "including a..." does not exclude the presence of other identical elements in the process, method, article or system including the element.

[0109] The serial numbers of the embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A buyer collaborative management system based on sales plan forecasting, characterized in that: The system comprises: A first acquisition module is used to obtain first associated data corresponding to a first sales plan performed by a supply object on a purchase object and to obtain second associated data corresponding to a second sales plan performed by the purchase object on the supply object; A first analysis module, configured to perform data association analysis on the first association data to obtain a first association feature corresponding to the first association data; A second analysis module, configured to perform data association analysis on the second association data to obtain a second association feature corresponding to the second association data; a model building module, configured to build a supply forecast model based on the first correlation data in combination with the first correlation feature, and to build a purchase forecast model based on the second correlation data in combination with the second correlation feature; The second acquisition module is used to obtain the third associated data corresponding to the third sales plan performed by the supply object on the purchase object and obtain the fourth associated data corresponding to the fourth sales plan performed by the purchase object on the supply object; a data prediction module, configured to perform supply prediction on the third related data according to the supply prediction model to obtain a first prediction result, and perform purchase prediction on the fourth related data according to the purchase prediction model to obtain a second prediction result; A result determination module is used to determine a target management result based on the first prediction result and the second prediction result, wherein the target management result includes a first management result corresponding to the third sales plan performed by the supplier to the purchasing object and a second management result corresponding to the fourth sales plan performed by the purchasing object to the supplier.

2. The system according to claim 1, characterized in that The first analysis module comprises: A first parameter screening module, used for obtaining a first target parameter from the first associated data, and removing any first parameter to be screened from the first target parameter to obtain a second target parameter; A first data determination module, configured to obtain fifth associated data corresponding to the second target parameter from the first associated data; A first reliability calculation module, used for calculating a first reliability value of the first parameter to be screened corresponding to the supply forecast model according to the first associated data and the fifth associated data; a first data screening module, configured to screen the first target parameter according to the first reliable value to obtain a third target parameter, and obtain third data corresponding to the third target parameter from the first associated data; The first relationship calculation module is used to perform data association analysis on any two parameters of the third target parameters according to the third data to obtain the first association feature.

3. The system according to claim 2, characterized in that The first reliable computing module comprises: A first segmentation module, used to determine a preset distance and a first target window, and perform data segmentation on the fifth associated data according to the first target window to obtain first window data; a first analysis module, configured to calculate a first data distance between any two of the second target parameters in the first window data, and determine a first data association degree corresponding to the first window data according to the first data distance and the preset distance; A first mean value calculation module, configured to perform mean calculation according to the first data association degrees of all the first window data to obtain a first mean association degree corresponding to the fifth association data; A second segmentation module, configured to segment the first associated data according to the first target window to obtain second window data; a second analysis module, configured to calculate a second data distance between any two of the first target parameters in the second window data, and determine a second data association degree corresponding to the second window data according to the second data distance and the preset distance; A second mean value calculation module, configured to perform mean calculation on the second data association degrees of all the second window data to obtain a second mean association degree corresponding to the first association data; A numerical calculation module, used for determining a first reliable value corresponding to the first parameter to be screened in the supply forecast model according to the first mean value correlation degree and the second mean value correlation degree; The first reliability value is obtained according to the following formula: ; in, represents the first reliable value corresponding to the kth first parameter to be screened; ln represents taking the natural logarithm, N represents the window size corresponding to the first target window, represents the first data association degree corresponding to the i-th first window data, represents the second data association degree corresponding to the j-th second window data, represents the first mean correlation, represents the second mean correlation.

4. The system according to claim 3, characterized in that The first relationship calculation module includes: A third segmentation module, used for determining a second target window, and performing data segmentation on the third data according to the second target window to obtain third window data; A data disassembly module, used for obtaining first analysis data corresponding to the first parameter and second analysis data corresponding to the second parameter from the third window data; a variance calculation module, used to calculate a first variance corresponding to the first analysis data and a second variance corresponding to the second analysis data; an association calculation module, configured to calculate the covariance data corresponding to the first analysis data and the second analysis data, and determine the window association degree between the first parameter and the second parameter under the third window data according to the covariance data combined with the first variance and the second variance; A fusion calculation module, used for fusing the window correlation of all the third window data under the second target window to obtain a target correlation value between the first parameter and the second parameter; A relationship determination module is used to determine the first association feature between the first parameter and the second parameter according to the target association value.

5. The system according to claim 1, characterized in that The second analysis module comprises: A second parameter screening module, used to obtain a fourth target parameter from the second associated data, and to remove any second parameter to be screened from the fourth target parameter to obtain a fifth target parameter; A second data determination module, configured to obtain sixth associated data corresponding to the fifth target parameter from the second associated data; A second reliability calculation module, used for calculating a second reliability value of the second parameter to be screened corresponding to the procurement forecast model according to the second associated data and the sixth associated data; a second data screening module, configured to screen the fourth target parameter according to the second reliable value to obtain a sixth target parameter, and obtain fourth data corresponding to the sixth target parameter from the second associated data; The second relationship calculation module is used to perform data association analysis on any two parameters of the sixth target parameter according to the fourth data to obtain the second association feature.

6. The system according to claim 2, characterized in that The model building module comprises: a first parameter determination layer, configured to determine a first random parameter corresponding to the first associated data according to the first reliable value using the first parameter determination layer of the supply forecasting model, and to determine a first associated parameter corresponding to the first associated data according to the first associated feature; A first regression prediction layer, configured to use a first fusion regression layer of the supply prediction model to perform fusion regression on the first association data according to the first random parameter and the first association parameter to obtain a third prediction result; a first loss calculation layer, configured to obtain a real supply result corresponding to the first associated data by using the first loss layer of the supply prediction model, and determine a first loss value according to the third prediction result and the real supply result; The first model training layer is used to perform model training on the supply prediction model according to the first loss value to obtain the supply prediction model.

7. The system according to claim 5, characterized in that The model building module comprises: A second parameter determination layer, configured to determine a second random parameter corresponding to the second associated data according to the second reliability value using the second parameter determination layer of the procurement forecast model, and to determine a second associated parameter corresponding to the second associated data according to the second associated feature; A second regression prediction layer, configured to use a second fusion regression layer of the procurement prediction model to perform fusion regression on the second associated data according to the second random parameter and the second associated parameter to obtain a fourth prediction result; a second loss calculation layer, configured to obtain a real purchase result corresponding to the second associated data by using the second loss layer of the purchase prediction model, and determine a second loss value according to the fourth prediction result and the real purchase result; The second model training layer is used to perform model training on the purchase prediction model according to the second loss value to obtain the purchase prediction model.

8. The system according to claim 1, characterized in that The result determination module comprises: A data comparison module, used for comparing the first prediction result with the second prediction result to obtain a target comparison result; A first adjustment module is used to adjust the first prediction result according to the target comparison result to obtain a first target result corresponding to when the supply object supplies the purchase object, and determine the first management result corresponding to the third sales plan performed by the supply object on the purchase object according to the first target result; A second adjustment module is used to adjust the second prediction result according to the target comparison result to obtain a second target result corresponding to when the purchasing object purchases the target object, and determine the second management result corresponding to the fourth sales plan performed by the purchasing object on the supply object according to the second target result; A result generating module is used to determine the target management result according to the first management result and the second management result.

9. The system according to any one of claims 1 to 8, characterized in that: The system further comprises: A real data acquisition module, used to obtain the real supply quantity of the supply object to the purchase object under the third associated data and to obtain the real purchase quantity of the purchase object to the supply object under the fourth associated data; A first optimization module, configured to optimize the supply forecasting model according to the third associated data and the actual supply quantity to obtain the optimized supply forecasting model; The second optimization module is used to optimize the purchase forecast model according to the fourth associated data and the actual purchase quantity to obtain the optimized purchase forecast model.

10. A method for collaborative management of buyers based on sales plan forecasting, characterized in that: include: Obtaining first associated data corresponding to a first sales plan performed by a supply object on a purchase object and obtaining second associated data corresponding to a second sales plan performed by the purchase object on the supply object; Performing data association analysis on the first associated data to obtain a first associated feature corresponding to the first associated data; Performing data association analysis on the second association data to obtain a second association feature corresponding to the second association data; Establishing a supply forecasting model based on the first associated data combined with the first associated feature, and establishing a procurement forecasting model based on the second associated data combined with the second associated feature; Obtaining third associated data corresponding to a third sales plan performed by the supplier on the purchaser and obtaining fourth associated data corresponding to a fourth sales plan performed by the purchaser on the supplier; Performing supply forecasting on the third related data according to the supply forecasting model to obtain a first forecasting result, and performing procurement forecasting on the fourth related data according to the procurement forecasting model to obtain a second forecasting result; The target management result is determined based on the first prediction result and the second prediction result, and the target management result includes the first management result corresponding to the third sales plan performed by the supplier to the purchasing object and the second management result corresponding to the fourth sales plan performed by the purchasing object to the supplier.

Images