Monitoring system for multi-level deviation accumulation quality risk in complex process
Through the multi-level deviation cumulative risk assessment model, the cross-process deviations in complex processes are monitored and evaluated, and the problem of difficulty in capturing cross-process quality fluctuations is solved in the existing technology, and effective monitoring and early warning of the quality risks of complex processes is achieved.
Patent Information
- Application Number
- CN202411533899.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-31
- Publication Date
- 2024-11-29
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing quality monitoring system is difficult to capture the comprehensive quality fluctuations across processes in complex processes, resulting in difficulty in traceability of quality problems.
Using a multi-level deviation cumulative risk assessment model, through data collection, deviation cumulative calculation, risk assessment, process correlation analysis and early warning modules, the deviation of each process is monitored and its impact on product quality is evaluated, and combined with the linkage effect of different processes. Comprehensive assessment.
It realizes effective monitoring of the accumulated quality risks of multi-level deviations in complex processes, can predict the joint effect of minor deviations in multiple processes on product quality, provides real-time early warning and traceability, and improves the visibility and traceability of quality problems.
Smart Images

Figure SMS_1 
Figure SMS_13 
Figure SMS_16
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of production quality management, and in particular to a monitoring system for cumulative quality risks of multi-level deviations in complex processes. Background Art
[0002] In the process of product quality management, there is a problem: even if the quality of each previous process meets the requirements, the accumulation of small deviations in multiple processes will lead to quality problems in a certain link of the final product. This situation is often very hidden, because most of the existing quality monitoring systems focus on examining the deviation of each process separately through data analysis. However, the current monitoring system has certain limitations in design. They often find it difficult to capture this cross-process, comprehensive quality fluctuation, making it neither intuitive nor difficult to trace the source of quality problems. Summary of the invention
[0003] In view of the deficiencies in the prior art, the present invention provides a monitoring system for the cumulative quality risks of multi-level deviations in complex processes. By establishing a risk assessment model based on multi-level deviation accumulation, it not only focuses on the deviation of each process, but also introduces a global perspective to monitor the deviations across processes layer by layer. By modeling and analyzing the correlation between each process, it is determined at which levels the deviation accumulates and affects the quality. The multi-level model can not only monitor a single process, but also conduct a comprehensive assessment in combination with the linkage effects of different processes, and estimate how small deviations of multiple processes work together to affect product quality.
[0004] In order to achieve the above object, the present invention adopts the following technical solutions: A monitoring system for multi-level deviation cumulative quality risk in complex processes, including the following modules: Data collection module, used to collect quality deviation data of each process; The deviation accumulation calculation module is used to gradually accumulate the deviations of each process and generate the cumulative deviation value of each process; Risk assessment module, used to calculate the overall quality risk index GRI based on the cumulative deviation value; Process correlation analysis module, used to analyze the deviation transmission path and mutual influence between processes; Early warning module, triggers an alarm when the cumulative deviation or quality risk index exceeds the preset threshold.
[0005] Preferably, the cumulative deviation calculation module calculates the cumulative deviation value of each process by the following formula:
[0006] in, For the Cumulative deviation of each process; For the Deviation weight of a process; For process The deviation For process The actual quality index value, For process target value.
[0007] Preferably, the weight of the process The importance of the process to the quality of the final product is determined through historical data analysis or criticality setting of the production process.
[0008] Preferably, the risk assessment module calculates the overall quality risk index GRI by the following formula:
[0009] Where n is the total number of processes, It is the risk function of accumulated deviations, reflecting the impact of accumulated deviations on product quality.
[0010] Preferably, the risk function Take the following form:
[0011] in, It is an adjustment parameter to control the nonlinear growth of risk.
[0012] Preferably, the process correlation analysis module is based on the Pearson correlation coefficient of each process deviation. The calculation formula of Pearson correlation coefficient is:
[0013] Among them, Cov(X,Y) is the covariance of process X and process Y, and are the standard deviations of process X and process Y respectively.
[0014] Preferably, the process correlation analysis module performs causal analysis through a Bayesian network to evaluate the dependencies between processes, conditional probability The calculation formula is:
[0015] in, For the process Under the condition that the deviation is known, the process The conditional probability of the deviation, For process and process The joint probability of For process The marginal probability of .
[0016] Preferably, when the cumulative deviation Ce or the quality risk index GRI exceeds a preset threshold, the early warning module triggers an alarm and traces the cause of the accumulated process deviation.
[0017] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a monitoring system for the cumulative quality risk of multi-level deviations in complex processes. By establishing a risk assessment model based on the cumulative multi-level deviations, it not only focuses on the deviation of each process, but also introduces a global perspective to monitor the deviations across processes layer by layer. By modeling and analyzing the correlation between each process, it is determined at which levels the deviation accumulates and affects the quality. The multi-level model can not only monitor a single process, but also conduct a comprehensive assessment in combination with the linkage effects of different processes, and estimate how small deviations of multiple processes work together on product quality. DETAILED DESCRIPTION
[0018] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations. The basic principles of the present invention defined in the following description can be used for other embodiments, variations, improvements, equivalents, and other technical solutions that do not deviate from the spirit and scope of the present invention.
[0019] The present invention provides a monitoring system for multi-level deviation cumulative quality risk in complex processes, comprising the following modules: Data collection module, used to collect quality deviation data of each process; The deviation accumulation calculation module is used to gradually accumulate the deviations of each process and generate the cumulative deviation value of each process; Risk assessment module, used to calculate the overall quality risk index GRI based on the cumulative deviation value; Process correlation analysis module, used to analyze the deviation transmission path and mutual influence between processes; Early warning module, triggers an alarm when the cumulative deviation or quality risk index exceeds the preset threshold.
[0020] In the implementation of this application, the data acquisition module needs to collect data for each process in the production process to ensure that key parameter data are obtained, including temperature, pressure, speed, size, etc., which represent the quality characteristics of the process; at this stage, sensors, quality monitoring systems and other equipment will be used for real-time collection.
[0021] The data acquisition module also preprocesses the collected data. The preprocessing steps include: Data cleaning: remove outliers and noise; Standardization: ensure that the deviation data of different processes are at the same level to facilitate subsequent processing; Time alignment: There may be time differences in data across processes. Ensure data alignment to analyze the linkage effects of multiple processes.
[0022] In this application, the deviation accumulation calculation module first performs single process deviation calculation, and then introduces a multi-level deviation accumulation risk assessment model to perform deviation accumulation calculation.
[0023] Specifically, the quality deviation of each process is monitored individually through conventional statistical control charts (such as mean-standard deviation control charts), and then the concept of "deviation accumulation" is introduced to analyze the cumulative effect of these deviations across processes. For single process deviation calculation, single process deviation is defined as:
[0024] in: For process The actual quality index value, For process Target value of It is a process deviation.
[0025] Specifically, this application introduces a multi-level deviation accumulation risk assessment model to monitor the deviations across processes layer by layer. Based on the cumulative effect of deviations, the model divides deviations into different levels and accumulates them step by step. The deviation accumulation calculation formula is:
[0026] in: For the Cumulative deviation of each process; It is a process The deviation It is the weight of each process deviation, which is used to indicate the influence of different processes on the final quality.
[0027] Among them, the weight of the process It can be set through historical data analysis or based on the importance of the process. For example, the deviation weight of the key process is higher, while the weight of the secondary process is lower.
[0028] In the implementation of this application, the risk assessment module monitors the cumulative deviation of each process. , it can be determined whether the deviation has accumulated abnormally at a certain stage, and then the potential risk of the process to product quality can be evaluated. The following formula is used to define the overall risk index (Global Risk Index, GRI): GRI calculation formula:
[0029] Where: n is the total number of processes, It is the risk function of accumulated deviations, reflecting the impact of accumulated deviations on product quality.
[0030] Specifically, the risk function Take the following form:
[0031] in, It is an adjustment parameter to control the nonlinear growth of risk.
[0032] In this application, the process correlation analysis module performs correlation analysis and causal analysis, where: Correlation analysis: Calculate the deviation correlation coefficient between processes and find highly correlated processes; Causal analysis: Use models such as Bayesian networks to build a causal relationship diagram between processes and analyze how certain process deviations affect subsequent processes through cumulative effects.
[0033] Therefore, the process correlation analysis module uses correlation analysis to evaluate the dependency between each process, further analyzes the linkage effect between processes, and determines which process deviations have a significant impact on subsequent processes by constructing a causal network between processes.
[0034] It should be noted that this application uses two methods, correlation analysis and cause-effect analysis, to conduct process correlation analysis. The specific process is as follows: 1. Data preparation: Collect the deviation data of each process and construct a data set in which each column represents the deviation value of a process and the rows correspond to different time points.
[0035] 2. Correlation analysis: Calculate the deviation correlation coefficient between each process, using the Pearson correlation coefficient For analysis, the formula is:
[0036] Among them, Cov(X,Y) is the covariance of variables X and Y, and the variables specifically refer to the deviation value of each process. and are the standard deviations of the variables.
[0037] in,
[0038] in, and are the means of variables X and Y respectively.
[0039] 3. Cause and effect analysis: To construct a Bayesian network to analyze the causal relationship between processes, we first define the nodes (processes) and their interdependencies, and then estimate the conditional probability distribution through maximum likelihood estimation or Bayesian inference.
[0040] For each pair of processes and , establish the conditional probability , indicating that in the process Under the condition of deviation, the process The probability of deviation. Conditional probability The calculation formula is: in, For the process Under the condition that the deviation is known, the process The conditional probability of the deviation, For process and process The joint probability of For process The marginal probability of .
[0041] 4. Model construction uses causal inference methods to build a causal relationship graph and identify direct and indirect effects. Apply algorithms such as DAG (directed acyclic graph) learning to determine the optimal path and node impact.
[0042] 5. Result analysis: For the analysis results, draw a correlation matrix and cause-and-effect diagram to identify highly correlated processes and major influencing paths. You can set a threshold to filter out processes with a correlation coefficient greater than a certain value for focus.
[0043] 6. Feedback and optimization: Based on the analysis results, optimize the production process and adjust the monitoring and intervention strategies of key processes to reduce overall quality risks.
[0044] In this application, the early warning module monitors the system in real time, provides an early warning mechanism for early warning and traces the cause. Real-time monitoring: Using edge computing technology, the deviation accumulation risk assessment model is embedded into the production line to calculate the deviation and accumulation value of each process in real time; Early warning mechanism: Once the accumulated deviation Or if the GRI exceeds the preset threshold, the system triggers an early warning and traces the cause, indicating which process or processes have accumulated deviations that lead to quality risks.
[0045] The system can also provide feedback and optimization, using historical data to iteratively optimize the model and adjust the weights. and the risk function parameters To improve the accuracy of prediction.
[0046] Those skilled in the art should understand that the embodiments of the present invention shown in the above description are only examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functions and structural principles of the present invention have been demonstrated and explained in the embodiments, and the embodiments of the present invention may be deformed or modified in any way without departing from the principles.
Claims
1. A monitoring system for multi-level deviation cumulative quality risks in complex processes, Features: Includes the following modules: Data collection module, used to collect quality deviation data of each process; The deviation accumulation calculation module is used to gradually accumulate the deviations of each process and generate the cumulative deviation value of each process; Risk assessment module, used to calculate the overall quality risk index GRI based on the cumulative deviation value; Process correlation analysis module, used to analyze the deviation transmission path and mutual influence between processes; Early warning module, which triggers an alarm when the cumulative deviation or quality risk index exceeds the preset threshold; The deviation accumulation calculation module calculates the cumulative deviation value of each process by the following formula: in, For the Cumulative deviation of each process; For the Deviation weight of a process; For process The deviation, For process The actual quality index value, For process Target value of Weight of process The importance of the process to the quality of the final product is determined through historical data analysis or criticality setting of the production process.
2. A monitoring system for cumulative quality risks of multi-level deviations in complex processes according to claim 1, Features: The risk assessment module calculates the overall quality risk index GRI by the following formula: Where n is the total number of processes, It is the risk function of accumulated deviations, reflecting the impact of accumulated deviations on product quality.
3. A monitoring system for cumulative quality risks of multi-level deviations in complex processes according to claim 4, Features: Risk Function Take the following form: in, It is an adjustment parameter to control the nonlinear growth of risk.
4. According to claim 1, a monitoring system for multi-level deviation cumulative quality risk in a complex process, Features: The process correlation analysis module is based on the Pearson correlation coefficient of each process deviation. The calculation formula of Pearson correlation coefficient is: Among them, Cov(X,Y) is the covariance of process X and process Y, and are the standard deviations of process X and process Y respectively.
5. A monitoring system for cumulative quality risks of multi-level deviations in complex processes according to claim 6, Features: The process correlation analysis module uses Bayesian network to perform causal analysis, evaluate the dependencies between processes, and conditional probability The calculation formula is: in, For the process Under the condition that the deviation is known, the process The conditional probability of the deviation, For process and process The joint probability of For process The marginal probability of .
6. A monitoring system for cumulative quality risks of multi-level deviations in complex processes according to claim 1, Features: When the cumulative deviation Ce or the quality risk index GRI exceeds a preset threshold, the early warning module triggers an alarm and traces the cause of the process deviation accumulation.
Citation Information
Patent Citations
Data processing method applied to cumulative sum control chart in statistical process control system
CN104216400A
Manufacturing process risk evaluation method based on direction weighted network
CN111815187A
Multi-source error analysis method in multi-process mechanical manufacturing process
CN113075907A
Construction quality control system and equipment for intelligent construction site management and storage medium
CN114662891A
Test risk early warning method and device, storage medium and computer equipment
CN115689046A
Cited By
Production and processing data collaborative management system and method for curtain wall single aluminum plate
CN119443742A
Food production information traceability management method and system
CN119990724A
Bridge construction progress monitoring method and system based on BIM
CN120047119A
A BIM-based Bridge Construction Progress Monitoring Method and System
CN120047119B
Electrophoretic coating water washing treatment method, device and system for aluminum profile
CN120556114A
