Maximization of manufacturing yield of systems with arbitrary distributions of component values

This paper presents a general method for maximizing manufacturing yield whe n the realizations of system components are independent random variables wi th arbitrary distributions. Design specifications define a feasible region which, in the nonlinear case, is linearized using a first-order approximati on. The method attempts to place the given tolerance hypercube of the uncer tain parameters such that the area with higher yield lies in the feasible r egion. The yield is estimated by using the joint cumulative density functio n over the portion of the tolerance hypercube that is contained in the feas ible region. A double-bounded density function is used to approximate vario us bounded distributions for which optimal designs are demonstrated on a tu torial example. Monte Carlo simulation is used to evaluate the actual yield s of optimal designs.