Year
1984
Abstract
A clarification and discussion on the controversial topic of bias corrections is given from a regulatory licensing viewooint. The emphasis is on bias corrections to Inventory Difference values, rather than the bias correcting of individual measurement values. The main objective behind NRC's requirement to determine bias corrections for each measurement system is to obtain the best estimate of the true ID. In the past, much attention has been given to the proper statistical treatment of measurement control data for deriving bias corrections. However, the quality of the data utilized for the estimation of biases is often neglected. The need to recognize the various parameters that can affect bias, and the methodology for representative monitoring of such parameters is discussed.