Determination of Optimum Data Points for Scaling Factor Determination

Scaling factors are calculated based on a database from radiochemical analyses of representative waste samples. Several data points are needed to derive a reliable scaling factor. The more the number of data points, the better is the correlation, but more costly because of number of needed radiochemical analyses. Therefore, optimization of data points should be considered to minimize the cost without compromising reliability and prediction of the scaling factor. Scaling factors for Ni-63, Sr-90, and C-14 were calculated using Co-60 and Cs-137 as Key nuclides based on the published data in EPRI-4037. Correlation coefficient, percent error and relative standard deviation were plotted against the number of data points used in the estimation of scaling factor. The optimum number of data points was obtained to where there was no significant improvement in the statistical uncertainties by using additional samples. Even though non-segregated (all data points) has greater effect in calculating scaling factor because of its more data points, this study showed that even segregated data points can also give good DTM characterization.