Abstract
The average run length (ARL) of a control chart is often used as the sole measure of the performance of the chart. However, the run length distribution of a cumulative sum (CUSUM) chart can be quite different from a geometric distribution especially when the process is not in control. Thus, the ARL may not be sufficient as the sole measure of performance. The probability function of run length provides a comprehensive understanding of the working of a CUSUM chart. A computer program for computing the probability function and percentiles of the run length for a CUSUM chart is presented. The program also provides the ARL and plots of the probability function and the corresponding cumulative distribution function. An underlying normal distribution is assumed.
Additional information
Notes on contributors
F. F. Gan
Dr. Gan is a Statistical Consultant and Senior Lecturer in the Department of Mathematics. He is a Member of ASQC.