Production abnormity early warning method and system based on knowledge graph

Abstract

The invention relates to the technical field of knowledge maps, and particularly discloses a production abnormity early warning method and system based on a knowledge map. According to the method, the timeliness of the to-do task, the unprocessed notification content and the monitoring rule triggering state are extracted from the enterprise management system, the knowledge graph is constructed to realize data integration and semantic association, flow abnormal links and node association are analyzed based on the unprocessed notification content, the node abnormality rate is calculated to accurately locate the blocked nodes, and the efficiency of the enterprise management system is improved. The method comprises the following steps: obtaining monitoring time efficiency and notification time efficiency through a monitoring rule triggering state, calculating abnormal delay rate quantitative response delay, generating a comprehensive abnormal rate in combination with an overstocked task rate, a node abnormal rate and an abnormal delay rate, identifying trend change through comparison with historical data, triggering a personalized early warning mechanism based on a preset threshold interval, and ensuring early intervention. According to the method, the overstocked state can be obtained in time, and the problems that the overstocked state cannot be obtained and early warning cannot be performed according to the corresponding overstocked state due to multi-task notification of window frame overstocked can be solved.

Description

Technical Field

[0001] The present invention relates to the technical field of knowledge graphs, and in particular to a production anomaly warning method and system based on a knowledge graph. Background Art

[0002] An enterprise management system refers to a software system that uses information technology to integrally manage various resources (such as human, material, financial, information, etc.) and business processes of an enterprise. It provides timely and accurate information support for the enterprise decision-making level and each department through data collection, processing, storage, and analysis, so as to achieve the strategic goals of the enterprise and the efficient management of daily operations. In an enterprise management system, a knowledge graph needs to be used. Among them, a knowledge graph is a structured data model that represents knowledge in the form of a semantic network. Its core is to abstract entities (such as people, things, concepts, etc.) and their relationships in the real world into "nodes" and "edges", and describe the characteristics of entities through attributes. It is essentially a semantic network containing entities, relationships, and attributes, which can store and express knowledge in a machine-understandable way, support complex knowledge reasoning and intelligent applications; When the existing enterprise management system processes task stacking, it usually adopts early timely response. In this way, during the process of timely response to tasks, there is a backlog of multi-task notification window frames, and the backlog status cannot be obtained, and there is a problem of warning according to the corresponding backlog status. For this reason, a production anomaly warning method based on a knowledge graph is needed to solve the above problems. Summary of the Invention

[0003] The purpose of the present invention is to provide a production anomaly warning method and system based on a knowledge graph to solve the technical problems raised in the above background art.

[0004] To achieve the above purpose, the present invention provides the following technical solutions: A production anomaly warning method based on a knowledge graph, comprising: Obtaining operation knowledge graph data in an enterprise management system, where the operation knowledge graph data includes to-be-done task timeliness data, unprocessed notification content, and monitoring rule trigger status; Obtaining the task backlog time according to the to-be-done task timeliness data, and obtaining the backlog task rate according to the task backlog time; Obtaining process anomaly link information and node association information according to the unprocessed notification content, and obtaining the node anomaly rate according to the process anomaly link information and the node association information; Obtaining monitoring timeliness information and notification timeliness information according to the monitoring rule trigger status, and obtaining the anomaly delay rate according to the monitoring timeliness information and the notification timeliness information; Obtain a comprehensive exception rate based on the backlog task rate, the node exception rate, and the exception delay rate; Warn about exceptions in the enterprise management system based on the comprehensive exception rate.

[0005] Preferably, the step of obtaining the task backlog time according to the to-do task timeliness data and obtaining the backlog task rate according to the task backlog time includes: Obtain the planned completion time and multiple actual completion times of single tasks according to the to-do task timeliness data, and obtain multiple single-task backlog times according to the planned completion time and the multiple actual completion times of single tasks; Successively determine whether multiple single-task backlog times exceed a preset threshold; If it exceeds the preset threshold, cumulatively calculate the backlog times of multiple single tasks that exceed the preset threshold to obtain the total cumulative backlog duration; Obtain the corresponding number of tasks according to the multiple actual completion times of single tasks, and calculate the standard total completion time according to the number of tasks and the planned completion time; Obtain the backlog task rate according to the standard total completion time and the total cumulative backlog duration.

[0006] Preferably, the step of obtaining the node exception rate according to the process exception link information and the node association information includes: Obtain multiple blocking nodes according to the node association information, and obtain multiple previous passing times of the corresponding blocking nodes according to the multiple blocking nodes; And obtain the current passing occupancy ratio according to the multiple previous passing times and the preset historical node passing, and use the current passing occupancy ratio as the node blocking deviation; Obtain the stay task identifier according to the process exception link information, and obtain the task path identifier according to the stay task identifier; Obtain the historical communication average number and the current communication number of the corresponding task identifier according to the task path identifier, and calculate the node communication frequency difference according to the historical node communication average number, the current communication number, and the preset communication time, and use the node communication frequency difference as the task dispersion coefficient; Obtain the node exception rate according to the node blocking deviation and the task dispersion coefficient.

[0007] Preferably, the step of obtaining the exception delay rate according to the monitoring timeliness information and the notification timeliness information includes: Obtain the monitoring average delay time according to the monitoring timeliness information; Obtain the monitoring delay ratio by obtaining the historical average delay time and the monitoring average delay time within a preset time; Obtain the notification response time according to the notification timeliness information, where the notification response time includes a lag response time and an early response time, and obtain the response average value according to the lag response time and the early response time; Obtain the historical average response time within a preset time, and obtain the notification delay ratio according to the historical average response time and the response average value; Perform weighted calculation according to the monitoring delay ratio and the notification delay ratio to obtain the abnormal delay rate.

[0008] Preferably, the step of obtaining the comprehensive abnormality rate according to the backlog task rate, the node abnormality rate, and the abnormal delay rate includes: Obtain the corresponding backlog task rate weight value according to the backlog task rate; Obtain the corresponding node abnormality rate weight value according to the node abnormality rate; Calculate the comprehensive abnormality rate according to the backlog task rate, the node abnormality rate, the abnormal delay rate, the backlog task rate weight value, and the node abnormality rate weight value, where the calculation formula is: ; Among them, represents the comprehensive abnormality rate, represents the backlog task rate, represents the node abnormality rate, represents the abnormal delay rate, represents the backlog task rate weight value, and b represents the node abnormality rate weight value.

[0009] Preferably, the step of warning about the abnormality in the enterprise management system according to the comprehensive abnormality rate includes: Obtain the historical abnormality rate in the enterprise management system in history, and calculate the abnormality rate occupancy ratio according to the comprehensive abnormality rate and the historical abnormality rate; Determine whether the abnormality rate occupancy ratio is greater than the preset abnormality rate occupancy ratio; If the comprehensive abnormality rate is greater than the preset abnormality rate occupancy ratio, determine the abnormal state in the current enterprise management system, and give a warning about the abnormality in the enterprise management system.

[0010] A production abnormality warning system based on a knowledge graph, including: A first acquisition module for acquiring operation knowledge graph data in the enterprise management system, where the operation knowledge graph data includes to-be-done task timeliness data, unprocessed notification content, and monitoring rule trigger status; A second acquisition module for obtaining the task backlog time according to the to-be-done task timeliness data, and obtaining the backlog task rate according to the task backlog time; A third acquisition module, configured to acquire process exception link information and node association information according to the unprocessed notification content, and acquire a node exception rate according to the process exception link information and the node association information; A fourth acquisition module, configured to acquire monitoring timeliness information and notification timeliness information according to the triggered status of the monitoring rule, and acquire an exception delay rate according to the monitoring timeliness information and the notification timeliness information; A fifth acquisition module, configured to acquire a comprehensive exception rate according to the backlog task rate, the node exception rate, and the exception delay rate; An early warning module, configured to give an early warning of exceptions in the enterprise management system according to the comprehensive exception rate.

[0011] Preferably, the second acquisition module includes: A first acquisition unit, configured to acquire a planned completion time and multiple single-task actual completion times according to the to-do task timeliness data, and acquire multiple single-task backlog times according to the planned completion time and the multiple single-task actual completion times; A judgment unit, configured to sequentially judge whether the multiple single-task backlog times exceed a preset threshold; If it exceeds the preset threshold, cumulatively calculate the multiple single-task backlog times that exceed the preset threshold to obtain a cumulative backlog total duration; A second acquisition unit, configured to acquire the corresponding number of tasks according to the multiple single-task actual completion times, and calculate a standard total completion time according to the number of tasks and the planned completion time; A third acquisition unit, configured to acquire a backlog task rate according to the standard total completion time and the cumulative backlog total duration.

[0012] This application also provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, the steps of the above method are implemented.

[0013] This application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0014] The beneficial effects of this application are as follows: The present invention extracts the timeliness of to-do tasks, the content of unprocessed notifications, and the triggering status of monitoring rules from the enterprise management system, constructs a knowledge graph to achieve data integration and semantic association, analyzes the abnormal process links and node associations based on the content of unprocessed notifications, calculates the node abnormal rate to accurately locate blocked nodes; obtains the monitoring timeliness and notification timeliness through the triggering status of monitoring rules, calculates the abnormal delay rate to quantify the response delay, combines the backlog task rate, node abnormal rate, and abnormal delay rate to generate a comprehensive abnormal rate, comprehensively evaluates the system status from three dimensions of task backlog, process blockage, and response delay, and finally issues a warning based on the comprehensive abnormal rate: identifies trend changes by comparing with historical data, and triggers a personalized warning mechanism based on a preset threshold range to ensure early intervention. This method can timely obtain the backlog status, solve the problem of the backlog of multi-task notification window frames, resulting in the inability to obtain the backlog status, and issue a warning according to the corresponding backlog status. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic flowchart of the method according to an embodiment of this application.

[0016] Figure 2 It is a schematic structural diagram of the system according to an embodiment of this application.

[0017] Figure 3 It is a schematic internal structure diagram of a computer device according to an embodiment of this application.

[0018] The realization, functional features, and advantages of the purpose of this application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0020] As Figures 1 - 3 shown, this application provides a production anomaly warning method based on a knowledge graph, including: S1. Obtain the running knowledge graph data in the enterprise management system, where the running knowledge graph data includes to-do task timeliness data, unprocessed notification content, and monitoring rule triggering status; S2. Obtain the task backlog time according to the to-do task timeliness data, and obtain the backlog task rate according to the task backlog time; S3. Obtain the process abnormal link information and node association information according to the unprocessed notification content, and obtain the node abnormal rate according to the process abnormal link information and the node association information; S4. Obtain the monitoring timeliness information and notification timeliness information according to the monitoring rule triggering status, and obtain the abnormal delay rate according to the monitoring timeliness information and the notification timeliness information; S5. Obtain a comprehensive exception rate based on the backlog task rate, the node exception rate, and the exception delay rate; S6. Issue a warning about exceptions in the enterprise management system based on the comprehensive exception rate.

[0021] As described in the above steps S1 - S6, since the existing enterprise management system usually responds in advance when dealing with task stacking, which causes backlogs in the multi - task notification window frames during the timely response process of tasks, and it is impossible to obtain the backlog status and issue a warning based on the corresponding backlog status. The present invention first obtains the operation knowledge graph data in the enterprise management system. Among them, the operation knowledge graph data includes to - do task timeliness data, unprocessed notification content, and monitoring rule trigger status. In this way, the to - do task timeliness data, unprocessed notification content, and monitoring rule trigger status are extracted from the enterprise management system to construct a structured knowledge graph, realizing the integration and semantic representation of production data, and providing multi - dimensional data support for subsequent exception analysis to ensure information comprehensiveness (such as task timeliness, process exceptions, monitoring status). Secondly, through the "node - edge" structure of the knowledge graph, the association relationships between data are clarified (such as the mapping between tasks and nodes, processes and monitoring); The specific acquisition process is as follows: First, deploy a data acquisition interface in the enterprise management system, and extract data from the to - do task module, notification center, and monitoring rule engine through API interfaces or database real - time synchronization methods. For example, obtain timeliness data such as the planned completion time and actual completion time of to - do tasks from the task table through SQL query statements, and then map the collected data to nodes and edges in the knowledge graph. To - do tasks, notification content, monitoring rules, etc. are used as entity nodes, and the association relationships between tasks and notifications, the trigger relationships between rules and tasks, etc. are used as edges, and attributes such as timeliness, content, and status are added to the nodes. Finally, the collected data is de - duplicated, corrected, and standardized to ensure data quality. For example, unify the time format, standardize task identifiers, and eliminate the problem of inconsistent data formats in different modules; Then, obtain the task backlog time based on the to - do task timeliness data, and obtain the backlog task rate based on the task backlog time. In this way, by comparing the planned completion time and actual completion time of tasks, the degree of task backlog is quantified, reflecting the abnormality of production efficiency, and identifying the scale (such as the total backlog duration) and proportion (backlog task rate) of task processing delays, locating the efficiency bottleneck. At the same time, by real - time tracking the difference between the actual completion time and planned time of tasks, the task delay trend can be discovered in a timely manner, providing data support for resource allocation; Then, obtain process exception link information and node association information according to the unprocessed notification content, and obtain the node exception rate according to the process exception link information and the node association information. By analyzing the process exception link information and the node association information in this way, blocked nodes in the process can be accurately identified, avoiding the fuzzy judgment of the exception location in the traditional method. Secondly, by combining the node block deviation and the task dispersion coefficient, node exceptions are evaluated from two dimensions of passing efficiency and communication frequency, improving the accuracy of exception judgment. And based on the node association relationship of the knowledge graph, dynamic association analysis between the exception node and the task path is realized, facilitating the tracing of the exception root cause; Secondly, obtain monitoring timeliness information and notification timeliness information according to the triggered status of the monitoring rule, and obtain the exception delay rate according to the monitoring timeliness information and the notification timeliness information. By monitoring the average delay time and the notification response time in this way, accurate quantification of the timeliness of the system monitoring and notification links is realized, avoiding the fuzzy evaluation of response delays; At the same time, obtain the comprehensive exception rate according to the backlog task rate, the node exception rate, and the exception delay rate. By combining the backlog task rate, the node exception rate, and the exception delay rate, system exceptions are comprehensively evaluated from three dimensions of task backlog, process blockage, and response delay, avoiding the one-sidedness of a single indicator. At the same time, multiple-dimensional indicators are synthesized into a single comprehensive exception rate, facilitating the management layer to intuitively understand the overall exception degree of the system and providing a clear basis for decision-making; Finally, give early warnings about exceptions in the enterprise management system according to the comprehensive exception rate. By comparing the current comprehensive exception rate with the historical exception rate in this way, the changing trend of the exception degree can be found, avoiding misjudgment based only on absolute thresholds. And by setting the preset exception rate occupancy ratio according to the actual situation of the enterprise, personalized customization of the early warning mechanism is realized, improving the accuracy of early warning. Secondly, when the exception rate occupancy ratio exceeds the threshold, the early warning is triggered immediately, ensuring that the enterprise can timely obtain the backlog status and can solve the problem that the backlog in the multi-task notification window causes the inability to obtain the backlog status and the early warning is carried out according to the corresponding backlog status.

[0022] In one embodiment, step S2 of obtaining the task backlog time according to the to-be-done task timeliness data and obtaining the backlog task rate according to the task backlog time includes: S201. Obtain the planned completion time and the actual completion times of multiple single tasks according to the to-be-done task timeliness data, and obtain the backlog times of multiple single tasks according to the planned completion time and the actual completion times of multiple single tasks; S202. Sequentially judge whether the backlog times of multiple single tasks exceed a preset threshold; If it exceeds the preset threshold, cumulatively calculate the backlog times of multiple single tasks that exceed the preset threshold to obtain the total cumulative backlog duration; S203. Obtain the corresponding number of tasks according to the actual completion times of multiple single tasks, and calculate the standard total completion time according to the number of tasks and the planned completion time; S204. Obtain the backlog task rate according to the standard total completion time and the cumulative backlog total duration.

[0023] As described in the above steps S201 - S204, the present invention first obtains the planned completion time and the actual completion times of multiple single tasks according to the to - do task timeliness data, and obtains the backlog times of multiple single tasks according to the planned completion time and the actual completion times of multiple single tasks. In this way, the planned completion time of the task and the actual completion time of each single task are extracted from the to - do task timeliness data to form a basic data pair in the time dimension. Secondly, establishing a comparison benchmark of "planned completion time - actual completion time" is the primary link for quantifying the degree of task backlog. By structurally collecting the time data of multiple tasks, it lays a foundation for batch - analyzing the overall delay trend of the task queue, ensuring the operability and accuracy of subsequent analysis; Next, sequentially determine whether the backlog times of multiple single tasks exceed a preset threshold; If it exceeds the preset threshold, then cumulatively calculate the backlog times of multiple single tasks that exceed the preset threshold to obtain the cumulative backlog total duration. In this way, tasks with serious delays can be screened out and their total delay durations can be counted, and it focuses on the task backlog situation that has a greater impact on the production process, avoiding the interference of minor delays, highlighting key issues, enabling the enterprise to prioritize tasks that seriously affect production efficiency. At the same time, the setting of the preset threshold can be determined according to the actual production requirements and historical data of the enterprise. By screening the backlog times that exceed the threshold, tasks that have a significant impact on the production progress can be accurately located, providing a clear direction for resource allocation and exception handling, and improving the pertinence of exception warning; Secondly, obtain the corresponding number of tasks according to the actual completion times of multiple single tasks, and calculate the standard total completion time according to the number of tasks and the planned completion time. In this way, a task completion time benchmark under normal circumstances can be established, and a comparison standard for evaluating the actual backlog situation can be provided, making the measurement of the backlog degree relative, facilitating comparison and analysis between task queues of different scales. At the same time, considering the relationship between the number of tasks and the planned completion time, by calculating the standard total completion time through standardization, tasks of different quantities can be converted into comparable time dimensions, avoiding evaluation biases caused by differences in the number of tasks, and ensuring the scientific nature of the calculation of the backlog task rate; Finally, obtain the backlog task rate based on the total time according to the standard and the total cumulative backlog duration. In this way, through the backlog task rate, the task backlog situation can be converted into a quantitative indicator, intuitively reflecting the overall task backlog level. At the same time, it can also provide a clear quantitative indicator for the enterprise management layer, facilitating a quick understanding of the severity of task backlog in the production system and providing data support for decision-making, such as whether to increase manpower, adjust the production plan, etc. Secondly, as a comprehensive indicator, the backlog task rate integrates two key factors: backlog duration and standard completion time, converting the qualitative backlog problem into a quantitative value, making the judgment of abnormal situations more objective and accurate, and meeting the needs of the enterprise's refined management.

[0024] In one embodiment, step S3 of obtaining the node abnormality rate according to the process abnormal link information and the node association information includes: S301. Obtain a plurality of blocking nodes according to the node association information, and obtain a plurality of previous passing times of the corresponding blocking nodes according to the plurality of blocking nodes; S302. Obtain the current passing occupancy ratio according to the plurality of previous passing times and the preset historical node passing, and use the current passing occupancy ratio as the node blocking deviation degree; S303. Obtain the stay task identifier according to the process abnormal link information, and obtain the task path identifier according to the stay task identifier; S304. Obtain the historical communication average number and the current communication number of the corresponding task identifier according to the task path identifier, calculate the node communication frequency difference according to the historical node communication average number, the current communication number and the preset communication time, and use the node communication frequency difference as the task dispersion coefficient; S305. Obtain the node abnormality rate according to the node blocking deviation degree and the task dispersion coefficient.

[0025] As described in the above steps S301 - S305, the present invention first obtains a plurality of blocking nodes according to the node association information, and obtains a plurality of previous passing times of the corresponding blocking nodes according to the plurality of blocking nodes. In this way, by positioning the blocking nodes and their passing data, it provides a direct basis for evaluating node abnormalities, facilitating the quick locking of bottleneck links in the production process. Secondly, the node association relationship in the knowledge graph can intuitively reflect the process logic, and obtaining the blocking nodes and passing times is the basis for quantifying node abnormalities. Through historical data comparison, the abnormal fluctuations of the current passing efficiency can be effectively identified, providing data support for subsequent deviation degree calculation; Next, obtain the current traffic occupancy ratio based on the multiple previous traffic times and the preset historical node traffic, and use the current traffic occupancy ratio as the node block deviation degree. In this way, by converting the current traffic occupancy ratio into the node block deviation degree, the abnormal degree of the node traffic efficiency can be quantified. Secondly, convert the change in the traffic times into a quantifiable deviation index, intuitively reflecting the severity of the node block, and providing a basis for abnormal classification. Immediately afterwards, obtain the stay task identifier according to the process abnormal link information, and obtain the task path identifier according to the stay task identifier, which can establish the mapping relationship between the abnormal task and the process path. Secondly, through the association of the task identifier and the path, the process traceability of the abnormal task is realized, which is convenient for analyzing the propagation path and influence range of the abnormality in the overall process. At the same time, obtain the historical communication average number and the current communication number of the corresponding task identifier according to the task path identifier, calculate the node communication frequency difference according to the historical node communication average number, the current communication number and the preset communication time, and use the node communication frequency difference as the task dispersion coefficient, which can form the task dispersion coefficient and quantify the abnormality of the cooperation efficiency between nodes. At the same time, the abnormal cooperation between nodes is reflected through the change in the communication frequency, such as communication delay or frequent communication, which can assist in judging whether the blocked node is abnormal due to cooperation problems. Finally, obtain the node abnormality rate according to the node block deviation degree and the task dispersion coefficient. In this way, the node abnormality rate is obtained through weighted calculation, forming a single quantifiable index to reflect the overall abnormal degree of the node, and integrating the abnormalities in the two dimensions of traffic efficiency and cooperation efficiency into a unified index, which is convenient for the management to quickly judge the health status of the node and provide a decision-making basis for the abnormal handling priority. At the same time, a single index (such as the block deviation degree) may not comprehensively reflect the node abnormality. Combining the task dispersion coefficient in the cooperation dimension can avoid misjudgment caused by a single factor. The weighted calculation can adjust the weights of each dimension according to the actual needs of the enterprise to improve the adaptability of the early warning system. Among them, the weight distribution logic of the node block deviation degree and the task dispersion coefficient is that node block will directly lead to task backlog and process interruption, which has a more significant impact on production efficiency. For example, in the system allocation of the material approval node, it causes the material to be unable to flow. At this time, the block deviation degree can quickly locate the problem node. At the same time, through the statistics of historical production data, the proportion of node block type abnormalities in the total abnormalities is relatively high (such as 60%), so a higher weight is given to prioritize the weight distribution of the task dispersion coefficient.

[0026] In one embodiment, step S4 of obtaining the abnormal delay rate according to the monitoring timeliness information and the notification timeliness information includes: S401. Obtain the monitoring average delay time according to the monitoring timeliness information; S402. Obtain the historical average delay time and the monitored average delay time within a preset time, and obtain the monitored delay ratio; S403. Obtain the notification response time according to the notification timeliness information, where the notification response time includes a lag response time and an early response time, and obtain the response mean value according to the lag response time and the early response time; S404. Obtain the historical average response time within a preset time, and obtain the notification delay ratio according to the historical average response time and the response mean value; S405. Perform weighted calculation according to the monitored delay ratio and the notification delay ratio to obtain the abnormal delay rate.

[0027] As described in the above steps S401 - S405, the present invention first obtains the monitored average delay time according to the monitored timeliness information, so as to extract the average delay time of the monitoring task from the monitored timeliness information, quantify the response lag degree of the monitoring link, and establish the benchmark data of the monitored timeliness, providing a quantitative basis for judging whether the current monitoring system can detect abnormalities in a timely manner. Secondly, as the "perception layer" of production abnormalities, the delay of monitoring directly affects the timeliness of early warning. The efficiency bottleneck of the monitoring module can be quickly located through the average delay time, providing a direction for subsequent optimization; Next, obtain the historical average delay time and the monitored average delay time within a preset time to obtain the monitored delay ratio, so as to obtain the monitored delay ratio, reflecting the deviation degree of the current delay relative to the historical level, and eliminating the time base difference of different monitoring tasks through historical data comparison, avoiding misjudgment caused by a single absolute value judgment, and improving the accuracy of abnormality recognition; Then, obtain the notification response time according to the notification timeliness information, where the notification response time includes a lag response time and an early response time, and obtain the response mean value according to the lag response time and the early response time. The response mean values of the two comprehensively evaluate the timeliness performance of the notification link, and avoid ignoring the abnormality of the early response while only focusing on the lag response (such as misjudgment caused by premature triggering of the notification). By balancing the two abnormal situations through the mean value, a more objective evaluation is achieved, and both "too early" or "too late" of the notification response may lead to production decision-making mistakes. For example, an early response may trigger ineffective operations, while a lag response may miss the processing opportunity, and both need to be equally included in the evaluation system; Secondly, obtain the historical average response time within a preset time, and obtain the notification delay ratio based on the historical average response time and the response mean value. This can quantify the timeliness anomaly degree of the notification link, calibrate the current response efficiency through historical data, identify whether there is systematic delay or advance in the notification system, and provide data support for process optimization. Secondly, the historical average response time reflects the normal processing ability of the notification system, and the ratio calculation can eliminate the processing duration differences of different notification types (such as urgent / ordinary notifications), making the index more universal; Finally, perform weighted calculation based on the monitoring delay ratio and the notification delay ratio to obtain the abnormal delay rate. In this way, through the abnormal delay rate, the integrated evaluation of multi-link anomalies can be realized, and the timeliness anomalies of the two key links of monitoring and notification are transformed into a single index, which is convenient for the management to quickly master the overall response efficiency of the system. Moreover, monitoring and notification have different weights in the abnormal warning chain - monitoring is the basis for "discovering problems", and notification is the key for "transmitting problems". The weighted calculation can adjust the importance of the two according to the actual needs of the enterprise (such as monitoring weight 0.6 and notification weight 0.4), improving the adaptability of the warning system. Among them, the monitoring delay ratio reflects the timeliness of "abnormal discovery" and is the basic link of the warning chain - if the monitoring fails to discover the anomaly in time, no matter how efficient the subsequent notification link is, it cannot make up for it. Secondly, the notification delay ratio reflects the efficiency of "abnormal transmission" and is the key link of the warning chain - after the monitoring discovers the anomaly, the notification delay will lead to a response lag. Therefore, the matching logic needs to follow the principle of "basis prior to transmission", that is, the weight of the monitoring delay ratio is usually higher than that of the notification delay ratio.

[0028] In one embodiment, step S5 of obtaining the comprehensive anomaly rate according to the backlog task rate, the node anomaly rate, and the abnormal delay rate includes: S501. Obtain the corresponding backlog task rate weight value according to the backlog task rate; S502. Obtain the corresponding node anomaly rate weight value according to the node anomaly rate; S503. Calculate the comprehensive anomaly rate according to the backlog task rate, the node anomaly rate, the abnormal delay rate, the backlog task rate weight value, and the node anomaly rate weight value, where the calculation formula is: ; Among them, represents the comprehensive anomaly rate, represents the backlog task rate, represents the node anomaly rate, represents the abnormal delay rate, represents the backlog task rate weight value, and b represents the node anomaly rate weight value.

[0029] As described in the above steps S501 - S503, the present invention first obtains the corresponding backlog task rate weight value according to the backlog task rate. Secondly, it obtains the corresponding node exception rate weight value according to the node exception rate. Finally, it calculates the comprehensive exception rate based on the backlog task rate, the node exception rate, the exception delay rate, the backlog task rate weight value, and the node exception rate weight value. In this way, by integrating the backlog task rate, the node exception rate, the exception delay rate, and their weight values through a formula, a single comprehensive exception rate index is output, which comprehensively reflects the overall exception degree of the system. Moreover, it converts multi - dimensional exception indicators (task backlog, process blockage, response delay) into unified quantified values, facilitating the management to intuitively grasp the health status of the production system and reducing the decision - making complexity. Secondly, a single indicator cannot cover all scenarios of production exceptions (such as task backlog and process blockage may occur simultaneously). The weighted formula can achieve the coupled analysis of multi - dimensional exceptions through mathematical modeling. The weight assignment logic for the backlog task rate, the node exception rate, and the exception delay rate is as follows: The backlog task rate directly reflects the bottleneck of production efficiency (such as task accumulation causing delivery delays), which is an exception at the "efficiency layer". The node exception rate reflects the risk of process continuity (such as production line node blockage), which is an exception at the "process layer". The exception delay rate reflects the monitoring and response timeliness (such as lag in exception discovery / notification), which is an exception at the "response layer". Furthermore, the weight assignment follows the priority of "efficiency layer > process layer > response layer", that is, the weight of the backlog task rate is usually the highest, and the weight of the exception delay rate is the lowest.

[0030] In one embodiment, step S6 of warning about exceptions in the enterprise management system according to the comprehensive exception rate includes: S601. Obtain the historical exception rate in the history of the enterprise management system, and calculate the exception rate occupancy ratio according to the comprehensive exception rate and the historical exception rate; S602. Determine whether the exception rate occupancy ratio is greater than the preset exception rate occupancy ratio; If the comprehensive exception rate is greater than the preset exception rate occupancy ratio, then determine the exception status in the current enterprise management system and give a warning about the exceptions in the enterprise management system.

[0031] As described in the above steps S601 - S602, the present invention first obtains the historical abnormal rate of the enterprise management system in history, calculates the ratio of the abnormal rate based on the comprehensive abnormal rate and the historical abnormal rate. Then, it determines whether the ratio of the abnormal rate is greater than the preset ratio of the abnormal rate. If the comprehensive abnormal rate is greater than the preset ratio of the abnormal rate, it determines the abnormal state in the current enterprise management system and gives an early warning of the abnormality in the enterprise management system. In this way, it can quantify the deviation amplitude of the current abnormal degree relative to the historical level, and can avoid the mechanical defect of judging abnormality only based on the absolute threshold. It dynamically calibrates the abnormal standard through historical data to adapt to the periodic changes of the enterprise production mode (such as the difference between peak and off - peak seasons). And the production characteristics and management standards of different enterprises vary significantly (such as the different abnormal tolerance levels of semiconductor factories and food processing factories). The historical abnormal rate can reflect the "normal fluctuation range" of the enterprise itself. Calculating the ratio can transform the abnormal assessment into a "relative change amount", improving the pertinence of the early warning (for example, if the historical abnormal rate of an enterprise is generally high, and although the current abnormal rate is high in absolute value but the ratio does not exceed the threshold, no early warning may be given temporarily), realizing the personalized customization of the early warning mechanism and improving the accuracy of the early warning. Secondly, when the ratio of the abnormal rate exceeds the threshold, the early warning is immediately triggered to ensure that the enterprise can timely obtain the backlog status and can solve the problem that the backlog of the multi - task notification window causes the inability to obtain the backlog status and give an early warning according to the corresponding backlog status.

[0032] This application also provides a production abnormal early warning system based on a knowledge graph, including: The first acquisition module is used to acquire the operation knowledge graph data in the enterprise management system, where the operation knowledge graph data includes the time - limit data of to - do tasks, the content of unprocessed notifications, and the monitoring rule trigger status; The second acquisition module is used to obtain the task backlog time according to the time - limit data of the to - do tasks, and obtain the backlog task rate according to the task backlog time; The third acquisition module is used to obtain the process abnormal link information and node association information according to the content of the unprocessed notifications, and obtain the node abnormal rate according to the process abnormal link information and the node association information; The fourth acquisition module is used to obtain the monitoring time - limit information and notification time - limit information according to the monitoring rule trigger status, and obtain the abnormal delay rate according to the monitoring time - limit information and the notification time - limit information; The fifth acquisition module is used to obtain the comprehensive abnormal rate according to the backlog task rate, the node abnormal rate, and the abnormal delay rate; The early warning module is used to give an early warning of the abnormality in the enterprise management system according to the comprehensive abnormal rate.

[0033] In one embodiment, the second acquisition module includes: A first acquisition unit, configured to acquire a planned completion time and multiple actual completion times of single tasks according to the to-do task timeliness data, and acquire multiple single-task backlog times according to the planned completion time and the multiple actual completion times of single tasks; A judgment unit, configured to sequentially judge whether multiple single-task backlog times exceed a preset threshold; If it exceeds the preset threshold, cumulatively calculate the multiple single-task backlog times that exceed the preset threshold to obtain a cumulative backlog total duration; A second acquisition unit, configured to acquire the corresponding number of tasks according to the multiple actual completion times of single tasks, and calculate a standard total completion time according to the number of tasks and the planned completion time; A third acquisition unit, configured to acquire a backlog task rate according to the standard total completion time and the cumulative backlog total duration.

[0034] This application also provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, the steps of the above method are implemented.

[0035] This application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0036] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium provided in this application and the embodiments can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be obtained in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0037] It should be noted that in this text, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article or method comprising a series of elements not only includes those elements but also other elements not expressly listed, or further includes elements inherent to such process, apparatus, article or method. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, apparatus, article or method comprising such element.

[0038] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be similarly included in the patent protection scope of the present invention.

Claims

1. A production anomaly warning method based on a knowledge graph, characterized in that, Including: Obtain the running knowledge graph data in the enterprise management system, where the running knowledge graph data includes to-be-done task timeliness data, unprocessed notification content, and monitoring rule trigger status; Obtain the task backlog time according to the to-be-done task timeliness data, and obtain the backlog task rate according to the task backlog time; Obtain the process exception link information and node association information according to the unprocessed notification content, and obtain the node exception rate according to the process exception link information and the node association information; Obtain the monitoring timeliness information and notification timeliness information according to the monitoring rule trigger status, and obtain the exception delay rate according to the monitoring timeliness information and the notification timeliness information; Obtain the comprehensive exception rate according to the backlog task rate, the node exception rate, and the exception delay rate; Warn about the exceptions in the enterprise management system according to the comprehensive exception rate.

2. The production anomaly warning method based on a knowledge graph according to claim 1, wherein The step of obtaining the task backlog time according to the to-be-done task timeliness data and obtaining the backlog task rate according to the task backlog time includes: Obtain the planned completion time and multiple single-task actual completion times according to the to-be-done task timeliness data, and obtain multiple single-task backlog times according to the planned completion time and the multiple single-task actual completion times; Judge in turn whether the multiple single-task backlog times exceed a preset threshold; If it exceeds the preset threshold, cumulatively calculate the multiple single-task backlog times that exceed the preset threshold to obtain the cumulative backlog total duration; Obtain the corresponding number of tasks according to the multiple single-task actual completion times, and calculate the standard completion total time according to the number of tasks and the planned completion time; Obtain the backlog task rate according to the standard completion total time and the cumulative backlog total duration.

3. The production anomaly warning method based on a knowledge graph according to claim 1, wherein The step of obtaining the node exception rate according to the process exception link information and the node association information includes: Obtain multiple blocking nodes according to the node association information, and obtain multiple previous passing times of the corresponding blocking nodes according to the multiple blocking nodes; And obtain the current passing occupancy ratio according to the multiple previous passing times and the preset historical node passing, and use the current passing occupancy ratio as the node blocking deviation; Obtain the stay task identifier according to the process exception link information, and obtain the task path identifier according to the stay task identifier; Obtain the historical communication average number and the current communication number of the corresponding task identifier according to the task path identifier, calculate the node communication frequency difference according to the historical node communication average number, the current communication number, and the preset communication time, and use the node communication frequency difference as the task dispersion coefficient; Obtain the node exception rate according to the node blocking deviation and the task dispersion coefficient.

4. The production anomaly warning method based on a knowledge graph according to claim 1, wherein The step of obtaining the exception delay rate according to the monitoring timeliness information and the notification timeliness information includes: Obtain the monitoring average delay time according to the monitoring timeliness information; Obtain the monitoring delay ratio by obtaining the historical average delay time and the monitoring average delay time within a preset time; Obtain the notification response time according to the notification timeliness information, where the notification response time includes the lag response time and the early response time, and obtain the response average value according to the lag response time and the early response time; Obtain the historical average response time within a preset time, and obtain the notification delay ratio according to the historical average response time and the response average value; Perform weighted calculation according to the monitoring delay ratio and the notification delay ratio to obtain the abnormal delay rate.

5. The production anomaly warning method based on a knowledge graph according to claim 1, wherein The step of obtaining the comprehensive abnormality rate according to the backlog task rate, the node abnormality rate and the abnormal delay rate includes: Obtain the corresponding backlog task rate weight value according to the backlog task rate; Obtain the corresponding node abnormality rate weight value according to the node abnormality rate; Calculate the comprehensive abnormality rate according to the backlog task rate, the node abnormality rate, the abnormal delay rate, the backlog task rate weight value and the node abnormality rate weight value.

6. The production anomaly warning method based on a knowledge graph according to claim 1, wherein The step of warning about the abnormality in the enterprise management system according to the comprehensive abnormality rate includes: Obtain the historical abnormality rate in the history of the enterprise management system, and calculate the abnormality rate occupancy ratio according to the comprehensive abnormality rate and the historical abnormality rate; Determine whether the abnormality rate occupancy ratio is greater than the preset abnormality rate occupancy ratio; If the comprehensive abnormality rate is greater than the preset abnormality rate occupancy ratio, then determine the abnormal state in the current enterprise management system, and warn about the abnormality in the enterprise management system.

7. A production anomaly warning system based on a knowledge graph, characterized in that, including: The first acquisition module is used to acquire the operation knowledge graph data in the enterprise management system, where the operation knowledge graph data includes to-be-done task timeliness data, unprocessed notification content, and monitoring rule trigger status; The second acquisition module is used to obtain the task backlog time according to the to-be-done task timeliness data, and obtain the backlog task rate according to the task backlog time; The third acquisition module is used to obtain the process abnormal link information and node association information according to the unprocessed notification content, and obtain the node abnormality rate according to the process abnormal link information and the node association information; The fourth acquisition module is used to obtain the monitoring timeliness information and notification timeliness information according to the monitoring rule trigger status, and obtain the abnormal delay rate according to the monitoring timeliness information and the notification timeliness information; The fifth acquisition module is used to obtain the comprehensive abnormality rate according to the backlog task rate, the node abnormality rate and the abnormal delay rate; The warning module is used to warn about the abnormality in the enterprise management system according to the comprehensive abnormality rate.

8. The production anomaly warning system based on a knowledge graph according to claim 7, characterized in that The second acquisition module includes: The first acquisition unit is used to obtain the planned completion time and multiple single-task actual completion times according to the to-be-done task timeliness data, and obtain multiple single-task backlog times according to the planned completion time and the multiple single-task actual completion times; The judgment unit is used to sequentially judge whether the multiple single-task backlog times exceed a preset threshold; If it exceeds the preset threshold, then cumulatively calculate the multiple single-task backlog times that exceed the preset threshold to obtain the cumulative backlog total duration; The second acquisition unit is used to obtain the corresponding number of tasks according to the multiple single-task actual completion times, and calculate the standard completion total time according to the number of tasks and the planned completion time; A third obtaining unit, configured to obtain a backlog task rate according to the total time of the standard and the total accumulated backlog duration.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images