Methods, devices, equipment and media for analyzing the execution status of planning schemes

Abstract

This invention discloses a method, apparatus, device, and medium for analyzing the execution status of a planning scheme, applicable to intelligent agent devices. The method includes: layer-by-layer decomposition of a target planning scheme to obtain multiple target sub-tasks; generation of corresponding execution path graphs based on the multiple target sub-tasks, and acquisition of real-time task data corresponding to the multiple target sub-tasks; determination of the execution status of the multiple target sub-tasks based on the execution path graphs and real-time task data; and generation of a visual monitoring view based on the execution status, mapping the visual monitoring view to a preset display area. This method constructs a multi-dimensional task association network through layer-by-layer deconstruction of the target planning scheme, enabling dynamic tracking of the progress, bottlenecks, and resource consumption of each sub-task. Mapping the execution status of each target sub-task to a preset display area achieves real-time visual monitoring of the complex planning execution process, facilitating real-time feedback on whether there are execution deviations or delays in the target sub-tasks.

Description

Technical Field

[0001] This invention relates to the field of intelligent agent technology, and in particular to a method, apparatus, device and medium for analyzing the execution status of a planning scheme. Background Technology

[0002] In current corporate management practices, strategic plans (such as annual strategic goals) formulated by senior management face significant challenges when being broken down and implemented at the department, team, and individual levels. Traditional decomposition methods (such as using Excel or basic project management tools) heavily rely on the personal experience of managers and manual transmission, which is cumbersome and prone to distortion and misrepresentation of target information during transmission, resulting in significant deviations between strategy and execution.

[0003] Meanwhile, tracking the progress of decomposed objectives generally relies on manual reporting by employees and regular meeting presentations. This approach leads to significant delays in data updates, preventing management from obtaining a real-time, comprehensive view of the company's strategic execution. When deviations occur, timely warnings are not available, making it difficult to quickly pinpoint the root cause and take corrective measures, thus missing optimal decision-making opportunities and risking the failure of strategic objectives. Therefore, there is an urgent need for solutions that can intelligently decompose planning schemes, dynamically track them, and display their status. Summary of the Invention

[0004] This invention provides a method, apparatus, device, and medium for analyzing the execution status of planning schemes, in order to solve the technical problems in related technologies such as reliance on manual decomposition of planning schemes, delayed perception of execution status, and difficulty in locating problems.

[0005] In a first aspect, the present invention provides a method for analyzing the execution status of a planning scheme, applied to an intelligent agent device, comprising: The target planning scheme is broken down layer by layer to obtain multiple target sub-tasks; Based on multiple target sub-tasks, generate corresponding execution path graphs and obtain real-time task data corresponding to multiple target sub-tasks; Based on the execution path graph and real-time task data, the execution status of multiple target sub-tasks is determined; Based on the execution status, a visual monitoring view is generated and mapped to a preset display area.

[0006] In some embodiments, the target planning scheme is broken down layer by layer to obtain multiple target sub-tasks, including: Obtain the historical sub-task breakdown results of the historical planning scheme, where the historical planning scheme and the target planning scheme have the same scheme type; Based on preset splitting rules, the target planning scheme is split into multiple levels of granularity to generate multiple candidate subtasks; Based on a pre-defined natural language model, semantic extraction is performed on multiple candidate subtasks to obtain semantic labels for multiple candidate subtasks. Based on semantic tags and historical subtask splitting results, candidate subtasks are split or merged to obtain the target subtask.

[0007] In some embodiments, based on multiple target subtasks, a corresponding execution path graph is generated, including: Based on the relationships and priorities of the target subtasks, the directed edges and path nodes of the execution path graph are determined. The directed edges are used to indicate the execution order of the target subtasks, and the path nodes are used to indicate the target subtasks. An execution path graph is generated based on directed edges and path nodes.

[0008] In some embodiments, the execution status of multiple target sub-tasks is determined based on the execution path graph and real-time task data, including: Based on real-time task data, determine the execution progress and resource consumption parameters of multiple target sub-tasks; Based on the execution path map and execution progress, determine the route deviation parameters for multiple target sub-tasks; Based on execution progress and resource consumption parameters, determine the abnormal resource parameters of multiple target subtasks; Based on resource anomaly parameters and route deviation parameters, the execution status of multiple target sub-tasks is determined.

[0009] In some embodiments, after generating a visual monitoring view based on the execution state and mapping the visual monitoring view to a preset display area, the method further includes: When the execution status meets the abnormal warning conditions, the cause of the abnormality is determined based on the execution path graph and the execution status; Based on the cause of the anomaly, a warning notification and an anomaly solution are sent to the corresponding user.

[0010] In some embodiments, determining the cause of an anomaly based on the execution path graph and execution status includes: Based on the execution status, identify the set of abnormal tasks among multiple target subtasks; Based on the abnormal task set and execution path graph, the abnormal starting subtask and the abnormal related subtasks adjacent to the abnormal starting subtask are determined. The abnormal starting subtask is the target subtask that first appears abnormal in the abnormal task set. Perform data analysis on the execution status of the abnormal initiating subtask and the abnormal associated subtask to determine the cause of the abnormality.

[0011] In some embodiments, based on the cause of the anomaly, a warning notification and anomaly solution are sent to the corresponding user, including: Based on the cause of the anomaly, determine the corresponding warning level and scope of impact; Based on the warning level and the scope of impact, a warning notification is generated, and the corresponding abnormal solution is matched from the preset solution library; Send early warning notifications and anomaly solutions to the relevant users.

[0012] Secondly, this disclosure provides a planning scheme execution status analysis device, comprising: The hierarchical decomposition module is used to decompose the target planning scheme layer by layer to obtain multiple target sub-tasks; The path determination module is used to generate corresponding execution path maps based on multiple target subtasks and to obtain real-time task data corresponding to multiple target subtasks. The status analysis module is used to determine the execution status of multiple target subtasks based on the execution path graph and real-time task data. The visualization module is used to generate visual monitoring views based on the execution status and map the visual monitoring views to a preset display area.

[0013] Thirdly, the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned planning scheme execution state analysis method.

[0014] Fourthly, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described planning scheme execution state analysis method.

[0015] The aforementioned planning scheme execution status analysis method, device, equipment, and medium can be applied to intelligent agent devices, including: decomposing the target planning scheme layer by layer to obtain multiple target sub-tasks; generating corresponding execution path graphs based on the multiple target sub-tasks and acquiring real-time task data corresponding to the multiple target sub-tasks; determining the execution status of the multiple target sub-tasks based on the execution path graphs and real-time task data; generating a visual monitoring view based on the execution status and mapping the visual monitoring view to a preset display area. This method, through layer-by-layer deconstruction of the target planning scheme, constructs a multi-dimensional task association network, enabling dynamic tracking of the progress, bottlenecks, and resource consumption of each sub-task, and mapping the execution status of each target sub-task to a preset display area. This achieves real-time visual monitoring of the complex planning execution process, helping to provide real-time feedback on whether there are execution deviations or delays in the target sub-tasks. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart of a planning scheme execution status analysis method according to an embodiment of the present invention; Figure 2 This is a flowchart of a planning scheme execution status analysis method according to an embodiment of the present invention; Figure 3 This is a flowchart of a planning scheme execution status analysis method according to an embodiment of the present invention; Figure 4 This is a flowchart of a planning scheme execution status analysis method according to an embodiment of the present invention; Figure 5 This is a flowchart of a planning scheme execution status analysis method according to an embodiment of the present invention; Figure 6 This is a flowchart of a planning scheme execution status analysis method according to an embodiment of the present invention; Figure 7 This is a schematic block diagram of a planning scheme execution status analysis device according to an embodiment of the present invention; Figure 8 This is a schematic block diagram of a computer device according to an embodiment of the present invention. Detailed Implementation

[0018] As an example, such as Figure 1 As shown, a method for analyzing the execution status of a planning scheme is provided, which can be applied to intelligent agent devices, and includes the following steps: S101, the target planning scheme is broken down layer by layer to obtain multiple target sub-tasks; S102, Based on multiple target sub-tasks, generate corresponding execution path graphs and obtain real-time task data corresponding to multiple target sub-tasks; S103, Based on the execution path graph and real-time task data, determine the execution status of multiple target sub-tasks; S104. Based on the execution status, generate a visual monitoring view and map the visual monitoring view to a preset display area.

[0019] As an example, in step S101, the intelligent agent device can perform semantic parsing of the target planning scheme through a natural language model, identify key nodes and hierarchical relationships, and thus decompose the target planning scheme layer by layer.

[0020] Optionally, the intelligent agent device can also analyze unstructured documents such as the company's annual report, departmental job descriptions, and project statements corresponding to the target planning scheme to generate the process steps and key nodes of the target planning scheme, thereby automatically decomposing the target planning scheme into multiple logically related target sub-tasks.

[0021] In one embodiment, the target planning scheme can be broken down layer by layer according to the enterprise's organizational structure and functional division of labor. For example, the target planning scheme can be decomposed into each business unit, functional department and specific project team in sequence. Alternatively, it can be broken down according to the stage characteristics of the business process, decomposing the target planning scheme into stages such as project initiation, execution, monitoring and closing, and further refining the actionable target sub-tasks within each stage.

[0022] The target planning scheme can be a corporate annual strategic plan, a major project implementation plan, or a work plan of various departments, etc., with a clear target orientation. This disclosure does not limit the specific form of the target planning scheme.

[0023] As an example, in step S102, the intelligent agent device can construct an execution path graph containing nodes and directed edges based on the dependencies and execution logic between multiple target sub-tasks. Nodes represent target sub-tasks, and directed edges represent the order of tasks and the direction of data flow. At the same time, the intelligent agent device can collect execution data of each target sub-task in real time, such as task progress, resource input, and key performance indicators, and determine this execution data as the aforementioned real-time task data for subsequent dynamic evaluation of execution status.

[0024] In other words, the execution path map can be used to indicate the logical connections and execution sequences between various target sub-tasks, and can be combined with real-time task data to dynamically reflect the actual progress of the planning scheme.

[0025] Among them, intelligent agents can obtain real-time task data of each target sub-task by calling the enterprise's internal project management tools and enterprise resource planning (ERP) system interfaces, ensuring the accuracy and timeliness of the data source.

[0026] As an example, in step S103, the intelligent agent device can determine the resource consumption and task completion status of each target sub-task based on the execution path graph and real-time task data, thereby determining whether each sub-task has resource overruns. It can also combine the dependencies in the path graph to determine whether each target sub-task has a progress lag, thereby determining the judgment result as the execution status of each target sub-task.

[0027] Optionally, when a target subtask has resource overruns or schedule delays, the intelligent agent device can determine that the target subtask is in an abnormal state and reflect the abnormal state in the execution status of the target subtask to issue a warning to the relevant responsible persons.

[0028] As an example, in step S104, the intelligent agent device maps the execution status of each target sub-task to a visual monitoring view and dynamically displays it in a preset display area, thereby realizing real-time monitoring and intuitive presentation of the entire planning scheme, which makes it easier for management to grasp the overall progress and identify potential risks in a timely manner.

[0029] The visualization monitoring view adopts a multi-dimensional dynamic dashboard format, which supports displaying data according to organizational level, project stage, key indicators, and other dimensions. It can also intuitively present the execution status of each target sub-task through color coding, progress bars, and warning icons. Furthermore, the graphical type of the visualization monitoring view can be one or more combinations of Gantt charts, tree diagrams, or dashboards to meet the viewing needs of users at different levels regarding the execution status of targets.

[0030] In summary, this disclosure proposes a method for analyzing the execution status of a planning scheme, applicable to intelligent agent devices. The method includes: layer-by-layer decomposition of the target planning scheme to obtain multiple target sub-tasks; generation of corresponding execution path graphs based on these sub-tasks, and acquisition of real-time task data for each sub-task; determination of the execution status of the sub-tasks based on the execution path graphs and real-time task data; and generation of a visual monitoring view based on the execution status, mapping the visual monitoring view to a preset display area. This method, through layer-by-layer deconstruction of the target planning scheme, constructs a multi-dimensional task association network, enabling dynamic tracking of the progress, bottlenecks, and resource consumption of each sub-task. Mapping the execution status of each sub-task to a preset display area achieves real-time visual monitoring of the complex planning execution process, facilitating real-time feedback on whether there are execution deviations or delays in the target sub-tasks.

[0031] like Figure 2 As shown, step S101 involves breaking down the target planning scheme layer by layer to obtain multiple target sub-tasks; including: S201, Obtain the historical sub-task breakdown results of historical planning schemes; S202, based on preset splitting rules, performs multi-level granular splitting of the target planning scheme to generate multiple candidate sub-tasks; S203, based on a preset natural language model, performs semantic extraction on multiple candidate sub-tasks to obtain semantic labels for multiple candidate sub-tasks; S204: Based on semantic tags and historical subtask splitting results, candidate subtasks are split or merged to obtain the target subtask.

[0032] As an example, in step S201, the intelligent agent device can retrieve historical planning schemes of the same type as the target planning scheme from the database and extract its historical subtask decomposition results as a reference template to ensure that the current decomposition logic is consistent with the historical decomposition mode, thereby improving the rationality and operability of the target decomposition.

[0033] The type of historical planning scheme can be identified by matching tags or keywords. The historical planning scheme should be the same type as the target planning scheme to ensure that the referenced historical sub-task decomposition results are comparable and applicable.

[0034] For example, if the target planning scheme is a "new product marketing plan", then the matching historical planning scheme should also belong to the marketing category, and its historical sub-task breakdown results may include structured nodes such as "brand promotion", "channel expansion" and "user conversion".

[0035] As an example, in step S202, the preset splitting rules can be set based on task complexity, execution cycle or organizational function dimension, etc., to refine the target planning scheme into operable candidate sub-tasks from top to bottom, ensuring that the granularity of each level of task is adapted to the actual execution scenario.

[0036] For example, taking the target planning scheme as "business growth strategy for next year" as an example, it can be broken down into three different levels of granularity according to the preset splitting rules: department-level tasks, project team-level tasks, and individual execution items. This generates candidate sub-tasks such as "marketing department revenue target achievement", "R&D department new product delivery", and "sales team customer signing volume", ensuring that the strategic intent is transmitted and implemented at each level.

[0037] As an example, in step S203, the intelligent agent device can perform semantic analysis on the text and image descriptions of candidate subtasks according to a preset natural language model, extract their semantic features, and generate corresponding semantic tags. These semantic tags can be used to indicate the business attributes, functional scope, and execution domain of the candidate subtasks, such as "market promotion," "product development," and "customer management," to support the subsequent determination of the target subtask.

[0038] The preset natural language model can be a pre-trained deep semantic understanding network, which combines the context to encode multi-dimensional features of candidate sub-tasks, thereby improving the accuracy and generalization ability of semantic label generation. The preset natural language model is, for example, the BERT model, the RoBERTa model, etc., and this disclosure does not limit it.

[0039] As an example, in step S204, the intelligent agent device can perform alignment analysis between semantic tags and historical subtask splitting results, identify reusable splitting structures by matching similar semantic patterns, and determine whether further splitting or merging is needed based on the actual context of the candidate subtasks.

[0040] Specifically, the intelligent agent device can calculate the similarity between the semantic tags of each candidate subtask and the similar tags in the historical subtask splitting results. If the semantic matching degree exceeds the preset semantic similarity threshold, the corresponding historical splitting method is used to obtain the target subtask. If the semantic matching degree does not exceed the preset semantic similarity threshold, the candidate subtasks can be dynamically split or merged based on the semantic tags and contextual information to generate a target subtask that fits the current planning scheme.

[0041] For example, if a candidate subtask has the semantic tag "channel expansion" and there is a refined structure of the same tag in the historical subtask splitting results, it can be further split by referring to its lower-level nodes such as "online channel construction" and "offline channel expansion", and the splitting result is determined as the target subtask; if multiple candidate subtasks have highly similar semantic tags and overlapping execution content, such as "user profile construction" and "target audience analysis", they can be merged into the same target subtask to avoid duplicate execution.

[0042] In one embodiment, such as Figure 3 As shown, step S102, which generates a corresponding execution path graph based on multiple target sub-tasks, includes: S301, Based on the association between target subtasks and task priorities, determine the directed edges and path nodes of the execution path graph; S302 generates an execution path graph based on directed edges and path nodes.

[0043] As an example, in step S301, the intelligent agent device can analyze the relationships between each target subtask (such as logical dependencies and resource call order), and determine the execution order of each target subtask by combining the task priority of each target subtask, thereby constructing directed edges to represent the dependency flow between tasks, and binding path nodes with the corresponding target subtasks to form a structured execution path graph.

[0044] Task priorities can be dynamically evaluated and determined using historical execution data or a preset rule base to ensure that target subtasks on the critical path are given priority scheduling.

[0045] For example, taking a target subtask comprising subtask A, subtask B, and subtask C, if the output of subtask A is used as the input of subtask B, then a directed edge from node A to node B is constructed in the execution path graph to represent their execution dependency. If the output of subtask A is also used as the input of subtask C, and the task priority of subtask C is higher than that of subtask B, then a directed edge from node A to node C is also constructed in the execution path graph. Priority weights are also marked on the directed edge between subtask C and subtask A to ensure that higher-priority tasks are processed first during scheduling. In this way, the execution path graph not only reflects the logical dependencies between tasks but also integrates priority information, enabling dynamic resource allocation and process optimization, and improving overall execution efficiency.

[0046] As an example, in step S302, the intelligent agent device can graphically model the generated directed edges and path nodes to construct an execution path graph that includes node-related attributes and the priority weights of directed edges.

[0047] In one embodiment, such as Figure 4 As shown, step S103, which involves determining the execution status of multiple target sub-tasks based on the execution path graph and real-time task data, includes: S401, based on real-time task data, determines the execution progress and resource consumption parameters of multiple target sub-tasks; S402, Based on the execution path map and execution progress, determine the route deviation parameters of multiple target sub-tasks; S403, Based on execution progress and resource consumption parameters, determine the abnormal resource parameters of multiple target subtasks; S404 determines the execution status of multiple target subtasks based on resource anomaly parameters and route deviation parameters.

[0048] As an example, in step S401, the intelligent agent device can collect parameters such as the completion rate, time consumption, and resource usage of the target sub-tasks in real time from each task execution end, thereby determining the execution progress and resource consumption parameters of each sub-task and providing data support for determining the execution status.

[0049] As an example, in step S402, the intelligent agent device can calculate the degree of deviation between the actual execution path and the expected theoretical path, i.e., the route deviation parameter, based on the order of directed edges in the execution path graph and the actual execution progress of each target subtask, in order to measure whether there is a disorder in the order or a delay in the critical path during the task execution process.

[0050] When the route deviation parameter exceeds the preset threshold, it indicates that there is a critical path delay or task dependency conflict. When the route deviation parameter does not exceed the preset threshold, it indicates that the task execution order conforms to the expected path planning.

[0051] For example, based on the execution path graph, it is determined that after the expected theoretical execution path is completed, subtasks C and B should be triggered in sequence. However, in actual execution, subtask B starts before subtask C. The system then determines that there is an execution order abnormality and identifies this abnormality as a route deviation parameter.

[0052] As an example, in step S403, the intelligent agent device can estimate the resource requirements required for each target sub-task to complete the remaining work at the current progress based on the execution progress and resource consumption parameters of each target sub-task, compare it with the preset resource quota, calculate the resource consumption deviation value, and then determine the abnormal resource parameters.

[0053] Specifically, when the resource consumption deviation exceeds the preset fluctuation range, it is determined that the target subtask has an abnormal situation of resource overload or resource idleness; when the resource consumption deviation is within the preset fluctuation range, it is determined that the resource usage of the target subtask is in a normal state.

[0054] For example, if the resource quota for subtask A is 100 units of man-hours, and the actual man-hours consumed when the current execution progress reaches 60%, then it is estimated that it will take about 54 more units of man-hours to complete the remaining 40% of the work, which exceeds the remaining quota by 14 units of man-hours. At this time, it can be determined that subtask A has a risk of resource overload, and the estimated overload of 14 units of man-hours is identified as an abnormal resource parameter.

[0055] As an example, in step S404, the intelligent agent device can combine resource anomaly parameters and route deviation parameters to determine the execution status of multiple target sub-tasks. In other words, the intelligent agent device can integrate resource anomaly parameters and route deviation parameters, and determine the integrated result as the comprehensive execution status of each target sub-task, which is used to determine whether there are any anomalies in each target sub-task.

[0056] In one embodiment, such as Figure 5 As shown, after step S104, which generates a visual monitoring view based on the execution state and maps the visual monitoring view to a preset display area, the process further includes: S501, When the execution status meets the abnormal warning conditions, determine the cause of the abnormality based on the execution path graph and the execution status; S502 sends an alert notification and an error solution to the corresponding user based on the cause of the error.

[0057] As an example, in step S501, the intelligent agent device can identify the abnormal target subtask based on the execution status and include it in the abnormal task set. Combining the dependencies in the execution path graph, it can determine the pre-tasks and subsequent tasks associated with the abnormal target subtask. Based on the execution status and execution order of the pre-tasks, subsequent tasks and the abnormal target subtask, it can analyze whether there are logical conflicts or resource allocation contradictions, and then determine the cause of the abnormality.

[0058] When the agent determines whether the execution status meets the abnormal warning conditions, it can determine whether the route deviation parameter or resource abnormal parameter in the execution status exceeds the normal range based on the preset threshold. When any parameter exceeds the preset threshold range, it is determined that the execution status of the target subtask meets the abnormal warning conditions.

[0059] For example, if the startup of target subtask A is delayed due to the delay of the preceding task, the intelligent agent device can determine that the abnormality is caused by the obstruction of progress transmission; if target subtask B is caused by the deviation of resource consumption exceeding the threshold, the abnormality is determined to be caused by the imbalance of resource allocation.

[0060] As an example, in step S502, the intelligent agent device can match the response strategy in the preset solution library according to the type of abnormal cause, generate an early warning notification including the warning level, the scope of impact, and the corresponding abnormal solution, and push it to the relevant responsible user through the corresponding channel.

[0061] For example, when an intelligent agent device detects that the progress of target subtask A is stalled due to the incomplete upstream dependencies, it automatically associates the execution status and expected completion time of its predecessor tasks, assesses the degree of impact, generates a warning notification of the corresponding level, and recommends adjusting resource scheduling or launching an emergency parallel solution.

[0062] Furthermore, the intelligent agent device can determine the corresponding warning level and scope of impact based on the cause of the anomaly; generate a warning notification based on the warning level and scope of impact, and match the corresponding anomaly solution from the preset solution library; and send the warning notification and anomaly solution to the corresponding user.

[0063] Specifically, based on a preset mapping table of anomaly causes and warning levels, the warning level corresponding to different anomaly types can be determined. For example, progress transmission obstruction corresponds to a high warning level, and resource allocation imbalance corresponds to a medium warning level. Furthermore, by combining the number of target sub-tasks of the anomaly with the correlation of the critical path, the breadth and depth of the impact range can be determined, thereby generating differentiated warning notifications.

[0064] For example, when an intelligent agent device identifies an anomaly as a delay in progress transmission, it can automatically match a dynamic scheduling plan, adjust the time window and resource quota of the relevant target tasks, and generate a high-priority early warning notification. This early warning notification and the dynamic scheduling plan are then pushed to the user terminals of the project management and execution layers. When an imbalance in resource allocation is identified, a resource reallocation protocol is triggered to optimize the resource allocation scheme of the current target sub-tasks and generate a medium-priority early warning notification. This early warning notification and resource reallocation suggestions are then pushed to the resource management department and relevant personnel.

[0065] In one embodiment, such as Figure 6 As shown, step S501, which involves determining the cause of the anomaly based on the execution path graph and execution status, includes: S601, Based on the execution status, determine the set of abnormal tasks among multiple target subtasks; S602, Based on the abnormal task set and execution path graph, determine the abnormal starting subtask and the abnormal associated subtasks adjacent to the abnormal starting subtask; S603 performs data analysis on the execution status of the abnormal initiating subtask and the abnormal associated subtask to determine the cause of the abnormality.

[0066] As an example, in step S601, the intelligent agent device can identify the abnormal target subtasks among multiple target subtasks by using the route deviation parameter and resource anomaly parameter in the execution state, and classify them into the abnormal task set.

[0067] Specifically, if the deviation of the target subtask from the route exceeds a preset threshold, or if its resource consumption rate deviates significantly from the plan, then the target subtask is determined to be an abnormal target subtask.

[0068] As an example, in step S602, the intelligent agent device locates the abnormal starting subtask and its adjacent abnormal associated subtasks based on the directed edge relationship between the abnormal task set and the execution path graph, wherein the abnormal starting subtask is the target subtask that first appears abnormal in the abnormal task set.

[0069] It should be understood that by locating the subtask that initiates the anomaly, the source of the anomaly can be traced, and the impact on related subtasks can be analyzed by combining the execution path dependency relationship. This allows for the accurate identification of the propagation path and scope of impact of the anomaly subtask, providing a basis for subsequent anomaly attribution and intervention strategies.

[0070] As an example, in step S603, the intelligent agent device can perform multi-dimensional analysis on the execution status data of the abnormal initiation subtask and the abnormal associated subtask, and identify the abnormal cause by combining parameters such as task progress, resource consumption, dependency relationship and environmental variables.

[0071] For example, when the progress of an abnormally initiated subtask is lagging while its associated abnormal subtasks have been completed, the cause of the anomaly can be determined to be insufficient execution efficiency of the abnormally initiated subtask itself, rather than external dependencies or resource limitations. Optionally, in this case, the intelligent agent device can automatically call historical data and compare the time difference of tasks of the same type as the abnormally initiated subtask in the historical data to further verify the cause of the anomaly.

[0072] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0073] In one embodiment, a planning scheme execution status analysis device is provided, which corresponds one-to-one with the planning scheme execution status analysis method described in the above embodiments. For example... Figure 7 As shown, the planning scheme execution status analysis device includes a hierarchical decomposition module 701, a path determination module 702, a status analysis module 703, and a visualization module 704. Detailed descriptions of each functional module are as follows: The hierarchical decomposition module 701 is used to decompose the target planning scheme layer by layer to obtain multiple target sub-tasks; The path determination module 702 is used to generate a corresponding execution path map based on multiple target sub-tasks and to obtain real-time task data corresponding to multiple target sub-tasks. The status analysis module 703 is used to determine the execution status of multiple target subtasks based on the execution path graph and real-time task data. The visualization module 704 is used to generate a visual monitoring view based on the execution status and map the visual monitoring view to a preset display area.

[0074] In one embodiment, the hierarchical splitting module 701 is further configured to obtain the historical subtask splitting results of the historical planning scheme, wherein the historical planning scheme and the target planning scheme have the same scheme type. Based on preset splitting rules, the target planning scheme is split into multiple levels of granularity to generate multiple candidate subtasks; Based on a pre-defined natural language model, semantic extraction is performed on multiple candidate subtasks to obtain semantic labels for multiple candidate subtasks. Based on semantic tags and historical subtask splitting results, candidate subtasks are split or merged to obtain the target subtask.

[0075] In one embodiment, the path determination module 702 is further configured to determine the directed edges and path nodes of the execution path graph based on the association relationship and task priority of the target sub-tasks, wherein the directed edges are used to indicate the execution order of the target sub-tasks and the path nodes are used to indicate the target sub-tasks. An execution path graph is generated based on directed edges and path nodes.

[0076] In one embodiment, the status analysis module 703 is further configured to determine the execution progress and resource consumption parameters of multiple target sub-tasks based on real-time task data. Based on the execution path map and execution progress, determine the route deviation parameters for multiple target sub-tasks; Based on execution progress and resource consumption parameters, determine the abnormal resource parameters of multiple target subtasks; Based on resource anomaly parameters and route deviation parameters, the execution status of multiple target sub-tasks is determined.

[0077] In one embodiment, the state analysis module 703 is further configured to determine the cause of the anomaly based on the execution path graph and the execution state when the execution state meets the anomaly warning conditions; Based on the cause of the anomaly, a warning notification and an anomaly solution are sent to the corresponding user.

[0078] In one embodiment, the state analysis module 703 is further configured to determine a set of abnormal tasks among multiple target subtasks based on the execution state; Based on the set of abnormal tasks and the execution path graph, the abnormal initiating subtask and the abnormal associated subtasks adjacent to the abnormal initiating subtask are determined. Perform data analysis on the execution status of the abnormal initiating subtask and the abnormal associated subtask to determine the cause of the abnormality.

[0079] In one embodiment, the status analysis module 703 is further configured to determine the corresponding warning level and scope of impact based on the cause of the anomaly; Based on the warning level and the scope of impact, a warning notification is generated, and the corresponding abnormal solution is matched from the preset solution library; Send early warning notifications and anomaly solutions to the relevant users.

[0080] This invention provides a planning scheme execution status analysis device, comprising: a hierarchical decomposition module for decomposing the target planning scheme layer by layer to obtain multiple target sub-tasks; a path determination module for generating corresponding execution path maps based on the multiple target sub-tasks and acquiring real-time task data corresponding to the multiple target sub-tasks; a status analysis module for determining the execution status of the multiple target sub-tasks based on the execution path maps and real-time task data; and a visualization module for generating a visual monitoring view based on the execution status and mapping the visual monitoring view to a preset display area. This device constructs a multi-dimensional task association network by deconstructing the target planning scheme layer by layer, enabling dynamic tracking of the progress, bottlenecks, and resource consumption of each sub-task, and mapping the execution status of each target sub-task to a preset display area. This achieves real-time visual monitoring of the complex planning execution process and helps to provide real-time feedback on whether there are execution deviations or delays in the target sub-tasks.

[0081] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 8 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database is used for data employed in the planning scheme execution state analysis method. The network interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program can implement a planning scheme execution state analysis method.

[0082] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a planning scheme execution state analysis method.

[0083] In one embodiment, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program that, when executed by a processor, implements a method for analyzing the execution status of a planning scheme.

[0084] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), IAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0085] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0086] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for analyzing the execution status of a planning scheme, characterized in that, Applied to intelligent agent devices, including: The target planning scheme is broken down layer by layer to obtain multiple target sub-tasks; Based on the multiple target sub-tasks, a corresponding execution path graph is generated, and real-time task data corresponding to the multiple target sub-tasks is obtained; Based on the execution path graph and the real-time task data, the execution status of the multiple target sub-tasks is determined; Based on the execution status, a visual monitoring view is generated and mapped to a preset display area.

2. The method according to claim 1, characterized in that, The goal planning scheme is broken down layer by layer to obtain multiple goal sub-tasks, including: Obtain the historical sub-task breakdown results of the historical planning scheme, wherein the historical planning scheme and the target planning scheme have the same scheme type; Based on preset splitting rules, the target planning scheme is split into multiple levels of granularity to generate multiple candidate subtasks; Based on a preset natural language model, semantic extraction is performed on the multiple candidate subtasks to obtain semantic labels for the multiple candidate subtasks; Based on the semantic tags and the historical subtask splitting results, the candidate subtasks are split or merged to obtain the target subtask.

3. The method according to claim 1, characterized in that, The step of generating a corresponding execution path graph based on the multiple target sub-tasks includes: Based on the association and priority of the target sub-tasks, the directed edges and path nodes of the execution path graph are determined. The directed edges are used to indicate the execution order of the target sub-tasks, and the path nodes are used to indicate the target sub-tasks. The execution path graph is generated based on the directed edges and the path nodes.

4. The method according to claim 1, characterized in that, Determining the execution status of the multiple target sub-tasks based on the execution path graph and the real-time task data includes: Based on the real-time task data, the execution progress and resource consumption parameters of the multiple target sub-tasks are determined; Based on the execution path map and the execution progress, the route deviation parameters of the multiple target sub-tasks are determined; Based on the execution progress and resource consumption parameters, determine the abnormal resource parameters of the multiple target subtasks; Based on the resource anomaly parameters and the route deviation parameters, the execution status of the multiple target sub-tasks is determined.

5. The method according to claim 1, characterized in that, After generating a visual monitoring view based on the execution state and mapping the visual monitoring view to a preset display area, the method further includes: When the execution status meets the abnormal warning conditions, the cause of the abnormality is determined based on the execution path graph and the execution status; Based on the stated cause of the anomaly, a warning notification and an anomaly solution will be sent to the corresponding user.

6. The method according to claim 5, characterized in that, The process of determining the cause of the anomaly based on the execution path graph and the execution status includes: Based on the execution status, determine the set of abnormal tasks among the plurality of target subtasks; Based on the set of abnormal tasks and the execution path graph, the abnormal starting subtask and the abnormal associated subtask adjacent to the abnormal starting subtask are determined. The abnormal starting subtask is the target subtask that first appears abnormal in the set of abnormal tasks. Data analysis is performed on the execution status of the abnormal initiating subtask and the abnormal associated subtask to determine the cause of the abnormality.

7. The method according to claim 5, characterized in that, Based on the cause of the anomaly, sending a warning notification and anomaly solution to the corresponding user includes: Based on the aforementioned causes of the anomaly, the corresponding warning level and scope of impact are determined; Based on the warning level and the scope of impact, the warning notification is generated, and the corresponding abnormal solution is matched from the preset solution library; The warning notification and the anomaly solution will be sent to the corresponding user.

8. A planning scheme execution status analysis device, characterized in that, include: The hierarchical decomposition module is used to decompose the target planning scheme layer by layer to obtain multiple target sub-tasks; The path determination module is used to generate a corresponding execution path map based on the multiple target sub-tasks, and to obtain real-time task data corresponding to the multiple target sub-tasks; The status analysis module is used to determine the execution status of the multiple target sub-tasks based on the execution path graph and the real-time task data. The visualization module is used to generate a visual monitoring view based on the execution status and map the visual monitoring view to a preset display area.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the planning scheme execution status analysis method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the planning scheme execution status analysis method as described in any one of claims 1 to 7.

Images