Production progress real-time monitoring method fusing multi-source sensing data
By using real-time data acquisition from multi-source sensors and conflict analysis of the production progress recognition framework, and dynamically updating weight parameters, the shortcomings of data acquisition and status determination in production progress monitoring are resolved, achieving high-precision closed-loop control throughout the entire process and improving the accuracy and adaptability of production progress monitoring.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Filing Date
- 2026-06-02
- Publication Date
- 2026-07-14
AI Technical Summary
Existing production progress monitoring technologies suffer from shortcomings in data acquisition, data conflicts and redundancy from multiple sensors, and a lack of effective preprocessing. Their status determination is inaccurate, making it impossible to achieve high-precision closed-loop control of the entire process. This results in the progress deviation calculation results deviating from the actual production situation.
A multi-source sensor fusion adaptive acquisition frequency dynamic adjustment strategy is adopted to acquire production progress data in real time. By constructing a production progress identification framework, evidence conflict analysis is performed, weight parameters are dynamically updated, and a hierarchical early warning mechanism is built to achieve accurate fusion and status determination of multi-source sensor data.
It significantly improves the accuracy and precision of production progress status determination, provides reliable data support, provides a scientific basis for production scheduling adjustments, ensures that weight allocation is in line with actual production conditions, and reduces misjudgments and omissions.
Smart Images
Figure CN122390427A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of industrial production monitoring technology, and more specifically to a method for real-time monitoring of production progress by integrating multi-source sensor data. Background Technology
[0002] Real-time production progress monitoring is the core foundation for modern manufacturing enterprises to achieve lean production and intelligent decision-making. In the past, production progress monitoring mainly relied on traditional manual statistics and simple equipment monitoring, which was a "post-event traceability" model. Although some enterprises introduced early data acquisition and monitoring systems to achieve simple centralized monitoring, the data acquisition frequency was low, the processing capacity was limited, and it was impossible to achieve data linkage across multiple stages. The monitoring scope was limited to a single piece of equipment or a local process. Under this model, the transmission of production progress information was delayed. Managers could only discover problems after significant delays occurred, and they could not quickly trace the root cause of the delays, resulting in a high order delivery delay rate and making it difficult to adapt to the needs of large-scale production.
[0003] With the rapid development of the Industrial Internet of Things (IIoT) and sensor technology, production progress monitoring has entered the initial intelligent stage of "data-driven" monitoring, realizing the real-time collection of basic data such as equipment, materials, and environment. Some leading enterprises have achieved progress visualization through industrial internet platforms, promoting the transformation of monitoring mode from "post-event traceability" to "in-process control," which has solved the problem of low efficiency in traditional monitoring to a certain extent and helped enterprises improve their production collaboration capabilities.
[0004] However, current production progress monitoring technologies still have significant shortcomings, making it difficult to meet the actual needs of high-precision, full-process, and closed-loop management. First, data acquisition has weaknesses; data from multiple sensors is prone to conflicts and redundancy, and the lack of effective preprocessing and synchronization mechanisms leads to insufficient data reliability. Second, status judgment and deviation quantification are not precise enough; most solutions do not fully consider the problem of multi-parameter conflicts, and the weights are set fixed and cannot be dynamically adjusted according to actual working conditions, easily leading to misjudgments or omissions, causing the progress deviation calculation results to deviate from the actual production situation. Summary of the Invention
[0005] To overcome the aforementioned deficiencies of the prior art, the present invention provides a method for real-time monitoring of production progress by integrating multi-source sensor data, thereby solving the problems existing in the background art.
[0006] This invention provides the following technical solution: a method for real-time monitoring of production progress by integrating multi-source sensor data, comprising the following steps: Step S01: Real-time acquisition of production progress data based on multi-source sensors: Real-time acquisition of production progress data is carried out using a multi-source sensor fusion adaptive acquisition frequency dynamic adjustment strategy, and a data preprocessing strategy is embedded synchronously during the acquisition process; Step S02: Determine the status of production progress data and obtain independent judgment evidence: Determine the status of equipment-type sensor data collected in step S01, obtain independent judgment evidence when the equipment is operating normally, and analyze the production progress deviation rate at the same time. Step S03: Construct a production progress identification framework and perform conflict analysis on the independent judgment evidence: Construct a production progress identification framework and perform conflict analysis on the independent judgment evidence obtained in step S02. Step S04: Update the weight parameters based on the results of the conflict analysis and output the updated production schedule deviation rate: Update the weight parameters based on the analysis results of the evidence body in step S03 and output the updated production schedule deviation rate. Step S05: Construct a hierarchical early warning mechanism to complete anomaly judgment and early warning: Based on the updated production progress deviation rate output in step S04, construct a hierarchical early warning mechanism and push it in a hierarchical manner.
[0007] Preferably, in step S01, the specific content of acquiring production progress data in real time based on multi-source sensors is as follows: The production progress data is collected in real time using a multi-source sensor fusion adaptive acquisition frequency dynamic adjustment strategy. The multi-source sensors adopt a layered deployment mode of core sensing and auxiliary sensing. The core sensing layer deploys three types of core sensors: equipment sensing, material sensing, and environmental sensing; the auxiliary sensing layer deploys vision sensing and personnel on-duty sensing. During the data acquisition process, a data preprocessing strategy is embedded synchronously. This strategy includes real-time noise removal, missing data completion, and time-series synchronization calibration.
[0008] Preferably, in step S02, the specific content of determining the status of the production progress data and obtaining independent determination evidence is as follows: Obtain the device-type sensor data collected by the device sensor in step S01, and analyze whether the device-type sensor data meets the preset data threshold range: when the device-type sensor data meets the preset data threshold range, it is determined that the device is operating normally; when the device-type sensor data does not meet the preset data threshold range, it is determined that the device is operating abnormally. When equipment malfunctions, the abnormal information is sent to the integrated early warning and correction mechanism; when equipment is operating normally, the progress deviation rate is analyzed based on the actual completed amount and the planned completion rate, and independent judgment evidence is obtained. The independent evidence includes environmental evidence and material evidence. Based on environmental sensing data, a process adaptability threshold is introduced, and differentiated environmental compliance standards are set to classify environmental status into three categories: environmental compliance, slight exceedance, and severe exceedance. Based on material sensing data, RFID positioning trajectory and material retention time are integrated to further classify material status into four states: smooth material flow, material accumulation, material shortage, and work-in-process retention.
[0009] Preferably, in step S03, the specific content of constructing the production progress identification framework to perform conflict analysis on the independently determined evidence is as follows: Construct a production progress identification framework, defined as: {Normal Production} Process delay abnormal material stagnation Insufficient environmental adaptability }; The corresponding content is: Normal production Equipment is processing normally, materials are flowing smoothly, and environmental parameters meet standards; process delays are not observed. The equipment is processing normally, but material flow is slow or environmental parameters are slightly out of standard; material is abnormally stagnant. The equipment is processing normally, but there is an accumulation or shortage of materials, or there is stagnation of in-process goods; the environment is not well adapted. The equipment was operating normally, but environmental parameters were severely exceeding the standards. Define the basic probability assignment function for environmental evidence as follows: The basic probability allocation function of the material evidence body is: ; Calculate the conflict coefficient of environmental evidence and material evidence based on any subset of the production progress identification framework. Based on the accuracy requirements of real-time production monitoring, a conflict threshold is set, and the conflict determination logic is as follows: when the conflict coefficient is less than the preset threshold, it is determined that there is no conflict; when the conflict coefficient is greater than or equal to the preset threshold, it is determined that there is a conflict. Output the conflict coefficients and corresponding conflict determination conclusions for the two types of evidence: environment and materials. Retain the basic probability allocation function assignments for the two types of evidence and the schedule deviation rate initially calculated in step S02, as the core inputs for updating the weight parameters and optimizing the deviation rate in step S04.
[0010] Preferably, in step S04, the specific content of updating the weight parameters based on the conflict analysis results and outputting the updated production schedule deviation rate is as follows: Weight initialization and update rules: Set initial weights for environmental parameters and material parameters. Maintain the initial weights when there are no conflicts, and automatically adjust the weights when there are conflicts. Automatic weight adjustment: Obtain the credibility scores of two types of parameters, calculate the update coefficients by combining the conflict coefficients, and obtain the updated weights based on the initial weights, credibility scores, and update coefficients; Deviation rate update and output: Based on the preliminary deviation rate output in step S02, and combined with the updated weights, the deviations caused by environmental and material parameters are weighted and corrected to obtain the updated production schedule deviation rate.
[0011] Preferably, in step S05, the specific content of constructing the integrated early warning and correction mechanism to complete the anomaly judgment and early warning is as follows: Anomaly detection and early warning classification based on deviation rate: A three-level early warning threshold is set based on the updated production schedule deviation rate to construct a graded early warning mechanism. Level 1 warning: The updated deviation rate is greater than or equal to the Level 1 warning threshold and less than the Level 2 warning threshold; Level 2 warning: The updated deviation rate is greater than or equal to the Level 2 warning threshold and less than the Level 3 warning threshold; Level 3 warning: The updated deviation rate is greater than or equal to the Level 3 warning threshold; Multi-terminal tiered push notifications: Level 1 warning: Pushed to workstation operators to remind them to pay attention to material flow and changes in environmental parameters, and to promptly investigate minor abnormalities; Level 2 warning: Pushed to operators and workshop dispatchers, along with an analysis of the root cause of the anomaly and preliminary handling suggestions. The dispatcher will coordinate and adjust the process rhythm, optimize material allocation or environmental parameters. Level 3 early warning: The warning is simultaneously pushed to operators, dispatchers and management to clarify the degree of urgency of the abnormality. Management is responsible for coordinating resources, material replenishment, environmental control or equipment backup plans.
[0012] The technical effects and advantages of this invention are as follows: This invention includes the following steps: S01: real-time acquisition of production progress data based on multi-source sensors; S02: status determination of production progress data and acquisition of independent determination evidence; S03: construction of a production progress identification framework to perform conflict analysis on the independent determination evidence; S04: updating weight parameters based on the conflict analysis results and outputting the updated production progress deviation rate; and S05: construction of a hierarchical early warning mechanism to complete anomaly determination and early warning, which greatly improves the accuracy of production progress anomaly determination and provides reliable data support for production scheduling adjustments. By constructing a clear production progress identification framework and conducting evidence conflict analysis, the pain points of ambiguous progress status determination and inability to effectively handle multi-parameter conflicts in existing technologies are effectively solved, and the accuracy of production progress status determination is greatly improved. By assigning values to the two types of evidence and calculating the conflict coefficient, the contradictions and conflicts between multi-source sensor data can be accurately identified, providing a scientific basis for subsequent weight adjustment. Dynamically updating weight parameters and optimizing production schedule deviation rates based on conflict analysis results further enhances the accuracy and adaptability of schedule monitoring, providing reliable data support for production scheduling adjustments. Combining credibility scores and conflict coefficients to calculate and update weights retains the core role of high-credibility data while reducing interference from low-credibility and conflicting data, ensuring that weight allocation aligns with actual production conditions. Simultaneously, by correcting weights and outputting optimized schedule deviation rates, the traditional deviation rate calculation ignores multi-parameter conflicts and credibility differences, making the deviation quantification results more accurate and better reflect actual production progress. Attached Figure Description
[0013] Figure 1 A flowchart illustrating a method for real-time monitoring of production progress by integrating multi-source sensor data. Detailed Implementation
[0014] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. In addition, the forms of the various structures described in the following embodiments are merely illustrative. The real-time monitoring method for production progress that integrates multi-source sensor data involved in the present invention is not limited to the structures described in the following embodiments. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0015] like Figure 1 As shown, this invention provides a method for real-time monitoring of production progress by fusing multi-source sensor data, including the following steps: Step S01: Real-time acquisition of production progress data based on multi-source sensors: Real-time acquisition of production progress data is carried out using a multi-source sensor fusion adaptive acquisition frequency dynamic adjustment strategy, and a data preprocessing strategy is embedded synchronously during the acquisition process; Step S02: Determine the status of production progress data and obtain independent judgment evidence: Determine the status of equipment-type sensor data collected in step S01, obtain independent judgment evidence when the equipment is operating normally, and analyze the production progress deviation rate at the same time. Step S03: Construct a production progress identification framework and perform conflict analysis on the independent judgment evidence: Construct a production progress identification framework and perform conflict analysis on the independent judgment evidence obtained in step S02. Step S04: Update the weight parameters based on the results of the conflict analysis and output the updated production schedule deviation rate: Update the weight parameters based on the analysis results of the evidence body in step S03 and output the updated production schedule deviation rate. Step S05: Construct a hierarchical early warning mechanism to complete anomaly judgment and early warning: Based on the updated production progress deviation rate output in step S04, construct a hierarchical early warning mechanism and push it in a hierarchical manner.
[0016] In this embodiment, it should be specifically explained that the specific content of acquiring production progress data in real time based on multi-source sensors in step S01 is as follows: A multi-source sensor fusion adaptive acquisition frequency dynamic adjustment strategy is used to collect production progress data in real time. The multi-source sensors adopt a layered deployment mode of core sensors and auxiliary sensors. The core sensor layer deploys three types of core sensors: equipment sensors, material sensors, and environmental sensors, covering three key dimensions: production equipment operation, material flow, and production environment. The auxiliary sensor layer deploys visual sensors and personnel on-duty sensors to supplement and verify the accuracy of the core data and avoid misjudgment based on single sensor data. During the data acquisition process, a pre-processing strategy is embedded synchronously. This strategy includes: real-time noise removal, employing an adaptive threshold denoising algorithm to remove abrupt noise data caused by industrial site interference, thus preventing noise from affecting the accuracy of subsequent judgments; missing data completion, based on the correlation characteristics of time-series data, using adjacent interpolation combined with historical data fitting to quickly complete missing data caused by temporary sensor offline status, ensuring data continuity; and time-series synchronization calibration, using timestamp synchronization technology to unify the time reference of all multi-source sensors, eliminating data misalignment caused by time differences in different sensor acquisition, ensuring the time-series consistency of equipment, environment, and material sensor data, and providing a unified data foundation for subsequent independent state judgments.
[0017] In this embodiment, it should be specifically explained that the specific content of step S02, which involves determining the status of production progress data and obtaining independent determination evidence, is as follows: Obtain the device-type sensor data collected by the device sensor in step S01, and analyze whether the device-type sensor data meets the preset data threshold range: when the device-type sensor data meets the preset data threshold range, it is determined that the device is operating normally; when the device-type sensor data does not meet the preset data threshold range, it is determined that the device is operating abnormally. When equipment malfunctions, the abnormal information is sent to the integrated early warning and correction mechanism; when equipment is operating normally, the progress deviation rate is analyzed based on the actual and planned completion amounts. ,in It represents the difference between the actual completed amount and the planned completed amount, and the ratio of the planned completed amount, while obtaining independent evidence for judgment; The independent evidence includes environmental evidence and material evidence. Based on environmental sensing data, a process adaptability threshold is introduced, and differentiated environmental compliance standards are set to classify environmental status into three categories: environmental compliance, slight exceedance, and severe exceedance. Based on material sensing data, RFID positioning trajectory and material retention time are integrated to further classify material status into four states: smooth material flow, material accumulation, material shortage, and work-in-process retention.
[0018] In this embodiment, it should be specifically explained that the specific content of constructing the production progress identification framework to perform conflict analysis on the independent judgment evidence in step S03 is as follows: Construct a production progress identification framework, defined as: {Normal Production} Process delay abnormal material stagnation Insufficient environmental adaptability }; The content corresponding to each subset in the production progress identification framework is: normal production. Equipment is processing normally, materials are flowing smoothly, and environmental parameters meet standards; process delays are not observed. The equipment is processing normally, but material flow is slow or environmental parameters are slightly out of standard; material is abnormally stagnant. The equipment is processing normally, but there is an accumulation or shortage of materials, or there is stagnation of in-process goods; the environment is not well adapted. The equipment was operating normally, but environmental parameters were severely exceeding the standards. The criteria for judging smooth material flow, slow material flow, material accumulation, shortage, or stagnation of work-in-process are as follows: Smooth material flow: Flow duration Standard processing time, no backlog, shortage, or delays; counting and processing rhythm matched; slow material flow: flow time exceeded, rhythm lagging behind equipment processing; no backlog or shortages; material backlog: workstation inventory. 120% preset standard, stay Preset time threshold; Material shortage: material input <95% of the preset planned value, or inventory below the safety threshold; Work-in-process inventory retention: duration of inventory dwell time. 150% of standard processing time, no flow signal; Among them, environmental parameters are matched with the preset environmental category classification range, and the environmental parameters are classified into environmental parameters that meet the standards, environmental parameters that slightly exceed the standards, and environmental parameters that seriously exceed the standards according to the environmental category classification range. Define the basic probability assignment function for environmental evidence as follows: The basic probability allocation function of the material evidence body is: The basic probability assignment function satisfies two core constraints: first, the probability of the empty set is 0; second, the sum of the probabilities of all states and uncertainties is 1. Calculate the conflict coefficient for environmental and material evidence based on any subset of the production progress identification framework. Where A and B are arbitrary subsets in the production progress identification framework, and The value of k ranges from [0, 1]. k=0 indicates that the judgment results of the two types of evidence are completely consistent, and k=1 indicates that the judgment results of the two types of evidence are completely conflicting. The larger the value of k, the more obvious the discrepancy between the two types of parameters in judging the progress status. Based on the accuracy requirements of real-time production monitoring, a conflict threshold is set. Based on the conflict judgment logic: when the conflict coefficient is less than the preset threshold, it is judged as no conflict, indicating that the two types of parameters, environment and material, have a high degree of consistency in judging the production progress status and the difference is small; when the conflict coefficient is greater than or equal to the preset threshold, it is judged as conflict, indicating that there is a serious difference in the judgment of the two types of parameters, and the impact of the difference needs to be weakened through subsequent weight adjustment. Output the conflict coefficients and corresponding conflict determination conclusions for the two types of evidence: environment and materials. Retain the basic probability allocation function assignments for the two types of evidence and the schedule deviation rate initially calculated in step S02, as the core inputs for updating the weight parameters and optimizing the deviation rate in step S04.
[0019] In this embodiment, it should be specifically noted that in step S04, the details of updating the weight parameters based on the conflict analysis results and outputting the updated production schedule deviation rate are as follows: Weight initialization and update rules: Set initial weights for environmental parameters Initial weights of material parameters When there is no conflict, the initial weight is maintained; when there is a conflict, the weight is automatically adjusted. The specific steps for automatically adjusting weights are as follows: The reliability scores for two types of parameters are obtained based on sensor accuracy, with environmental reliability being... Material reliability is ; The update coefficient is calculated by combining the conflict coefficient. The update coefficient is: ,in Indicates the update coefficients. Indicates the conflict coefficient; The updated weights are obtained based on the initial weights, credibility score, and update coefficient. The updated weights are as follows: , ,in Indicates the weights after the environmental parameters are updated. Indicates the weight after the material parameters are updated; Deviation rate update and output: Based on the preliminary deviation rate output in step S02, and combined with the updated weights, the deviations caused by environmental and material parameters are weighted and corrected to obtain the updated production schedule deviation rate. ,in This indicates the updated production schedule deviation rate. This indicates the schedule deviation rate.
[0020] In this embodiment, it should be specifically explained that the specific content of constructing the integrated early warning and correction mechanism to complete the anomaly judgment and early warning in step S05 is as follows: Anomaly detection and early warning classification based on deviation rate: A three-level early warning threshold is set based on the updated production schedule deviation rate to construct a graded early warning mechanism. Level 1 warning: The updated deviation rate is greater than or equal to the Level 1 warning threshold and less than the Level 2 warning threshold; Level 2 warning: The updated deviation rate is greater than or equal to the Level 2 warning threshold and less than the Level 3 warning threshold; Level 3 warning: The updated deviation rate is greater than or equal to the Level 3 warning threshold; Multi-terminal tiered push notifications: Level 1 warning: Pushed to workstation operators to remind them to pay attention to material flow and changes in environmental parameters, and to promptly investigate minor abnormalities; Level 2 warning: Pushed to operators and workshop dispatchers, along with an analysis of the root cause of the anomaly and preliminary handling suggestions. The dispatcher will coordinate and adjust the process rhythm, optimize material allocation or environmental parameters. Level 3 early warning: The warning is simultaneously pushed to operators, dispatchers and management to clarify the degree of urgency of the abnormality. Management is responsible for coordinating resources, material replenishment, environmental control or equipment backup plans to minimize downtime.
[0021] The main difference between this implementation and the prior art is that this embodiment includes the following steps: S01: real-time acquisition of production progress data based on multi-source sensors; S02: status determination of production progress data and acquisition of independent determination evidence; S03: construction of a production progress identification framework to perform conflict analysis on the independent determination evidence; S04: updating weight parameters based on the results of conflict analysis and outputting the updated production progress deviation rate; and S05: construction of a hierarchical early warning mechanism to complete the anomaly determination and early warning, which greatly improves the accuracy of production progress anomaly determination and provides reliable data support for production scheduling adjustments. By constructing a clear production progress identification framework and conducting evidence conflict analysis, the pain points of ambiguous progress status determination and inability to effectively handle multi-parameter conflicts in existing technologies are effectively solved, and the accuracy of production progress status determination is greatly improved. By assigning values to the two types of evidence and calculating the conflict coefficient, the contradictions and conflicts between multi-source sensor data can be accurately identified, providing a scientific basis for subsequent weight adjustment. Dynamically updating weight parameters and optimizing production schedule deviation rates based on conflict analysis results further enhances the accuracy and adaptability of schedule monitoring, providing reliable data support for production scheduling adjustments. Combining credibility scores and conflict coefficients to calculate and update weights retains the core role of high-credibility data while reducing interference from low-credibility and conflicting data, ensuring that weight allocation aligns with actual production conditions. Simultaneously, by correcting weights and outputting optimized schedule deviation rates, the traditional deviation rate calculation ignores multi-parameter conflicts and credibility differences, making the deviation quantification results more accurate and better reflect actual production progress.
[0022] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
[0023] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
Claims
1. A method for real-time monitoring of production progress by integrating multi-source sensor data, characterized in that: Includes the following steps: Step S01: Real-time acquisition of production progress data based on multi-source sensors: Real-time acquisition of production progress data is carried out using a multi-source sensor fusion adaptive acquisition frequency dynamic adjustment strategy, and a data preprocessing strategy is embedded synchronously during the acquisition process; Step S02: Determine the status of production progress data and obtain independent judgment evidence: Determine the status of equipment-type sensor data collected in step S01, obtain independent judgment evidence when the equipment is operating normally, and analyze the production progress deviation rate at the same time. Step S03: Construct a production progress identification framework and perform conflict analysis on the independent judgment evidence: Construct a production progress identification framework and perform conflict analysis on the independent judgment evidence obtained in step S02. Step S04: Update the weight parameters based on the results of the conflict analysis and output the updated production schedule deviation rate: Update the weight parameters based on the analysis results of the evidence body in step S03 and output the updated production schedule deviation rate. Step S05: Construct a hierarchical early warning mechanism to complete anomaly judgment and early warning: Based on the updated production progress deviation rate output in step S04, construct a hierarchical early warning mechanism and push it in a hierarchical manner.
2. The method for real-time monitoring of production progress by fusing multi-source sensor data according to claim 1, characterized in that: In step S01, the specific details of acquiring production progress data in real time based on multi-source sensors are as follows: The production progress data is collected in real time using a multi-source sensor fusion adaptive acquisition frequency dynamic adjustment strategy. The multi-source sensors adopt a layered deployment mode of core sensing and auxiliary sensing. The core sensing layer deploys three types of core sensors: equipment sensing, material sensing, and environmental sensing. The auxiliary sensing layer deploys visual sensors and personnel presence sensors; During the data acquisition process, a data preprocessing strategy is embedded synchronously. This strategy includes real-time noise removal, missing data completion, and time-series synchronization calibration.
3. The method for real-time monitoring of production progress by fusing multi-source sensor data according to claim 1, characterized in that: In step S02, the specific content of determining the status of production progress data and obtaining independent determination evidence is as follows: Obtain the device-type sensor data collected by the device sensor in step S01, and analyze whether the device-type sensor data meets the preset data threshold range: when the device-type sensor data meets the preset data threshold range, it is determined that the device is operating normally; when the device-type sensor data does not meet the preset data threshold range, it is determined that the device is operating abnormally. When an abnormality is detected in the equipment, the abnormality information is sent to the integrated early warning and correction mechanism; When it is determined that the equipment is operating normally, the progress deviation rate is analyzed based on the actual completed amount and the planned completion rate, and independent evidence is obtained at the same time. The independent evidence includes environmental evidence and material evidence. Based on environmental sensing data, a process adaptability threshold is introduced, and differentiated environmental compliance standards are set to classify environmental status into three categories: environmental compliance, slight exceedance, and severe exceedance. Based on material sensing data, RFID positioning trajectory and material retention time are integrated to further classify material status into four states: smooth material flow, material accumulation, material shortage, and work-in-process retention.
4. The method for real-time monitoring of production progress by fusing multi-source sensor data according to claim 1, characterized in that: In step S03, the specific content of constructing the production progress identification framework to perform conflict analysis on independently determined evidence is as follows: Construct a production progress identification framework, defined as: {Normal Production} Process delay abnormal material stagnation Insufficient environmental adaptability }; The corresponding content is: Normal production Equipment is processing normally, materials are flowing smoothly, and environmental parameters meet standards; process delays are not observed. The equipment is processing normally, but material flow is slow or environmental parameters are slightly out of standard; material is abnormally stagnant. The equipment is processing normally, but there is an accumulation or shortage of materials, or there is stagnation of in-process goods; the environment is not well adapted. The equipment was operating normally, but environmental parameters were severely exceeding the standards. Define the basic probability assignment function for environmental evidence as follows: The basic probability allocation function of the material evidence body is: ; Calculate the conflict coefficient of environmental evidence and material evidence based on any subset of the production progress identification framework. Based on the accuracy requirements of real-time production monitoring, a conflict threshold is set, and the conflict determination logic is: when the conflict coefficient is less than the preset threshold, it is determined to be without conflict; When the conflict coefficient is greater than or equal to the preset threshold, it is determined that there is a conflict; Output the conflict coefficients and corresponding conflict determination conclusions for the two types of evidence: environment and materials. Retain the basic probability allocation function assignments for the two types of evidence and the schedule deviation rate initially calculated in step S02, as the core inputs for updating the weight parameters and optimizing the deviation rate in step S04.
5. The method for real-time monitoring of production progress by fusing multi-source sensor data according to claim 1, characterized in that: In step S04, the specific details of updating the weight parameters based on the conflict analysis results and outputting the updated production schedule deviation rate are as follows: Weight initialization and update rules: Set initial weights for environmental parameters and material parameters. Maintain the initial weights when there are no conflicts, and automatically adjust the weights when there are conflicts. Automatic weight adjustment: Obtain the credibility scores of two types of parameters, calculate the update coefficients by combining the conflict coefficients, and obtain the updated weights based on the initial weights, credibility scores, and update coefficients; Deviation rate update and output: Based on the preliminary deviation rate output in step S02, and combined with the updated weights, the deviations caused by environmental and material parameters are weighted and corrected to obtain the updated production schedule deviation rate.
6. The method for real-time monitoring of production progress by fusing multi-source sensor data according to claim 1, characterized in that: In step S05, the specific details of constructing the integrated early warning and correction mechanism to complete the anomaly judgment and early warning are as follows: Anomaly detection and early warning classification based on deviation rate: A three-level early warning threshold is set based on the updated production schedule deviation rate to construct a graded early warning mechanism. Level 1 warning: The updated deviation rate is greater than or equal to the Level 1 warning threshold and less than the Level 2 warning threshold; Level 2 warning: The updated deviation rate is greater than or equal to the Level 2 warning threshold and less than the Level 3 warning threshold; Level 3 warning: The updated deviation rate is greater than or equal to the Level 3 warning threshold; Multi-terminal tiered push notifications: Level 1 warning: Pushed to workstation operators to remind them to pay attention to material flow and changes in environmental parameters, and to promptly investigate minor abnormalities; Level 2 warning: Pushed to operators and workshop dispatchers, along with an analysis of the root cause of the anomaly and preliminary handling suggestions. The dispatcher will coordinate and adjust the process rhythm, optimize material allocation or environmental parameters. Level 3 early warning: The warning is simultaneously pushed to operators, dispatchers and management to clarify the degree of urgency of the abnormality. Management is responsible for coordinating resources, material replenishment, environmental control or equipment backup plans.