A Study on Quality Prediction for Smart Manufacturing Based on the Optimized BP-AdaBoost Model

To accurately predict the product quality in smart manufacturing, this paper designs the BP-AdsysBoost model on the basis of BP neural network and AdaBoost algorithm. The BP-AdsysBoost model considers both the data characteristics and the technology advantages, which pays more attentions to the unqualified products wrongly predicted. To further examine the model, the 110560 data of smart manufacturing from German BOSCH company is used for this research. The proposed BP-AdsysBoost model is compared with the BP neural network and the unmodified BP-AdaBoost model according to prediction performance. The results show that the BP-AdsysBoost model has significant advantages in prediction accuracy and FDR, which proves its satisfied prediction ability for product quality in smart manufacturing.