This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, tor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not Infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do net necessarily state or reflect those of the United States Government or any agency thereof. To be presented in proceedings of 1985 Mg*C Topical o,i Advances in Nuclear Engineering Computational Methods. By •ccapunc* of thl» trticla. tha publithar or raciplant acknowladgat tha U.S. Govarnmant'i right to retain a nonaxclutiva. royalty-fraa flcanta in and to any copyright covaring ch« artlcla. *Rosearch sponsored by the Defense Nuclear Agency, U.S. Department of Energy, under contract DE-AC05-840R21400 with the Martin Marietta Energy Systems, Inc. DISTRIBUTE Cr-r^. .., rBT |MlTEn (c J MONTE CARLO TECHNIQUES FOR ANAL SZ ING DEEP PENETRATION PROBLEMS S. N. Cramer ~ J. Gonnord Qafejiidge,.National.^Laboratory. (i .,^l rtl ,^, h Centre d'Etudes Nucleaires de Saclay J. S. Hendricks _ Los Alamos National^LaboratoryL os Alamos, New Mexicb r B7545 1 ," USA '*' A review of current methods and difficulties in Monte Carlo deep-penetration calculations is presented. Statistical uncertainty is discussed, and recent adjoint optimization of splitting, Russian roulette, and exponential-transformation biasing is reviewed. Other aspects of the random walk and estimation processes are covered, including the relatively new DXANG angular biasing technique. Specific items summarized ai*e albedo scattering, Monte Carlo coupling techniques with discrete ordinates and other methods, adjoint solutions, and multi-group Monte Carlo. The topic of code-generated biasing parameters is presented, including the creation of adjoint importance functions from forward calculations. Finally, current and future work in the area of computer learning and artificial intelligence is discussed in connection with Monte Carlo applications.