Product quality risk early warning method driven by large language model
By using a large language model-driven approach and combining internal and external data features of the production management system, a ridge regression model is used to construct a quality risk index prediction value. This solves the problem of low accuracy in product quality risk early warning in existing technologies and achieves more efficient product quality risk early warning.
CN121391010APending Publication Date: 2026-01-23JIANGSU AEROSPACE DAWEI TECH CO LTD
View PDF 0 Cites 0 Cited by
Patent Information
- Application Number
- CN202511479733.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-16
- Publication Date
- 2026-01-23
AI Technical Summary
Technical Problem
Existing product quality risk early warning methods are unable to capture the interactions between complex modern industrial product factors, resulting in poor accuracy and reliability of early warnings.
Method used
A large language model-driven approach is adopted, which combines internal and external data features of the production management system to extract product elements, personnel and external risk features. The feature weights are solved by ridge regression model to construct the predicted value of quality risk index.
Benefits of technology
It significantly improves the accuracy, reliability, and real-time performance of product quality risk warnings, enabling better capture of dynamic and interactive impacts.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN121391010A_ABST
Abstract
The invention discloses a large language model-driven product quality risk early warning method, which relates to the field of product quality risk early warning, and comprises the following steps of: considering influence of product elements and production processing to extract element risk characteristics and personnel risk characteristics of a target product; in addition, a large language model is adopted to realize automatic conversion of non-standard data of after-sales quality abnormal events of reference similar products of a target product, external risk features are extracted, and then internal and external data features of a production management system are fused to perform quality risk prediction. And the feature weight vector used when the features are fused is obtained based on a ridge regression model, so that compared with a common linear regression algorithm, the quality prediction early warning capability is remarkably improved while high precision is maintained, and the accuracy, the reliability and the real-time performance of predicting the product quality risk by the method are relatively good.
Need to check novelty before this filing date? Find Prior Art