Intelligent agent service platform based on real-time supply chain situation risk awareness

Abstract

The invention relates to the technical field of supply chain risk analysis, in particular to an agent service platform based on real-time supply chain situation risk awareness, which is provided with a data acquisition module, a node feature recognition module, a node feature analysis module and a node regulation and control module. A transmission fluctuation curve is determined through a node feature analysis module, whether a transmission risk exists in a next monitoring period is judged, and a risk node tendency coefficient of a supply chain is determined through a node regulation and control module so as to select to adjust a profit sharing data bandwidth proportion in transmission data for each risk node on the supply chain or send out an early warning. According to the method, the time period with the transmission risk is predicted according to the data transmission fluctuation characteristics of the nodes, each risk node on the supply chain is adaptively adjusted according to the prediction result, and the reliability and the processing efficiency of the intelligent agent service platform for supply chain situation risk perception are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of supply chain risk analysis, and in particular to an intelligent agent service platform based on real-time supply chain situation risk perception. BACKGROUND

[0002] In the real-time supply chain situation risk perception system, the data transmission stability between the link nodes such as manufacturers, logistics companies, warehousing centers, and distributors is the core basis for supporting the intelligent agent service platform to achieve accurate risk perception and dynamic distribution. However, there is a problem of unstable data transmission in the current supply chain network, which affects the efficiency of distribution. The risk monitoring of traditional supply chains relies on manual inspection, post-data statistics, or single-dimensional indicators, such as only monitoring transmission delay, which is difficult to capture the dynamic changes of each node in real time. Therefore, improving the distribution efficiency of the supply chain and improving the reliability and processing efficiency of the intelligent agent service platform for supply chain situation risk perception are technical problems that need to be solved.

[0003] For example, Chinese Patent No. CN119808082B discloses a software supply chain risk detection and protection method and system, which obtains third-party components from the software supply chain, analyzes their multi-level dynamic behavior, uses a Bayesian network algorithm to predict risk behavior patterns, constructs a behavior pattern library, monitors the behavior of the third-party components during runtime, records behavior data streams, applies a deep learning time series anomaly detection algorithm to identify abnormal behavior that deviates from normal behavior patterns, uses a graph neural network to evaluate the risk level of abnormal behavior, generates an abnormal behavior list by combining the relevance and propagation path between behaviors, and dynamically adjusts the response strategy in the security control measure library to develop security control measures for the third-party components.

[0004] The existing technology also has the following problems: The existing technology does not consider that the data transmission fluctuations of each node on the supply chain will affect the efficiency of supply chain distribution, thereby affecting the reliability and processing efficiency of the intelligent agent service platform. The existing technology cannot predict the time periods with transmission risks based on the data transmission fluctuation characteristics of the nodes, and cannot adaptively adjust each risk node on the supply chain, which affects the reliability and processing efficiency of the intelligent agent service platform for supply chain situation risk perception. SUMMARY

[0005] Therefore, the present application provides an intelligent agent service platform based on real-time supply chain situation risk perception to overcome the problem that the existing technology cannot predict the time periods with transmission risks based on the data transmission fluctuation characteristics of the nodes, and cannot adaptively adjust each risk node on the supply chain, which affects the reliability and processing efficiency of the intelligent agent service platform for supply chain situation risk perception.

[0006] To achieve the above object, the application provides an intelligent agent service platform based on real-time supply chain situation risk perception, comprising: A data acquisition module is used to acquire the transmission delay jitter frequency of each node on the supply chain, data verification return parameters, and resource idle parameters, wherein the data verification return parameters include data verification return rate and data verification return time; A node feature recognition module is connected with the data acquisition module and used to determine risk representation parameters according to the link jitter frequency and data verification return rate of the node in a preset monitoring period, and screen risk nodes based on the risk representation parameters of the node; A node feature analysis module is connected with the data acquisition module and the node feature recognition module respectively, and used to determine transmission fluctuation tendency parameters of each monitoring period in the preset monitoring period based on the data verification return time and resource idle parameters of the risk node, determine a transmission fluctuation curve according to the transmission fluctuation tendency parameters, and determine whether there is a transmission risk in the next monitoring period in time sequence based on the transmission disorder coefficient; A node regulation module is connected with the node feature recognition module and the node feature analysis module respectively, and used to determine a risk node tendency coefficient of the supply chain according to the risk representation parameters of the next monitoring period and the determination result of the transmission risk, so as to select to adjust the bandwidth proportion of the distribution data in the transmission data of each risk node on the supply chain or issue a warning to each risk node on the supply chain.

[0007] Further, the node feature recognition module is used to determine the risk representation parameters, wherein, The node feature recognition module determines the weighted sum of the first risk tendency feature and the second risk tendency feature as the risk representation parameters; The first risk tendency feature is the ratio of the link jitter frequency of the node in the preset monitoring period to the link jitter frequency threshold; The second risk tendency feature is the ratio of the data verification return rate of the node in the preset monitoring period to the data verification return rate threshold.

[0008] Further, the node feature recognition module is used to screen the risk nodes, wherein, The node feature recognition module screens the node as a risk node based on the determination result that the risk representation parameters of the node meet the risk node screening condition; The risk node screening condition is that the risk representation parameters exceed the preset risk representation parameter threshold.

[0009] Further, the node feature analysis module is used to determine the transmission fluctuation tendency parameters, wherein, The node feature analysis module determines a sum of the first transmission fluctuation coefficient and the second transmission fluctuation coefficient as the transmission fluctuation tendency parameter. The first transmission fluctuation coefficient is a ratio of the backhaul interval fluctuation parameter to a backhaul interval fluctuation parameter threshold, and the second transmission fluctuation coefficient is a ratio of the resource idle parameter to a resource idle parameter threshold.

[0010] Further, the backhaul interval fluctuation parameter is an average of interval durations between adjacent data check backhaul time points. The resource idle parameter is a ratio of a CPU idle duration of the node in a monitoring period to a total duration of the monitoring period.

[0011] Further, the node feature analysis module is configured to determine a transmission disorder coefficient, wherein The node feature analysis module is configured to obtain slopes of each monitoring period on a transmission fluctuation curve, calculate absolute values of differences between adjacent period slopes, and determine an average of the absolute values as the transmission disorder coefficient.

[0012] Further, the node feature analysis module is configured to determine whether a next monitoring period in time sequence has a transmission risk, wherein The node feature analysis module determines that the next monitoring period in time sequence has a transmission risk based on a determination result that the transmission fluctuation tendency parameter and the transmission disorder coefficient of the transmission fluctuation curve meet a risk period condition. The risk period condition is that the transmission fluctuation tendency parameter of a last monitoring period on the transmission fluctuation curve exceeds a preset transmission fluctuation tendency parameter threshold, and the transmission disorder coefficient exceeds a preset transmission disorder coefficient threshold.

[0013] Further, the node regulation module is configured to mark a feature node and determine a risk node tendency coefficient, wherein The node regulation module marks the risk node as a feature node based on a determination result that a risk representation parameter of the risk node in the next monitoring period and a determination result of the transmission risk meet a feature node condition. The feature node condition is that the risk representation parameter exceeds a preset risk representation parameter threshold, and the determination result is that there is a transmission risk. The risk node tendency coefficient is a ratio of a number of the feature nodes on the supply chain to a number of the risk nodes.

[0014] Further, the node regulation module adjusts a share data bandwidth proportion in transmission data of each risk node on the supply chain based on a determination result that a risk node tendency coefficient of the supply chain does not meet a warning condition. The node regulation module sends a warning to each risk node on the supply chain based on the determination result of whether the risk node tendency coefficient of the supply chain meets the warning condition. The warning condition is that the risk node tendency coefficient of the supply chain is not more than a preset risk node tendency coefficient threshold of the supply chain.

[0015] Further, the increase in the share data bandwidth proportion in the transmission data of the risk node is positively correlated with the risk representation parameter.

[0016] Compared with the prior art, the present application has the beneficial effects that the data acquisition module, the node feature identification module, the node feature analysis module, and the node regulation module are provided, the transmission delay jitter frequency, the data verification return parameter, and the resource idle parameter are acquired through the data acquisition module, the risk representation parameter is determined through the node feature identification module, the risk node is screened, the transmission fluctuation tendency parameter of each monitoring period in the preset monitoring period is determined through the node feature analysis module, the transmission fluctuation curve is determined, whether there is a transmission risk in the next monitoring period in time sequence is determined based on the transmission disorder coefficient, the risk node tendency coefficient of the supply chain is determined through the node regulation module, and the share data bandwidth proportion in the transmission data of each risk node on the supply chain is adjusted or a warning is sent to each risk node on the supply chain, thereby realizing rapid screening of the risk node according to the dynamic parameter of the node, predicting the period with the transmission risk according to the data transmission fluctuation characteristics of the node, adaptively adjusting each risk node on the supply chain according to the prediction result, and improving the reliability and processing efficiency of the intelligent agent service platform for supply chain situation risk perception.

[0017] Especially, the risk representation parameter is determined by the node feature identification module according to the link jitter frequency and the data verification return rate of the node to screen the risk node, and it can be understood that the link jitter frequency can represent the transmission stability, and the data verification return rate can represent the data reliability. By combining the link jitter frequency and the data verification return rate, the traditional method of relying on a single index, such as a single index identification of only looking at the transmission interruption, is converted into multi-dimensional evaluation, the implicit risk is quantified, the identification accuracy is improved, for example, although a certain node does not completely interrupt the transmission, the link jitter is frequent and the data verification return rate is low, the traditional method may ignore the risk of the node, and through the high risk representation parameter, the node can be identified as a risk node in time, before the node completely fails, such as when the link jitter intensifies but does not interrupt, the node is identified as a risk node in time, the risk representation parameter is determined by the node feature identification module according to the link jitter frequency and the data verification return rate of the node to screen the risk node, thereby realizing rapid screening of the risk node according to the dynamic parameter of the node, and improving the reliability and processing efficiency of the intelligent agent service platform for supply chain situation risk perception.

[0018] Especially, the application can determine whether there is a transmission risk in the next monitoring period in time sequence based on the transmission disorder coefficient through the node feature analysis module. It can be understood that in actual scenarios, it is easy to misjudge the data transmission risk by using a single indicator such as transmission delay or resource occupation. Resource is tight but transmission timing is stable, that is, the node load is high but the scheduling is orderly, which may not be risky. The resource is idle but the transmission timing is disordered, which may be an abnormal fluctuation before the node failure. At the same time, traditional supply chain risk monitoring relies on single-point data threshold judgment, such as monitoring whether the current delay is excessive, while ignoring the trend deterioration, such as gradually increasing delay but not reaching the threshold, and finally suddenly exploding. Through the transmission fluctuation curve and the transmission disorder coefficient, the accumulation process of the risk can be captured in advance, the risk judgment of the next period is more forward-looking, the transmission interruption caused by risk burst is avoided, the misjudgment rate is reduced, and the stability of data transmission of each node on the supply chain directly affects the core business such as distribution, for example, data transmission interruption will cause distribution data to be unable to synchronize, and cause settlement disputes. By determining whether there is a transmission risk in the next monitoring period in time sequence through the transmission disorder coefficient, the data real-time transmission efficiency is ensured, thereby improving the smoothness of the supply chain distribution business, and further, the period with transmission risk is predicted according to the data transmission fluctuation characteristics of the node, and the reliability and processing efficiency of the intelligent agent service platform for supply chain situation risk perception are improved.

[0019] Especially, the application can determine whether there is a transmission risk in the next monitoring period in time sequence based on the transmission disorder coefficient through the node feature analysis module. It can be understood that in actual scenarios, it is easy to misjudge the data transmission risk by using a single indicator such as transmission delay or resource occupation. Resource is tight but transmission timing is stable, that is, the node load is high but the scheduling is orderly, which may not be risky. The resource is idle but the transmission timing is disordered, which may be an abnormal fluctuation before the node failure. At the same time, traditional supply chain risk monitoring relies on single-point data threshold judgment, such as monitoring whether the current delay is excessive, while ignoring the trend deterioration, such as gradually increasing delay but not reaching the threshold, and finally suddenly exploding. Through the transmission fluctuation curve and the transmission disorder coefficient, the accumulation process of the risk can be captured in advance, the risk judgment of the next period is more forward-looking, the transmission interruption caused by risk burst is avoided, the misjudgment rate is reduced, and the stability of data transmission of each node on the supply chain directly affects the core business such as distribution, for example, data transmission interruption will cause distribution data to be unable to synchronize, and cause settlement disputes. By determining whether there is a transmission risk in the next monitoring period in time sequence through the transmission disorder coefficient, the data real-time transmission efficiency is ensured, thereby improving the smoothness of the supply chain distribution business, and further, the period with transmission risk is predicted according to the data transmission fluctuation characteristics of the node, and the reliability and processing efficiency of the intelligent agent service platform for supply chain situation risk perception are improved.

[0020] Especially, the application determines the risk node tendency coefficient of the supply chain according to the risk characterization parameter of the next monitoring period and the determination result of the transmission risk by the node regulation module, so as to select the risk nodes on the supply chain to issue a warning. It can be understood that the risk node tendency coefficient of the supply chain meets the warning condition, that is, there is data transmission risk in the supply chain, the proportion of the number of nodes meeting the prediction conclusion is small, the risk assessment prediction is unreliable, the network environment has a sudden abnormality, and it is difficult to improve the data transmission by adjusting the bandwidth proportion of the profit sharing data. The risk nodes are warned, and blind adjustment is avoided, and the passive regulation is changed to active tracing, triggering manual or automatic investigation, finding the source of prediction deviation, and specifically regulating, and then, the application realizes adaptive adjustment of each risk node on the supply chain according to the prediction result, improves the reliability and processing efficiency of the intelligent agent service platform for supply chain situation risk perception. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 The functional block diagram of the intelligent agent service platform for real-time supply chain situation risk perception based on the embodiments of the application is shown in the figure; Figure 2 The logical flowchart of the node feature identification module of the embodiments of the application for screening risk nodes is shown in the figure; Figure 3 The logical flowchart of the node feature analysis module of the embodiments of the application for determining whether there is transmission risk in the next monitoring period in time sequence is shown in the figure; Figure 4 The logical flowchart of the node regulation module of the embodiments of the application for marking feature nodes is shown in the figure. DETAILED DESCRIPTION

[0022] In order to make the purpose and advantages of the application more clear and obvious, the application will be further described below in combination with embodiments; it should be understood that the specific embodiments described herein are only used to explain the application, and do not limit the application.

[0023] The preferred embodiments of the application will be described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the application, and are not intended to limit the protection scope of the application.

[0024] It should be noted that in the description of the application, the terms "up", "down", "in", "out" and the like indicate the direction or positional relationship of the terms based on the direction or positional relationship shown in the drawings, which is only for the convenience of description, and does not indicate or imply that the device or element must have a particular orientation, be constructed and operated in a particular orientation, therefore it cannot be understood as a limitation of the application.

[0025] Moreover, it needs to be explained that, in the description of the present application, unless explicitly defined and limited, the terms "mounting", "connection" should be understood broadly, for example, it can be fixed connection, or detachable connection, or integral connection, it can be mechanical connection, or electrical connection, it can be direct connection, or indirect connection through intermediate medium, it can be internal communication of two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0026] Please refer to Figure 1 As shown in the figure, it is a function block diagram of the embodiment of the present application based on real-time supply chain situation risk perception agent service platform, the agent service platform based on real-time supply chain situation risk perception of the present application comprises: The data acquisition module is used to acquire the transmission delay jitter frequency, data verification return parameter and resource idle parameter of each node on the supply chain, and the data verification return parameter comprises data verification return rate and data verification return time; Specifically, the structure of the data acquisition module is not limited in the embodiment of the present application, preferably, it can acquire the transmission delay jitter frequency, data verification return parameter and resource idle parameter through a lightweight background process, which will not be repeated here.

[0027] The node feature recognition module is connected with the data acquisition module, and is used to determine the risk representation parameter according to the link jitter frequency and data verification return rate of the node in the preset monitoring period, and screen the risk node based on the risk representation parameter of the node; Specifically, the structure of the node feature recognition module is not limited in the embodiment of the present application, preferably, it can be a microprocessor, which is used to determine the risk representation parameter and screen the risk node, which will not be repeated here.

[0028] Specifically, the preset monitoring period can be set by those skilled in the art according to the accuracy requirement of the agent service platform, the higher the accuracy requirement, the shorter the preset monitoring period, the value range of the preset monitoring period can be [10, 30] with min as the interval unit, preferably, the preset monitoring period can be 20 min.

[0029] The node feature analysis module is connected with the data acquisition module and the node feature recognition module respectively, and is used to determine the transmission fluctuation tendency parameter of each monitoring period in the preset monitoring period based on the data verification return time and resource idle parameter of the risk node, determine the transmission fluctuation curve according to the transmission fluctuation tendency parameter, and determine whether there is transmission risk in the next monitoring period in time sequence based on the transmission disorder coefficient; Specifically, the next monitoring period in time sequence is the next monitoring period in time sequence of the current monitoring period.

[0030] Specifically, the structure of the node feature analysis module is not limited in the embodiment of the application, and preferably, it can be a processor used in a computer to determine the transmission fluctuation tendency parameter, the transmission fluctuation curve, and whether there is a transmission risk in the next monitoring period in the timing.

[0031] Specifically, the duration of each monitoring period in the preset monitoring period is the product of the duration of the preset monitoring period and a monitoring period value factor, the monitoring period value factor can be set by a person skilled in the art according to the accuracy requirement of the intelligent agent service platform, the higher the accuracy requirement, the smaller the monitoring period value factor, and the value range of the monitoring period value factor can be [0.1, 0.3], preferably, the monitoring period value factor can be 0.2.

[0032] Specifically, the way of constructing the transmission fluctuation curve is not limited, for example, the transmission fluctuation curve can be fitted by matlab related fitting software, which will not be repeated.

[0033] Specifically, the transmission fluctuation curve establishes a rectangular coordinate system with time as the horizontal axis and the numerical value of the transmission fluctuation tendency parameter as the vertical axis, and connects each transmission fluctuation point with a smooth curve, and the transmission fluctuation point is determined according to the transmission fluctuation tendency parameter and the midpoint time of the monitoring time period in which the transmission fluctuation tendency parameter is located.

[0034] The node regulation module is connected with the node feature recognition module and the node feature analysis module respectively, and is used to determine the risk node tendency coefficient of the supply chain according to the risk representation parameter of the next monitoring period and the determination result of the transmission risk, so as to select to adjust the bandwidth proportion of the distribution data in the transmission data of each risk node on the supply chain, or to issue a warning to each risk node on the supply chain.

[0035] Specifically, the structure of the node regulation module is not limited in the embodiment of the application, and preferably, it can be a processor used in a computer to determine the risk node tendency coefficient of the supply chain, select to adjust the bandwidth proportion of the distribution data in the transmission data of each risk node on the supply chain, or issue a warning to each risk node on the supply chain, which will not be repeated.

[0036] Specifically, it can be understood that the distribution data includes key data related to transaction settlement and benefit distribution in the supply chain, such as order confirmation, fund reconciliation, inventory allocation, etc. The stability of the transmission of such data directly affects the efficiency of supply chain collaboration. There is a data transmission risk in the supply chain. The node number conforming to the prediction conclusion accounts for a large proportion, indicating that the risk assessment prediction is relatively reliable, and the risk is basically in a predictable state. By adjusting the bandwidth proportion of the distribution data of the risk node, the greater the risk representation parameter, the higher the data transmission risk, the more bandwidth the distribution data is allocated, and the transmission resources of the key data are preferentially guaranteed. Further, the prediction result is adaptively adjusted to each risk node on the supply chain, and the reliability and processing efficiency of the intelligent agent service platform for supply chain situation risk perception are improved.

[0037] Specifically, the node feature recognition module is configured to determine a risk representation parameter, wherein, The node feature recognition module determines a weighted sum of the first risk tendency feature and the second risk tendency feature as the risk representation parameter. The first risk tendency feature is a ratio of a link jitter frequency of the node in a preset monitoring period to a link jitter frequency threshold. The second risk tendency feature is a ratio of a data verification return rate of the node in the preset monitoring period to a data verification return rate threshold.

[0038] Specifically, the link jitter frequency is a ratio of a number of times that a time delay fluctuation of the node in a preset monitoring period exceeds a preset time delay fluctuation threshold to the preset monitoring period. The preset time delay fluctuation threshold is a product of a time delay fluctuation reference value and a time delay fluctuation factor. The time delay fluctuation reference value is an average value of the preset time delay fluctuation of the node in the historical data. The time delay fluctuation factor can be set by a person skilled in the art according to the accuracy requirement of the intelligent agent service platform. The higher the accuracy requirement, the smaller the time delay fluctuation factor. The value range of the time delay fluctuation factor can be [1.1, 1.3], and preferably, the time delay fluctuation factor can be 1.2. The data verification return rate is a product of a ratio of a number of times of successfully returning the verification result to a total number of received data and 100%.

[0039] Specifically, in the data transmission process of the actual supply chain node, the data check return rate can more directly reflect the reliability and integrity of the node data transmission, and compared with the link jitter frequency, can better reflect the core influence on the node transmission risk, and according to the influence degree of the data check return rate and the link jitter frequency in the historical data on the calculation result, the size of the weight value is selected, so in the implementation, the reliability characteristics of data transmission are preferentially considered, so a slightly higher weight is given to the second risk tendency feature calculated based on the data check return rate, therefore, when the weighted sum is performed, the weight of the first risk tendency feature can be set to 0.4, and the weight of the second risk tendency feature can be set to 0.6. Specifically, in the embodiment, the purpose of setting the link jitter frequency threshold and the data check return rate threshold is to represent the case that the node transmission stability and the data interaction reliability have a greater influence on the supply chain data transmission risk, by calling historical transmission data of node operation for several times, the link jitter frequency historical data and the data check return rate historical data of the node in the same preset monitoring period are obtained, the link jitter frequency mean value and the data check return rate mean value are solved, based on the purpose of setting the two thresholds, the link jitter frequency threshold is determined as the product of the link jitter frequency mean value and the first adjustment coefficient, and the data check return rate threshold is determined as the product of the data check return rate mean value and the second adjustment coefficient, wherein the first adjustment coefficient can be selected in [1.2, 1.3], and the second adjustment coefficient can be selected in [1.3, 1.4], preferably, the first adjustment coefficient can be 1.25, and the second adjustment coefficient can be 1.35.

[0040] Referring to Figure 2 As shown in FIG. 1, which is a logic flow chart of the node feature identification module screening a risk node according to the embodiment of the present application, the node feature identification module is used to screen a risk node, wherein, The node feature identification module screens the node as a risk node based on the determination result that the risk representation parameter of the node meets the risk node screening condition; If the risk representation parameter of the node does not meet the risk node screening condition, the node feature identification module does not screen the node; The risk node screening condition is that the risk representation parameter exceeds the preset risk representation parameter threshold.

[0041] Specifically, the preset risk representation parameter threshold is a product of a risk representation parameter reference value and a preset risk representation regulation factor, the risk representation parameter reference value is an average value of the risk representation parameters of the nodes in the historical data, the risk representation regulation factor can be set by a person skilled in the art according to the accuracy requirement of the intelligent agent service platform, the higher the accuracy requirement, the smaller the risk representation regulation factor, and the value range of the risk representation regulation factor can be [1.1, 1.35], preferably, the risk representation regulation factor can be 1.2.

[0042] Specifically, the node feature identification module determines the risk representation parameter according to the link jitter frequency and the data verification return rate of the node to screen the risk nodes. It can be understood that the link jitter frequency can represent the transmission stability, and the data verification return rate can represent the data reliability. By combining the link jitter frequency and the data verification return rate, the traditional single indicator identification, such as only looking at the single indicator of transmission interruption, is converted into multi-dimensional evaluation, the implicit risk is quantified, and the identification accuracy is improved. For example, although a certain node does not completely interrupt the transmission, the link jitter is frequent and the data verification return rate is low, the traditional method can ignore the risk of the node, and through the high risk representation parameter, the node can be identified as a risk node in time. Before the node completely fails, such as when the link jitter intensifies but does not interrupt, the node is identified as a risk node in time. The node feature identification module determines the risk representation parameter according to the link jitter frequency and the data verification return rate of the node to screen the risk nodes, and then, the risk nodes are quickly screened according to the dynamic parameters of the nodes, and the reliability and processing efficiency of the intelligent agent service platform for supply chain situation risk perception are improved.

[0043] Specifically, it can be understood that the link jitter frequency can represent the network layer transmission quality, and high frequency jitter can cause data packet loss, thereby affecting the upper layer application, i.e., the distribution data transmission delay. The data verification return rate can represent the reliability of the application layer data, and the retransmission causes the data transmission efficiency to decrease, thereby affecting the upper layer application, i.e., the distribution data transmission delay. When both of them are abnormal, such as high link jitter and low return rate, the risk representation parameter is larger, and the transmission risk of the node is higher. The data transmission risk of the node can be more sensitively captured. The node feature identification module determines the risk representation parameter according to the link jitter frequency and the data verification return rate of the node to screen the risk nodes, and then, the risk nodes are quickly screened according to the dynamic parameters of the nodes, and the reliability and processing efficiency of the intelligent agent service platform for supply chain situation risk perception are improved.

[0044] Specifically, the node feature analysis module is configured to determine a transmission fluctuation tendency parameter, wherein, The node feature analysis module determines a sum of the first transmission fluctuation coefficient and the second transmission fluctuation coefficient as the transmission fluctuation tendency parameter. The first transmission fluctuation coefficient is a ratio of the backhaul interval fluctuation parameter to a backhaul interval fluctuation parameter threshold value, and the second transmission fluctuation coefficient is a ratio of the resource idle parameter to a resource idle parameter threshold value.

[0045] In this embodiment, the purpose of setting the backhaul interval fluctuation parameter threshold value and the resource idle parameter threshold value is to represent a situation in which the node transmission fluctuation is more significant and the interference on the data transmission stability is stronger. By calling historical transmission data of node operation for several times, the backhaul interval fluctuation parameter historical data and the resource idle parameter historical data of the node are obtained, the mean value of the backhaul interval fluctuation parameter and the mean value of the resource idle parameter are solved, and based on the purpose of setting the backhaul interval fluctuation parameter threshold value and the resource idle parameter threshold value, the backhaul interval fluctuation parameter threshold value is determined as a product of the mean value of the backhaul interval fluctuation parameter and a first deviation coefficient, and the resource idle parameter threshold value is determined as a product of the mean value of the resource idle parameter and a second deviation coefficient. The value range of the first deviation coefficient can be [1.2, 1.4], the value range of the second deviation coefficient can be [0.8, 0.9], preferably, the first deviation coefficient can be 1.3, and the second deviation coefficient can be 0.85.

[0046] Specifically, the backhaul interval fluctuation parameter is an average value of interval durations between adjacent data check backhaul moments. The resource idle parameter is a ratio of the CPU idle duration of the node in the monitoring period to the total duration of the monitoring period.

[0047] Specifically, the node feature analysis module is used to determine a transmission disorder coefficient, wherein, The node feature analysis module is used to obtain the slope of each monitoring period on the transmission fluctuation curve, calculate the absolute value of the difference of the slopes of adjacent periods, and determine the average value of the absolute value as the transmission disorder coefficient.

[0048] Referring to Figure 3 FIG. 2 is a logic flow diagram of the node feature analysis module of the embodiment of the present application for determining whether the next monitoring period in the time sequence has transmission risk, as shown in the figure, the node feature analysis module is used to determine whether the next monitoring period in the time sequence has transmission risk, wherein, The node feature analysis module determines that the next monitoring period in the time sequence has transmission risk based on the determination result that the transmission fluctuation tendency parameter and the transmission disorder coefficient of the transmission fluctuation curve meet the risk period condition; If the transmission fluctuation tendency parameter and the transmission disorder coefficient of the transmission fluctuation curve do not meet the risk period condition, the node feature analysis module determines that the next monitoring period in the time sequence does not have transmission risk; The risk period condition is that the transmission fluctuation trend parameter of the last monitoring period on the transmission fluctuation curve exceeds a preset transmission fluctuation trend parameter threshold, and the transmission disorder coefficient exceeds a preset transmission disorder coefficient threshold.

[0049] Specifically, the preset transmission fluctuation trend parameter threshold is a product of a transmission fluctuation trend parameter reference value and a transmission fluctuation factor, the preset transmission disorder coefficient threshold is a product of a transmission disorder coefficient reference value and a transmission disorder factor, the transmission fluctuation trend parameter reference value is an average value of the transmission disorder coefficient in historical data, the transmission disorder coefficient reference value is an average value of the transmission disorder coefficient in historical data, the transmission fluctuation factor and the transmission disorder factor can be set by a person skilled in the art according to the accuracy requirement of the intelligent agent service platform, the higher the accuracy requirement, the smaller the transmission fluctuation factor and the transmission disorder factor are set, the value range of the transmission fluctuation factor can be [1.2, 1.35], the value range of the transmission disorder factor can be [1.15, 1.3], preferably, the transmission fluctuation factor can be 1.3, and the transmission disorder factor can be 1.2.

[0050] Specifically, the embodiment of the application determines whether there is a transmission risk in the next monitoring period in time sequence based on the transmission disorder coefficient through the node feature analysis module. It can be understood that in actual scenarios, it is easy to misjudge the data transmission risk by using a single index such as transmission delay or resource occupation. For example, there is no risk when resources are scarce but the transmission time sequence is stable, that is, the node load is high but the scheduling is orderly, or there is abnormal fluctuation before node failure when resources are idle but the transmission time sequence is disordered. At the same time, traditional supply chain risk monitoring relies on single-point data threshold judgment, such as monitoring whether the current delay exceeds the threshold, while ignoring the trend deterioration, such as gradually increasing delay but not reaching the threshold, and finally suddenly exploding. Through the transmission fluctuation curve and the transmission disorder coefficient, the accumulation process of the risk can be captured in advance, the risk judgment of the next period is more forward-looking, the transmission interruption caused by risk burst is avoided, the misjudgment rate is reduced, and the stability of data transmission of each node on the supply chain directly affects core businesses such as distribution, for example, data transmission interruption will cause distribution data to be unable to be synchronized, and settlement disputes are caused. By determining whether there is a transmission risk in the next monitoring period in time sequence through the transmission disorder coefficient, adjustment is made in time to ensure the real-time transmission efficiency of data, thereby improving the smoothness of the supply chain distribution business, and further, the period with a transmission risk is predicted according to the data transmission fluctuation characteristics of the node, and the reliability and processing efficiency of the intelligent agent service platform for supply chain situation risk perception are improved.

[0051] Specifically, it can be understood that the transmission fluctuation tendency parameter is determined according to the data check return moment and the resource idle parameter, the data check return moment reflects the time synchronization of data transmission, the resource idle parameter, that is, the CPU idle time length proportion reflects the processing capacity of the node, the transmission fluctuation tendency parameter can comprehensively quantify the transmission stability of the node from the time dimension (transmission sequence) and the capacity dimension (processing resource), avoid the judgment deviation caused by a single parameter, determine the transmission fluctuation curve according to the transmission fluctuation tendency parameters of several monitoring periods, convert discrete single-point data into continuous time sequence trend, intuitively reflect the data transmission fluctuation trend of the node, the slope of each period can represent the change rate of fluctuation, the transmission disorder coefficient represents the stability of fluctuation, the data transmission risk is predicted through the stability of fluctuation trend, the risk is judged through the double conditions of the fluctuation tendency parameter and the transmission disorder coefficient of the transmission fluctuation curve in the last period, both the current state and the change trend are considered, and accidental fluctuation or potential deterioration risk is avoided. Misjudgment or missed judgment, and then, the period with transmission risk is predicted according to the data transmission fluctuation characteristics of the node, the reliability and processing efficiency of the intelligent agent service platform for supply chain situation risk perception are improved.

[0052] Referring to Figure 4 As shown in the figure, it is a logic flow chart of the node regulation module of the embodiment of the application, the node regulation module is used to mark the characteristic node, and determine the risk node tendency coefficient, wherein, The node regulation module marks the risk node as a characteristic node based on the determination result that the risk representation parameter of the risk node in the next monitoring period and the determination result of the transmission risk meet the characteristic node condition; If the risk representation parameter of the risk node in the next monitoring period and the determination result of the transmission risk do not meet the characteristic node condition, the node regulation module does not mark the risk node; The characteristic node condition is that the risk representation parameter exceeds the preset risk representation parameter threshold, and the determination result is that there is transmission risk; The risk node tendency coefficient is the ratio of the number of characteristic nodes on the supply chain to the number of risk nodes.

[0053] Specifically, the risk node is in the next monitoring cycle, and the risk characterization parameter is the risk characterization parameter actually obtained in the next monitoring cycle. The preset risk characterization parameter threshold is the product of the risk characterization parameter reference value and the preset risk characterization control factor. The risk characterization parameter reference value is the average value of the risk characterization parameter of the node in the historical data. The risk characterization control factor can be set by the person skilled in the art according to the accuracy requirement of the intelligent agent service platform. The higher the accuracy requirement, the smaller the risk characterization control factor. The value range of the risk characterization control factor can be [1.1, 1.35], and preferably, the risk characterization control factor can be 1.2.

[0054] Specifically, the node control module adjusts the bandwidth proportion of the distribution data in the transmission data of each risk node on the supply chain based on the determination result that the risk node tendency coefficient of the supply chain does not meet the early warning condition. Specifically, in the embodiment of the application, the determination method of the distribution data in the transmission data is not specifically limited. A distribution feature library can be established by a pre-set keyword or field library specific to the distribution data. A text analysis tool is used to scan the content of the transmission data, and the content of the transmission data is matched and identified. This will not be repeated here.

[0055] Specifically, the node control module determines the risk node tendency coefficient of the supply chain according to the risk characterization parameter of the next monitoring cycle and the determination result of the transmission risk, so as to select the bandwidth proportion of the distribution data in the transmission data of each risk node on the supply chain. It can be understood that the risk node tendency coefficient of the supply chain does not meet the early warning condition, that is, there is a data transmission risk in the supply chain, and the proportion of the number of nodes meeting the prediction conclusion is large, and the risk assessment prediction is more reliable. The bandwidth proportion of the distribution data in the transmission data of each risk node on the supply chain can preferentially guarantee the transmission of core data, reduce the risk of supply chain cooperation, and the transmission interruption of the distribution data may cause order delay, fund reconciliation error and other chain problems. By adjusting the bandwidth proportion of the distribution data of the risk node, the stability of the core data can be preferentially guaranteed when the node transmission capacity fluctuates, and the supply chain cooperation failure caused by loss or delay of key data is reduced. Further, the reliability and processing efficiency of the intelligent agent service platform for supply chain situation risk perception are improved by adaptively adjusting each risk node on the supply chain according to the prediction result.

[0056] The node control module issues an early warning to each risk node on the supply chain based on the determination result that the risk node tendency coefficient of the supply chain meets the early warning condition. Specifically, the embodiment of the present application determines the risk node tendency coefficient of the supply chain according to the risk characterization parameter of the next monitoring period and the determination result of the transmission risk through the node regulation module, so as to select the risk nodes on the supply chain to issue a warning. It can be understood that the risk node tendency coefficient of the supply chain meets the warning condition, that is, there is a data transmission risk in the supply chain, the proportion of the number of nodes meeting the prediction conclusion is small, the risk assessment prediction is unreliable, the network environment has a sudden abnormality, and it is difficult to improve the data transmission by adjusting the bandwidth proportion of the profit data. The risk nodes are warned, and it is not blindly adjusted, from passive regulation to active tracing, triggering manual or automatic investigation, finding the root cause of the prediction deviation, and adjusting the nodes on the supply chain according to the prediction result, thereby improving the reliability and processing efficiency of the intelligent agent service platform for supply chain situation risk perception.

[0057] The warning condition is that the risk node tendency coefficient of the supply chain does not exceed a preset risk node tendency coefficient threshold of the supply chain.

[0058] Specifically, the preset risk node tendency coefficient threshold of the supply chain is the product of the risk node tendency coefficient reference value and the risk node tendency regulation factor. The risk node tendency coefficient reference value is the average value of the risk node tendency coefficient in the historical data. The risk node tendency regulation factor can be set by the person skilled in the art according to the accuracy requirement of the intelligent agent service platform. The higher the accuracy requirement, the smaller the risk node tendency regulation factor. The value range of the risk node tendency regulation factor can be [1.15, 1.35], and preferably, the risk node tendency regulation factor can be 1.2.

[0059] Specifically, the increase of the bandwidth proportion of the profit data in the transmission data of the risk node has a positive correlation with the risk characterization parameter.

[0060] Specifically, the increase of the bandwidth proportion of the profit data in the transmission data is the product of the current bandwidth proportion of the profit data and the bandwidth regulation coefficient. The bandwidth regulation coefficient is the risk characterization parameter / risk characterization parameter reference value x bandwidth regulation factor. The risk characterization parameter reference value is the average value of the risk characterization parameter of the node in the historical data. The bandwidth regulation factor can be set by the person skilled in the art according to the accuracy requirement of the intelligent agent service platform. The higher the accuracy requirement, the smaller the bandwidth regulation factor. The value range of the bandwidth regulation factor can be [0.2, 0.45] to avoid too large or too small bandwidth adjustment each time. Preferably, the bandwidth regulation factor can be 0.3, and the maximum adjustment proportion of the bandwidth proportion is not more than 70%.

[0061] So far, the technical solutions of the present application have been described in combination with the preferred embodiments shown in the drawings, but it is easy for those skilled in the art to understand that the protection scope of the present application is obviously not limited to these specific embodiments. Those skilled in the art can make equivalent changes or replacements to the related technical features without departing from the principles of the present application, and the technical solutions after the changes or replacements will all fall within the protection scope of the present application.

[0062] The above only describes the preferred embodiments of the present application and is not intended to limit the present application; the present application can have various changes and variations for those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. An intelligent agent service platform based on real-time supply chain situation risk perception, characterized in that, include: The data acquisition module is used to acquire the transmission delay jitter frequency, data verification return parameters, and resource idle parameters of each node in the supply chain. The data verification return parameters include the data verification return rate and the data verification return time. The node feature identification module, which is connected to the data acquisition module, is used to determine risk characterization parameters based on the link jitter frequency and data verification return rate of the node within a preset monitoring period, and to screen risk nodes based on the risk characterization parameters of the node. The node feature analysis module is connected to the data acquisition module and the node feature identification module respectively. It is used to determine the transmission fluctuation tendency parameters of each monitoring period within the preset monitoring period based on the data verification and feedback time and resource idle parameters of the risk node, determine the transmission fluctuation curve based on the transmission fluctuation tendency parameters, and determine whether there is a transmission risk in the next monitoring period in the time sequence based on the transmission misalignment coefficient. The node control module is connected to the node feature identification module and the node feature analysis module respectively. It is used to determine the risk node tendency coefficient of the supply chain based on the risk characterization parameters and the judgment result of transmission risk in the next monitoring cycle, so as to select the adjustment of the proportion of revenue sharing data bandwidth in the transmission data of each risk node in the supply chain, or to issue an early warning to each risk node in the supply chain.

2. The intelligent agent service platform based on real-time supply chain situation risk perception according to claim 1, characterized in that, The node feature recognition module is used to determine risk characterization parameters, wherein, The node feature recognition module determines the risk characterization parameter by weighted summation of the first risk propensity feature and the second risk propensity feature; The first risk tendency feature is the ratio of the link jitter frequency of the node to the link jitter frequency threshold within a preset monitoring period; The second risk tendency feature is the ratio of the data verification return rate of the node to the data verification return rate threshold within a preset monitoring period.

3. The intelligent agent service platform based on real-time supply chain situation risk perception according to claim 2, characterized in that, The node feature recognition module is used to screen for risky nodes, wherein... The node feature recognition module filters the node as a risk node based on the judgment result that the node's risk characterization parameters meet the risk node screening conditions. The risk node screening condition is that the risk characterization parameter exceeds the preset risk characterization parameter threshold.

4. The intelligent agent service platform based on real-time supply chain situation risk perception according to claim 3, characterized in that, The node feature analysis module is used to determine transmission fluctuation tendency parameters, wherein... The node feature analysis module determines the sum of the first transmission fluctuation coefficient and the second transmission fluctuation coefficient as the transmission fluctuation tendency parameter; The first transmission fluctuation coefficient is the ratio of the backhaul interval fluctuation parameter to the backhaul interval fluctuation parameter threshold, and the second transmission fluctuation coefficient is the ratio of the resource idle parameter to the resource idle parameter threshold.

5. The intelligent agent service platform based on real-time supply chain situation risk perception according to claim 4, characterized in that, The backhaul interval fluctuation parameter is the average of the interval duration between adjacent data verification backhaul times; The resource idle parameter is the ratio of the CPU idle time of the node during the monitoring period to the total duration of the monitoring period.

6. The intelligent agent service platform based on real-time supply chain situation risk perception according to claim 5, characterized in that, The node feature analysis module is used to determine the transmission offset coefficient, wherein... The node feature analysis module is used to obtain the slope of each monitoring period on the transmission fluctuation curve, calculate the absolute value of the difference between the slopes of adjacent periods, and determine the average value of the absolute value of the difference as the transmission misalignment coefficient.

7. The intelligent agent service platform based on real-time supply chain situation risk perception according to claim 6, characterized in that, The node feature analysis module is used to determine whether there is a transmission risk in the next monitoring period in terms of timing. The node feature analysis module determines that there is a transmission risk in the next monitoring cycle based on the judgment results that the transmission fluctuation tendency parameter and transmission misalignment coefficient of the transmission fluctuation curve meet the risk period conditions. The risk period condition is that the transmission fluctuation tendency parameter of the last monitoring period on the transmission fluctuation curve exceeds the preset transmission fluctuation tendency parameter threshold, and the transmission misalignment coefficient exceeds the preset transmission misalignment coefficient threshold.

8. The intelligent agent service platform based on real-time supply chain situation risk perception according to claim 7, characterized in that, The node control module is used to mark characteristic nodes and determine the risk node tendency coefficient, wherein... The node control module marks the risk node as a characteristic node based on the judgment result that the risk characterization parameters and transmission risk judgment result of the risk node in the next monitoring cycle meet the characteristic node conditions. The condition for the feature node is that the risk characterization parameter exceeds the preset risk characterization parameter threshold, and the determination result is that there is a transmission risk. The risk node tendency coefficient is the ratio of the number of characteristic nodes to the number of risk nodes in the supply chain.

9. The intelligent agent service platform based on real-time supply chain situation risk perception according to claim 8, characterized in that, The node control module adjusts the proportion of revenue-sharing data bandwidth in the transmitted data for each risk node in the supply chain based on the judgment result that the risk node tendency coefficient of the supply chain does not meet the early warning conditions. The node control module issues early warnings to each risk node in the supply chain based on the judgment result that the risk node tendency coefficient of the supply chain meets the early warning conditions. The warning condition is that the risk node tendency coefficient of the supply chain does not exceed the preset risk node tendency coefficient threshold of the supply chain.

10. The intelligent agent service platform based on real-time supply chain situation risk perception according to claim 9, characterized in that, The increase in the proportion of bandwidth allocated to revenue sharing in the transmitted data of risk nodes is positively correlated with the risk characterization parameters.

Images