USE OF MEDIAN BASED ESTIMATOR TO MITIGATE OUTLIER’S EFFECT THROUGH 𝑺𝟐 CHART

In this paper, we consider an upper-sided Phase II variance chart with probability limits in case of unknown parameter because the quality practitioner interested in monitoring increased variance of the process parameter. It is well established that when the Phase I data are contaminated with spurious observations, performance of the chart is suspected to deviate from what is normally expected. Therefore, we propose an improved performance of one-sided variance chart under the exceedance probability criterion for a fixed in-control average run length using the absolute deviation from median estimator. Under the exceedance probability criteria, the chart is designed so that the user can get more confidence in their in-control average run length values. The proposed chart is compared with the existing chart in case of contaminated and non-contaminated observations. Result shows that performance of variance chart shows robust performance when using absolute deviation from median estimator. Finally, an example has been provided in the favour of our proposed study.