DESIGN FOR NONSYMMETRICAL STATISTICAL DISTRIBUTIONS

This paper presents a method for approximation of the constraint regio n and design centering for statistically distributed design parameters . Arbitrary uncorrelated probability density functions (p.d.f's) are c onsidered, including truncated ones, The paper extends results publish ed in [1], The approximation problem is formulated deterministically a nd solved by optimization, The constraint region is approximated by a family of second-order ellipsoidal type functions, Formulation of the approximation problem takes into account statistical distributions of parameters, to be used in subsequent design centering, A special metri c replaces evaluation of a multidimensional (yield) integral over the constraint region by simple integrals over fixed intervals. An additio nal outcome of the approximation process is location of a nominal desi gn that increases yield.