Abstract
A large number of researches addressed the problem of monitoring statistical processes using complete data. Nevertheless, in engineering applications, especially reliability engineering and lifetime test, we often observe the incomplete or censored sample. In this paper, we introduce three EWMA schemes for monitoring exponentially distributed processes based on type-II censored data. Our proposed procedure is capable of detecting a shift in both the origin (location) and scale parameters of a shifted (two-parameter) exponential distribution. As one-parameter exponential distribution is a particular case of the two-parameter exponential with the origin equals zero, the present paper will be useful to detect a shift in the location (origin) or scale or both from traditional one-parameter exponential processes. Also, one can achieve a complete sample situation as a particular case. Therefore, the current article is very general and inclusive. We provide the design parameters, such as control limits of the three schemes and evaluate the performance in terms of average run length characteristics. We investigate the effects of sample size and censoring design on the performance of the proposed schemes. We discuss some realistic examples to illustrate the implementation strategies of the proposed schemes. • Three Phase-II EWMA control charts are developed for two-parameter exponential distribution based on type-II censored data. • The notions of both the maximum likelihood method of estimation and uniformly minimum variance unbiased estimation are used. • Simulation studies based on Monte-Carlo are conducted to compare the three control charts. • Three case studies are also presented to illustrate the effectiveness of our proposed method.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call