A new method of estimating the process spread using confidence interval of sample range

This paper deals with a novel method of estimating the process spread (σ) in the construction of Shewhart control chart for means, basing ona new estimatederived from the confidence interval (CI) of sample range. The classical estimate (R /d2,n)of σ for a normally distributed data is utilized to arrive at anew estimateproposed as the weighted sum of the lower, middle and upper values of the 100(1-α) %CI for the process rangebasing on sample range.The weights are defined as inversely proportional to the absolute bias from the target spread. It is shown by simulation that the new estimate is more consistent than the classical point estimate based on R /d2,n. It is also shown that the x chart performs better in terms of B-risk when the new estimate is used.