Abstract
This article describes a simulation to compare the effectiveness of Variable Sampling Intervals (VSI) with Fixed Sampling Intervals (FSI). This comparison is made using both techniques in conjunction with Shewhart X-bar charts and data that are unknowingly positively autocorrelated. Simulation was essential to providing these research results because to date no theoretical solution addresses the performance of the VSI Shewhart X-bar chart when the data are autocorrelated. Control limits, calculated from initial sample observations, are utilized. The use of VSI is found to cause a quicker time to signal for an out of control process. We conclude that while the use of variable sampling intervals is advantageous when a large shift in the mean is expected, the VSI procedure is more sensitive to the impact of autocorrelation.
Get full access to this article
View all access options for this article.
