CN-121979149-A - Quality control system based on multivariate exponential weighting moving average control chart

CN121979149ACN 121979149 ACN121979149 ACN 121979149ACN-121979149-A

Abstract

The invention provides a quality control system based on a multivariate exponential weighting moving average control chart, and belongs to the technical field of quality management and statistical process control. The system comprises a quality data acquisition module, a data preprocessing module, a statistical process control module, an abnormality discrimination module and a quality evaluation and improvement module, and is used for monitoring and controlling the production process of the product with a plurality of quality characteristics in real time. The system comprises a system, a system and a method, wherein the system acquires quality data of a plurality of quality characteristics in a production process, pre-processes the quality data to construct a multi-element quality characteristic data set, estimates a mean vector and a covariance matrix in a controlled state based on historical quality data, performs weighted update on the multi-element quality data by adopting a multi-element index weighted moving average method, calculates corresponding MEWMA statistics and generates a MEWMA control chart, compares the MEWMA statistics with a preset control limit to judge whether the production process is in a runaway state or not, and outputs early warning information, a corresponding quality evaluation result and a corresponding improvement suggestion when abnormality is detected. The method can comprehensively analyze the related information among the multiple quality characteristics, improve the sensitivity and accuracy of abnormal detection in the production process, and effectively improve the quality control level of the production process in the manufacturing industry.

Inventors

WU XIAOFANG
Lu Pengcan
MA ZIYU

Assignees

昆明理工大学

Dates

Publication Date: 20260505
Application Date: 20260204

Claims (7)

1. The quality control system based on the multivariate exponential weighting moving average control chart is characterized by comprising a quality data acquisition module, a data preprocessing module, a statistical process control module, an anomaly discrimination module and a quality evaluation and improvement module; The quality data acquisition module is used for acquiring quality data of a plurality of quality characteristics in the production process; the data preprocessing module is used for sorting the acquired quality data to form a multi-element quality characteristic data set; The statistical process control module is used for carrying out weighted updating on the multi-element quality data according to the multi-element index weighted moving average method based on the multi-element quality characteristic data set, calculating MEWMA statistics and constructing a corresponding MEWMA control chart; The abnormality judging module is used for comparing the MEWMA statistic with a preset control limit to judge whether abnormality occurs in the production process; the quality evaluation and improvement module is used for outputting corresponding quality evaluation results and quality improvement information when the production process is judged to be abnormal.
2. The quality control system based on a multivariate exponentially weighted moving average control graph of claim 1, wherein the statistical process control module estimates a mean vector and covariance matrix under a controlled state based on historical quality data for use in the calculation of the MEWMA statistic.
3. The quality control system of claim 1 wherein the statistical process control module generates a univariate exponentially weighted moving average control chart when the quality data comprises only a single quality characteristic and generates a multivariate exponentially weighted moving average control chart when the quality data comprises a plurality of quality characteristics.
4. The quality control system based on a multivariate exponentially weighted moving average control graph of claim 1, wherein the control limits are determined based on a multivariate statistical distribution model, and wherein the anomaly discrimination module determines that the production process is in a runaway state when the MEWMA statistic exceeds the control limits.
5. The quality control system based on a multivariate exponentially weighted moving average control chart according to claim 1, wherein the quality evaluation and improvement module generates early warning information after determining that an abnormality occurs in the production process, and outputs a quality analysis result and an improvement suggestion corresponding to the abnormality.
6. The quality control system based on a multivariate exponentially weighted moving average control chart of claim 1, wherein the statistical process control module performs exponentially weighted updating of the multivariate quality data based on a predetermined smoothing coefficient.
7. The quality control system based on a multivariate exponentially weighted moving average control graph of claim 1, wherein the quality control system is used for product quality monitoring during manufacturing.

Description

Quality control system based on multivariate exponential weighting moving average control chart Technical Field The invention relates to the technical field of quality control and statistical process control, in particular to a quality control system based on a multivariate exponential weighting moving average control chart, which is suitable for quality monitoring and anomaly detection of a product production process with a plurality of quality characteristics. Background Quality control is an important component of manufacturing quality management, and monitors and controls the quality formation process of products through the collection and analysis of quality data in the production process. Conventional quality control methods typically employ Statistical Process Control (SPC) techniques to monitor quality characteristics of the production process to determine whether the production process is in a controlled state. In the existing statistical process control method, a univariate control chart is widely applied to quality monitoring of a production process. However, in the actual manufacturing process, the product quality is often determined by a plurality of quality characteristics associated with each other, and the single variable control chart is difficult to reflect the correlation between the plurality of quality characteristics, which easily causes abnormal detection lag or erroneous judgment, thereby reducing the effectiveness of quality control. In order to solve the problem of multi-quality characteristic monitoring, the prior art provides a multi-element statistical process control method, but the prior multi-element control method still has the defects of insufficient sensitivity to small changes in the production process and difficulty in timely identifying potential quality risks in practical application, and on the other hand, part of multi-element control methods have complex calculation and lower visualization degree, so that visual judgment and quick decision-making of the production process state by production management staff are not facilitated. Exponentially Weighted Moving Average (MEWMA) control charts are gradually introduced into the field of quality control because of their higher sensitivity to process minor shifts. Furthermore, the multi-element index weighted moving average control chart can comprehensively consider a plurality of quality characteristics and correlation thereof, and has a good application prospect in multi-variable quality monitoring. However, the prior art has few systematic implementations for multivariate exponentially weighted moving average control charts, and in particular lacks a quality control system that combines multivariate quality data collection, statistical analysis, control chart construction, and anomaly discrimination with quality improvement. Therefore, a quality control system based on a multi-element index weighted moving average control chart is needed to effectively monitor multi-element quality characteristics, improve timeliness and accuracy of abnormal detection in a production process, and further improve quality control level in a manufacturing process. Disclosure of Invention Aiming at the problems of abnormal recognition lag and insufficient consideration of multivariate correlation in the process of multi-element quality characteristic monitoring of the existing quality control system, the invention provides a quality control system based on a multi-element index weighted moving average control chart. According to the invention, by constructing the statistical process control module taking the multi-element index weighted moving average statistical model as a core, the real-time monitoring and abnormal discrimination of multi-dimensional quality data are realized, and the multi-element quality data acquisition, statistical analysis, process control and quality improvement functions are integrated into a unified system platform, so that the quality control efficiency and accuracy of the multi-quality characteristic production process are improved. In order to achieve the above purpose, the present invention provides the following technical solutions: The quality control system based on the multivariate exponential weighting moving average control chart is characterized by comprising a quality data acquisition module, a data preprocessing module, a statistical process control module, an anomaly discrimination module and a quality evaluation and improvement module. Specifically, the statistical process control module builds a MEWMA control chart based on a multivariate exponential weighted moving average method, and is used for carrying out real-time monitoring and anomaly detection on multivariate quality data collected in the production process. After the quality data acquisition is completed, the quality control system automatically starts a statistical analysis flow, calculates corresponding exponentially weighted moving aver