Discrete Event Simulation Modeling and Analysis of Key Leader Engagements.

摘要

The Cultural Geography (CG) Model is a low-resolution, agent-based discrete event social simulation tailored to specific operational environments. It is based on doctrine and social theory that are designed to represent the behavioral response of civilian populations in conflict environments. The current version of the CG Model does not represent key leader engagements (KLEs), which are activities between coalition military forces and host nation civilian personnel as means of obtaining information, influencing behavior, and building an indigenous base of support for coalition and government objectives. These capabilities are needed for additional tactical-level representation of the operational environment. This research develops a simulation model using Simkit to explore the feasibility of modeling KLEs using discrete event simulation. A total of 32 dynamic input factors are varied using a 512-design point design. Second-order regression metamodels and partition tree models are developed for simulation model output responses that track numbers of engagements, numbers of times knowledge is provided, numbers of campaigns, and numbers of captures and kills. These analytical models are used to verify the proper execution of the simulation model. Summary statistics are analyzed to gain further insights about the simulation model's behavior.