Using PCA and ANN to identify significant factors and modeling customer satisfaction for the complex service processes

This paper proposes a PCA and ANN based approach to identify significant influential quality factors and modeling customer satisfaction for complex service processes. Firstly, the performance evaluation index system includes initial factors and customer satisfaction degree is proposed, and then the measurement data are collected by questionnaires. Secondly, by using PCA, several preceding principal components (PCs) are extracted, which present about 90% contributions of the whole variations of initial factors. Thirdly, the extracted PCs are converted to new significant factors according to the corresponding coefficients of initial factors in each PC. Finally, BP network is applied to modeling the nonlinear relationship between the significant factors and customer satisfaction degree. The case study of the maintenance service process of an automobile 4S store shows that, the proposed approach can extracted the significant factors from lots of initial factors, and can exactly modeling the complex nonlinear relationship between influential factors and customer satisfaction as well.