Game-theoretic methods for analysis of tactical decision-making using agent-based combat simulations

In this paper we provide a motivation and methods for using game-theoretic analysis for combat simulations. Application to three previously studied scenarios using the MANA (Map-Aware Non-Uniform Automata) simulation tool demonstrates theefficacyofour approach, and we provide a concrete example of the differences between the game-theoretic and the more standard decision-theoretic approaches. We believe that the game-theoretic simulation analysis framework can enhance the understanding oftacticalcombat landscape, especially in cases where intelligence efforts are ineffective or provide unreliable information, because in those cases game-theoretic analysis may help to corroborate or challenge intelligence-limited beliefs by assessing the rationality of the opponent's apparent strategy. We suggest that game-theoretic simulation analysiscould often bestbeused during the low-resolution, exploratory-analysis phase of a multiresolution analysis.