The Korea Atomic Energy Research Institute (KAERI) employs a methodology for evaluating the concentration of radionuclides, dividing them into volatile and non-volatile nuclides based on their characteristics, to ensure the permanent disposal of internally generated radioactive waste. Gamma spectroscopy enables the detection and radiation concentration determination of individual nuclides in samples containing multiple gamma-emitting nuclides. Due to the stochastic nature of radioactive decay, the generated radiation signal can interact with the detector faster than the detected signal processing time, causing dead time in the gamma spectroscopy process. Radioactive waste samples typically exhibit higher radiation levels than environmental samples, leading to long dead times during the measurement process, consequently reducing the accuracy of the analysis. Therefore, dead time must be considered when analyzing radioactive waste samples. During the measurement process, dead time may vary between a few seconds to several tens of thousands of seconds. More long dead time may also result in a temporal loss in the analysis stage, requiring more time than the actual measurement time. Long dead time samples undergo re-measurement after dilution to facilitate the analysis. As the prepared solution is also utilized in the nuclide separation processes, minimizing sample loss during dilution is crucial. Hence, predicting the possibility of dead time exceeding the target sample in advance and determining the corresponding dilution factor can prevent delays in the analysis process and the loss of samples due to dilution. In this study, to improve the issues related to gamma analysis, by using data generated during the analysis process, investigated methods to predict long dead time samples in advance and determining criteria for dilution factors. As a result of comparing the dead time data of 5% or long with the dose of the solution sample, it was concluded that analysis should be performed after dilution when it is about 0.4 μSv/h or high. However, some samples required dilution even at doses below 0.4 μSv/h. Also, re-measurement after dilution, the sample with a dead time of less than 32% was measured with less than 5% when diluted 10 times, and more than 32% required more than 10 times dilution. We suppose that with additional data collection for analyzing these samples in the future, if we can establish clearer criteria, we can predict long dead time samples in advance and solve the problem of analysis delay and sample loss.