Automatic Variance Reduction for Monte Carlo Shielding Calculations with the Discrete Ordinates Adjoint Function

Although the Monte Carlo method is considered to be the most accurate method available for solving radiation transport problems its applicability is limited by its exorbitant computational expense. Thus, variance reduction techniques, which require intuition, guess work, and iterations involving manual adjustments, are employed to make reactor calculations feasible. Responding to this difficulty, we have developed methodology for automatic variance reduction of Monte Carlo shielding calculations using the discrete ordinates adjoint function for source biasing and consistent transport biasing with the weight window technique. We describe the implementation of this method into the standard production Monte Carlo code MCNP and its application to a realistic calculation; namely the reactor cavity dosimetry calculation. The effectiveness of the method, as demonstrated through the increase in calculations efficiency, is quantified. Moreover, important issues associated with this method and its efficient use are discussed. The increases in performance associated with this method can substantially increase the practical applicability of the Monte Carlo method for large real-world applications.