A novel quality prediction method based on feature selection considering high dimensional product quality data

Abstract

Product quality is the lifeline of enterprise survival and development. With the rapid development of information technology, the semiconductor manufacturing process produces multitude of quality features. Due to the increasing quality features, the requirement on the training time and classification accuracy of quality prediction methods becomes increasingly higher. Aiming at realizing the quality prediction for semiconductor manufacturing process, this paper proposes a modified support vector machine (SVM) model based on feature selection, considering the high dimensional and nonlinear characteristics of data. The model first improves the Radial Basis Function (RBF) in SVM, and then combines the Duelist algorithm (DA) and variable neighborhood search algorithm (VNS) for feature selection and parameters optimization. Compared with some other SVM models that are based on DA, genetic algorithm (GA), and Information Gain algorithm (IG), the experiment results show that our DA-VNS-SVM can obtain higher classification accuracy rate with a smaller feature subset. In addition, we compare the DA-VNS-SVM with some common machine learning algorithms such as logistic regression, naive Bayes, decision tree, random forest, and artificial neural network. The results indicate that our model outperform these machine learning algorithms for the quality prediction of semiconductor.

Product Quality Prediction Based on RBF Optimized by Firefly Algorithm

“…To remove the bad influence of unknown disturbance on the accuracy of product quality prediction, Yang et al [10] proposed a novel hybrid denoising-linear-nonlinear quality prediction model, which also introduces the support vector machine (SVM) to process nonlinear data. Ren et al [11] introduced the semisupervised learning mechanism and the manifold regularization to construct a new product quality prediction model, which can deal with the unlabeled samples issue. To overcome the problem that dataset from the semiconductor manufacturing process is of high dimension and nonlinear, Hu et al [12] put forward a modified SVM model on the basis of feature selection.…”

IoT Framework Application to SHM Systems

“…0, the fourth industrial revolution characterised by the integration of advanced technologies to create smart factories, cyber-physical systems and IIoT (Industrial Internet of Things) [5]. The implementation of IoT in industry is becoming more and more common due to its numerous advantages, generating several industry-beneficial applications: (1) Predictive analytics and preventive maintenance improving by enabling data collection and subsequent analysis and helping predict machine failures and perform preventive maintenance before they occur, which in turn reduces costs and downtime [6], (2) Improving product quality by monitoring and controlling critical factors during the production process allowed for greater control over processes and reduction of defects [7], (3) Edge computing [8], technology responsible for processing data in real time at the point, where it is generated rather than sending it to a central location for processing, reducing processing time and load on central networks and servers and (4) Digital twin generation [9], enabling simulation and testing of different scenarios virtually, which reduces the costs and risks associated with testing and validation processes.…”