Abstract—The observations are not always identically and independently distributed (i.i.d.) in continuous manufacturing which most observations is skewed away normal distribution. In this research aim to study the performance of Tukey's control chart for detecting a change in parameter when observation are from skew distributions such as exponential and Laplace distributions. Also, the performance of Tukey’s control chart is compared with Shewhart and Exponentially Weighted Moving Average (EWMA) charts. The Average Run Length is commonly measured the performance of control chart. The ARL0 is usually used when the process is in-control (sufficient large) and the ARL1 is correctly signaled to be out-of-control (minimum). The Tukey's control chart is superior to other charts for large shift; however, the EWMA performs better for small to moderate shifts. The Tukey's control chart is very easy to setup the control limits and use a simple statistical concept.

Index Terms—Average Run Length, Skew Distributions, EWMA, Tukey's Control Chart, Robustness.

S. Sukparungsee is with the Department of Applied Statistics, Faculty of Applied Science, King Mongkut’s University of Technology North Bangkok, Bangkok 10800, Thailand (tel.: +662-6538802; fax: + 662-6538802; e-mail address: swns@kmutnb.ac.th).

Cite: Saowanit Sukparungsee, "Robustness of Tukey’s Control Chart in Detecting a Changes of Parameter of Skew Distributions," International Journal of Applied Physics and Mathematics  vol. 2, no. 5, pp. 379-382, 2012.