OPTIMAL PROCESS SETTING FOR UNBALANCED TOLERANCE DESIGN WITH LINEAR LOSS FUNCTION

ABSTRACT Taguchi developed more than 68 loss functions. Among these functions, the quadratic function is the most widely applied one. However, using a quadratic loss function when the actual loss function is nonquadratic might be choosing the incorrect input parameter levels. In certain situations, the linear loss function is more appropriate in industrial applications. If tolerance design is unbalanced, setting the process mean at the nominal will not minimize the expected quality loss. In this paper, the asymmetrical linear loss function is applied in unbalanced tolerance design. The results show that the process mean should be shifted a little from the nominal such that the expected quality loss is minimized.