Data processing system and method of e-commerce platform

Abstract

The invention discloses a data processing system and method for an e-commerce platform, relates to the technical field of big data processing, and solves the problems that in the prior art, supplier information collection is limited, and problems of suppliers on different product batches are not fed back and updated in time, so that suppliers make misjudgments in the purchasing process; the method comprises the steps of obtaining various index data of suppliers; constructing and training a prediction probability model to obtain the qualification rate of a commodity sampling inspection result; calling default conditions of the suppliers on the platforms, and calculating default rates of the suppliers; judging whether the current batch of products of the corresponding supplier are allowed to enter the collaboration platform or not; if yes, scoring grades are set for the suppliers entering the collaboration platform, and the corresponding suppliers are selected according to the scoring grades; if not, continuing to judge; predicting the daily sales volume by constructing a sales volume analysis model, and determining the purchase time and purchase volume of the suppliers according to the daily sales volume; according to the invention, the reliability and performance capability of the supplier are evaluated, and the cooperation risk is reduced.

Description

Technical Field

[0001] The present invention belongs to the field of big data processing and relates to data management technology, and specifically to a data processing system and method for an e-commerce platform. Background Art

[0002] The application of the collaborative platform of smart supply chain SaaS e-commerce has brought significant benefits to enterprises. First of all, due to the introduction of cloud computing technology, enterprises do not need to invest a lot of hardware equipment and maintenance costs in the e-commerce business segment; instead, enterprises can obtain computing and storage resources on demand through the cloud computing platform according to their actual needs. This not only greatly improves the efficiency of resource utilization, but also realizes flexible cost control, allowing enterprises to be more comfortable in resource allocation.

[0003] The existing technology has certain regional characteristics and limitations in the collection of supplier information. There is no comprehensive and in-depth understanding of the market, which makes it difficult for excellent national suppliers to participate. The procurement process is not competitive enough, and buyers cannot choose high-quality and low-cost products. At the same time, the existing technology lacks a dynamic tracking information management mechanism and cannot conduct multi-dimensional comprehensive inspections in a timely manner. Many suppliers do not provide timely feedback and updates on problems with different product batches, which leads to buyers making wrong judgments during the procurement process.

[0004] The present invention provides a data processing system and method for an e-commerce platform to solve the above technical problems. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a data processing system and method for an e-commerce platform, which are used to solve the problems that the prior art has certain regionality and limitations in the collection of supplier information, lacks a comprehensive and in-depth understanding of the market, makes it difficult for excellent suppliers across the country to participate, the procurement process is not competitive enough, and buyers cannot choose high-quality and low-cost products; at the same time, the prior art lacks a dynamic tracking information management mechanism, and cannot conduct multi-dimensional comprehensive inspections in a timely manner. Many suppliers do not provide timely feedback and updates on problems with different product batches, which leads to buyers making wrong judgments during the procurement process.

[0006] To achieve the above-mentioned purpose, a first aspect of the present invention provides a data processing system for an e-commerce platform, comprising: a data collection module, a supplier selection module, and an inventory management module;

[0007] Data collection module: obtain various indicator data of suppliers; build and train the prediction probability model to obtain the qualified rate of commodity sampling results; retrieve the default situation of suppliers on various platforms and calculate the default rate of suppliers;

[0008] Supplier selection module: Determine whether the current batch of products of the corresponding supplier is allowed to enter the collaborative platform of the smart supply chain SaaS e-commerce; if yes, set the rating level for the supplier entering the collaborative platform, and the purchaser selects the corresponding supplier according to the rating level; if no, continue to judge;

[0009] Inventory management module: predict daily sales by building a sales analysis model, and determine the purchase time and quantity of the supplier based on the daily sales.

[0010] Preferably, the obtaining of various indicator data of the supplier includes:

[0011] Determine whether the supplier has abnormal operation; if yes, refuse the current batch of products of the corresponding supplier to enter the collaborative platform of smart supply chain SaaS e-commerce; if no, obtain various indicators of the supplier from multiple platforms, and conduct irregular spot checks on the supplier's products; eliminate abnormal spot check data to form a spot check report; among them, abnormal operation includes: the supplier is on the list of dishonest enterprises, the supplier operates in violation of regulations; various indicators of the supplier include: product situation, supplier situation; product situation includes: sample number, quantity, spot check time, and spot check results of each spot check; supplier situation includes: supplier information, supplier default rate; supplier information includes: supplier name, contact information, and supply batch.

[0012] By judging whether a supplier has abnormal operation, the present invention can avoid cooperating with bad suppliers, thereby ensuring the quality and safety of the goods on the platform; by irregularly sampling goods, potential quality problems can be discovered and dealt with in a timely manner, further optimizing the selection of suppliers.

[0013] Preferably, the construction and training of the prediction probability model to obtain the qualified rate of the commodity sampling results includes:

[0014] The historical sampling results of each platform are retrieved as input, imported into the prediction probability model, and the prediction probability model is used to predict and output the pass rate of the next sampling inspection, while the actual pass rate of the goods is manually counted; it is determined whether the difference between the pass rate of the next sampling inspection predicted by the prediction probability model and the actual pass rate of the goods manually counted is within the set error range; if yes, the result predicted by the prediction probability model this time is accurate; if no, the result predicted by the prediction probability model this time is inaccurate;

[0015] The accuracy of the prediction probability model prediction is obtained by dividing the number of accurate predictions by the total number of predictions;

[0016] Determine whether the prediction accuracy is higher than 95%; if yes, use the result predicted by the prediction probability model as the actual qualified rate of the next random inspection of goods; if no, discard the prediction result and manually calculate the actual qualified rate of the goods.

[0017] The present invention helps to timely discover model errors by comparing the prediction results of the prediction probability model with the actual pass rate manually counted, and adjust the model parameters accordingly, thereby improving the accuracy of the prediction; when the accuracy of the prediction probability model is high, a sampling strategy can be formulated based on the prediction results of the model, such as giving priority to sampling commodities with lower predicted pass rates, or reducing the sampling frequency of commodities with higher predicted pass rates, thereby improving efficiency while ensuring quality; at the same time, by using the prediction probability model, unnecessary sampling times are reduced and sampling costs are reduced.

[0018] Preferably, the calculation of the default rate of the supplier includes:

[0019] The number of breaches of contract due to insufficient quantity of delivered products is marked as a1, the number of times the qualified rate of delivered products is lower than the qualified threshold is marked as a2, and the number of times the delivery is not made on time is marked as a3; the number of breaches of contract due to individual factors of the supplier is marked as a4; A1 is the total number of times the supplier has signed contracts on other platforms; A2 is the total number of times the supplier has signed contracts on this platform; individual factors include: contract factors, funding issues, and force majeure factors;

[0020] By formula The default rate is calculated; where b i 、c j is the corresponding scoring coefficient; i represents the i-th default of the supplier on other platforms, and j represents the j-th default of the supplier on this platform.

[0021] The present invention can quantitatively evaluate the supplier's default risk by calculating the default rate, so that the purchaser can more intuitively understand the supplier's reliability and stability, thereby optimizing supplier selection: the default rate can be used as one of the important reference indicators for selecting suppliers, and the purchaser can give priority to suppliers with lower default rates, thereby reducing cooperation risks.

[0022] Preferably, the step of determining whether the current batch of products of the corresponding supplier is allowed to enter the collaborative platform of the smart supply chain SaaS e-commerce includes:

[0023] The number of times the actual qualified rate of historical batches exceeds the set qualified rate threshold is counted and scored to obtain the qualified rate number score; the supplier default score is obtained based on the supplier default rate;

[0024] The supplier score is calculated by summing the pass rate score and the supplier default score and marked as GYP.

[0025] Set a scoring threshold and mark it as PFY to determine whether the supplier's score exceeds the scoring threshold; if yes, the current batch of products of the corresponding supplier is allowed to enter the collaborative platform of the smart supply chain saas e-commerce; if no, the current batch of products of the corresponding supplier is denied to enter the collaborative platform of the smart supply chain saas e-commerce.

[0026] The present invention can ensure the quality of suppliers on the platform by rejecting suppliers whose indicators are all lower than the qualified rate, thereby improving the service level and credibility of the entire platform; by scoring the actual qualified rate and breach of contract of the supplier, the cooperation risk of the supplier can be quantitatively evaluated, so that the platform can judge the reliability of the supplier more accurately.

[0027] Preferably, the purchaser selects the corresponding supplier according to the rating level, including:

[0028] Retrieve the supplier ratings of the collaborative platform of the smart supply chain SaaS e-commerce, and mark the suppliers whose GYP is in the range of (PFY+30, PFY+40] as A-level; mark the suppliers whose GYP is in the range of (PFY+20, PFY+30] as B-level; mark the suppliers whose GYP is in the range of (PFY+10, PFY+20] as C-level; mark the suppliers whose GYP is in the range of [PFY, PFY+10] as D-level;

[0029] Choose the supplier with the lowest price within the same scoring range as the cooperative supplier.

[0030] The present invention helps the platform to have a clearer understanding of the pros and cons of suppliers by grading them according to their scores, thereby optimizing the supplier structure and ensuring the overall quality of suppliers on the platform; selecting the supplier with the lowest price within the same scoring range stimulates price competition among suppliers, prompting them to lower prices and improve cost-effectiveness, thereby bringing more benefits to consumers; and through grading and price screening, it helps buyers to quickly locate suppliers that meet the requirements, thereby improving procurement efficiency and shortening the procurement cycle.

[0031] Preferably, predicting daily sales volume by constructing a sales volume analysis model includes:

[0032] Integrate temperature, weather, market dynamics, and historical sales volume for the corresponding time period into a sequence of influencing factors;

[0033] Retrieve the sales analysis model, input the influencing factor sequence into the sales analysis model, obtain the corresponding influencing factor label, and match the corresponding influencing factor type according to the influencing factor label; wherein, the sales analysis model is built based on the artificial intelligence model; the influencing factor label is set to a positive integer; obtain the daily sales volume through the sales analysis model and mark it as RX.

[0034] The present invention helps to avoid buyers from choosing products with unreasonable prices by considering the historical sales volume in the corresponding time period. At the same time, it integrates multiple influencing factors to form an influencing factor sequence, which helps to more comprehensively reflect the changes in the market environment, thereby improving the accuracy of sales volume and price forecasts. There are complex correlations between these influencing factors and sales volume, and the sales analysis model can capture these relationships and make more accurate predictions.

[0035] Preferably, the sales analysis model is constructed based on an artificial intelligence model, including:

[0036] Acquire standard training data; wherein the standard training data includes standard input data consistent with the content attributes of the influencing factor sequence, and standard output data consistent with the content attributes of the influencing factor label;

[0037] The artificial intelligence model is trained using standard training data, and the trained artificial intelligence model is marked as a sales analysis model; wherein the artificial intelligence model includes a convolutional neural network model or a long short-term memory neural network model.

[0038] Preferably, the determining of the purchase time and purchase quantity of the purchaser includes:

[0039] Retrieve the daily sales volume RX, and obtain the corresponding supplier's n-period historical arrival time, marked as LDHn;

[0040] By formula Calculate the safety stock;

[0041] Determine whether the daily inventory of the purchaser is greater than the safety inventory; if yes, the purchaser does not need to purchase goods; if no, the date corresponding to the purchaser's daily inventory being less than or equal to the safety inventory is used as the purchase time;

[0042] The purchaser determines the purchase quantity according to the set purchasing standards.

[0043] By calculating the safety stock, the present invention enables enterprises to set a reasonable inventory threshold to ensure that the inventory level is neither too high, resulting in capital occupation and increased storage costs, nor too low, resulting in out-of-stock risks; by real-time judgment of the relationship between daily inventory and safety stock, it helps enterprises to take replenishment measures in a timely manner to avoid inventory shortages or surpluses.

[0044] To achieve the above object, a second aspect of the present invention provides a data processing method for an e-commerce platform, comprising:

[0045] Obtain various indicator data of suppliers; build and train prediction probability models to obtain the qualified rate of commodity sampling results; retrieve the default information of suppliers on various platforms and calculate the default rate of suppliers;

[0046] Determine whether the current batch of products of the corresponding supplier is allowed to enter the collaborative platform of the smart supply chain SaaS e-commerce; if yes, set the rating level for the supplier entering the collaborative platform, and the purchaser selects the corresponding supplier according to the rating level; if no, continue to judge;

[0047] By building a sales analysis model, we can predict daily sales volume, and use the daily sales volume to determine the purchase time and quantity of the supplier.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] 1. The present invention helps enterprises to fully understand the comprehensive strength and service level of suppliers by obtaining various indicator data of suppliers; accurately predicts the qualified rate of random inspection results of different batches of goods by constructing and training a prediction probability model to ensure the quality of goods; calculates the default rate of purchasers, helps enterprises to evaluate the reliability and performance ability of suppliers and reduce cooperation risks; determines whether suppliers are allowed to enter the collaborative platform of smart supply chain SaaS e-commerce, and forms a scoring table to help enterprises screen out high-quality suppliers, and can intuitively compare the advantages and disadvantages of different suppliers, so as to facilitate enterprises to make wise choices; constructs a sales analysis model to predict daily sales, which helps enterprises to grasp market demand and consumer behavior more accurately; accurate sales forecasts can provide strong support for enterprises to formulate sales strategies and inventory management plans.

[0050] 2. The present invention determines the purchase time and quantity of the supplier through daily sales volume, which helps the enterprise to achieve optimal inventory management; reasonable purchase time and quantity can ensure that the enterprise has sufficient inventory to meet market demand, while avoiding inventory backlog and capital occupation; the entire process of the present invention is based on data for decision-making and analysis, which helps the enterprise to achieve a data-driven operation and management model; by continuously accumulating and analyzing data, the enterprise can continuously optimize the supply chain management strategy and sales strategy, improve the overall operation efficiency and market response speed; high-quality supplier selection and accurate sales forecast can reduce the operating cost of the enterprise; reasonable purchase management can reduce inventory backlog and out-of-stock costs, and improve the economic benefits of the enterprise; information sharing and collaborative operation of each link in the supply chain can be achieved through the collaborative platform of smart supply chain SaaS e-commerce. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0052] Figure 1A schematic diagram showing the relationship between the modules included in the present invention;

[0053] Figure 2 It is a schematic diagram of the specific steps of data collection of the present invention;

[0054] Figure 3 A schematic diagram of the specific steps of supplier selection of the present invention;

[0055] Figure 4 It is a schematic diagram of the specific steps of inventory management of the present invention;

[0056] Figure 5 The figure is a schematic diagram of a specific flow of data processing of the present invention. DETAILED DESCRIPTION

[0057] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than 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.

[0058] See also Figure 1 , the first aspect of the present invention provides a data processing system for an e-commerce platform, including: a data collection module, a supplier selection module, and an inventory management module;

[0059] Data collection module: obtain various indicator data of suppliers; build and train the prediction probability model to obtain the qualified rate of commodity sampling results; retrieve the default situation of suppliers on various platforms and calculate the default rate of suppliers;

[0060] Supplier selection module: Determine whether the current batch of products of the corresponding supplier is allowed to enter the collaborative platform of the smart supply chain SaaS e-commerce; if yes, set the rating level for the supplier entering the collaborative platform, and the purchaser selects the corresponding supplier according to the rating level; if no, continue to judge;

[0061] Inventory management module: predict daily sales by building a sales analysis model, and determine the purchase time and quantity of the supplier based on the daily sales.

[0062] See also Figure 2 , the specific steps of data collection are to retrieve the historical sampling results of each platform as input, import the prediction probability model, and predict the pass rate of the next sampling inspection through the prediction probability model, and at the same time manually count the actual pass rate of the goods; judge whether the difference between the pass rate of the next sampling inspection predicted by the prediction probability model and the actual pass rate of the goods manually counted is within the set error range; if yes, the result predicted by the prediction probability model this time is accurate; if no, the result predicted by the prediction probability model this time is inaccurate;

[0063] The accuracy of the prediction probability model prediction is obtained by dividing the number of accurate predictions by the total number of predictions;

[0064] Determine whether the prediction accuracy is higher than 95%; if yes, use the prediction result of the prediction probability model as the actual qualified rate of the next random inspection; if no, discard the prediction result and manually calculate the actual qualified rate of the goods;

[0065] The number of breaches of contract due to insufficient quantity of delivered products is marked as a1, the number of times the qualified rate of delivered products is lower than the qualified threshold is marked as a2, and the number of times the delivery is not made on time is marked as a3; the number of breaches of contract due to individual factors of the supplier is marked as a4; A1 is the total number of times the supplier has signed contracts on other platforms; A2 is the total number of times the supplier has signed contracts on this platform; individual factors include: contract factors, funding issues, and force majeure factors;

[0066] By formula The default rate is calculated; where b i 、c j is the corresponding scoring coefficient; i represents the number of times the supplier has defaulted on other platforms, and j represents the number of times the supplier has defaulted on the collaborative platform of the smart supply chain SaaS e-commerce;

[0067] For example, we evaluate the suppliers A, B, and C of a clothing factory that have entered the collaborative platform of the smart supply chain SaaS e-commerce. None of the three clothing factory suppliers has any abnormal operation.

[0068] Now we retrieve the historical sampling results of supplier A on various platforms as input, import the prediction probability model, and use the prediction probability model to predict that the pass rate of the next sampling inspection of clothing factory supplier A is 98%. At the same time, the actual pass rate of the goods is 97.5% according to manual statistics; the error range is set to 0-2%;

[0069] It is judged that the difference of 0.5% between the qualified rate of random inspection predicted by the prediction probability model (98%) and the actual qualified rate of manually counted commodities (97.5%) is within the set error range, and the result predicted by the prediction probability model is accurate.

[0070] The total number of predictions obtained by the prediction probability model is 35, and the number of accurate predictions is 34. The accuracy of the prediction probability model is 97.14%. Since 97.14% is higher than 95%, the result of 98% predicted by the prediction probability model is used as the actual qualified rate of the next random inspection of the goods;

[0071] The statistics show that the number of breaches of contract caused by insufficient product delivery on other platforms by supplier A is 1, the number of times the qualified rate of delivered products is lower than the qualified threshold is 2, and the number of times the delivery is not made on time as agreed is 1; the number of breaches of contract due to individual factors by the supplier is 0; the total number of contracts signed by supplier A on other platforms is 12;

[0072] The statistics show that the number of breaches of contract due to insufficient number of delivered products on the collaborative platform of smart supply chain saas e-commerce is 0, the number of times the qualified rate of delivered products is lower than the qualified threshold is 1, and the number of times the delivery is not made according to the agreed time is 1; the number of breaches of contract due to individual factors of the supplier is 0; the total number of times supplier A signs contracts on the collaborative platform of smart supply chain saas e-commerce is 6;

[0073] Obtain the corresponding scoring coefficient through the default scoring coefficient table (such as Table 1);

[0074] By formula The calculated default rate is {[(1×0.4+2×0.25+1×0.3+0×0.05) / 12]+[(0×0.35+1×0.2+1×0.35+0×0.1) / 6]}×100%=19.1%.

[0075] Table 1 Default score coefficient table

[0076] Other platform rating coefficients b Collaborative platform scoring coefficient c a1 0.4 0.35 a2 0.25 0.2 a3 0.3 0.35 a4 0.05 0.1

[0077] Through the above steps, it can be obtained that the default rates of suppliers B and C are 12.5% ​​and 8.33% respectively.

[0078] See also Figure 3 , the specific steps of supplier selection, count the number of times the actual qualified rate of historical batches exceeds the set qualified rate threshold, and score the qualified rate number score; obtain the supplier default score based on the supplier default rate;

[0079] The supplier score is calculated by summing the pass rate score and the supplier default score and marked as GYP.

[0080] Set a scoring threshold and mark it as PFY to determine whether the supplier's score exceeds the scoring threshold; if yes, the current batch of products of the corresponding supplier is allowed to enter the collaborative platform of the smart supply chain SaaS e-commerce; if no, the current batch of products of the corresponding supplier is denied to enter the collaborative platform of the smart supply chain SaaS e-commerce;

[0081] Retrieve the supplier ratings of the collaborative platform of the smart supply chain SaaS e-commerce, and mark the suppliers whose GYP is in the range of (PFY+30, PFY+40] as A-level; mark the suppliers whose GYP is in the range of (PFY+20, PFY+30] as B-level; mark the suppliers whose GYP is in the range of (PFY+10, PFY+20] as C-level; mark the suppliers whose GYP is in the range of [PFY, PFY+10] as D-level;

[0082] Choose the supplier with the lowest price within the same scoring range as the cooperative supplier.

[0083] For example, the pass rate threshold is set to 95%, and the pass rate of supplier A is counted 35 times, of which the actual pass rate exceeds the set pass rate threshold 34 times; the pass rate times exceeding the set pass rate threshold account for 97.14%; the pass rate score of supplier A obtained from Table 2 is 50;

[0084] The number of times the qualified rate of supplier B is counted is 12 times, of which the number of times the actual qualified rate exceeds the set qualified rate threshold is 12 times; the number of qualified rates exceeding the set qualified rate threshold accounts for 100%; the qualified rate score of supplier B obtained from Table 2 is 50;

[0085] The number of times the qualified rate of supplier C is counted is 15 times, of which the actual qualified rate exceeds the set qualified rate threshold 12 times; the number of qualified rates exceeding the set qualified rate threshold accounts for 80%; the qualified rate score of supplier C obtained from Table 2 is 30;

[0086] Obtain the corresponding supplier's pass rate score through the pass rate score table (such as Table 2);

[0087] Table 2 Corresponding supplier's pass rate scores

[0088] Percentage of qualified times Corresponding supplier's pass rate score 95％-100％ 50 85％-94.99％ 40 75％-84.99％ 30 Less than 74.99% 0

[0089] Obtain the corresponding supplier default score through the supplier default score table (such as Table 3);

[0090] Table 3 Supplier Default Score Sheet

[0091] Default rate WYL Supplier Default Score Less than 9.99% 50 10％-14.99％ 40 15％-24.99％ 30 Above 25% 0

[0092] The default rates of suppliers A, B, and C are 19.1%, 12.5%, and 8.33%, respectively, so the corresponding supplier default scores are 30, 40, and 50, respectively;

[0093] The sum of the pass rate score and the supplier default score shows that supplier A scores 80 points; supplier B scores 90 points; and supplier C scores 80 points.

[0094] If the scoring threshold is set to 60 points, the current batches of products from the three suppliers A, B, and C are all allowed to enter the collaborative platform of the smart supply chain SaaS e-commerce;

[0095] Suppliers with GYP in the range of (90, 100] are marked as Class A; suppliers with GYP in the range of (80, 90] are marked as Class B; suppliers with GYP in the range of (70, 80] are marked as Class C; suppliers with GYP in the range of [60, 70] are marked as Class D; among them, suppliers A and C belong to Class C, and supplier B belongs to Class B;

[0096] The purchaser sets the selected suppliers with a score range of (70, 80]. Among them, the wholesale price of supplier A is 30 yuan / piece, and the wholesale price of supplier C is 29 yuan / piece; therefore, supplier C is selected.

[0097] See also Figure 4 ,The specific steps of inventory management are to integrate temperature, weather, market dynamics, and historical sales volume of the corresponding time period into a sequence of influencing factors;

[0098] Retrieve the sales analysis model, input the influencing factor sequence into the sales analysis model, obtain the corresponding influencing factor label, and match the corresponding influencing factor type according to the influencing factor label; wherein the sales analysis model is built based on the artificial intelligence model; the influencing factor label is set as a positive integer; obtain the daily sales volume through the sales analysis model and mark it as RX;

[0099] The time required to obtain the n-period historical arrival time of the corresponding supplier is marked as LDHn;

[0100] By formula Calculate the safety stock;

[0101] Determine whether the daily inventory of the purchaser is greater than the safety inventory; if yes, the purchaser does not need to purchase goods; if no, the date corresponding to the purchaser's daily inventory being less than or equal to the safety inventory is used as the purchase time;

[0102] The purchaser determines the purchase quantity according to the set purchasing standards.

[0103] For example, a sales analysis model based on an artificial intelligence model is constructed to obtain a daily sales volume of 25 pieces; the time required to obtain the corresponding supplier's 5 historical arrivals is marked as 3 days, 2 days, 4 days, 3 days, and 4 days respectively;

[0104] By formula The calculated safety stock is 3.2×25=80 pieces;

[0105] Determine whether the daily inventory of the purchaser is greater than the safety stock of 80 pieces; if yes, the purchaser does not need to purchase goods; if not, the date corresponding to the purchaser's daily inventory being less than or equal to the safety stock is used as the purchase time;

[0106] The purchaser determines the purchase quantity according to the set purchasing standards.

[0107] See also Figure 5 The first aspect of the present invention provides a data processing method for an e-commerce platform, comprising:

[0108] Obtain various indicator data of suppliers; build and train prediction probability models to obtain the qualified rate of commodity sampling results; retrieve the default information of suppliers on various platforms and calculate the default rate of suppliers;

[0109] Determine whether the current batch of products of the corresponding supplier is allowed to enter the collaborative platform of the smart supply chain SaaS e-commerce; if yes, set the rating level for the supplier entering the collaborative platform, and the purchaser selects the corresponding supplier according to the rating level; if no, continue to judge;

[0110] By building a sales analysis model, we can predict daily sales volume, and use the daily sales volume to determine the purchase time and quantity of the supplier.

[0111] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0112] The working principle of the present invention is as follows: the present invention obtains various indicator data of suppliers; constructs and trains a prediction probability model to obtain the qualified rate of commodity sampling results; retrieves the default situation of suppliers on various platforms and calculates the default rate of suppliers; determines whether the current batch of products of the corresponding supplier is allowed to enter the collaborative platform of the smart supply chain SaaS e-commerce; if yes, sets a scoring level for the suppliers entering the collaborative platform, and the suppliers select the corresponding suppliers according to the scoring level; if no, continues to judge; predicts daily sales volume by constructing a sales analysis model, and determines the purchase time and purchase volume of the suppliers according to the daily sales volume.

[0113] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A data processing system for an e-commerce platform, characterized in that: include: Data collection module, supplier selection module, inventory management module; Data collection module: obtain various indicator data of suppliers; build and train the prediction probability model to obtain the qualified rate of commodity sampling results; retrieve the default situation of suppliers on various platforms and calculate the default rate of suppliers; Supplier selection module: determines whether the current batch of products of the corresponding supplier is allowed to enter the collaborative platform of smart supply chain SaaS e-commerce; If yes, the rating level will be set for the suppliers entering the collaborative platform, and the purchaser will select the corresponding supplier according to the rating level; if no, the judgment will continue; Inventory management module: predict daily sales by building a sales analysis model, and determine the purchase time and quantity of the supplier based on the daily sales.

2. The data processing system of an e-commerce platform according to claim 1, characterized in that: The obtaining of various indicator data of suppliers includes: Determine whether the supplier has abnormal operation; if yes, refuse the current batch of products of the corresponding supplier to enter the collaborative platform of smart supply chain SaaS e-commerce; if no, obtain various indicators of the supplier from multiple platforms, and conduct irregular spot checks on the supplier's products; eliminate abnormal spot check data to form a spot check report; among them, abnormal operation includes: the supplier is on the list of dishonest enterprises, the supplier operates in violation of regulations; various indicators of the supplier include: product situation, supplier situation; product situation includes: sample number, quantity, spot check time, and spot check results of each spot check; supplier situation includes: supplier information, supplier default rate; supplier information includes: supplier name, contact information, and supply batch.

3. The data processing system of an e-commerce platform according to claim 1, characterized in that: The construction and training of the prediction probability model to obtain the qualified rate of the commodity sampling results includes: The historical sampling results of each platform are retrieved as input, imported into the prediction probability model, and the prediction probability model is used to predict and output the pass rate of the next sampling inspection, while the actual pass rate of the goods is manually counted; it is determined whether the difference between the pass rate of the next sampling inspection predicted by the prediction probability model and the actual pass rate of the goods manually counted is within the set error range; if yes, the result predicted by the prediction probability model this time is accurate; if no, the result predicted by the prediction probability model this time is inaccurate; The accuracy of the prediction probability model prediction is obtained by dividing the number of accurate predictions by the total number of predictions; Determine whether the prediction accuracy is higher than 95%; if yes, use the result predicted by the prediction probability model as the actual qualified rate of the next random inspection of goods; if no, discard the prediction result and manually calculate the actual qualified rate of the goods.

4. The data processing system of an e-commerce platform according to claim 1, characterized in that: The calculation of the default rate of the purchaser includes: The number of breaches of contract due to insufficient quantity of delivered products is marked as a1, the number of times the qualified rate of delivered products is lower than the qualified threshold is marked as a2, and the number of times the delivery is not made on time is marked as a3; the number of breaches of contract due to individual factors of the supplier is marked as a4; A1 is the total number of times the supplier has signed contracts on other platforms; A2 is the total number of times the supplier has signed contracts on this platform; individual factors include: contract factors, funding issues, and force majeure factors; By formula The default rate is calculated; where b i 、c j is the corresponding scoring coefficient; i represents the i-th default of the supplier on other platforms, and j represents the j-th default of the supplier on this platform.

5. The data processing system of an e-commerce platform according to claim 1, characterized in that: The determination of whether the current batch of products of the corresponding supplier is allowed to enter the collaborative platform of the smart supply chain SaaS e-commerce includes: The number of times the actual qualified rate of historical batches exceeds the set qualified rate threshold is counted and scored to obtain the qualified rate number score; the supplier default score is obtained based on the supplier default rate; The supplier score is calculated by summing the pass rate score and the supplier default score and marked as GYP. Set a scoring threshold and mark it as PFY to determine whether the supplier's score exceeds the scoring threshold; if yes, the current batch of products of the corresponding supplier is allowed to enter the collaborative platform of the smart supply chain saas e-commerce; if no, the current batch of products of the corresponding supplier is denied to enter the collaborative platform of the smart supply chain saas e-commerce.

6. The data processing system of an e-commerce platform according to claim 1, characterized in that: The purchaser selects the corresponding supplier according to the rating level, including: Retrieve the supplier ratings of the collaborative platform of the smart supply chain SaaS e-commerce, and mark the suppliers whose GYP is in the range of (PFY+30, PFY+40] as A-level; mark the suppliers whose GYP is in the range of (PFY+20, PFY+30] as B-level; mark the suppliers whose GYP is in the range of (PFY+10, PFY+20] as C-level; mark the suppliers whose GYP is in the range of [PFY, PFY+10] as D-level; Choose the supplier with the lowest price within the same scoring range as the cooperative supplier.

7. The data processing system of an e-commerce platform according to claim 1, characterized in that: The method of predicting daily sales volume by building a sales volume analysis model includes: Integrate temperature, weather, market dynamics, and historical sales volume for the corresponding time period into a sequence of influencing factors; Retrieve the sales analysis model, input the influencing factor sequence into the sales analysis model, obtain the corresponding influencing factor label, and match the corresponding influencing factor type according to the influencing factor label; wherein, the sales analysis model is built based on the artificial intelligence model; the influencing factor label is set to a positive integer; obtain the daily sales volume through the sales analysis model and mark it as RX.

8. The data processing system of an e-commerce platform according to claim 7, characterized in that: The sales analysis model is constructed based on an artificial intelligence model, including: Acquire standard training data; wherein the standard training data includes standard input data consistent with the content attributes of the influencing factor sequence, and standard output data consistent with the content attributes of the influencing factor label; The artificial intelligence model is trained using standard training data, and the trained artificial intelligence model is marked as a sales analysis model; wherein the artificial intelligence model includes a convolutional neural network model or a long short-term memory neural network model.

9. The data processing system of an e-commerce platform according to claim 1, characterized in that: Determining the purchase time and purchase quantity of the purchaser includes: Retrieve the daily sales volume RX, and obtain the corresponding supplier's n-period historical arrival time, marked as LDHn; By formula Calculate the safety stock; Determine whether the daily inventory of the purchaser is greater than the safety inventory; if yes, the purchaser does not need to purchase goods; if no, the date corresponding to the purchaser's daily inventory being less than or equal to the safety inventory is used as the purchase time; The purchaser determines the purchase quantity according to the set purchasing standards.

10. A data processing method for an e-commerce platform, applied to a data processing system for an e-commerce platform according to any one of claims 1 to 9, characterized in that: include: Obtain various indicator data of suppliers; build and train prediction probability models to obtain the qualified rate of commodity sampling results; retrieve the default information of suppliers on various platforms and calculate the default rate of suppliers; Determine whether the current batch of products of the corresponding supplier is allowed to enter the collaborative platform of smart supply chain SaaS e-commerce; If yes, the rating level will be set for the suppliers entering the collaborative platform, and the purchaser will select the corresponding supplier according to the rating level; if no, the judgment will continue; By building a sales analysis model, we can predict daily sales volume, and use the daily sales volume to determine the purchase time and quantity of the supplier.

Images