Propagating Uncertainty in Hierarchical Combat Models

Abstract

The Office of the Chief of Naval Operations (OPNAV) uses a hierarchy of simulation models as part of scenario-based planning to help decide how the Navy should be equipped, organized, and employed. Simulation is used throughout the acquisition process, from platform design to force structure analysis. In hierarchical combat modeling, the mean outputs of lower-level, higher-resolution models are typically used as inputs to higher-level, lower-resolution models. The objective of this process is to inform Navy leadership on how detailed design changes ultimately impact campaign effectiveness. Unfortunately, by ignoring variability in linkages between layers in the hierarchy, the results may bias campaign-level outcomes or understate the final variability (or risk) estimated by the campaign-level model. The primary goal of this research was to design and run experiments to better understand the impacts on the hierarchical modeling process associated with error propagation methods and design of experiments techniques. Another objective was to developed new algorithms that improve upon our ability to explore high-dimensional combat models. To empirically explore a host of different error propagation approaches, this research conducted thousands of experiments using a two-model hierarchical structure in an air- to-air warfare setting. The results indicate that the manner in which the engagement and campaign models are linked significantly affects the estimates of operational effectiveness and risk. In addition, this research advanced our ability to explore combat models and fit response surfaces through the development and testing of new metamodeling approaches and sequential Latin hypercube methods.