Managing Uncertainty in Preliminary Aeroshell Design Analysis

Manuscript prepared for submission to the
ASME 1997 Design Theory and Methodology Conference.

Michael J. Scott, Rolf W. Kaiser, Matthew Dilligan
Engineering Design Research Laboratory
California Institute of Technology
Pasadena, California 91125

Robert J. Glaser
Mechanical Systems Engineering and Research Division
Jet Propulsion Laboratory
Pasadena, California 91109

Erik K. Antonsson
Engineering Design Research Laboratory
Division of Engineering and Applied Science
California Institute of Technology
Pasadena, California 91125


An analysis tool to aid in preliminary design of re-entry aeroshells for interplanetary exploration is presented. The computational tool addresses a common feature of design analysis, the need to make use of results from several analysis regimes and from several sources. This paper formalizes for computation what has historically been a matter of informal engineering judgment. The automation of design analysis allows for more extensive search of the design space, and thus supports a design process that is at once more thorough and more efficient.

The human designer uses experience and intuition to combine results from different analyses. Here, this combination is formalized by recognizing that each analysis tool is valid in particular regimes of the design problem; where regimes overlap, a combination based on participation factors (and utilizing the mathematics of fuzzy sets) is employed. An aggregated level of confidence for each solution point is also calculated.