This paper presents a best estimate plus uncertainty (BEPU) methodology applied to dryout, or critical channel power (CCP), modeling based on a Monte Carlo approach. This method involves the identification of the sources of uncertainty and the development of error models for the characterization and separation of epistemic and aleatory uncertainties associated with the CCP parameter. Furthermore, the proposed method facilitates the use of actual operational data leading to improvements over traditional methods, such as sensitivity analysis, which assume parametric models that may not accurately capture the possible complex statistical structures in the system input and responses.