Defining Fidelity for Tradespace Application

This research proposes a definition of tradespace fidelity and presents a study on its effectiveness in the tradespace. Tradespace exploration helps evaluate various design options, allowing engineers to make informed decisions in early design stages. Fidelity generally refers to how closely a model approximates reality. However, the definition of fidelity has been inconsistently defined and interpreted in various contexts concerning particular problems. In some cases, it is a measure of accuracy; in others, information; in another, uncertainty; and in others, computational complexity. Fidelity is poorly and inconsistently defined in the literature. In general, higher fidelity models tend to be more accurate with respect to reality but can be computationally expensive to evaluate and less robust under different evaluation scenarios. Conversely, lower fidelity models are quicker and more flexible but may lack accuracy. The study provides a comprehensive literature review to understand the attributes used by various researchers to define fidelity over time. It also introduces a methodology for a novel approach to defining and quantifying tradespace fidelity using vectors and norms, which allow for a more objective comparison of relatively similar models. This definition allows for trade-offs between model uncertainty and computational complexity to be evaluated to determine the relative fidelity of different models within the tradespace. Effectively, this method evaluates fidelity as a relative metric between models rather than an absolute measure. The study aims to bridge the gap between subjective assessments and objective measurements by quantifying tradespace fidelity, enabling more consistent and reliable trade-offs in tradespace.