Continuous Quality Control Evaluation During Manufacturing Using Supervised Learning Algorithm for Industry 4.0

Smart industries use modern technologies such as machine learning and big data to maintain supply chain management and increase productivity but still the main challenge faced during quality control as this might affect the production rate. Smart industries are completely based on supervised learning that enables better inspection and effectively controls the parameter involved in the production process. Smart industries choose the mechanism that improves production and assures maximum quality. The various kernel function is initially used to select and extract a parameter. Support vector machine (SVM) is a supervised learning approach used in manufacturing industries to evaluate quality control. The SVM model uses the kernel function, namely RBF, along with Neural Networks, in identifying the parameter involved in quality management and undergoes the classification process. SVM consists of C-SVM and V-SVM classifier models involved in the classification process and undergoes training to handle the multiple numbers of consequence aroused during manufacturing. The performance of SVM classifiers and RBF NNs is evaluated. Different kernel functions, such as polynomial, linear, sigmoid, RBF, and over-varying gamma coefficient values, are tested in the experimental evaluation concerned with the comparative analysis of the continuous quality control function of the SVM classifier. Experimental results demonstrate the superiority of the SVM classifier in terms of the estimated computational time (88.1%), F1-measure (89.4%), ROC (65%), and accuracy (94.6%). The goal of the proposed model is to monitor the manufacturing process and control fault occurrence.