CN-122022589-A - Method and device for analyzing adverse effect factors of products, computer device and medium

CN122022589ACN 122022589 ACN122022589 ACN 122022589ACN-122022589-A

Abstract

The present disclosure provides a method and apparatus for analyzing product adverse effects, computer apparatus, and medium. The analysis method of the product adverse effect factors comprises the steps of obtaining multiple sets of product information, wherein the product information comprises at least two product parameters, calculating importance scores of each of the at least two product parameters by utilizing at least two adverse analysis algorithms, calculating weights of the importance scores of the at least two product parameters under each adverse analysis algorithm by utilizing information entropy, and weighting each product parameter by the weights to obtain an evaluation result of adverse effect of each product parameter in the multiple sets of product information. The analysis method of the adverse effect factors of the product can accurately identify the adverse root cause of the product, and improves the analysis efficiency and accuracy of the adverse factors.

Inventors

GAO CHAO
Wen Jinxiao
HUANG WEIXUE
ZHOU XIBO
ZHAO XINGXING

Assignees

京东方科技集团股份有限公司

Dates

Publication Date: 20260512
Application Date: 20260204

Claims (12)

1. A method for analyzing adverse effects of a product, comprising: obtaining product information of a plurality of groups of products and bad data of the products, wherein the product information comprises at least two product parameters; Calculating an importance score for each of the at least two product parameters based on the product information and the failure data, respectively, using at least two failure analysis algorithms; calculating the weight of importance scores of the at least two product parameters under each bad analysis algorithm by utilizing information entropy; and weighting each product parameter by the weight to obtain an evaluation result of the adverse effect of each product parameter in the plurality of sets of product information.
2. The method of claim 1, wherein after calculating the importance scores for each of the at least two product parameters using at least two poor analysis algorithms, the method further comprises: and sequencing importance scores calculated by the same bad analysis algorithm in each group of product information to obtain an importance sequence under the condition of each bad analysis algorithm.
3. The method of analysis of claim 2, further comprising: integrating the importance sequences of the product parameters in each set of product information in the same sequence table to obtain an importance score matrix for poor analysis of the at least two product parameters.
4. The method of claim 3, further comprising normalizing the importance scores of the same failure analysis algorithm in the importance score matrix to obtain normalized importance scores.
5. The method of claim 3, wherein calculating the weight of the importance scores of the at least two product parameters under each of the poor analysis algorithms using information entropy further comprises: Calculating the probability of each product parameter under each bad analysis algorithm based on the importance scores of the same bad analysis algorithm in the importance score matrix to obtain a probability distribution matrix; And calculating the weight of the importance scores of the at least two product parameters under each bad analysis algorithm based on the probability distribution matrix.
6. The method of claim 5, wherein weighting each product parameter with the weight to obtain an evaluation result of the adverse effect of each product parameter in the plurality of sets of product information further comprises: each product parameter is weighted with the weight based on the following formula: Wherein, the Represent the first Line product parameters are at The importance score of the failure analysis algorithm is listed, Representing weighted importance scores of the at least two poor analysis algorithms for the product parameters, w j representing the weight of the importance score of the j-th column of poor analysis algorithms.
7. The method according to claim 4, wherein the normalization process comprises at least one of a polar difference normalization, a mean normalization, a maximum absolute value normalization, and a standard deviation normalization.
8. The method of claim 1, wherein the at least two bad analysis algorithms are at least two selected from the group consisting of a first correlation algorithm, a second correlation algorithm, a first machine learning algorithm, a second machine learning algorithm, a first deep learning algorithm, and a second deep learning algorithm.
9. The method of claim 1, wherein the method comprises the steps of, The at least two product parameters are at least two selected from process time information, equipment information, environment variable information, and operating process parameter information.
10. An apparatus for analyzing adverse effects of a product, comprising: the acquisition unit acquires a plurality of groups of product information, wherein the product information comprises at least two product parameters; An algorithm analysis unit that calculates an importance score for each of the at least two product parameters using at least two bad analysis algorithms, respectively; the calculating unit is used for calculating the weight of the importance score of the at least two product parameters under each bad analysis algorithm by utilizing the information entropy; And the weighting integration unit weights each product parameter by the weight to obtain an evaluation result of the adverse effect of each product parameter in the plurality of sets of product information.
11. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that, The processor, when executing the program, implements a method of analyzing adverse effects of the product according to any one of claims 1 to 9.
12. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program, when executed by a processor, implements a method for analyzing adverse effects of the product according to any one of claims 1 to 9.

Description

Method and device for analyzing adverse effect factors of products, computer device and medium Technical Field The present disclosure relates to the field of display technology. And more particularly, to a method and apparatus for analyzing adverse effects of a product, a computer apparatus, and a computer-readable storage medium. Background Root cause analysis of product failure is a structured approach aimed at identifying the root cause of the problem and solving it, not just surface phenomena. In high-end manufacturing industries such as display panels, the production process is complex and numerous, hundreds of process stations are needed from raw materials to finished products, different equipment is used at each station, and parameter configurations of the equipment are different, so that determining the root cause of poor products is very challenging. At present, even though root cause analysis is performed by means of an analysis algorithm, all potential factors cannot be covered accurately and comprehensively and risks are estimated. Therefore, it is necessary to provide a solution capable of accurately and efficiently analyzing the root cause of adverse influence factors of products. Disclosure of Invention The invention aims to provide an analysis method of adverse effect factors of products, which comprises the following steps: obtaining product information of a plurality of groups of products and bad data of the products, wherein the product information comprises at least two product parameters; Based on the product information and the bad data, calculating an importance score of each of at least two product parameters by using at least two bad analysis algorithms respectively; calculating the weight of importance scores of at least two product parameters under each bad analysis algorithm by utilizing information entropy; And weighting each product parameter by a weight to obtain an evaluation result of the adverse effect of each product parameter in the plurality of groups of product information. Optionally, after calculating the importance scores for each of the at least two product parameters using the at least two failure analysis algorithms, the method further comprises: and sequencing importance scores calculated by the same bad analysis algorithm in each group of product information to obtain an importance sequence under the condition of each bad analysis algorithm. Optionally, the analysis method further comprises: And integrating the importance sequences of the product parameters in each group of product information into the same sequence table to obtain an importance score matrix of poor analysis of at least two product parameters. Optionally, the analysis method further comprises the step of carrying out standardization processing on the importance scores of the same bad analysis algorithm in the importance score matrix to obtain standardized importance scores. Optionally, calculating the weight of the importance scores of the at least two product parameters under each of the poor analysis algorithms using the information entropy further comprises: calculating the probability of each product parameter under each poor analysis algorithm based on the importance scores of the same poor analysis algorithm in the importance score matrix to obtain a probability distribution matrix; the weight of the importance scores of at least two product parameters under each of the poor analysis algorithms is calculated based on the probability distribution matrix. Optionally, weighting each product parameter with a weight to obtain an evaluation result of the adverse effect of each product parameter in the plurality of sets of product information further includes: each product parameter is weighted with a weight based on: Wherein, the Represent the firstLine product parameters are atThe importance score of the failure analysis algorithm is listed,Representing weighted importance scores of at least two of the poor analysis algorithms to the product parameters, w j represents the weight of the importance score of the j-th column of the poor analysis algorithm. Optionally, the normalization process includes at least one of a polar error normalization, a mean normalization, a maximum absolute value normalization, and a standard deviation normalization. Optionally, the at least two bad analysis algorithms are at least two selected from a first correlation algorithm, a second correlation algorithm, a first machine learning algorithm, a second machine learning algorithm, a first deep learning algorithm, and a second deep learning algorithm. Optionally, the at least two product parameters are at least two selected from process time information, equipment information, environmental variable information, and operating process parameter information. A second aspect of the present disclosure provides an analysis apparatus for product adverse influence factors, including: the acquisition unit acquires a pluralit