A supply chain management method based on industrial Internet identity resolution

Abstract

The present invention discloses a supply chain management method based on industrial Internet identification resolution, which relates to the field of supply chain management technology, including deploying intelligent identification of supply chain items, and the supply chain platform collecting supply chain data through the industrial Internet; monitoring anomalies in the supply chain network in real time according to the supply chain data, and processing the anomalies; and dynamically adjusting the supply chain network in combination with the supply chain data. The beneficial effects of the present invention are as follows: the present invention collects supply chain network data to construct a decision tree model and anomaly scoring function to detect anomalies in the supply chain network, which helps to monitor the supply chain network in real time and reduce losses caused by anomalies. At the same time, the supply chain network is dynamically adjusted in real time, which improves the flexibility and adaptability of the supply chain network, and provides assistance for the future development of the supply chain network by predicting the future state of the supply chain network, effectively helping the supply chain network to grow and develop.

Description

Technical Field

[0001] The present invention relates to the technical field of supply chain management, and in particular to a supply chain management method based on industrial Internet identification and resolution. Background Art

[0002] With the rapid development and popularization of the industrial Internet, supply chain management, as an important part of an enterprise's logistics, production, sales and other links, is facing unprecedented challenges and opportunities. Traditional supply chain management relies on linear and static processes and information systems, and often has problems such as information silos, data delays and low processing efficiency, making it difficult to meet the management needs of modern enterprises for high efficiency, transparency and flexibility. In recent years, supply chain management methods based on the industrial Internet have gradually become a research and application hotspot. By combining intelligent identification and Internet technology, real-time tracking and data collection of items in the supply chain are realized, providing new possibilities for the optimized management of the supply chain. However, the existing technologies lack the comprehensive insight and dynamic adjustment capabilities for the entire supply chain network, especially in the identification and processing of abnormal events in the real-time monitoring of the supply chain and the prediction of the supply chain status based on big data analysis, and are unable to effectively respond to complex and changeable market demands and supply chain risks. Summary of the Invention

[0003] In view of the problems existing in the above-mentioned existing supply chain management methods based on industrial Internet identification and resolution, the present invention is proposed.

[0004] Therefore, the problem to be solved by the present invention is that the existing technologies lack the comprehensive insight and dynamic adjustment capabilities for the entire supply chain network, especially in the identification and processing of abnormal events in the real-time monitoring of the supply chain and the prediction of the supply chain status based on big data analysis, and are unable to effectively respond to complex and changeable market demands and supply chain risks.

[0005] To solve the above technical problems, the present invention provides the following technical solutions: A supply chain management method based on industrial Internet identification and resolution, which includes deploying intelligent identification of supply chain items, and the supply chain platform collects supply chain data through the industrial Internet; real-time monitoring of abnormalities in the supply chain network according to the supply chain data, and processing the abnormalities; dynamically adjusting the supply chain network in combination with the supply chain data, predicting the future status of the supply chain and displaying it through the supply chain platform; the database collects data for storage and implements data security protection measures.

[0006] As a preferred solution of the supply chain management method based on industrial Internet identification and resolution according to the present invention, wherein: deploying intelligent identification for supply chain items, and the supply chain platform collects supply chain data through the industrial Internet refers to deploying intelligent identification on supply chain items to store item information, including RFID tags, two-dimensional codes, and NFC tags, using sensors and reading devices to collect data in the intelligent identification, and transmitting the data to the supply chain platform through the industrial Internet for data cleaning, filtering, and standardization.

[0007] As a preferred solution of the supply chain management method based on industrial Internet identification and resolution according to the present invention, wherein: the real-time monitoring of anomalies in the supply chain network based on the supply chain data includes:

[0008] Obtaining a feature vector from the collected supply chain item information data :

[0009] ,

[0010] where is the feature vector, is the logistics speed at time t, is the inventory level at time t, is the order volume at time t, is the order delivery time, is the item quality score, is the supplier reliability score;

[0011] Calculating the supply chain comprehensive status value:

[0012] ,

[0013] where is the supply chain comprehensive status value;

[0014] Defining the decision tree feature judgment threshold:

[0015] ,

[0016] where is the feature judgment threshold at time t, is the considered time length, is the considered number of time points, is the logistics speed at the i-th time point, is the inventory level at the i-th time point, is the order volume at the i-th time point, is the order delivery time at the i-th time point, is the item quality score at the i-th time point, is the supplier reliability score at the i-th time point, a and b are adjustment parameters;

[0017] Construct a decision tree model:

[0018] ,

[0019] where is the decision tree model, is the input value of the decision tree, is the number of feature vectors, is the comprehensive supply chain status value of the j-th feature vector, is the weight coefficient of the j-th feature vector, is the feature judgment threshold of the j-th feature vector, is the indicator function, which takes the value of 1 when and 0 otherwise;

[0020] Use a regularization term to train the decision tree model, and define the optimization objective function as:

[0021] ,

[0022] where is the number of training features, is the true label of the i-th feature, is the output of the i-th decision tree in the decision tree model, is the loss function, is the regularization coefficient, is the L2 norm of the i-th feature judgment threshold;

[0023] Define an anomaly scoring function based on the decision tree model as:

[0024] ,

[0025] where is the supply chain anomaly score, is the number of decision trees, is the weight of the i-th decision tree. Input the collected supply chain data into the anomaly scoring function to obtain the supply chain anomaly score.

[0026] As a preferred solution of the supply chain management method based on industrial Internet identification and resolution according to the present invention, wherein: the processing of anomalies refers to obtaining the supply chain anomaly score and comparing it with the preset anomaly threshold to judge the supply chain status:

[0027] If < , it indicates that the supply chain is operating well. Continue to collect and process supply chain data, regularly detect anomalies in the supply chain, generate periodic detection reports, and update the parameters of the decision tree model and anomaly scoring function regularly;

[0028] If ≥ , it indicates that there are anomalies in the operation of the supply chain. Notify the staff to analyze the location of the anomaly based on the collected supply chain data and investigate and handle the supply chain anomaly. If the staff analyzes that there is no anomaly in the supply chain, feedback a misjudgment notice to the supply chain platform, mark the supply chain data as misjudged data, and the supply chain platform retrains the decision tree model and anomaly scoring function, and uses the misjudged data to verify the decision tree model and anomaly scoring function after training until the anomaly analysis result is correct.

[0029] As a preferred solution of the supply chain management method based on industrial Internet identification and resolution described in the present invention, wherein: the dynamic adjustment of the supply chain network in combination with supply chain data refers to the real-time dynamic adjustment of the supply chain network through the collected supply chain data:

[0030] ,

[0031] where is the supply chain adjustment strategy value at time t, and are adjustment parameters, is the weight value of the logistics speed at time t, is the weight of the order volume and order delivery time at time t. Iteratively calculate until the value reaches the maximum, and use the input parameters at this time as the adjustment value to adjust the supply chain network.

[0032] As a preferred solution of the supply chain management method based on industrial Internet identification and resolution described in the present invention, wherein: predicting the future state of the supply chain refers to predicting the future state of the supply chain network after detecting and dynamically adjusting the supply chain network anomalies:

[0033] ,

[0034] where is the supply chain feature vector at time , is the i-th feature at time , and are the weights of the feature , , and are parameters for adjusting the shape of the prediction function, is a normalization function, is the total number of features, is the time delay, and after obtaining the supply chain feature vector of time the supply chain data in the feature vector is combined to form the supply chain state data of time .

[0035] As a preferred solution of the supply chain management method based on industrial Internet identification resolution described in the present invention, wherein: the display through the supply chain platform means that the supply chain platform displays the collected and analyzed data to the participants in the supply chain network through a visual control interaction panel, including suppliers, distributors, logistics providers, and supply chain platform staff. The supply chain platform provides interactive data services for the participants in the supply chain network and allows supply chain query and management through the industrial Internet.

[0036] As a preferred solution of the supply chain management method based on industrial Internet identification resolution described in the present invention, wherein: the database collects data for storage and implements data security protection measures, which means that the database stores the collected supply chain network data, decision tree model, anomaly evaluation function, anomaly analysis results, anomaly handling measures process, supply chain dynamic adjustment measures, and future state prediction results. The database sets security access permissions and access passwords for the stored data, and monitors the stored data when there is data access. After the data access is completed, the database scans the accessed data to generate a detection record for synchronous storage.

[0037] A computer device includes: a memory and a processor; the memory stores a computer program, and it is characterized in that: when the processor executes the computer program, the steps of the above-mentioned supply chain management method based on industrial Internet identification resolution are realized.

[0038] A computer-readable storage medium stores a computer program on it, and it is characterized in that: when the computer program is executed by a processor, the steps of the above-mentioned supply chain management method based on industrial Internet identification resolution are realized.

[0039] The beneficial effects of the present invention are as follows: The present invention constructs a decision tree model and an anomaly scoring function by collecting supply chain network data to detect anomalies in the supply chain network, which helps to monitor the supply chain network in real time, reduce losses caused by anomalies, and at the same time dynamically adjust the supply chain network in real time, improving the flexibility and adaptability of the supply chain network. By predicting the future state of the supply chain network, it provides assistance for the future development of the supply chain network and effectively helps the supply chain network to grow and expand. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0041] Figure 1 It is a schematic flow diagram of a supply chain management method based on industrial Internet identification and resolution.

[0042] Figure 2 It is a schematic diagram of a supply chain platform collecting data.

[0043] Figure 3 It is a schematic structural diagram of anomaly detection of a supply chain platform. Specific embodiments

[0044] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will specifically describe the specific embodiments of the present invention in conjunction with the accompanying drawings of the specification.

[0045] In the following description, many specific details are set forth to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0046] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation manner of the present invention. The "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selectively exclusive embodiment from other embodiments.

[0047] Embodiment 1

[0048] Refer to Figures 1 to 3 , which is the first embodiment of the present invention. This embodiment provides a supply chain management method based on industrial Internet identification and resolution. The supply chain management method based on industrial Internet identification and resolution includes the following steps:

[0049] S1. Deploy intelligent identification of supply chain items, and the supply chain platform collects supply chain data through the industrial Internet;

[0050] Specifically, deploy intelligent identification for supply chain items. The supply chain platform collects supply chain data through the industrial Internet, which means deploying intelligent identification on supply chain items to store item information, including RFID tags, QR codes, and NFC tags. Use sensors and reading devices to collect data from the intelligent identification, and transmit the data to the supply chain platform through the industrial Internet for data cleaning, filtering, and standardization.

[0051] By deploying RFID, QR codes, or NFC tags on items, rapid identification and information acquisition of items can be achieved, greatly reducing the error rate and time consumption of traditional manual information input. In addition, intelligent identification also supports real-time tracking, which helps to promptly detect anomalies in the supply chain, such as delays, losses, or thefts. After collecting the data in the intelligent identification through sensors and reading devices, use industrial Internet technology for efficient data transmission. This process not only ensures the real-time and integrity of the data but also makes it possible for cross-regional supply chain management. Through the high-speed and stable transmission of the industrial Internet, enterprises can easily manage supply chain nodes distributed in different regions. On the supply chain platform, the collected data will undergo cleaning, filtering, and standardization processes to ensure the quality and consistency of the data. The beneficial effect of this step is to improve the availability and reliability of the data, providing a solid foundation for subsequent data analysis and decision-making. By removing irrelevant data, correcting errors, and unifying data formats, enterprises can more accurately analyze supply chain performance, identify efficiency bottlenecks, and predict future trends of the supply chain.

[0052] S2. Real-time monitor anomalies in the supply chain network based on supply chain data and handle the anomalies;

[0053] Specifically, real-time monitoring of anomalies in the supply chain network based on supply chain data includes:

[0054] Derive a feature vector from the collected supply chain item information data :

[0055] ,

[0056] where is the feature vector, is the logistics speed at time t, is the inventory level at time t, is the order volume at time t, is the order delivery time, is the item quality score, is the supplier reliability score. By forming a feature vector from the supply chain item information data, the supply chain state can be effectively reflected;

[0057] Calculate the comprehensive supply chain status value:

[0058] ,

[0059] where is the comprehensive status value of the supply chain, which is obtained by synthesizing the collected supply chain network data;

[0060] Define the decision tree feature judgment threshold:

[0061] ,

[0062] where is the feature judgment threshold at time t, is the considered time length, is the number of considered time points, is the logistics speed at the i-th time point, is the inventory level at the i-th time point, is the order volume at the i-th time point, is the order delivery time at the i-th time point, is the item quality score at the i-th time point, is the supplier reliability score at the i-th time point, and b are adjustment parameters;

[0063] Construct a decision tree model:

[0064] ,

[0065] where is the decision tree model, is the decision tree input value, is the number of feature vectors, is the comprehensive status value of the supply chain of the j-th feature vector, is the weight coefficient of the j-th feature vector, is the feature judgment threshold of the j-th feature vector, is the indicator function, when takes the value of 1, otherwise 0;

[0066] Use a regularization term to train the decision tree model, and define the optimization objective function as:

[0067] ,

[0068] where is the number of training features, is the true label of the i-th feature, is the output of the i-th decision tree in the decision tree model, is the loss function, is the regularization coefficient, is the L2 norm of the i-th feature judgment threshold;

[0069] Based on the decision tree model, the anomaly scoring function is defined as:

[0070] ,

[0071] where is the supply chain anomaly score, is the number of decision trees, is the weight of the i-th decision tree. The collected supply chain data is input into the anomaly scoring function to obtain the supply chain anomaly score.

[0072] By collecting and analyzing supply chain data in real time, the present invention can timely detect and handle anomalies in the supply chain, such as logistics delays, overstock or shortage of inventory, abnormal order processing, etc., thereby improving the operational efficiency and response speed of the supply chain. By converting supply chain data into feature vectors, the present invention can more accurately reflect the current state of the supply chain, providing a more scientific and systematic analysis method for supply chain management. This method helps to identify key influencing factors in the supply chain and provides support for decision-making. By constructing a decision tree model, the present invention can automatically identify abnormal states in the supply chain and make accurate judgments according to the rules learned from historical data. This method improves the accuracy and efficiency of anomaly detection. By introducing a regularization term in the decision tree model training process, the present invention can effectively avoid model overfitting and ensure that the model has better generalization ability, which means that the model not only performs well on the training data but also has strong predictive ability for unknown data. By defining an anomaly scoring function, the present invention can quantify the degree of anomalies in the supply chain, providing an intuitive anomaly indicator for supply chain managers. This helps managers quickly identify and focus on high-risk supply chain links and take corresponding measures for intervention.

[0073] Furthermore, processing the anomaly means obtaining the supply chain anomaly score and comparing it with the preset anomaly threshold to judge the supply chain state:

[0074] If < , it indicates that the supply chain is operating well. Keep collecting and processing supply chain data, and regularly conduct anomaly detection on the supply chain to generate periodic detection reports. At the same time, regularly update the parameters of the decision tree model and the anomaly scoring function;

[0075] If ≥ , it indicates that there are abnormalities in the operation of the supply chain. Notify the staff to analyze the location where the abnormality occurs based on the collected supply chain data and conduct investigations and handling of the supply chain abnormality. If the staff analyzes that there is no abnormality in the supply chain, feedback a misjudgment notice to the supply chain platform, mark the supply chain data as misjudged data, and the supply chain platform retrains the decision tree model and the anomaly scoring function. After the training is completed, use the misjudged data to verify the decision tree model and the anomaly scoring function until the anomaly analysis result is correct.

[0076] S3. Dynamically adjust the supply chain network in combination with supply chain data, predict the future state of the supply chain and display it through the supply chain platform;

[0077] Specifically, dynamically adjusting the supply chain network in combination with supply chain data means making real-time dynamic adjustments to the supply chain network through the collected supply chain data:

[0078] ,

[0079] where is the supply chain adjustment strategy value at time t, and are adjustment parameters, is the weight value of the logistics speed at time t, is the weight of the order volume and order delivery time at time t. Iteratively calculate until the value reaches the maximum, and use the input parameters at this time as the adjustment value to adjust the supply chain network.

[0080] By comparing with the preset anomaly threshold, regularly conduct anomaly detection of the supply chain and generate a detection report, which helps to timely discover and record potential problems and change trends in the operation of the supply chain. The beneficial effects of this step include improving the transparency of the supply chain, promoting decision-makers' in-depth understanding of the health status of the supply chain, and formulating improvement measures based on the analysis and suggestions in the report. When the supply chain anomaly score reaches or exceeds the preset threshold, trigger the anomaly handling mechanism and require the staff to conduct a detailed analysis and investigation of the anomaly. If it is confirmed as a misjudgment, mark the relevant data as misjudged data and feedback it to the supply chain platform. This process not only helps to reduce the future misjudgment rate but also provides important information for the continuous optimization of the decision tree model and the anomaly scoring function. The collection and analysis of misjudged data are crucial for improving the accuracy and robustness of the model. By regularly updating the parameters of the decision tree model and the anomaly scoring function, and using misjudged data for model training and verification, the present invention can continuously improve the accuracy and efficiency of anomaly detection. This process ensures that the model can adapt to changes and new data patterns in the operation of the supply chain, thereby more effectively identifying and handling anomaly situations.

[0081] Further, predicting the future state of the supply chain means predicting the future state of the supply chain network after abnormal detection and dynamic adjustment of the supply chain network:

[0082] ,

[0083] where is the supply chain feature vector at time , is the i-th feature at time , and are the weights of feature , , and are parameters for adjusting the shape of the prediction function, is the normalization function, is the total number of features, is the time delay. After obtaining the supply chain feature vector at time , the supply chain data in the feature vector is combined to form the supply chain state data at time .

[0084] By precisely defining and quantifying the key features of the supply chain, the current state of the supply chain can be more comprehensively understood. This step helps to identify which factors have a significant impact on supply chain performance, thus providing support for management decisions. Assigning weights to different supply chain features means that their importance can be adjusted according to their actual impact on the supply chain state. This method ensures that the prediction model can focus on those key features, improving the accuracy and reliability of the prediction. Normalization is an important step in data preprocessing, which ensures that data of different magnitudes can be compared and analyzed under the same standard. This is particularly important for comprehensive analysis using multiple features with different dimensions and for building prediction models. Inputting the normalized supply chain feature vector into the prediction model can predict the future state of the supply chain based on current and historical data. This step is of great significance for early identification of potential supply chain risks, optimization of resource allocation, and improvement of supply chain strategies.

[0085] Furthermore, displaying through the supply chain platform means that the supply chain platform displays the collected and analyzed data to the participants in the supply chain network through a visual control interaction panel, including suppliers, distributors, logistics providers, and supply chain platform staff. The supply chain platform provides interactive data services for the participants in the supply chain network and allows supply chain queries and management through the industrial Internet.

[0086] Collecting and analyzing data from suppliers, distributors, and logistics providers through a supply chain platform enables real-time monitoring and analysis of the entire supply chain status. The beneficial effects of this step include improving the timeliness and accuracy of decision-making, as well as being able to promptly detect and address potential risks and issues. The interactive data services provided by the supply chain platform enable all parties involved in the supply chain network to query, analyze, and manage supply chain data according to their own needs. The provision of such services enhances the collaboration and communication among the parties involved in the supply chain network, contributing to jointly optimizing supply chain management and operations. Through industrial Internet technologies, the supply chain platform allows users to remotely query and manage the supply chain, significantly enhancing the flexibility and efficiency of supply chain management, which is particularly important for supply chain networks operating across regions, ensuring the circulation of information and the timeliness of management decisions. Displaying the analysis data to the personnel involved in the supply chain network through a visual control interaction panel not only enhances the transparency of the supply chain but also enables each party involved in the supply chain to more actively participate in supply chain management. This enhancement of transparency and participation helps to establish a closer and more efficient supply chain cooperation relationship.

[0087] S4. The database collects data for storage and implements data security protection measures.

[0088] Specifically, the database collecting data for storage and implementing data security protection measures means that the database stores the collected supply chain network data, decision tree models, anomaly evaluation functions, anomaly analysis results, anomaly handling measure processes, supply chain dynamic adjustment measures, and future state prediction results. The database sets security access permissions and access passwords for the stored data and monitors the stored data when there is data access. After the data access is completed, the database scans the accessed data to generate detection records for synchronous storage.

[0089] By centrally storing all key information in the database, the efficiency and reliability of data management can be improved. This centralized storage facilitates data retrieval, update, and backup, providing a solid data foundation for supply chain management. Setting secure access permissions and passwords is a basic measure to protect the database from unauthorized access. By authenticating and authorizing users, it can be ensured that only users with corresponding permissions can access sensitive data, thus protecting the security of supply chain information. Real-time monitoring of access to the data stored in the database can promptly detect and respond to potential data leakage or abuse behaviors. This measure helps to strengthen data security management and prevent data from being accessed by unauthorized individuals or programs. After the data is accessed, scanning it and generating detection records can further enhance data security. This can not only track the history of data access but also detect whether the data has been tampered with or damaged during the access process, ensuring the protection of data integrity and accuracy.

[0090] Embodiment 2

[0091] If the described function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs, etc., which can store program codes of various kinds.

[0092] The logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch instructions from the instruction execution system, apparatus, or device and execute the instructions), or in combination with these instruction execution systems, apparatuses, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

[0093] More specific examples (non-exhaustive list) of computer-readable media include the following: electrical connection parts with one or more wirings (electronic devices), portable computer disk cartridges (magnetic devices), random access memories (RAMs), read-only memories (ROMs), erasable programmable read-only memories (EPROMs or flash memories), fiber optic devices, and portable compact disc read-only memories (CDROMs). Additionally, a computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or, if necessary, other suitable processing, and then storing it in a computer memory.

[0094] It should be understood that each part of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits with logic gate circuits for implementing logic functions on data signals, application specific integrated circuits with appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0095] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A supply chain management method based on industrial Internet identity resolution, characterized by: include, Deploy intelligent identification of supply chain items, and the supply chain platform collects supply chain data through the industrial Internet; Monitor anomalies in the supply chain network in real time based on supply chain data and handle the anomalies; Dynamically adjust the supply chain network based on supply chain data, predict the future status of the supply chain and display it through the supply chain platform; The database collects data for storage and implements data security measures; The real-time monitoring of anomalies in the supply chain network based on supply chain data includes: The feature vector is obtained by collecting supply chain item information data : ； in is the feature vector, is the logistics speed at time t, is the inventory level at time t, is the order quantity at time t, For order delivery time, Rate the quality of the item, Rating supplier reliability; Calculate the comprehensive status value of the supply chain: ； in is the comprehensive status value of the supply chain; Define the decision tree feature judgment threshold: ； in is the feature judgment threshold at time t, For the length of time considered, is the number of time points considered, is the logistics speed at the i-th time point, is the inventory level at the i-th time point, is the order quantity at the i-th time point, is the order delivery time at the i-th time point, Score the quality of the item at the i-th time point, is the supplier reliability score at the i-th time point, and b are adjustment parameters; Build a decision tree model: ； in is the decision tree model, Enter values ​​for the decision tree, is the number of eigenvectors, is the comprehensive supply chain status value of the jth eigenvector, is the weight coefficient of the jth eigenvector, is the feature judgment threshold of the jth feature vector, is the indicator function, when The value is 1 when it is, otherwise it is 0; The regularization term is used to train the decision tree model, and the optimization objective function is defined as: ； in is the number of training features, is the true label of the i-th feature, is the output of the i-th decision tree in the decision tree model, is the loss function, is the regularization coefficient, is the L2 norm of the i-th feature judgment threshold; The anomaly scoring function is defined based on the decision tree model as: ； in Score supply chain anomalies, is the number of decision trees, is the weight of the i-th decision tree; Input the collected supply chain data into the anomaly scoring function to obtain a supply chain anomaly score; The processing of the anomaly refers to obtaining a supply chain anomaly score After that, the preset abnormal threshold Make comparisons to determine the status of the supply chain.

2. The supply chain management method based on industrial Internet identity resolution according to claim 1, characterized in that: The deployment of intelligent identification of supply chain items and the collection of supply chain data by the supply chain platform through the Industrial Internet refers to the deployment of intelligent identification on supply chain items to store item information, including RFID tags, QR codes and NFC tags, using sensors and reading devices to collect data in the intelligent identification, and transmitting the data to the supply chain platform through the Industrial Internet for data cleaning, filtering and standardization.

3. The supply chain management method based on industrial Internet identity resolution as claimed in claim 2, characterized in that: The processing of the anomaly refers to obtaining a supply chain anomaly score After that, the preset abnormal threshold Compare and judge the supply chain status: like < , it means that the supply chain is in good operation. We should keep collecting and processing supply chain data, regularly detect anomalies in the supply chain, generate periodic detection reports, and regularly update the parameters of the decision tree model and anomaly scoring function. like ≥ , it means that there are anomalies in the operation of the supply chain. The staff will be notified to analyze the location of the anomaly based on the collected supply chain data and investigate and handle the supply chain anomaly. If the staff analyzes that no anomaly has occurred in the supply chain, a misjudgment notification will be fed back to the supply chain platform, and the supply chain data will be marked as misjudged data. The supply chain platform will re-train the decision tree model and anomaly scoring function, and after the training is completed, the misjudgment data will be used to verify the decision tree model and anomaly scoring function until the anomaly analysis results are correct.

4. The supply chain management method based on industrial Internet identity resolution as claimed in claim 3, characterized in that: The dynamic adjustment of the supply chain network in combination with the supply chain data refers to the real-time dynamic adjustment of the supply chain network through the collected supply chain data: ； in is the supply chain adjustment policy value at time t, and To adjust the parameters, is the weight value of logistics speed at time t, is the weight of the order quantity and order delivery time at time t; Iterate until When the value reaches the maximum, the input parameters at this time are used as adjustment values ​​to adjust the supply chain network.

5. The supply chain management method based on industrial Internet identity resolution as claimed in claim 4, characterized in that: The prediction of the future state of the supply chain refers to the prediction of the future state of the supply chain network after abnormality detection and dynamic adjustment of the supply chain network: ； in For time The supply chain feature vector of For time The i-th feature of and Features The weight of , as well as To adjust the parameters of the prediction function shape, is the normalization function, is the total number of features, For time delay; Get time After obtaining the supply chain feature vector, the supply chain data in the feature vector are combined to form a time Supply chain status data.

6. The supply chain management method based on industrial Internet identity resolution as claimed in claim 5, characterized in that: The so-called display through the supply chain platform means that the supply chain platform displays the collected and analyzed data to the supply chain network participants, including suppliers, sellers, logistics providers and supply chain platform staff, through a visual control interactive panel. The supply chain platform provides interactive data services to the supply chain network participants and allows supply chain query and management through the Industrial Internet.

7. The supply chain management method based on industrial Internet identity resolution according to claim 6 is characterized in that: The database collects data for storage and implements data security protection measures, which means that the database stores the collected supply chain network data, decision tree models, anomaly assessment functions, anomaly analysis results, anomaly handling measures, supply chain dynamic adjustment measures and future state prediction results. The database sets security access rights and access passwords for the stored data, and keeps monitoring the stored data when the data is accessed. After the data access is completed, the database scans the accessed data to generate detection records and stores them synchronously.

8. A computer device comprising: Memory and processor; The memory stores a computer program, wherein the processor implements the steps of any one of claims 1 to 7 when executing the computer program.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images