Publication:
Reducing non-monotonicities in combat models

Loading...
Thumbnail Image
Authors
Vinyard, William C.
Subjects
Advisors
Lucas, Thomas W.
Borges, Carlos F.
Sanchez, Paul J.
Date of Issue
2001-09
Date
Publisher
Monterey, California. Naval Postgraduate School
Language
Abstract
Non-monotonic behavior in combat models is an important topic to those using the output of such models as a basis for decision making. These decisions can be complicated by non-monotonic behavior in the combat models. This paper examines the Dewar model which exhibits non- monotonic behavior caused by the chaos inherent in its structure. Previous papers have examined only small subsets of this 18 dimensional combat model. The combinatorial possibilities of main effects and interactions among the 18 dimensions are too great to examine en masse. Consequently, we have three goals. First, systematically explore the Dewar model for additional non-monotonic behavior. Second, determine the effect of stochastic modeling on the non-monotonic behavior of the Dewar model response surface. Third, we develop a method for measuring non- monotonicity in the response surface generated by the model. Latin Hypercube Sampling discovers non-monotonicity across broad regions of the model's phase space, and in multiple measures of effectiveness. Stochastic perturbation of model parameters has a dramatic effect on the non-monotonicity of the response surface. Stochastic perturbation can both reduce and exacerbate the non-monotonic behavior of the response surface. If done properly, stochastic modeling can significantly improve the interpretability of the response surface.
Type
Thesis
Description
Series/Report No
Department
Operations Research
Organization
Identifiers
NPS Report Number
Sponsors
Funder
Format
xxvi, 112 p. ;
Citation
Distribution Statement
Rights
This publication is a work of the U.S. Government as defined in Title 17, United States Code, Section 101. Copyright protection is not available for this work in the United States.
Collections