Many multivariate control charts have been proposed for monitoring several related quality characteristics simultaneously. However, even when an out-of-control signal is detected, the employed multivariate control charts generally do not provide any interpretable information associated with that signal. That is, the contributors of the out-of-control event can not be identified by the charts. Hence, how to tackle this interpretation problem effectively is an important issue in multivariate process control. One rarely addressed but very crucial property of this interpretation problem is that the number of possible outcomes can be very large. According to this key property, a nonparametric discriminant analysis (NDA)-based hierarchical classification scheme is proposed in this paper. A simulation experiment including several popular classification methods was conducted for evaluating the performance of the proposed method. The result shows that our proposed scheme is very competitive when measured against these popular methods.