Automating a study question methodology to enhance analysis in high level architecture

摘要

The Department of Defense (DoD) uses simulation for many purposes. Early computer based distributed simulation support environments allowed individual models to communicate with each other but fell short of providing a general distributed simulation solution until the advent of High Level Architecture (HLA). HLA allows users to combine sub-models into one simulation, but it employs a subscription based communications scheme that did not exist in previous support environments. Analysts often use a decompositional approach to identify measures of effectiveness (MOE) measures of performance (MOP), and data requirements for studies and tests. Fundamental study questions or operational requirements are decomposed until supporting data from tests and simulations are identified. This thesis formalizes this decompositional process, calling it the Study Question Methodology (SQM) and procedurally describes the steps all analysts should use to establish a clear audit trail from question to data inputs. It applies the SQM process to a study question relating to attack helicopters to demonstrate the dendritic (tree like decomposition) approach. This thesis also provides a general solution for automating the SQM (ASQM) for use in distributed simulations that use the HLA. The ASQM enhances the analyst's pre, during, and post exercise analysis. It provides the ability to answer study questions, establishes a clear audit trail, and helps fill an analysis tool void that presently exists in HLA.