Discrete manufacturing machinery product quality tracing method based on outlier detection

Abstract

The invention discloses a discrete manufacturing machinery product quality tracing method based on outlier detection, which belongs to the technical field of informationization of manufacturing industries. In view of the problem that the quality tracing method in the current discrete manufacturing process can not accurately analyze the causes of the quality problems of machinery products, a discrete manufacturing machinery product quality tracing method based on outlier detection is put forward. Firstly, m causes that possibly cause quality problems under various production conditions are extracted; then, according to the tracking data in the production process, n quality index data of the products produced under the potential causes are respectively counted, the data are used to build anm*n quality problem tracing data set, an outlier factor for the quality index data under each potential cause is then calculated, and a quality anomaly index is calculated; and finally, by comparing the sizes of the quality anomaly indexes, the possibility of a cause of a quality problem is analyzed, and important information is provided for product quality tracing.

Description

technical field [0001] The invention belongs to the technical field of manufacturing informatization, and relates to a quality traceability method for discrete manufacturing machinery products based on outlier point detection. Background technique [0002] With the rapid development of the manufacturing industry, manufacturing companies pay more and more attention to product quality traceability. How to solve the problem of product quality traceability puts forward new requirements for the information construction of manufacturing enterprises. The core part of realizing quality traceability is quality traceability. [0003] In recent years, with the improvement of quality awareness of enterprises, some simple quality traceability methods have gradually been popularized in enterprises, but most of them are based on statistical methods using single quality index data. These methods generally use the average value of indicators such as product qualification rate as data to an...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is mainly aimed at the problem that the current quality traceability method in the discrete manufacturing process cannot accurately analyze the cause of the quality problem of the mechanical product, and proposes a method for the quality traceability of the discrete manufacturing mechanical product based on outlier detection. Extract m types of reasons that may cause quality problems in various production situations from the aspects of people, machines, materials, methods, and environment, and then calculate the n types of quality of products produced under these potential reasons according to the tracking data in the production process Index data, using these data to construct an m*n quality problem traceability data set, and then calculate the outlier factors of the quality index data under various potential reasons, and calculate the quality anomaly index, and finally compare the quality anomaly index Size to analyze the possibility of the cause of quality problems, so as to provide important information for product quality traceability

Images