Integrating response surface methods and uncertainty analysis into ship concept exploration
Price, Shelly L. (Shelly Loustaunau)
MetadataShow full item record
The concept design phase of any type of ship determines the hull form, baseline capabilities, and a large portion of the total program cost. The complexity of the ship design process leads to numerous assumptions and a great deal of uncertainty in the point designs during the concept exploration phase. While it is not feasible to eliminate this uncertainty, it is useful to explore how it affects the overall design. An analysis of the uncertainty associated with each point design provides the designer with additional information for comparing designs. It is important to consider all options and choose the baseline design that best meets the customer's requirements. Current trade-off studies tend to examine a few point designs that may or may not cover the entire design space. This approach relies heavily on designer experience, is inefficient, and may not lead to an optimum baseline design. Response Surface Methods (RSM) provide statistical tools for determining the relationships between factors (inputs) and responses (outputs). When combined with Design of Experiments (DOE), this approach allows the designer to thoroughly investigate the design space using relatively few point designs. The benefit of this analysis is the ability to efficiently examine the effects of changing factors on the overall design. Finally, the combination of RSM and an uncertainty analysis gives the designer a tremendous understanding of the design space. The thesis develops a method allowing the designer to make important decisions, such as hull form or basic mission capabilities of the ship, explicitly showing the uncertainty associated with key design parameters.
CIVINS (Civilian Institutions) Thesis document
Showing items related by title, author, creator and subject.
Bergman, Mark; Lyytinen, Kalle; Mark, Gloria (2007);Traditionally, Systems Analysis and Design (SAD) research has focused on ways of working and ways of modeling. Design ecology – the task, organizational and political context surrounding design – is less well understood. ...
Giachetti, Ronald E. (Monterey, California. Naval Postgraduate School, 2014-03-18); NPS-AM-14-015; NPS-AM-14-015This research addresses the problem of how to (1) value architectural options that deliver capabilities to the warfighter not inherently measured in dollar values and (2) conduct a trade study of architectural options, ...
Kleijnen, Jack P.C.; Cioppa, Thomas M.; Sanchez, Susan M.; Lucas, Thomas W. (2005);Many simulation practitioners can get more from their analyses by using the statistical theory on design of experiments (DOE) developed specifically for exploring computer models. In this paper, we discuss a toolkit of ...