Study of product production quality risk identification based on fault data

In this paper, within the framework of intelligent transformation for product production quality risk control, a risk identification model based on fault data from the production link is developed. Drawing upon the analytical principles of Quality Function Deployment (QFD), the proposed method for identifying production quality risks using fault data is structured into three stages of knowledge extraction. The initial stage consolidates disparate raw data into semi-structured electronic document data. The subsequent stage generates fault abstraction data employing text structuring and regular expression techniques among other methods. The final stage leverages the Biterm Topic Model (BTM) and K-Means algorithm for extracting knowledge elements and conducting association clustering analysis. An empirical analysis is conducted using product data to perform association analysis of product faults and cluster analysis of quality management factors, thereby demonstrating the effectiveness of the proposed model and methodology.