Background: The ASTRUM (Automated Statistical Treatment of Uncertainty Method) methodology is developed by Westinghouse. In this method

some important parameters for large break loss of coolant accident (LBLOCA) analysis are treated conservatively to get a conservative set

which is called confirmatory study. And the value set of parameters is then used to generate the reference transient input deck of ASTRUM calculation. The value set may be varied in different sampling cases. However

the ASTRUM is executed based on one value set of some parameters

although other parameters are sampled. The result of the ASTRUM calculation may be not very exact. Purpose: The aim of this study is to reduce the uncertainty and conservatism

and optimize the ASTRUM. Methods: The parameters concerned in the confirmatory study of ASTRUM were sampled by BE_SAMPLE code developed by this paper for each ASTRUM case. The optimized ASTRUM is executed in company with other sampled parameters. Results: The results of the original and optimized ASTRUM calculations are compared

and the peak cladding temperature (PCT) calculated by new ASTRUM is lower than the original result. Conclusion: The study on ASTRUM optimization reduces the uncertainty caused by confirmatory study

which makes the ASTRUM calculation much more precise. Abstract