Premium
A stochastic version of the concepts evaluation model (CEM)
Author(s) -
Johnson Ralph E.,
Isensee Ernst K.,
Allison William T.
Publication year - 1995
Publication title -
naval research logistics (nrl)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.665
H-Index - 68
eISSN - 1520-6750
pISSN - 0894-069X
DOI - 10.1002/1520-6750(199503)42:2<233::aid-nav3220420207>3.0.co;2-g
Subject(s) - operations research , computer science , range (aeronautics) , stochastic modelling , agency (philosophy) , sample (material) , government (linguistics) , attrition , domain (mathematical analysis) , stochastic process , container (type theory) , statistics , engineering , mathematics , medicine , mechanical engineering , mathematical analysis , philosophy , chemistry , linguistics , dentistry , epistemology , chromatography , aerospace engineering
The concepts evaluation model (CEM) is a computer simulation model of ground and air warfare operations that is used by the U.S. Army Concepts Analysis Agency (CAA) to conduct analysis of the capabilities and requirements of forces engaged in warfare at theater level. The CEM has been applied to campaign analyses for numerous scenarios since the early 1970s, including Central Europe, Korea, Iran, and Iraq theaters of operation. The standard CEM is fully automated and deterministic, yielding a single outcome for any situation simulated, providing no confidence intervals, range, nor distribution of possible outcomes. Modern, faster computers have now reduced CEM execution time to a level that makes multiple replications of the CEM feasible. In this project a stochastic version of the CEM has been developed that makes use of individual replications of stochastic attrition input data, rather than averaged sample data, and that simulates commanders' decisions, the disposition of casualties and of combat‐damaged vehicles, and certain other functions based on statistical distributions rather than on expected values. This article reports the methodology and results of an analysis of this stochastic version of CEM, indicating which stochastic features most influenced the variability among replications of one simulated campaign and outlining costs and benefits of using a stochastic version of the CEM. © 1995 John Wiley & Sons, Inc. This article is a US Government work and, as such, is in the public domain in the United States of America.