Designing Exponentially Weighted Moving Average Control Charts under Failure Censoring Reliability Tests

One of the most important quality characteristics in a production process is the product lifetime. The production of highly reliable products is a concern of manufacturers. Since it is time-consuming and costly to measure lifetime data, designing a control chart seems difficult. To solve the problem, lifetime tests are employed. In the present study, one-sided and two-sided EWMA control charts are designed under a type II censoring (failure censoring) life test. Product lifetime is a quality characteristic dealt with in this study. It is assumed to follow the Weibull distribution with a fixed shape parameter and a variable scale parameter. In order to design a control chart, first, the control chart limits are calculated for different parameters, and then the Average Run Length (ARL) in the out-of-control state is used to evaluate the performance of the proposed control chart. Next, a comprehensive sensitivity analysis is performed for the different parameters involved. The computational results show that the one-sided control chart has better performance to detect the shift of lifetime data than the two-sided control chart. The average run length curve of the two-sided control chart is biased, while that of the one-sided control chart is unbiased. A very effective parameter that increases the performance of a control chart is found to be the number of failures in the failure censoring process. Finally, simulated and real examples are provided to show the performance of the proposed control chart.