E-commerce commodity supply chain risk monitoring analysis method and system based on big data

Through the big data-based e-commerce commodity supply chain risk monitoring and analysis method, the clustered neural network model is used to divide the supply chain and deploy monitoring solutions, which solves the problem that the existing technology cannot effectively monitor multi-source supply chain data, and achieves efficient risks Monitor and analyze.

CN118941079AInactive Publication Date: 2024-11-12XINGHANG (SHENZHEN) TRADING CO LTD

Patent Information

Application Number
CN202410978106.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-12
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing technology cannot comprehensively and timely monitor multi-source supply chain data, resulting in the improvement and efficiency of monitoring and analysis.

Method used

Using the big data-based e-commerce commodity supply chain risk monitoring and analysis method, the risk data and supply chain characteristics are analyzed through the clustering neural network model, and the supply chain is divided into clustered sets with similar characteristics, and deployed on it. Monitoring solutions, including big data analysis, Internet of Things and artificial intelligence technology, conduct real-time data monitoring and analysis.

Benefits of technology

Accurate risk monitoring and analysis of the supply chain is achieved, monitoring efficiency and perfection is improved, monitoring and analysis can be completed quickly, and usage needs can be met.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118941079A_ABST
    Figure CN118941079A_ABST
Patent Text Reader

Abstract

The invention discloses an e-commerce commodity supply chain risk monitoring analysis method and system based on big data, and relates to the technical field of supply chain management. Analyzing and extracting commodity supply chain features; training and generating a clustering neural network model; dividing commodity supply chains into clustering sets with similar features; selecting a corresponding monitoring scheme for each divided clustering set; monitoring point positions in the commodity supply chain are set; performing preliminary analysis on the real-time data of the commodity supply chain to obtain real-time data features; forming at least one judgment range based on the commodity supply chain features; and combining the real-time data features with the at least one judgment range to obtain a commodity supply chain risk analysis result. By arranging the risk acquisition module, the feature extraction module, the clustering division module, the monitoring deployment module and the data analysis module, monitoring analysis can be rapidly completed, the monitoring result is perfect enough, and the monitoring efficiency can also meet the use requirements.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of supply chain management, and in particular to a method and system for monitoring and analyzing the risk of an e-commerce commodity supply chain based on big data. Background Art

[0002] Supply chain risk refers to potential threats or events that may cause losses or other adverse effects during the operation of the supply chain. In today's era of globalization and informatization, the supply chain has become an important part of corporate competition, but it is also accompanied by various potential risks. Some companies or individuals in the supply chain may become bottlenecks that affect the normal operation of the entire supply chain due to natural disasters and other events. At this time, supply chain managers are often required to conduct risk assessments on the entire supply chain and formulate corresponding measures for the bottlenecks to ensure the normal operation of the supply chain.

[0003] However, the supply chain covers a wide range of areas and involves a large amount of data with various types, which will cause great interference to monitoring. Existing technologies are unable to comprehensively and timely monitor multi-source supply chain data, resulting in the need to improve the completeness of monitoring and analysis and the monitoring efficiency. Summary of the invention

[0004] In order to solve the above technical problems, a method and system for monitoring and analyzing the risk of e-commerce commodity supply chain based on big data are provided. This technical solution solves the problem proposed in the above background technology that the supply chain involves a wide range, involves a large amount of data and a wide variety of types, which will cause great interference to the monitoring. The existing technology is unable to comprehensively and timely monitor the supply chain data from multiple sources, resulting in the problem that the completeness of the monitoring and analysis and the monitoring efficiency need to be improved.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] E-commerce commodity supply chain risk monitoring and analysis methods based on big data include:

[0007] Based on big data, conduct detailed analysis of the risks in the commodity supply chain and obtain risk data;

[0008] Combined with risk data, analyze and extract commodity supply chain characteristics, which include external environment risk, supplier risk, quality risk, logistics risk and data risk;

[0009] Based on big data, train and generate clustering neural network models;

[0010] Using a clustering neural network model, we analyze risk data and commodity supply chain feature data, and divide the commodity supply chain into cluster sets with similar features.

[0011] According to the results of the clustering neural network model analysis, a corresponding monitoring solution is selected for each divided cluster set, and the corresponding monitoring solution is deployed in the cluster set, wherein the monitoring solution includes a big data analysis technology solution, an Internet of Things technology solution, and an artificial intelligence technology solution;

[0012] Set up monitoring points in the commodity supply chain;

[0013] Collect real-time data of the commodity supply chain through the deployed monitoring solutions at monitoring points in the commodity supply chain;

[0014] Conduct preliminary analysis on the real-time data of the commodity supply chain to obtain the real-time data characteristics;

[0015] Based on the product supply chain characteristics, at least one judgment range is formed;

[0016] The real-time data feature is combined with at least one judgment range to obtain a commodity supply chain risk analysis result.

[0017] Preferably, the detailed analysis of the risks of the commodity supply chain and obtaining risk data comprises the following steps:

[0018] Based on big data, during the supply chain operation process, obtain at least one uncertain event that poses a potential or actual threat to the supply chain from the external environment, internal management, and partners;

[0019] At least one uncertainty event is classified to form risk data.

[0020] Preferably, the step of analyzing and extracting commodity supply chain characteristics in combination with risk data includes the following steps:

[0021] Extract the risks caused by external factors such as economy and natural disasters from the risk data to form external environmental risks;

[0022] Extract the risks caused by supplier bankruptcy and delayed delivery from the risk data to form supplier risks;

[0023] Extract the risks caused by raw material quality and production process factors from the risk data to form quality risks;

[0024] Extract the risks caused by transportation delays and cargo loss from the risk data to form logistics risks;

[0025] Extract the risks caused by network security and multiple data leakage factors from the risk data to form data risks;

[0026] External environmental risks, supplier risks, quality risks, logistics risks and data risks are summarized to obtain the commodity supply chain characteristics.

[0027] Preferably, the training and generating of the clustering neural network model based on big data comprises the following steps:

[0028] An input layer, a hidden layer and an output layer are set, and the input layer, the hidden layer and the output layer constitute a clustering neural network model;

[0029] Based on big data, obtain large sample data sets;

[0030] Establish preliminary input neurons in the input layer, preliminary hidden neurons in the hidden layer, and preliminary output neurons in the output layer;

[0031] Transfer the data in the sample big data set to the input layer and obtain the output results of the output layer;

[0032] Determine whether the output result of the output layer is reasonable. If so, no processing is performed. If not, obtain the path through which the sample large data set flows, adjust the neurons in the path, and retransmit the data in the sample large data set to the input layer until the data output results in the sample large data set are reasonable.

[0033] Preferably, dividing the commodity supply chain into cluster sets with similar characteristics comprises the following steps:

[0034] Decompose the commodity supply chain into at least one basic supply link;

[0035] Obtain at least one abnormal event occurring in each of the basic supply links;

[0036] Abnormal events are classified according to external environment risks, supplier risks, quality risks, logistics risks and data risks to obtain abnormal event categories;

[0037] The category of abnormal events with the highest proportion in each of the basic supply links is used as the category of the basic supply link;

[0038] The basic supply links of the same category are divided into cluster sets with similar characteristics, and the basic supply links of different categories are divided into different cluster sets.

[0039] Preferably, selecting a corresponding monitoring scheme for each divided cluster set comprises the following steps:

[0040] Use a big data analysis technical solution, an Internet of Things technical solution, and an artificial intelligence technical solution to pre-monitor at least one basic supply link in the cluster set, and obtain at least one pre-monitoring result;

[0041] Compare the preliminary monitoring results with the actual situation of at least one basic supply link to obtain the monitoring gap;

[0042] The technical solution corresponding to the pre-monitoring result with the smallest monitoring gap is used as the monitoring solution corresponding to the cluster set, where the pre-monitoring result corresponds to one of the big data analysis technical solution, the Internet of Things technical solution and the artificial intelligence technical solution.

[0043] Preferably, the setting of monitoring points in the commodity supply chain includes the following steps:

[0044] Use Dijkstra algorithm to plan the monitoring path and obtain the monitoring path;

[0045] Monitoring points are selected at the monitoring nodes of the monitoring path, and the monitoring points are arranged at equal intervals.

[0046] Preferably, the preliminary analysis of the real-time data of the commodity supply chain to obtain the real-time data characteristics comprises the following steps:

[0047] Preprocessing the collected data, wherein the preprocessing includes cleaning, denoising and filling;

[0048] Convert the preprocessed data;

[0049] Feature extraction is performed from data through machine learning and pattern recognition to obtain real-time data features.

[0050] Preferably, the step of combining the real-time data feature with at least one judgment range to obtain a commodity supply chain risk analysis result comprises the following steps:

[0051] Obtain at least one abnormal value in the commodity supply chain characteristics;

[0052] Classifying at least one outlier according to external environment risk, supplier risk, quality risk, logistics risk, and data risk to obtain at least one outlier classification set;

[0053] Obtain the value range of the outlier in the outlier classification set, and use the value range of the outlier as the judgment range;

[0054] When the real-time data feature does not fall within the corresponding judgment range, there is no risk in the commodity supply chain;

[0055] When the real-time data feature belongs to the corresponding judgment range, the risk corresponding to the judgment range will be taken as the risk of the commodity supply chain, wherein the risk corresponding to the judgment range is one of the external environment risk, supplier risk, quality risk, logistics risk and data risk.

[0056] The e-commerce commodity supply chain risk monitoring and analysis system based on big data is used to implement the above-mentioned e-commerce commodity supply chain risk monitoring and analysis method based on big data, including:

[0057] A risk acquisition module, which performs a detailed analysis of the risks of the commodity supply chain based on big data to acquire risk data;

[0058] A feature extraction module, wherein the feature extraction module combines risk data to analyze and extract product supply chain features;

[0059] A model training module, wherein the model training module is based on big data and trains and generates a clustering neural network model;

[0060] A clustering module, which analyzes the risk data and commodity supply chain feature data and divides the commodity supply chain into cluster sets with similar features;

[0061] A monitoring deployment module, which selects a corresponding monitoring scheme for each divided cluster set, and deploys the corresponding monitoring scheme in the cluster set to set monitoring points in the commodity supply chain;

[0062] A real-time monitoring module, which collects real-time data of the commodity supply chain at monitoring points in the commodity supply chain through a deployed monitoring solution;

[0063] A data analysis module performs preliminary analysis on the real-time data of the commodity supply chain to obtain real-time data characteristics, forms at least one judgment range based on the commodity supply chain characteristics, and combines the real-time data characteristics with the at least one judgment range to obtain commodity supply chain risk analysis results.

[0064] Compared with the prior art, the present invention has the following beneficial effects:

[0065] By setting up risk acquisition modules, feature extraction modules, clustering modules, monitoring deployment modules and data analysis modules, it is possible to extract features from a large amount of data involved in the supply chain and classify them according to the features, so that the commodity supply chain can be divided into cluster sets with similar features, and then targeted monitoring can be carried out according to the attributes of the cluster sets, thereby ensuring that accurate detection results can be obtained. The classification monitoring method it uses can effectively cope with multi-source supply chain data, and then when analyzing and judging the monitoring data, it can quickly complete the monitoring analysis, and the monitoring results are complete enough, and the monitoring efficiency can meet the usage requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 This is a flow chart of the big data-based e-commerce commodity supply chain risk monitoring and analysis method of the present invention;

[0067] Figure 2 A schematic diagram of a process for obtaining risk data by performing a detailed analysis of the risks of a commodity supply chain according to the present invention;

[0068] Figure 3 This is a schematic diagram of the process of combining risk data to analyze and extract commodity supply chain characteristics in the present invention;

[0069] Figure 4 A schematic diagram of the process of training and generating a clustering neural network model based on big data in the present invention;

[0070] Figure 5 This is a schematic diagram of the process of dividing a commodity supply chain into cluster sets with similar characteristics according to the present invention;

[0071] Figure 6 A schematic diagram of a flow chart of selecting a corresponding monitoring solution for each divided cluster set of the present invention;

[0072] Figure 7 A schematic diagram of the process of setting monitoring points in a commodity supply chain according to the present invention;

[0073] Figure 8 The present invention performs a preliminary analysis on the real-time data of the commodity supply chain to obtain a schematic flow chart of the real-time data characteristics;

[0074] Fig. 9 The present invention combines real-time data features with at least one judgment range to obtain a flow chart of commodity supply chain risk analysis results. DETAILED DESCRIPTION

[0075] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0076] Reference Figure 1 As shown in the figure, the risk monitoring and analysis method of e-commerce commodity supply chain based on big data includes:

[0077] Based on big data, conduct detailed analysis of the risks in the commodity supply chain and obtain risk data;

[0078] Combined with risk data, analyze and extract commodity supply chain characteristics, which include external environment risk, supplier risk, quality risk, logistics risk and data risk;

[0079] Based on big data, train and generate clustering neural network models;

[0080] Using a clustering neural network model, we analyze risk data and commodity supply chain feature data, and divide the commodity supply chain into cluster sets with similar features.

[0081] According to the results of the clustering neural network model analysis, a corresponding monitoring solution is selected for each divided cluster set, and the corresponding monitoring solution is deployed in the cluster set, wherein the monitoring solution includes a big data analysis technology solution, an Internet of Things technology solution, and an artificial intelligence technology solution;

[0082] Set up monitoring points in the commodity supply chain;

[0083] Collect real-time data of the commodity supply chain through the deployed monitoring solutions at monitoring points in the commodity supply chain;

[0084] Conduct preliminary analysis on the real-time data of the commodity supply chain to obtain the real-time data characteristics;

[0085] Based on the product supply chain characteristics, form at least one judgment range;

[0086] The real-time data feature is combined with at least one judgment range to obtain a commodity supply chain risk analysis result.

[0087] Reference Figure 2 As shown in the figure, a detailed analysis of the risks in the commodity supply chain and obtaining risk data include the following steps:

[0088] Based on big data, during the supply chain operation process, obtain at least one uncertain event that poses a potential or actual threat to the supply chain from the external environment, internal management, and partners;

[0089] At least one uncertainty event is classified to form risk data.

[0090] The purpose of using risk data is to analyze the data that has occurred in the commodity supply chain, and then extract subsequent commodity supply chain characteristics based on the existing data, and conduct monitoring deployment and data analysis based on the extracted commodity supply chain characteristics.

[0091] Reference Figure 3 As shown, combining risk data to analyze and extract commodity supply chain characteristics includes the following steps:

[0092] Extract the risks caused by external factors such as economy and natural disasters from the risk data to form external environmental risks;

[0093] Extract the risks caused by supplier bankruptcy and delayed delivery from the risk data to form supplier risks;

[0094] Extract the risks caused by raw material quality and production process factors from the risk data to form quality risks;

[0095] Extract the risks caused by transportation delays and cargo loss from the risk data to form logistics risks;

[0096] Extract the risks caused by network security and multiple data leakage factors from the risk data to form data risks;

[0097] External environmental risks, supplier risks, quality risks, logistics risks and data risks are summarized to obtain the commodity supply chain characteristics.

[0098] Since the commodity supply chain involves multi-source data, there are many risks. Therefore, if classification analysis is not performed during analysis, the information will interfere with each other, which may easily lead to large analysis errors. Therefore, it is necessary to classify risks according to the causes of the risks, and then form the characteristics of the commodity supply chain. Therefore, monitoring and analysis based on the characteristics of the commodity supply chain can make the monitoring process more orderly, thereby improving monitoring efficiency and accuracy.

[0099] Reference Figure 4 As shown, based on big data, training and generating a clustering neural network model includes the following steps:

[0100] An input layer, a hidden layer and an output layer are set, and the input layer, the hidden layer and the output layer constitute a clustering neural network model;

[0101] Based on big data, obtain large sample data sets;

[0102] Establish preliminary input neurons in the input layer, preliminary hidden neurons in the hidden layer, and preliminary output neurons in the output layer;

[0103] Transfer the data in the sample big data set to the input layer and obtain the output results of the output layer;

[0104] Determine whether the output result of the output layer is reasonable. If so, no processing is performed. If not, obtain the path through which the sample large data set flows, adjust the neurons in the path, and retransmit the data in the sample large data set to the input layer until the data output results in the sample large data set are reasonable.

[0105] The purpose of training and generating a clustering neural network model is to automate the clustering process of the commodity supply chain. In this way, the clustering can be completed automatically, which can save operation time. Because the operations at each location in the commodity supply chain are repetitive and the risks they generate are also repetitive, they can be clustered and trained to form a clustering neural network model based on their repetitive patterns.

[0106] Reference Figure 5 As shown in Figure 1, dividing the commodity supply chain into cluster sets with similar characteristics includes the following steps:

[0107] Decompose the commodity supply chain into at least one basic supply link;

[0108] Obtain at least one abnormal event occurring in each of the basic supply links;

[0109] Abnormal events are classified according to external environment risks, supplier risks, quality risks, logistics risks and data risks to obtain abnormal event categories;

[0110] The category of abnormal events with the highest proportion in each of the basic supply links is used as the category of the basic supply link;

[0111] The basic supply links of the same category are divided into cluster sets with similar characteristics, and the basic supply links of different categories are divided into different cluster sets.

[0112] After dividing the commodity supply chain into cluster sets with similar characteristics, each basic supply link in the commodity supply chain is included in the cluster set. At least one basic supply link in the cluster set has the same attribute. The attributes of the basic supply links are classified according to the abnormal event category with the highest proportion in the basic supply link. Then, the basic supply links with the same attributes can be aggregated in the same cluster set, and then a targeted monitoring plan can be formed for the cluster set.

[0113] Reference Figure 6 As shown, selecting a corresponding monitoring scheme for each divided cluster set includes the following steps:

[0114] Use a big data analysis technical solution, an Internet of Things technical solution, and an artificial intelligence technical solution to pre-monitor at least one basic supply link in the cluster set, and obtain at least one pre-monitoring result;

[0115] Compare the preliminary monitoring results with the actual situation of at least one basic supply link to obtain the monitoring gap;

[0116] The technical solution corresponding to the pre-monitoring result with the smallest monitoring gap is used as the monitoring solution corresponding to the cluster set, where the pre-monitoring result corresponds to one of the big data analysis technical solution, the Internet of Things technical solution and the artificial intelligence technical solution.

[0117] Due to the differences in cluster sets, if a single monitoring scheme is used for monitoring, there will be differences in matching degree, which will lead to poor monitoring effect and different analysis results. Therefore, in order to ensure the monitoring effect, it is necessary to assign a monitoring scheme with a higher matching degree to the cluster set.

[0118] Reference Figure 7 As shown, setting monitoring points in the commodity supply chain includes the following steps:

[0119] Use Dijkstra algorithm to plan the monitoring path and obtain the monitoring path;

[0120] Monitoring points are selected at the monitoring nodes of the monitoring path, and the monitoring points are arranged at equal intervals.

[0121] The setting of monitoring points must ensure that there are no omissions in the monitoring, and at the same time, ensure that the number of monitoring points is relatively small, thereby reducing the computing power consumption of the system operation. Using the Dijkstra algorithm for monitoring path planning can make the path as short as possible, and thus the number of monitoring points can be reduced accordingly.

[0122] Reference Figure 8 As shown, a preliminary analysis of the real-time data of the commodity supply chain and the real-time data characteristics include the following steps:

[0123] Preprocessing the collected data, wherein the preprocessing includes cleaning, denoising and filling;

[0124] Convert the preprocessed data;

[0125] Feature extraction is performed from data through machine learning and pattern recognition to obtain real-time data features.

[0126] The classification of real-time data characteristics is consistent with that of commodity supply chain characteristics, and is classified according to external environmental risks, supplier risks, quality risks, logistics risks and data risks. When making comparisons, they are also compared according to external environmental risks, supplier risks, quality risks, logistics risks and data risks.

[0127] Reference Fig. 9 As shown, combining the real-time data feature with at least one judgment range to obtain the commodity supply chain risk analysis result includes the following steps:

[0128] Obtain at least one abnormal value in the commodity supply chain characteristics;

[0129] Classifying at least one outlier according to external environment risk, supplier risk, quality risk, logistics risk, and data risk to obtain at least one outlier classification set;

[0130] Obtain the value range of the outlier in the outlier classification set, and use the value range of the outlier as the judgment range;

[0131] When the real-time data feature does not fall within the corresponding judgment range, there is no risk in the commodity supply chain;

[0132] When the real-time data feature belongs to the corresponding judgment range, the risk corresponding to the judgment range will be taken as the risk of the commodity supply chain, wherein the risk corresponding to the judgment range is one of the external environment risk, supplier risk, quality risk, logistics risk and data risk.

[0133] During the analysis, the cause of the risk can be attributed to one of the external environment risk, supplier risk, quality risk, logistics risk and data risk, and targeted risk management can be carried out, thereby improving the management effect and efficiency.

[0134] The e-commerce commodity supply chain risk monitoring and analysis system based on big data is used to implement the above-mentioned e-commerce commodity supply chain risk monitoring and analysis method based on big data, including:

[0135] A risk acquisition module, which performs a detailed analysis of the risks of the commodity supply chain based on big data to acquire risk data;

[0136] A feature extraction module, wherein the feature extraction module combines risk data to analyze and extract product supply chain features;

[0137] A model training module, wherein the model training module is based on big data and trains and generates a clustering neural network model;

[0138] A clustering module, which analyzes the risk data and commodity supply chain feature data and divides the commodity supply chain into cluster sets with similar features;

[0139] A monitoring deployment module, which selects a corresponding monitoring scheme for each divided cluster set, and deploys the corresponding monitoring scheme in the cluster set to set monitoring points in the commodity supply chain;

[0140] A real-time monitoring module, which collects real-time data of the commodity supply chain at monitoring points in the commodity supply chain through a deployed monitoring solution;

[0141] A data analysis module performs preliminary analysis on the real-time data of the commodity supply chain to obtain real-time data characteristics, forms at least one judgment range based on the commodity supply chain characteristics, and combines the real-time data characteristics with the at least one judgment range to obtain commodity supply chain risk analysis results.

[0142] Furthermore, the present solution also proposes a storage medium on which a computer-readable program is stored, and when the computer-readable program is called, the above-mentioned e-commerce product supply chain risk monitoring and analysis method based on big data is executed.

[0143] It is understandable that the storage medium may be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid state drive (SSD).

[0144] To sum up, the advantages of the present invention are: by setting up a risk acquisition module, a feature extraction module, a clustering module, a monitoring deployment module and a data analysis module, it is possible to perform feature extraction on a large amount of data involved in the supply chain and classify them according to the features, so that the commodity supply chain can be divided into cluster sets with similar characteristics, and then targeted monitoring can be performed according to the attributes of the cluster sets, thereby ensuring that accurate detection results can be obtained. The classification monitoring method used can effectively cope with multi-source supply chain data, and then when analyzing and judging the monitoring data, the monitoring analysis can be completed quickly, and the monitoring results are sufficiently complete, and the monitoring efficiency can also meet the usage requirements.

[0145] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. E-commerce commodity supply chain risk monitoring and analysis method based on big data, It is characterized in that include: Based on big data, conduct detailed analysis of the risks in the commodity supply chain and obtain risk data; Combined with risk data, analyze and extract commodity supply chain characteristics, which include external environment risk, supplier risk, quality risk, logistics risk and data risk; Based on big data, train and generate clustering neural network models; Using a clustering neural network model, we analyze risk data and commodity supply chain feature data, and divide the commodity supply chain into cluster sets with similar features. According to the results of the clustering neural network model analysis, a corresponding monitoring solution is selected for each divided cluster set, and the corresponding monitoring solution is deployed in the cluster set, wherein the monitoring solution includes a big data analysis technology solution, an Internet of Things technology solution, and an artificial intelligence technology solution; Set up monitoring points in the commodity supply chain; Collect real-time data of the commodity supply chain through the deployed monitoring solutions at monitoring points in the commodity supply chain; Conduct preliminary analysis on the real-time data of the commodity supply chain to obtain the real-time data characteristics; Based on the product supply chain characteristics, at least one judgment range is formed; The real-time data feature is combined with at least one judgment range to obtain a commodity supply chain risk analysis result.

2. According to the big data-based e-commerce commodity supply chain risk monitoring and analysis method of claim 1, It is characterized in that The detailed analysis of the risks in the commodity supply chain and the acquisition of risk data include the following steps: Based on big data, during the supply chain operation process, obtain at least one uncertain event that poses a potential or actual threat to the supply chain from the external environment, internal management, and partners; At least one uncertainty event is classified to form risk data.

3. According to the big data-based e-commerce commodity supply chain risk monitoring and analysis method of claim 2, It is characterized in that The step of analyzing and extracting commodity supply chain characteristics by combining risk data includes the following steps: Extract the risks caused by external factors such as economy and natural disasters from the risk data to form external environmental risks; Extract the risks caused by supplier bankruptcy and delayed delivery from the risk data to form supplier risks; Extract the risks caused by raw material quality and production process factors from the risk data to form quality risks; Extract the risks caused by transportation delays and cargo loss from the risk data to form logistics risks; Extract the risks caused by network security and multiple data leakage factors from the risk data to form data risks; External environmental risks, supplier risks, quality risks, logistics risks and data risks are summarized to obtain the commodity supply chain characteristics.

4. According to the big data-based e-commerce commodity supply chain risk monitoring and analysis method of claim 3, It is characterized in that The training and generating of the clustering neural network model based on big data includes the following steps: An input layer, a hidden layer and an output layer are set, and the input layer, the hidden layer and the output layer constitute a clustering neural network model; Based on big data, obtain large sample data sets; Establish preliminary input neurons in the input layer, preliminary hidden neurons in the hidden layer, and preliminary output neurons in the output layer; Transfer the data in the sample big data set to the input layer and obtain the output results of the output layer; Determine whether the output result of the output layer is reasonable. If so, no processing is performed. If not, obtain the path through which the sample large data set flows, adjust the neurons in the path, and retransmit the data in the sample large data set to the input layer until the data output results in the sample large data set are reasonable.

5. According to the big data-based e-commerce commodity supply chain risk monitoring and analysis method of claim 4, It is characterized in that The step of dividing the commodity supply chain into cluster sets with similar characteristics comprises the following steps: Decompose the commodity supply chain into at least one basic supply link; Obtain at least one abnormal event occurring in each of the basic supply links; Abnormal events are classified according to external environment risks, supplier risks, quality risks, logistics risks and data risks to obtain abnormal event categories; The category of abnormal events with the highest proportion in each of the basic supply links is used as the category of the basic supply link; The basic supply links of the same category are divided into cluster sets with similar characteristics, and the basic supply links of different categories are divided into different cluster sets.

6. According to the big data-based e-commerce commodity supply chain risk monitoring and analysis method of claim 5, It is characterized in that The selecting a corresponding monitoring scheme for each divided cluster set comprises the following steps: Use a big data analysis technical solution, an Internet of Things technical solution, and an artificial intelligence technical solution to pre-monitor at least one basic supply link in the cluster set, and obtain at least one pre-monitoring result; Compare the preliminary monitoring results with the actual situation of at least one basic supply link to obtain the monitoring gap; The technical solution corresponding to the pre-monitoring result with the smallest monitoring gap is used as the monitoring solution corresponding to the cluster set, where the pre-monitoring result corresponds to one of the big data analysis technical solution, the Internet of Things technical solution and the artificial intelligence technical solution.

7. According to the big data-based e-commerce commodity supply chain risk monitoring and analysis method of claim 6, It is characterized in that The setting of monitoring points in the commodity supply chain includes the following steps: Use Dijkstra algorithm to plan the monitoring path and obtain the monitoring path; Monitoring points are selected at the monitoring nodes of the monitoring path, and the monitoring points are arranged at equal intervals.

8. According to the big data-based e-commerce commodity supply chain risk monitoring and analysis method of claim 7, It is characterized in that The preliminary analysis of the real-time data of the commodity supply chain to obtain the real-time data characteristics includes the following steps: Preprocessing the collected data, wherein the preprocessing includes cleaning, denoising and filling; Convert the preprocessed data; Feature extraction is performed from data through machine learning and pattern recognition to obtain real-time data features.

9. According to the big data-based e-commerce commodity supply chain risk monitoring and analysis method of claim 8, It is characterized in that The step of combining the real-time data feature with at least one judgment range to obtain a commodity supply chain risk analysis result comprises the following steps: Obtain at least one abnormal value in the commodity supply chain characteristics; Classifying at least one outlier according to external environment risk, supplier risk, quality risk, logistics risk, and data risk to obtain at least one outlier classification set; Obtain the value range of the outlier in the outlier classification set, and use the value range of the outlier as the judgment range; When the real-time data feature does not fall within the corresponding judgment range, there is no risk in the commodity supply chain; When the real-time data feature belongs to the corresponding judgment range, the risk corresponding to the judgment range will be taken as the risk of the commodity supply chain, wherein the risk corresponding to the judgment range is one of the external environment risk, supplier risk, quality risk, logistics risk and data risk.

10. An e-commerce commodity supply chain risk monitoring and analysis system based on big data, used to implement the e-commerce commodity supply chain risk monitoring and analysis method based on big data as described in any one of claims 1 to 9, It is characterized in that include: A risk acquisition module, which performs a detailed analysis of the risks of the commodity supply chain based on big data to acquire risk data; A feature extraction module, wherein the feature extraction module combines risk data to analyze and extract product supply chain features; A model training module, wherein the model training module is based on big data and trains and generates a clustering neural network model; A clustering module, which analyzes the risk data and commodity supply chain feature data and divides the commodity supply chain into cluster sets with similar features; A monitoring deployment module, which selects a corresponding monitoring scheme for each divided cluster set, and deploys the corresponding monitoring scheme in the cluster set to set monitoring points in the commodity supply chain; A real-time monitoring module, which collects real-time data of the commodity supply chain at monitoring points in the commodity supply chain through a deployed monitoring solution; A data analysis module performs preliminary analysis on the real-time data of the commodity supply chain to obtain real-time data characteristics, forms at least one judgment range based on the commodity supply chain characteristics, and combines the real-time data characteristics with the at least one judgment range to obtain commodity supply chain risk analysis results.

Citation Information

Patent Citations

  • Supply chain integrated management method and system

    CN116645021A

  • Cross-border e-commerce risk control method and system based on dynamic neural network

    CN117422306A

Cited By

  • Data set management method oriented to purchase supply chain brain model training

    CN120181590A