Office of Maral Research (Code 462) DepartoMflt of the Navy ArUivtoa, Virginia 22217 Under Contract N00014-72-C-0293 ^[^riCU SUTi^KT Appiov.d for public rel«M«. P'tribuaoo ÜBlimif^ '
I«. DEVOnT tICURITV CkAttiriCATION UNCLASSIFIED lb. OKOUP > mtpomr TITLI Techniques for Efficient Monte Carlo Simulation Volume I: Selecting Probability Distributions 4. 01 I (Trp* el ttpon end Inelintwe dele») Final Report s *uTHO"(tJ (Plfl naaw. middle Inlilel, leti nmrn»
DISTRIBUTION STATEMENTReproduction in whole or in part is permitted for any purpose of the U.S. Government.
EXECUTIVE SUMMARYMonte Carlo simulation is one of the most powerful and commonly used techniques for analyzing complex physical problems. Applications can be found in many diverse areas from radiation transport to river basin modeling. Important Navy applications include analysis of antisubmarine warfare exercises and operations, prediction of aircraft or sensor performance, tactical analysis, and matrix gaine solutions where random processes are considered to be of particular importance. The range of applications has been broadening and the size, complexity, and computational effort required have been increasing. However, such developments are expected and desirable since increased realism is concomitant with more complex and extensive problem descriptions.In recognition of such trends, the requirements for improved simulation techniques are becoming more pressing. Unfortunately, methods for achieving greater efficiency are frequently overlooked in developing simulations. This can generally be attributed to one or more of the following reasons:• Analysts usually seek advanced computer systems to perform more complex simulation studies by exploiting increased speed and/or storage capabilities. This is often achieved at a considerably increased expense.• Many efficient simulation methods have evolved for specialized applic itions. For example, some of the most impressive Monte Carlo techniques have been developed in radiation transport, a discipline that does not overlap into areas where even a small number of simulation analysts are working.• Known techniques are not developed to the point where they can be easily understood or applied by even a small fraction of the analysts who are performing simulation studies or developing simulation models.
The use of advanced techniques can greatly improve the effectiveness of Monte Carlo simulation calculations. As a demonstration model, the Navy's Antisubmarine Warfare Air Engagement Model, APAIR, which simulates a single aircraft hunting and destroying a submarine, was selected. Possible improvements in random number generation are presented; however, the study centers on implementation of variance reduction techniques.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.