Distributed data monitoring method of supply chain intelligent collaborative management platform

Abstract

The invention discloses a distributed data monitoring method for a supply chain intelligent collaborative management platform, and relates to the technical field of distributed data analysis, and the method specifically comprises the following steps: constructing the supply chain intelligent collaborative management platform; installing data monitoring equipment, collecting feature data of different intermediate links, arranging the feature data into time sequence vectors, and respectively calculating dynamic correlation of the different intermediate links in the time dimension; constructing a dynamic association graph, and performing causal testing on the basis of the constructed dynamic association graph to form a directed dynamic association graph; collecting feature data of different nodes in real time, predicting the influence of the feature data change of the nodes on other nodes according to the constructed directed dynamic association graph, and sequencing high-bottleneck risk nodes in real time; according to the method, the dynamic correlation is calculated based on the time sequence vector, the time delay parameter is introduced to construct the directed dynamic association graph, the time sequence dependence and the causal relationship among the links are accurately captured, and the defect that link connection mining is insufficient in traditional monitoring is overcome.

Description

Technical Field

[0001] The present invention relates to the technical field of distributed data analysis, and particularly to a distributed data monitoring method for a supply chain intelligent collaborative management platform. Background Art

[0002] At present, with the booming development of the global economy, the scale of the supply chain system continues to expand and the structure becomes increasingly complex, covering many intermediate links such as raw material supply, product manufacturing, transportation and distribution, warehousing management, and sales and delivery. Each link is closely related and interacts with each other. A fluctuation in one link may quickly spread along the supply chain chain, triggering a chain reaction, and then affecting the efficiency, cost, and stability of the entire supply chain.

[0003] Traditional supply chain data monitoring methods mostly adopt a centralized architecture, where data is uniformly transmitted to a central server for processing and analysis. However, this method exposes many drawbacks when facing large-scale distributed data. On the one hand, the centralized transmission of a large amount of data is likely to cause network congestion, resulting in data transmission delays and affecting the timeliness of information. On the other hand, once the central server fails, the entire monitoring system will face the risk of paralysis, and the reliability and fault tolerance of the system are poor. In addition, traditional monitoring methods often can only simply monitor a single link or local data, and it is difficult to effectively capture the complex temporal dependence relationships and dynamic associations between each link, and it is impossible to conduct collaborative management and risk warning of the supply chain from a global perspective.

[0004] With the continuous development and application of emerging technologies such as the Internet of Things, big data, and artificial intelligence, supply chain management is moving towards the direction of intelligence and collaboration. In order to realize data interconnection and intelligent analysis among the intermediate links of the supply chain and improve the overall efficiency and risk resistance ability of the supply chain, there is an urgent need for an efficient and reliable distributed data monitoring method. This method should be able to collect the characteristic data of each link in real time, deeply mine the dynamic correlation and causal relationship between the data, accurately predict the impact of changes in each link on the overall supply chain, so as to timely identify high bottleneck risk nodes and provide strong support for the intelligent collaborative management of the supply chain. Summary of the Invention

[0005] The purpose of the present invention is to provide a distributed data monitoring method for a supply chain intelligent collaborative management platform to solve the problems raised in the prior art.

[0006] To achieve the above purpose, the present invention provides the following technical solution: A distributed data monitoring method for a supply chain intelligent collaborative management platform, the distributed data monitoring method specifically includes the following steps: Step S100, construct a supply chain intelligent collaborative management platform; used to realize data interconnection and intelligent analysis among intermediate links.

[0007] Step S200: Install data monitoring devices at all intermediate links of the supply chain intelligent collaborative management platform. The data monitoring devices are used to collect characteristic data of different intermediate links, and arrange the characteristic data of different intermediate links into a time series vector according to the collection time; Step S300: Based on the collected time series vector, calculate the dynamic correlation of different intermediate links in the time dimension respectively; to reflect the time series dependence relationship between each link.

[0008] Step S400: Construct a dynamic association graph with the intermediate links as nodes and the dynamic correlation of different intermediate links in the time dimension as edge weights, and perform a causal test on the constructed dynamic association graph to form a directed dynamic association graph; form a graph structure data model for modeling the relationship between intermediate links.

[0009] Step S500: Collect the characteristic data of different nodes in real time, and predict the impact of the change of the characteristic data of a node on the remaining nodes according to the constructed directed dynamic association graph, and sort the high bottleneck risk nodes in real time; In step S100, constructing a supply chain intelligent collaborative management platform specifically includes: The supply chain intelligent collaborative management platform includes a basic layer, an application layer and a collaborative layer; The basic layer includes a warehouse local server, a logistics center edge gateway, a supplier hosting server and a supplier API gateway; The application layer includes an order processing microservice, an in-vehicle Internet of Things device, and an API for a logistics status and supplier contract management system; The collaborative layer includes a quality inspection system, a transportation plan management system and a collaborative control system.

[0010] Specifically: The intermediate links include a number of supply ends, manufacturing ends, transportation ends, a number of storage ends, a number of sales ends and a number of delivery ends.

[0011] Specifically: Among them, preferably, Warehouse local server: Deployed at the storage end (such as a regional warehouse), and collect data such as temperature, humidity and inventory of the storage end in real time; Logistics center edge gateway: Installed at the transportation end, and collect data such as transportation routes, cargo status and fuel consumption through in-vehicle Internet of Things devices (GPS, temperature and humidity sensors); Supplier hosting server: Connected to the supply end (raw material supplier), and synchronize data such as raw material batches, price fluctuations and delivery cycles through the supplier API gateway; Provide distributed data monitoring devices; The data monitoring devices operate independently; The data monitoring devices cooperate and interconnect with each other.

[0012] In step S200, data monitoring devices are installed at all intermediate links of the supply chain intelligent collaborative management platform. The data monitoring devices are used to collect characteristic data of different intermediate links, and arrange the characteristic data of different intermediate links into a time series vector according to the collection time, specifically: Screen the characteristic data based on statistical analysis; Preferably, screening the characteristic data based on statistical analysis is specifically: Principal Component Analysis (PCA): Perform dimensionality reduction processing to extract the main component data explaining the supply chain fluctuations (such as supplier defaults, manufacturing machine anomalies, transportation overtime, abnormal data updates at the storage end, etc.); Mutual information: Quantify the non-linear correlation between the main component data and the target (delivery delay), filter low-correlation features, and screen to obtain the characteristic data; The specific steps of arranging the characteristic data of different intermediate links into a time series vector are as follows: Step S211: Based on the screened characteristic data, collect the characteristic data of different intermediate links at a certain frequency; Step S212: Perform data preprocessing; the data preprocessing includes data cleaning and data standardization; Among them, preferably, the data standardization uses the Z-score method; The mean value and standard deviation in the Z-score method can be selected as the mean value and standard deviation within a certain period of historical data; Step S213: Align different frequency data to the same time unit; Step S214: Generate time series vectors of different intermediate links.

[0013] Specifically: Set a timestamp; the timestamp is used to record the time of collecting the characteristic data and is used to align different frequency data to the same time unit.

[0014] In step S300, based on the collected time series vectors, calculate the dynamic correlations of different intermediate links in the time dimension, specifically: Step S301: Real-time collect the time series vectors of different intermediate links; Step S302: Determine the window length according to the dimension of the time series vector, and calculate the mean value of the characteristic data of different windows; Step S303: According to the determined window length, calculate the dynamic correlations of different intermediate links in the time dimension according to the method of windowed Pearson correlation coefficient.

[0015] Among them, preferably, the calculation formula for the dynamic correlation of different intermediate links in the time dimension is characterized as: ; Among them, represents the dynamic correlation between intermediate link X and intermediate link Y; k represents the window index; W represents the window length; S represents the time step, that is, the same time unit; x t represents the t-th feature data in the time series vector of intermediate link X; represents the t-th feature data in the time series vector of intermediate link Y; represents the mean value of the feature data of the k-th window in the time series vector of intermediate link X; represents the mean value of the feature data of the k-th window in the time series vector of intermediate link Y; t represents the quantity label of the feature data.

[0016] In step S400, taking the intermediate link as a node and the dynamic correlation of different intermediate links in the time dimension as the edge weight, construct a dynamic association graph, and perform a causality test on the constructed dynamic association graph to form a directed dynamic association graph, specifically: Step S401: Take the intermediate link as a node, the feature data transmission relationship between the intermediate links as an edge, and the dynamic correlation of different intermediate links in the time dimension as the edge weight to construct a dynamic association graph; Step S402: Perform a causality test between different nodes in the dynamic association graph to determine the dominant direction between the nodes and optimize the edge weight; Step S403: Optimize the constructed dynamic association graph into a directed dynamic association graph; Preferably, on the basis of the original dynamic correlation calculation, introduce a time delay parameter τ to perform a causality test between different nodes, and calculate the correlations in different directions, specifically: Positive correlation (XY), that is, it means predicting the time series vector of node Y through the historical time series vector before the time delay of node X: ; Reverse correlation (YX): that is, it means predicting the time series vector of node X through the historical time series vector before the time delay of node Y: ; Among them, τ represents the time delay parameter; represents the directed dynamic correlation from intermediate link X to intermediate link Y; represents the directed dynamic correlation from intermediate link Y to intermediate link X; It represents the mean value of the feature data of the k-th window in the historical time series before the time lag of the intermediate link X; It represents the mean value of the feature data of the k-th window in the historical time series before the time lag of the intermediate link Y; Among them, when τ = 1, it means using the historical time series vector of X to predict the current time series vector of Y; Determine the maximum time lag τmax according to the business scenario; Calculate different and ; Take the one with the largest absolute value of the directed dynamic correlation as the dominant direction, and update the edge weight to the directed dynamic correlation with the maximum absolute value; Update the directed dynamic association graph regularly.

[0017] In step S500, the feature data of different nodes are collected in real time. According to the constructed directed dynamic association graph, predict the impact of the change of the feature data of a node on the remaining nodes, and sort the high bottleneck risk nodes in real time. The specific steps are as follows: Step S501: Collect the feature data of different nodes in real time and generate a time series vector; Step S502: According to the change amount of the time series vector of the nodes in the constructed directed dynamic association graph and the edge weight, construct an impact propagation formula to predict the impact of the change of the feature data of a node on the remaining nodes; Preferably, the impact propagation formula is specifically: ; Among them, represents the change amount of the time series vector of node X; represents the impact propagation value on the time series vector of node Y; represents the value after normalizing the in-degree edge weight of node Y; among them, the normalization method selects the Max-Min normalization method; Step S503: According to the current feature data of the predicted node, combined with the predicted impact propagation value, sort the high bottleneck risk nodes in real time; Among them, preferably, the prediction formula of the bottleneck risk is specifically: ; Among them, represents the bottleneck risk of node i; N represents the number of nodes; d represents the damping coefficient, indicating the probability of node influence transmission; j represents all nodes pointing to node i; represents the edge weight from node j to node i; represents the influence delay time from node j to i; represents the value after normalizing the sum of all out-edge weights of node j; Indicates the bottleneck risk of the initial node j, and the initial value is the average distribution of the basic risk 1 / N; Calculate the bottleneck risks of the nodes respectively, sort them in descending order, and check the nodes according to the sorting according to the actual situation.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: The distributed data monitoring method of the supply chain intelligent collaborative management platform of the present invention, by installing independent and collaborative monitoring devices in each intermediate link of the supply chain, and using statistical analysis methods to screen and preprocess feature data, effectively solves the drawbacks of data transmission and processing in the traditional centralized architecture, and improves the data processing efficiency; calculates the dynamic correlation based on the time series vector and introduces the time delay parameter to construct a directed dynamic association graph, accurately captures the temporal dependence and causal relationship between each link, and makes up for the deficiency of the traditional monitoring in mining the link connection; according to the constructed graph model and related formulas, predicts the impact of node changes in real time and sorts the high-bottleneck risk nodes to achieve dynamic risk control, overcoming the problem that the traditional monitoring is difficult to globally supervise; finally, promotes the collaborative cooperation of each link through the graph structure data model, optimizes the resource allocation, enhances the system fault tolerance and dynamic adaptability, effectively improves the overall operation efficiency, stability and response speed of the supply chain, and enhances the market competitiveness of enterprises. Brief Description of the Drawings

[0019] Figure 1 It is a schematic structural diagram of a distributed data monitoring method for a supply chain intelligent collaborative management platform of the present invention. Detailed Embodiment

[0020] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a 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 those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0021] Embodiment: As Figure 1 shown, the present invention provides a technical solution, a distributed data monitoring method for a supply chain intelligent collaborative management platform, and the distributed data monitoring method specifically includes the following steps: Step S100, construct a supply chain intelligent collaborative management platform; used to realize data interconnection and intelligent analysis between each intermediate link.

[0022] In step S100, constructing a supply chain intelligent collaborative management platform specifically includes: The supply chain intelligent collaborative management platform includes a basic layer, an application layer and a collaborative layer; The basic layer includes a local warehouse server, a logistics center edge gateway, a supplier-hosted server, and a supplier API gateway; The application layer includes an order processing microservice, in-vehicle Internet of Things devices, and an API for a logistics status and supplier contract management system; The collaboration layer includes a quality inspection system, a transportation plan management system, and a collaboration control system.

[0023] Specifically: The intermediate links include a number of supply ends, manufacturing ends, transportation ends, a number of storage ends, a number of sales ends, and a number of delivery ends.

[0024] Specifically: Local warehouse server: Deployed at the storage end (such as a regional warehouse), it collects data such as temperature, humidity, and inventory levels at the storage end in real time; Logistics center edge gateway: Installed at the transportation end, it collects data such as transportation routes, cargo status, and fuel consumption through in-vehicle Internet of Things devices (GPS, temperature and humidity sensors); Supplier-hosted server: Connected to the supply end (raw material suppliers), it synchronizes data such as raw material batches, price fluctuations, and delivery cycles through the supplier API gateway; Provide distributed data monitoring devices; The data monitoring device is designed based on an embedded system architecture, integrating a high-performance processor, a large-capacity storage module, multiple sensor interfaces, and a communication module, and has independent data acquisition, processing, and storage capabilities. Taking the logistics center edge gateway deployed at the transportation end as an example, it is equipped with an ARM processor and can process in real time data such as transportation routes, cargo status, and fuel consumption collected by in-vehicle Internet of Things devices (GPS, temperature and humidity sensors, etc.); the built-in large-capacity flash memory can cache the collected data for a certain period of time, and even when the network is interrupted, it can continuously record data to ensure the continuity of data acquisition and achieve independent operation. At the same time, each device is equipped with an independent power management system, supporting dual modes of battery power supply and external power supply, and can ensure the stable operation of the device in complex environments.

[0025] The data monitoring device operates independently; The data monitoring devices are collaboratively interconnected.

[0026] Step S200: Install data monitoring devices at all intermediate links of the supply chain intelligent collaboration management platform. The data monitoring devices are used to collect characteristic data of different intermediate links, and arrange the characteristic data of different intermediate links into a time series vector according to the collection time; In step S200, data monitoring devices are installed at all intermediate links of the supply chain intelligent collaborative management platform. The data monitoring devices are used to collect characteristic data of different intermediate links, and arrange the characteristic data of different intermediate links into a time series vector according to the collection time, specifically: Screen the characteristic data based on statistical analysis; Preferably, screening the characteristic data based on statistical analysis is specifically: Principal component analysis (PCA): Perform dimensionality reduction processing to extract the main component data explaining the supply chain fluctuations (such as supplier defaults, manufacturing machine anomalies, transportation overtime, abnormal data updates at the storage end, etc.); Mutual information: Quantify the non-linear correlation between the main component data and the target (delivery delay), filter out low-correlation features, and obtain the screened characteristic data; Use principal component analysis (PCA) to perform dimensionality reduction processing on the data of each link, and extract the main component data explaining the supply chain fluctuations, such as supplier delivery delay, supplier price fluctuation, manufacturing machine failure downtime, transportation overtime times, abnormal data update delay duration of storage end inventory data, abnormal fluctuation of order volume at the sales end, etc.; then quantify the non-linear correlation between these main component data and the delivery delay target through mutual information, and filter out low-correlation features to obtain the screened characteristic data.

[0027] Raw material related data: Raw material batch number, used to trace the source and quality of raw materials; Raw material component content, reflecting the quality of raw materials; Raw material price fluctuation data, reflecting the market price change trend; Raw material inventory quantity, to master the stocking situation at the supply end.

[0028] Supplier delivery cycle, that is, the time from order placement to raw material delivery; Supplier default times, such as the occurrence frequency of situations like delayed delivery and unqualified delivery quality; Supplier production equipment failure rate, affecting the stability of raw material supply.

[0029] Production process data: Production plan completion rate, measuring the execution of production tasks; Manufacturing machine operation status data, including startup duration, shutdown times, and fault alarm information; Product qualification rate, reflecting the production quality level; Production line capacity utilization rate, reflecting the utilization degree of production resources.

[0030] Raw material input quantity, counting the quantity of raw materials used in the production process; Spare part loss rate, analyzing the consumption of spare parts in the production process; Energy consumption data, such as the usage amounts of electricity, water, natural gas, etc.

[0031] Transport status data: transportation route information, which records the goods transportation route in real time; the location of the transportation vehicle (GPS coordinates), which accurately locates the goods; the real-time temperature and humidity of the goods (for goods with temperature and humidity requirements), which ensures a suitable transportation environment for the goods; the transportation speed, which reflects the transportation efficiency.

[0032] Inventory management data: inventory quantity, which shows the quantity of goods in stock currently; inventory turnover rate, which measures the turnover speed of the goods in stock; the inbound and outbound records of the goods in stock, including information such as time, quantity, and operators; the warning data of the shelf life of the goods in stock, which prevents the goods from expiring.

[0033] Sales business data: order quantity, which reflects the market demand; the order delivery time requirement, which guides the production and distribution of the supply chain; the customer return rate, which analyzes the market acceptance of the product.

[0034] Distribution execution data: the location information of the delivery staff, which is convenient for customers to query in real time; the estimated arrival time of the distribution, which improves the customer experience; the goods receipt status, which confirms the completion of the delivery; the record of distribution abnormal events; The specific steps of arranging the characteristic data of different intermediate links into a time series vector according to the collection time are as follows: Step S211: Based on the screened characteristic data, collect the characteristic data of different intermediate links at a certain frequency; Step S212: Perform data preprocessing; data preprocessing includes data cleaning and data standardization; Among them, the Z-score method is used for data standardization; The average value and standard deviation in the Z-score method can be selected as the average value and standard deviation within a certain period of historical data; Step S213: Align different frequency data to the same time unit uniformly; Step S214: Generate time series vectors of different intermediate links.

[0035] Specifically: Set a timestamp; the timestamp is the time when the characteristic data is collected, which is used to align different frequency data to the same time unit uniformly.

[0036] Step S300: Based on the collected time series vectors, calculate the dynamic correlations of different intermediate links in the time dimension respectively; to reflect the time series dependence relationship between each link.

[0037] In step S300, based on the collected time series vectors, calculate the dynamic correlations of different intermediate links in the time dimension respectively, specifically as follows: Step S301: Collect the time series vectors of different intermediate links in real time; Step S302: Determine the window length according to the dimension of the time series vector, and calculate the mean of the characteristic data of different windows; Step S303: According to the determined window length, calculate the dynamic correlation of different intermediate links in the time dimension according to the method of windowed Pearson correlation coefficient.

[0038] Among them, preferably, the calculation formula for the dynamic correlation of different intermediate links in the time dimension is characterized as: ; Among them, represents the dynamic correlation between intermediate link X and intermediate link Y; k represents the window index; W represents the window length; S represents the time step, that is, the same time unit; x t represents the t-th characteristic data in the time series vector of intermediate link X; represents the t-th characteristic data in the time series vector of intermediate link Y; represents the mean of the characteristic data of the k-th window in the time series vector of intermediate link X; represents the mean of the characteristic data of the k-th window in the time series vector of intermediate link Y; t represents the quantity label of the characteristic data.

[0039] Step S400: Construct a dynamic association graph with intermediate links as nodes and the dynamic correlation of different intermediate links in the time dimension as edge weights. Perform a causal test on the constructed dynamic association graph to form a directed dynamic association graph; form a graph structure data model for modeling the relationships between intermediate links.

[0040] In step S400, construct a dynamic association graph with intermediate links as nodes and the dynamic correlation of different intermediate links in the time dimension as edge weights. Perform a causal test on the constructed dynamic association graph to form a directed dynamic association graph, specifically: Step S401: Use intermediate links as nodes, the characteristic data transmission relationship between intermediate links as edges, and the dynamic correlation of different intermediate links in the time dimension as edge weights to construct a dynamic association graph; Step S402: Perform a causal test between different nodes in the dynamic association graph to determine the dominant direction between nodes and optimize the edge weights; Step S403: Optimize the constructed dynamic association graph into a directed dynamic association graph; On the basis of the original dynamic correlation calculation, introduce a time delay parameter τ to perform a causal test between different nodes, and calculate the correlations in different directions respectively, specifically: Positive correlation (XY), that is, it means predicting the time series vector of node Y through the historical time series vector before the time delay of node X: ; Reverse correlation (YX): It means predicting the time series vector of node X through the historical time series vector before the time lag of node Y: ; where τ represents the time lag parameter; represents the directed dynamic correlation from intermediate link X to intermediate link Y; represents the directed dynamic correlation from intermediate link Y to intermediate link X; represents the mean value of the feature data of the k-th window in the historical time series before the time lag of intermediate link X; represents the mean value of the feature data of the k-th window in the historical time series before the time lag of intermediate link Y; where, when τ = 1, it means predicting the current time series vector of Y with the historical time series vector of X; Determine the maximum time lag τmax according to the business scenario; Calculate different and ; Take the one with the largest absolute value of the directed dynamic correlation as the dominant direction, and update the edge weight to the directed dynamic correlation with the maximum absolute value; Update the directed dynamic association graph regularly.

[0041] Step S500: Real-time collect the feature data of different nodes, and according to the constructed directed dynamic association graph, predict the impact of the change of the feature data of a node on the other nodes, and sort the high bottleneck risk nodes in real time; In step S500, real-time collect the feature data of different nodes, and according to the constructed directed dynamic association graph, predict the impact of the change of the feature data of a node on the other nodes, and sort the high bottleneck risk nodes in real time. The specific steps are as follows: Step S501: Real-time collect the feature data of different nodes and generate a time series vector; Step S502: According to the change amount of the time series vector of the nodes in the constructed directed dynamic association graph and the edge weight, construct an impact propagation formula to predict the impact of the change of the feature data of a node on the other nodes; Preferably, the impact propagation formula is specifically: ; where, represents the change amount of the time series vector of node X; represents the impact propagation value on the time series vector of node Y; represents the value after normalizing the in-degree edge weight of node Y; where, the normalization method selects the Max-Min normalization method; Step S503: Based on the current feature data of the predicted nodes and in combination with the predicted influence propagation values, sort the high bottleneck risk nodes in real time; Preferably, the prediction formula for the bottleneck risk is specifically: ; Wherein, represents the bottleneck risk of node i; N represents the number of nodes; d represents the damping coefficient, indicating the probability of influence transmission of the node; j represents all nodes pointing to node i; represents the edge weight of node j pointing to node i; represents the influence delay time from node j to i; represents the value after normalization of the sum of all out-edge weights of node j; represents the bottleneck risk of the initial node j, and the initial value is the average distribution of the basic risk 1 / N; Calculate the bottleneck risks of the nodes respectively, perform a descending order sort, and check the nodes according to the sort according to the actual situation.

[0042] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present invention. Any reference signs in the claims should not be regarded as limiting the claimed rights.

Claims

1. A distributed data monitoring method for a supply chain intelligent collaborative management platform, characterized in that: The distributed data monitoring method specifically comprises the following steps: Step S100: Building a supply chain intelligent collaborative management platform; Step S200: installing data monitoring equipment in all intermediate links of the supply chain intelligent collaborative management platform, wherein the data monitoring equipment is used to collect characteristic data of different intermediate links, and arrange the characteristic data of different intermediate links into time series vectors according to the collection time; Step S300: taking the collected time series vector as a reference, respectively calculating the dynamic correlation of different intermediate links in the time dimension; Step S400: construct a dynamic association graph with the intermediate links as nodes and the dynamic correlations of the different intermediate links in the time dimension as edge weights, and perform causal testing on the basis of the constructed dynamic association graph to form a directed dynamic association graph; Step S500: collect feature data of different nodes in real time, predict the impact of feature data changes of nodes on other nodes based on the constructed directed dynamic association graph, and sort high bottleneck risk nodes in real time.

2. The distributed data monitoring method of a supply chain intelligent collaborative management platform according to claim 1 is characterized by: In step S100, a supply chain intelligent collaborative management platform is constructed, specifically: The supply chain intelligent collaborative management platform includes a basic layer, an application layer and a collaborative layer; The basic layer includes warehouse local servers, logistics center edge gateways, supplier hosting servers, and supplier API gateways; The application layer includes order processing microservices, vehicle-mounted IoT devices, logistics status and supplier contract management system APIs; The collaborative layer includes a quality inspection system, a transportation plan management system and a collaborative control system.

3. The distributed data monitoring method of a supply chain intelligent collaborative management platform according to claim 1 is characterized by: Specifically: The intermediate links include several supply ends, manufacturing ends, transportation ends, several storage ends, several sales ends and several delivery ends.

4. The distributed data monitoring method of a supply chain intelligent collaborative management platform according to claim 3 is characterized by: Specifically: Provide distributed data monitoring equipment; The data monitoring device operates independently; The data monitoring devices are collaboratively interconnected.

5. The distributed data monitoring method of a supply chain intelligent collaborative management platform according to claim 4 is characterized by: In step S200, data monitoring equipment is installed in all intermediate links of the supply chain intelligent collaborative management platform. The data monitoring equipment is used to collect feature data of different intermediate links and arrange the feature data of different intermediate links into time series vectors according to the collection time, specifically: Screening of characteristic data based on statistical analysis; The specific steps of arranging the characteristic data of different intermediate links into time series vectors according to the collection time are as follows: Step S211: Based on the filtered feature data, feature data of different intermediate links are collected at a certain frequency; Step S212, performing data preprocessing; the data preprocessing includes data cleaning and data standardization; Step S213, aligning data of different frequencies to the same time unit; Step S214: Generate time series vectors of different intermediate links.

6. The distributed data monitoring method of a supply chain intelligent collaborative management platform according to claim 5 is characterized by: Specifically: Set a timestamp; the timestamp is the time when the characteristic data is collected, and is used to align data of different frequencies to the same time unit.

7. The distributed data monitoring method of a supply chain intelligent collaborative management platform according to claim 6 is characterized by: In step S300, the dynamic correlations of different intermediate links in the time dimension are calculated based on the collected time series vectors, specifically: Step S301, real-time collection of time series vectors of different intermediate links; Step S302: determining the window length according to the dimension of the time series vector, and calculating the mean of the characteristic data of different windows; Step S303: According to the determined window length, the dynamic correlation of different intermediate links in the time dimension is calculated according to the windowed Pearson correlation coefficient method.

8. The distributed data monitoring method of a supply chain intelligent collaborative management platform according to claim 7 is characterized by: In step S400, a dynamic association graph is constructed with the intermediate links as nodes and the dynamic correlations of the different intermediate links in the time dimension as edge weights. A causal test is performed on the basis of the constructed dynamic association graph to form a directed dynamic association graph, specifically: Step S401: construct a dynamic association graph by taking the intermediate links as nodes, the characteristic data transmission relationships between the intermediate links as edges, and the dynamic correlations of different intermediate links in the time dimension as edge weights; Step S402, performing causal tests between different nodes in the dynamic association graph, determining the dominant direction between the nodes, and optimizing edge weights; Step S403, optimizing the constructed dynamic association graph into a directed dynamic association graph; The directed dynamic association graph is updated periodically.

9. The distributed data monitoring method of a supply chain intelligent collaborative management platform according to claim 8, characterized in that: In step S500, feature data of different nodes are collected in real time, and the influence of feature data changes of nodes on other nodes is predicted according to the constructed directed dynamic association graph, and high bottleneck risk nodes are sorted in real time. The specific steps are as follows: Step S501, collecting feature data of different nodes in real time to generate a time series vector; Step S502: construct an influence propagation formula based on the change amount of the time series vector of the node in the constructed directed dynamic association graph and the edge weight, and predict the impact of the change of the characteristic data of the node on the other nodes; Step S503: Sort high bottleneck risk nodes in real time based on the current feature data of the prediction node and the predicted impact propagation value.

Images