Design-of-experiment methods for computational parametric studies in electronic packaging

Abstract

Design-of-experiment methods are utilized to improve the tractability of numerically intensive design problems in reliable electronic packaging. Fractional factorial parametric studies, commonly used by experimentalists to economize on resources, are shown to provide similar benefits in design simulations also. Response surface analysis is used to develop pseudo-closed-form models from the computational parametric studies. An example is provided to illustrate the use of these concepts and tools in electronic packaging design.