Multiple Runs of a Deterministic Combat Model
Abstract
In the past, combat simulations of corps and division level operations required several days to model a few days of combat. Driven by the speed, size, and cost of the computers of that time, it was infeasible, if not impossible, to replicate the runs in a run design. Thus, the investigation of that scale of combat used deterministic models. Today, computer technology offers performance improvements of two to three orders of magnitude, all packaged to fit on a corner of a desk at less than a tenth of the cost of the computers ten years ago. We now have the ability to consider making significantly more runs of large combat simulations. One reason to consider making multiple runs of a deterministic model may be the concept offered by Carl von Clausewitz, who identified chance events as explicit sources of 'general friction', which can potentially turn success into failure and vice versa. This paper presents a deterministic method that produces multiple runs for analysis of the non-monotonic results of Clausewitzian war by varying the initial states of information systems within the model. The results of recent studies serve as case studies to explore the utility of this approach.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 2001
- Accession Number
- ADA406335
Entities
People
- Timothy J. Bailey