Combining independent component analysis and support vector machine for identifying fault quality variables in the multivariate process

Abstract

This study proposes a hybrid approach which is composed of independent component analysis (ICA) and support vector machine (SVM) to identify the fault quality variables when a step-change disturbance existed in a multivariable process. The multivariate statistical process control (MSPC) chart plays an important role in monitoring a multivariate process. However, the use of MSPC chart encounters a difficulty in practice. This difficult issue involves which quality variable or which set of the quality variables is responsible for the generation of the out-of-control signal. The proposed hybrid ICA-SVM scheme first uses ICA to the Hotelling T <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> statistics generating independent components (ICs). The hidden useful information of the fault quality variables could be discovered in these ICs. The ICs are then used as the input variables of the SVM for building the classification model. Experimental results shows that the proposed ICA-SVM method can effective detect the fault quality variables in the multivariable process.