Prediction of quality in production using optimized Hyper-parameter tuning based deep learning model

Large volumes of manufacturing data may now be collected because to the growing popularity of smart Industry 4.0. Product quality may be predicted from manufacturing data acquired during production using machine learning approaches such as classification. A supply chain can benefit from eliminating uncertainty by precise forecasting at any point in the process. As a result, knowing the quality of a product batch early on can save money on recalls, packaging, and shipping. Classification methods have been extensively studied for forecasting the quality of certain manufacturing processes, but the overall obedience of production batches has not been carefully studied. Classification methods based on deep learning (Convolutional Neural Network) and optimal hyper-parameter tuning are the focus of this article, which aims to evaluate the suggested appliance production process. Existing approaches for classifying unit batches are compared to the proposed classification model in terms of several quality parameters for compliance. As a result, a model for predicting compliance quality may be built using the new method. Features and dataset knowledge are also critical in training classification models, according to this study.