Identification of Critical-to-quality Characteristics Based on Improved XGBoost

Product quality is the lifeline of product manufacturing enterprises. Quality control of product manufacturing process is the fundamental way to ensure product quality. The development and integration of the two fields of information systems and machine learning have a huge impetus for the quality control of enterprises. Identification of critical-to-quality characteristics (CTQ) is the process of identifying the factors that affect product quality, which is an essential step in quality control. A large amount of quality characteristic data collected from the production line provides the basis for CTQ identification. In this paper, for the problem of lithium battery data imbalance, a CTQ recognition method based on xgboost feature selection algorithm is established.