Statistical design of individual and integrated multivariate hotelling T 2 control chart for a multi-component load-sharing system subject to dependent failure modes

In this paper, we plan to propose individual and integrated multivariate Hotelling T2 control chart for a multi component load sharing system. In fact, the multivariate Hotelling T2 control chart for monitoring process under assumptions that the vector of signals follows a multivariate normal distribution is considered. It has been assumed the system is subject to hard failure and soft failure and failure modes have been considered dependent. Also soft failures increase probability of event of hard failures. The degradation states is described by discrete time Markov chain (DTMC). To model the dependency between hard failures and soft failures, the proportional intensity model (PIM) is applied. The objective is to compare individual multivariate T2 Hotelling control chart performance with integrated multivariate T2 Hotelling control chart performance by computing Average run length (ARL) and standard deviation run length (SDRL). Also sensitivity analysis technique is applied to evaluate effect of load sharing parameters to ARL. Computational results shows that the proposed control chart is sensitive to model parameters.