Production line visual management platform based on dynamic task data driving

Abstract

The invention discloses a production line visual management platform based on dynamic task data driving, and relates to the technical field of intelligent manufacturing and industrial informatization, and the platform comprises a dynamic task data collection module which is used for collecting the operation state data of production line stations, equipment, personnel and materials in real time; the task evolution modeling module is used for constructing a task evolution process according to the collected data; the task granularity self-adaptive refining module is used for dynamically adjusting the task management granularity according to the task state change; the abnormal fluctuation trend detection module is used for detecting abnormal fluctuation in the task state data; the visual management module is used for generating a production line visual interface based on the task evolution and the anomaly detection result; and the scheduling decision linkage module is used for optimizing a production line scheduling strategy according to the anomaly detection result and the task state. The problems of static data updating, task granularity fixing and anomaly response lagging existing in an existing production line visual management platform are solved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing and industrial information technology, and specifically to a production line visualization management platform driven by dynamic task data. Background Art

[0002] As the manufacturing industry evolves toward high flexibility, small batches, and a wide variety of products, the complexity of production line management is rapidly increasing. To improve capacity utilization and production response speed, more and more production lines are introducing automated monitoring and visual management technologies, hoping to optimize operational processes through real-time data collection, dynamic modeling, and intelligent scheduling.

[0003] Existing production line visualization management systems generally use fixed-cycle data updates or near-real-time data push mechanisms, which are unable to adjust data collection and update strategies in real time based on changes in task status. In highly dynamic production line environments, frequent fluctuations in task status can easily lead to data consistency issues and lack accurate status traceability support after anomalies occur.

[0004] In addition, traditional task management models usually divide task units at a fixed granularity (such as by work order, process, etc.), and are unable to dynamically refine or coarsen tasks according to actual production changes, resulting in insufficient granularity for abnormal location and increased response delays.

[0005] Therefore, it is necessary to design a dynamic task data-driven production line visualization management platform that supports dynamic task evolution modeling and granularity adaptive adjustment. Summary of the Invention

[0006] The purpose of the present invention is to provide a production line visualization management platform driven by dynamic task data to solve the problems raised in the above background technology.

[0007] In order to solve the above technical problems, the present invention provides the following technical solution: a production line visualization management platform driven by dynamic task data, comprising a dynamic task data acquisition module, a task evolution modeling module, a task granularity adaptive refinement module, an abnormal fluctuation trend detection module, a visualization management module, and a scheduling decision linkage module, wherein the dynamic task data acquisition module, the task evolution modeling module, the task granularity adaptive refinement module, the abnormal fluctuation trend detection module, the visualization management module, and the scheduling decision linkage module are interconnected and communicated with each other, wherein:

[0008] Dynamic task data collection module, used to collect real-time operating status data of production line workstations, equipment, personnel and materials;

[0009] The task evolution modeling module is used to construct the task evolution process based on the collected data;

[0010] Task granularity adaptive refinement module, used to dynamically adjust task management granularity according to task status changes;

[0011] Abnormal fluctuation trend detection module, used to detect abnormal fluctuations in task status data;

[0012] Visual management module, used to generate a production line visualization interface based on task evolution and anomaly detection results;

[0013] The scheduling decision linkage module is used to optimize the production line scheduling strategy based on anomaly detection results and task status.

[0014] According to the above technical solution, the dynamic task data acquisition module includes a group of dynamic acquisition units distributed at each operation node of the production line. The dynamic acquisition unit can evaluate the node status in real time based on the node's current task load, task execution time and operation type;

[0015] The evaluation indicators of node status include load factor, response delay and abnormal frequency. The sampling frequency F is determined in real time according to the following dynamic sampling frequency adjustment formula: F = F0 × (1 + α·L + β·D + γ·E), where F0 is the initial frequency, L is the node load ratio, D is the delay ratio, E is the abnormality rate, and α, β, and γ are the corresponding empirical adjustment coefficients.

[0016] And when the node state suddenly changes, it switches to high-frequency sampling mode to refine the data change trajectory, wherein the node state suddenly changes when the state indicator fluctuation exceeds the set threshold δ.

[0017] According to the above technical solution, the dynamic sampling frequency adjustment process further includes:

[0018] Based on the data change rate ΔV in the continuous time window T, the node activity A is calculated, and its calculation expression is:

[0019] For nodes whose activity A is higher than the activity threshold θ1, the data sampling density is increased to form a hotspot sampling area;

[0020] For nodes whose activity A is lower than the silent threshold θ2, the sampling frequency is reduced to save system bandwidth and computing resources;

[0021] The hotspot sampling area is further refined to form a secondary sampling level. The calculation expression of the secondary sampling frequency F2 is: F2 = F×(1+λ·(A-θ1)), where λ is the hotspot sensitivity coefficient.

[0022] According to the above technical solution, the method further includes dynamically generating task node priorities according to node activity, and the task node priorities are specifically:

[0023] As activity A increases, priority P increases according to P = P0-k·(A-θ1), where P0 is the default priority and k is the activity impact factor;

[0024] Nodes with high priority will receive higher processing priority in subsequent task scheduling, exception response, and visualization presentation;

[0025] When the node activity continues to exceed the set upper limit θ3, the node abnormality warning mark is triggered, providing a basis for subsequent scheduling and evolution modeling.

[0026] According to the above technical solution, the specific workflow of the task evolution modeling module is as follows:

[0027] Based on the data flow relationship between job nodes, a dynamic directed acyclic graph (DAG) model is constructed;

[0028] Each job node acts as a DAG node, and its node attributes include node ID, load, latency, exception rate, and activity;

[0029] Edge relationships represent data and task dependencies, and edge weights are determined by the node output data packet size and latency.

[0030] During dynamic monitoring of task processes, DAG node and edge attributes are updated in real time, and the visualization layout is dynamically adjusted based on changing trends.

[0031] Based on the DAG structure, confidence calculation is further introduced to quantitatively evaluate the stability of node output. The calculation expression of confidence C is: Where σ is the standard deviation of node output fluctuation, μ' is the expected output stability level, and μ is the scaling factor;

[0032] Low-confidence nodes trigger early warnings and are highlighted in the visual interface, significantly improving the efficiency of anomaly location.

[0033] According to the above technical solution, the method further includes a lagging node identification method for low-confidence nodes, specifically:

[0034] In the DAG model, nodes whose node confidence C is continuously lower than the warning threshold θ4 in three consecutive time windows are counted and marked as suspected lagging nodes;

[0035] Combining the upstream input flow and downstream response delay of the lagging node, the hysteresis index H is established: Where N is the number of associated nodes;

[0036] When the H value is higher than the hysteresis determination threshold θ5, the node is confirmed as a hysteresis node;

[0037] The system automatically generates local optimization suggestions for lagging nodes.

[0038] According to the above technical solution, an auxiliary scheduling mechanism for lagging nodes is further included, specifically including:

[0039] Prioritize adjusting the task start times of upstream and downstream nodes of the lagging node, avoiding large-scale global changes through local fine-tuning;

[0040] When local adjustments are ineffective, the task flow rerouting algorithm is triggered to dynamically change some task execution paths to achieve load balancing.

[0041] According to the above technical solution, the visualization management module generates a presentation method for the production line visualization interface, including automatically generating a multi-layer display interface according to the node priority and activity, wherein hot spots are marked with high-brightness flashing, low-activity areas automatically reduce the display granularity, abnormal nodes are marked with red warnings, and support is provided for replaying the node change trajectory along the timeline to facilitate subsequent auditing and analysis.

[0042] According to the above technical solution, the method for presenting the visual interface further includes:

[0043] Automatically extract key nodes in the task chain;

[0044] Dynamically generate animated flow diagrams to display the task evolution process, and train self-learning based on historical data to continuously optimize sampling strategies and modeling parameters to improve system adaptability.

[0045] Compared with the existing technology, the beneficial effect achieved by the present invention is: the production line visualization management platform based on dynamic task data drive proposed in the present invention can effectively overcome the shortcomings of the existing technology in data collection and update delays, fixed task granularity and abnormal response lag, and has the advantages of strong real-time performance, high degree of intelligence, timely abnormality handling and high resource utilization efficiency. It is suitable for complex production line management needs in highly flexible, small batch and multi-variety production environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0047] In the attached figure:

[0048] Figure 1 This is a schematic diagram of the module composition of a production line visualization management platform driven by dynamic task data in the present invention. DETAILED DESCRIPTION

[0049] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0050] See also Figure 1 The present invention provides a technical solution: a production line visualization management platform driven by dynamic task data, including a dynamic task data acquisition module, a task evolution modeling module, a task granularity adaptive refinement module, an abnormal fluctuation trend detection module, a visualization management module, and a scheduling decision linkage module. The dynamic task data acquisition module, the task evolution modeling module, the task granularity adaptive refinement module, the abnormal fluctuation trend detection module, the visualization management module, and the scheduling decision linkage module are interconnected, wherein:

[0051] Dynamic task data collection module, used to collect real-time operating status data of production line workstations, equipment, personnel and materials;

[0052] The dynamic task data acquisition module includes a set of dynamic acquisition units distributed at each operation node of the production line. The dynamic acquisition unit can evaluate the node status in real time based on the node's current task load, task execution time and operation type;

[0053] The evaluation indicators of node status include load factor, response delay and abnormal frequency. The sampling frequency F is determined in real time according to the following dynamic sampling frequency adjustment formula: F = F0 × (1 + α·L + β·D + γ·E), where F0 is the initial frequency, L is the node load ratio, D is the delay ratio, E is the abnormality rate, and α, β, and γ are the corresponding empirical adjustment coefficients.

[0054] When the node state suddenly changes, it switches to high-frequency sampling mode to refine the data change trajectory, thereby reducing system resource consumption while ensuring data integrity. The node state suddenly changes when the state indicator fluctuation exceeds the set threshold δ;

[0055] The dynamic sampling frequency adjustment process further includes:

[0056] Based on the data change rate ΔV in the continuous time window T, the node activity A is calculated, and its calculation expression is:

[0057] For nodes whose activity A is higher than the activity threshold θ1, the data sampling density is increased to form a hotspot sampling area;

[0058] For nodes whose activity A is lower than the silent threshold θ2, the sampling frequency is reduced to save system bandwidth and computing resources;

[0059] The hotspot sampling area is further refined to form a secondary sampling level. The calculation expression for the secondary sampling frequency F2 is: F2 = F × (1 + λ·(A-θ1)), where λ is the hotspot sensitivity coefficient. Through these steps, the data granularity of high-value nodes is improved without significantly increasing the overall data volume. It can also quickly capture abnormal trends when sudden anomalies occur, further realizing the function of early prediction and intervention of sudden production line failures.

[0060] In an embodiment of the present invention, a method for dynamically generating task node priorities according to node activity is further included, wherein the task node priorities are specifically:

[0061] As activity A increases, priority P increases according to P = P0-k·(A-θ1), where P0 is the default priority and k is the activity impact factor;

[0062] Nodes with high priority will receive higher processing priority in subsequent task scheduling, exception response, and visualization presentation;

[0063] When the node activity continues to exceed the set upper limit θ3, the node abnormality warning mark is triggered, providing a basis for subsequent scheduling and evolution modeling.

[0064] The task evolution modeling module is used to construct the task evolution process based on the collected data;

[0065] In an embodiment of the present invention, the specific workflow of the task evolution modeling module is as follows:

[0066] Based on the data flow relationship between job nodes, a dynamic directed acyclic graph (DAG) model is constructed;

[0067] Each job node acts as a DAG node, and its node attributes include node ID, load, latency, exception rate, and activity;

[0068] Edge relationships represent data and task dependencies, and edge weights are determined by the node output data packet size and latency.

[0069] During dynamic monitoring of task processes, DAG node and edge attributes are updated in real time, and the visualization layout is dynamically adjusted based on changing trends.

[0070] Based on the DAG structure, confidence calculation is further introduced to quantitatively evaluate the stability of node output. The calculation expression of confidence C is: Where σ is the standard deviation of node output fluctuation, μ' is the expected output stability level, and μ is the scaling factor;

[0071] Low-confidence nodes trigger early warnings and are highlighted in the visual interface, significantly improving the efficiency of anomaly location.

[0072] It further includes a method for identifying lagging nodes for low-confidence nodes, specifically:

[0073] In the DAG model, nodes whose node confidence C is continuously lower than the warning threshold θ4 in three consecutive time windows are counted and marked as suspected lagging nodes;

[0074] Combining the upstream input flow and downstream response delay of the lagging node, the hysteresis index H is established: Where N is the number of associated nodes. Considering that downstream latency is more indicative of node bottlenecks than upstream latency, H is designed to measure lag.

[0075] When the H value is higher than the hysteresis determination threshold θ5, the node is confirmed as a hysteresis node;

[0076] The system automatically generates local optimization suggestions for lagging nodes, including reallocating task loads and adjusting execution order. This allows for early identification of potential bottlenecks, optimizing the overall production process, and further increasing production line throughput.

[0077] It further includes an auxiliary scheduling mechanism for lagging nodes, specifically including:

[0078] Prioritize adjusting the task start times of upstream and downstream nodes of the lagging node, avoiding large-scale global changes through local fine-tuning;

[0079] When local adjustments are ineffective, the task flow rerouting algorithm is triggered to dynamically change some task execution paths to achieve load balancing;

[0080] This auxiliary scheduling mechanism is linked with the overall scheduling system of the production line, and can smoothly transition abnormal nodes without interrupting the production process to avoid production delays.

[0081] Task granularity adaptive refinement module, used to dynamically adjust task management granularity according to task status changes;

[0082] Abnormal fluctuation trend detection module, used to detect abnormal fluctuations in task status data;

[0083] Visual management module, used to generate a production line visualization interface based on task evolution and anomaly detection results;

[0084] In an embodiment of the present invention, the visualization management module generates a production line visualization interface presentation method, which includes automatically generating a multi-layer display interface based on node priority and activity. Hot spots are highlighted and flashed, low-activity areas are automatically displayed at a reduced granularity, abnormal nodes are marked with red warnings, and the node change trajectory can be replayed along the timeline to facilitate post-audit and analysis.

[0085] The method for presenting the visual interface further includes:

[0086] Automatically extract key nodes in the task chain;

[0087] Dynamically generate animated flow diagrams to display the task evolution process, and train self-learning based on historical data. Based on historical node change data, use simple regression prediction to optimize sampling frequency and hotspot thresholds, continuously optimize sampling strategies and modeling parameters, and improve system adaptability.

[0088] The scheduling decision linkage module is used to optimize the production line scheduling strategy based on anomaly detection results and task status.

[0089] This invention aims to address the problems of static data updates, fixed task granularity, and delayed exception response in existing production line visualization management systems by providing a production line visualization management platform driven by dynamic task data. By introducing dynamic task data collection, task evolution modeling, adaptive task granularity refinement, abnormal fluctuation trend detection, and a scheduling decision-making linkage mechanism, this platform achieves real-time perception of production line operating status, rapid identification of abnormal trends, and intelligent optimization of task management, thereby improving the real-time performance, accuracy, and intelligence of the production line.

[0090] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0091] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0092] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0093] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which are all protected by the present invention.

Claims

1. A production line visualization management platform driven by dynamic task data, characterized by: It includes a dynamic task data acquisition module, a task evolution modeling module, a task granularity adaptive refinement module, an abnormal fluctuation trend detection module, a visualization management module and a scheduling decision linkage module. The dynamic task data acquisition module, the task evolution modeling module, the task granularity adaptive refinement module, the abnormal fluctuation trend detection module, the visualization management module and the scheduling decision linkage module are interconnected, wherein: Dynamic task data collection module, used to collect real-time operating status data of production line workstations, equipment, personnel and materials; The task evolution modeling module is used to construct the task evolution process based on the collected data; Task granularity adaptive refinement module, used to dynamically adjust task management granularity according to task status changes; Abnormal fluctuation trend detection module, used to detect abnormal fluctuations in task status data; Visual management module, used to generate a production line visualization interface based on task evolution and anomaly detection results; The scheduling decision linkage module is used to optimize the production line scheduling strategy based on anomaly detection results and task status.

2. The production line visualization management platform based on dynamic task data drive according to claim 1, characterized in that: The dynamic task data acquisition module includes a set of dynamic acquisition units distributed at each operation node of the production line. The dynamic acquisition unit can evaluate the node status in real time based on the node's current task load, task execution time and operation type; The evaluation indicators of node status include load factor, response delay and abnormal frequency. The sampling frequency F is determined in real time according to the following dynamic sampling frequency adjustment formula: F = F0 × (1 + α·L + β·D + γ·E), where F0 is the initial frequency, L is the node load ratio, D is the delay ratio, E is the abnormality rate, and α, β, and γ are the corresponding empirical adjustment coefficients. And when the node state suddenly changes, it switches to high-frequency sampling mode to refine the data change trajectory, wherein the node state suddenly changes when the state indicator fluctuation exceeds the set threshold δ.

3. The production line visualization management platform based on dynamic task data drive according to claim 2, characterized in that: The dynamic sampling frequency adjustment process further includes: Based on the data change rate ΔV in the continuous time window T, the node activity A is calculated, and its calculation expression is: For nodes whose activity A is higher than the activity threshold θ1, the data sampling density is increased to form a hotspot sampling area; For nodes whose activity A is lower than the silent threshold θ2, the sampling frequency is reduced to save system bandwidth and computing resources; The hotspot sampling area is further refined to form a secondary sampling level. The calculation expression of the secondary sampling frequency F2 is: F2 = F×(1+λ·(A-θ1)), where λ is the hotspot sensitivity coefficient.

4. The production line visualization management platform based on dynamic task data drive according to claim 3, characterized in that: It further includes a method for dynamically generating task node priorities based on node activity, wherein the task node priorities are specifically: As activity A increases, priority P increases according to P = P0-k·(A-θ1), where P0 is the default priority and k is the activity impact factor; Nodes with high priority will receive higher processing priority in subsequent task scheduling, exception response, and visualization presentation; When the node activity continues to exceed the set upper limit θ3, the node abnormality warning mark is triggered, providing a basis for subsequent scheduling and evolution modeling.

5. The production line visualization management platform based on dynamic task data drive according to claim 1, characterized in that: The specific workflow of the task evolution modeling module is as follows: Based on the data flow relationship between job nodes, a dynamic directed acyclic graph (DAG) model is constructed; Each job node acts as a DAG node, and its node attributes include node ID, load, latency, exception rate, and activity; Edge relationships represent data and task dependencies, and edge weights are determined by the node output data packet size and latency. During dynamic monitoring of task processes, DAG node and edge attributes are updated in real time, and the visualization layout is dynamically adjusted based on changing trends. Based on the DAG structure, confidence calculation is further introduced to quantitatively evaluate the stability of node output. The calculation expression of confidence C is: Where σ is the standard deviation of node output fluctuation, μ' is the expected output stability level, and μ is the scaling factor; Low-confidence nodes trigger early warnings and are highlighted in the visual interface, significantly improving the efficiency of anomaly location.

6. The production line visualization management platform based on dynamic task data drive according to claim 5, characterized in that: It further includes a method for identifying lagging nodes for low-confidence nodes, specifically: In the DAG model, nodes whose node confidence C is continuously lower than the warning threshold θ4 in three consecutive time windows are counted and marked as suspected lagging nodes; Combining the upstream input flow and downstream response delay of the lagging node, the hysteresis index H is established: Where N is the number of associated nodes; When the H value is higher than the hysteresis determination threshold θ5, the node is confirmed as a hysteresis node; The system automatically generates local optimization suggestions for lagging nodes.

7. The production line visualization management platform based on dynamic task data drive according to claim 6, characterized in that: It further includes an auxiliary scheduling mechanism for lagging nodes, specifically including: Prioritize adjusting the task start times of upstream and downstream nodes of the lagging node, avoiding large-scale global changes through local fine-tuning; When local adjustments are ineffective, the task flow rerouting algorithm is triggered to dynamically change some task execution paths to achieve load balancing.

8. The production line visualization management platform based on dynamic task data drive according to claim 5, characterized in that: The visualization management module generates a presentation method for a production line visualization interface, including automatically generating a multi-layer display interface according to node priority and activity, wherein hot spots are highlighted and flashed, low-activity areas are automatically reduced in display granularity, abnormal nodes are marked with red warnings, and support is provided for replaying node change trajectories along the timeline to facilitate subsequent auditing and analysis.

9. The production line visualization management platform based on dynamic task data drive according to claim 8, characterized in that: The method for presenting the visual interface further includes: Automatically extract key nodes in the task chain; Dynamically generate animated flow diagrams to display the task evolution process, and train self-learning based on historical data to continuously optimize sampling strategies and modeling parameters to improve system adaptability.

Images