Monitoring Lognormal Reliability Data in a Two-Stage Process Using Accelerated Failure Time Model

The reliability data is getting used to monitor and improve the quality of products or services. Nowadays, most of products or services are the results of processes with dependent stages referred to as multi-stage process. In these processes, the quality characteristics are affected by the quality characteristics in the previous stages, called as cascade property. In some cases, it is not possible to collect all the lifetime data due to resource limitations. Thus, the control charts have been compared under two different scenarios; censored and non-censored data. In this paper, the accelerated failure time (AFT) model is used and two control charts are presented to monitor the quality characteristic in the second stage under the censored and non-censored reliability data. The exponentially weighted moving average (EWMA) and cumulative sum (CUSUM) control charts are used based on the proposed residuals. The performance of the proposed control schemes is evaluated in terms of zero-state and steady-state average run length criteria through extensive simulation studies. The results generally show that CUSUM control chart performs better than the EWMA control chart for monitoring lognormal reliability data in a two-stage process. However, the EWMA control chart outperforms the CUSUM control chart under small shifts when there is no censoring in reliability data or in the censoring rate of 20%.