Abstract: To expand the coverage of data points used for fitting the relative efficiency curve in plutonium Isotopic Composition analysis via gamma spectrometry, and to enhance the accuracy of the relative efficiency curve fitting, the block coordinate descent method is employed to normalize the relative detection efficiencies at different energy points of distinct nuclides. Outlier data points are removed based on the 3σ criterion, and the remaining points are used to fit the relative detection efficiency curve. The energy spectra of plutonium samples with three different abundances were analyzed using the approximation method, the interpolation method, the block coordinate descent method, and the two energy regions defined in FRAM. The results show that the maximum deviation of the ²³⁹Pu content obtained by the block coordinate descent method from the nominal value is 3.4%, which is a significant improvement compared to the approximation method and the interpolation method. The root mean square error of the relative detection efficiency curves fitted by the block coordinate descent method under three shielding conditions is calculated. The result with 2 mm cadmium shielding is the best and 2 mm cadmium is suitable for the analysis of sample abundance.