Comparison of different control charts for a Weibull process with type-I censoring

ABSTRACT

Some control charts have been proposed to monitor the mean of a Weibull process with type-I censoring. One type of control charts is to monitor changes in the scale parameter because it indicates changes in the mean. With this approach, we compare different control charts such as Shewhart-type and exponentially weighted moving average (EWMA) charts based on conditional expected value (CEV) and cumulative sum (CUSUM) chart based on likelihood-ratio. A simulation approach is employed to compute control limits and average run lengths. The results show that the CUSUM chart has the best performance. However, the EWMA-CEV chart is recommendable for practitioners with its competitive performance and ease of use advantage. An illustrative example is also provided.

KEYWORDS:

• Weibull process
• Conditional expected value
• Type-I censoring
• EWMA
• CUSUM

MATHEMATICAL SUBJECT CLASSIFICATION:

• 62N01
• 62P30

Acknowledgements

The authors are grateful to reviewer's valuable suggestions to improve the presentation.