Chan, L.K.; Cheng, S.W.; Wang, Z.M. (1991, ASQC) The University of Manitoba, Winnipeg, Manitoba, Canada
This abstract is an edited version of the author's original.
This paper demonstrates how the ideas of Loss Function and Tolerance Design in Dr. G. Taguchi's Quality Engineering may be applied to different types of quality improvement problems. This paper introduces several types of the loss function and tolerance design and illustrates each through a practical example.
Quality engineering consists of: (1) System Design; (2) Parameter Design, and (3) Tolerance Design. System Design involves the setup of product prototype for performing a required function, selections of materials, parts and components, and determination of factors affecting the quality characteristics. The Parameter Design then selects the optimal levels of the component or factor parameters that improve the product quality and make the product robust against the inferences of the imperfection of manufacturing process and changes in usage environment. The Tolerance Design can further improve the quality of those characteristics that still have substantial variations after the Parameter Design through considering the trade-off between the gain due to variation reduction and the increase in manufacturing cost.
The Loss Function, which consolidates the costs to the customer and the producer due to quality loss, provides a numerical evaluation of quality. It is a useful analytical tool for comparing product qualities, evaluating quality improvement, and determining tolerances.
Statistics,Taguchi, Genichi,Tolerances,Loss functions