| Complex systems design is currently undergoing a paradigm shift toward Design for Capability. In this new paradigm, fewer vehicles are called on to perform a greater number of missions than ever before. As a result, solutions must be more robust to operational uncertainties while maintaining the ability to perform a greater number of tasks. Due to the nature of this goal, top-level needs are well known while specific vehicle requirements are poorly defined. This presents a combinatorial problem in which there are unlimited potential solutions from which to choose a subset of assets that can meet the stated needs. In order to downselect from the vast number of alternative solutions, designers often rely on qualitative methods because there are simply not enough resources available to thoroughly investigate all the potential solutions. However, qualitative information is often based on preconceived notions about what the design should look like, or partial derivatives. With this kind of static information, there is no reliable way to extrapolate how a particular solution might behave in a different environment or in uncertain operating conditions.; For this reason, the ideal is to base concept selection on parametric, quantitative data so that informed, unbiased decisions can be made. However, this kind of information can be expensive and difficult to obtain, which is one reason quantitative analyses are traditionally reserved for optimization or more detailed design after a concept has been selected.; This thesis presents a method for streamlining the process of obtaining and interpreting quantitative data for the purpose of creating a low-fidelity modeling and simulation environment. By providing a more efficient means for obtaining such information, quantitative analyses become much more practical for decision-making in the very early stages of design. However, in capability-based design, where the solution space is essentially unrestricted, we are faced with several common challenges to the creation of quantitative modeling and simulation environments. Namely, a greater number of alternative solutions imply a greater number of design variables as well as larger ranges on those variables. This translates to a high-dimension combinatorial problem. As the size and dimensionality of the solution space gets larger, the number of physically impossible solutions within that space greatly increases. Thus, the ratio of feasible design space to infeasible space decreases, making it much harder to not only obtain a good quantitative sample of the space, but to also make sense of that data. This is especially the case in the early stages of design, where it is not practical to dedicate a great deal of resources to performing thorough, high-fidelity analyses on all the potential solutions. To make quantitative analyses feasible in these early stages of design, a method is needed that allows for a relatively sparse set of information to be collected quickly and efficiently, and yet, that information needs to be meaningful enough with which to base a decision.; The method developed to address this need uses a Systematic Process for Adaptive Concept Exploration (SPACE). In the SPACE method, design space exploration occurs in a sequential fashion; as data is acquired, the sampling scheme adapts to the specific problem at hand. Previously gathered data is used to make inferences about the nature of the problem so that future samples can be taken from the more interesting portions of the design space. Furthermore, the SPACE method identifies those analyses that have significant impacts on the relationships being modeled, so that effort can be focused on acquiring only the most pertinent information.; The SPACE method uses a four-part sampling scheme to efficiently uncover the parametric relationships between the design variables and responses. Step 1 aims to identify the location of infeasible space within the region of interest using an initi... |
