Development Approaches Coupled with Verification and Validation Methodologies for Agent-Based Mission-Level Analytical Combat Simulations
Date of Award
Doctor of Philosophy (PhD)
Department of Operational Sciences
Raymond R. Hill, PhD
This research investigated the applicability of agent-based combat simulations to real-world combat operations. An agent-based simulation of the Allied offensive search for German U-Boats in the Bay of Biscay during World War II was constructed, extending the state-of-the-art in agent-based combat simulations, bridging the gap between the current level of agent-like combat simulations and the concept of agent-based simulations found in the broader literature. The proposed simulation advances agent-based combat simulations to “validateable” mission-level military operations. Simulation validation is a complex task with numerous, diverse techniques available and levels of validation differing significantly among simulations and applications. This research presents a verification and validation taxonomy based on face validity, empirical validity, and theoretical validity, extending the verification and validation knowledge-base to include techniques specific to agent-based models. The verification and validation techniques are demonstrated in a Bay of Biscay case study. Validating combat operations pose particular problems due to the infrequency of real-world occurrences to serve as simulation validation cases; often just a single validation comparison can be made. This means comparisons to the underlying stochastic process are not possible without significant loss of statistical confidence. This research also presents a statistical validation methodology based on re-sampling historical outcomes, which when coupled with the traditional nonparametric sign test, allows comparison between a simulation and historic operation providing an improved validation indicator beyond the single pass or fail test.
DTIC Accession Number
Champagne, Lance E., "Development Approaches Coupled with Verification and Validation Methodologies for Agent-Based Mission-Level Analytical Combat Simulations" (2004). Theses and Dissertations. 3901.