Author

Jung-Dae Kim

Date of Graduation

1998

Document Type

Dissertation/Thesis

Abstract

The quadratic loss function has been used in the statistical and control theory literature. Taguchi applied quadratic loss function to quality improvement. His quality loss function has been known to be effective in assessing quality loss in monetary terms, while most of the traditional quality control techniques, such as tolerance design and control chart design, use estimated cost factors. Taguchi's quality loss function concept is different from traditional methods in which all the values within the specification limits are equally acceptable. However, his idea is that for many products, parts elements and components, a target or nominal size or value is preferred. Therefore, by introducing Taguchi's quality loss function, it is possible to develop more tight and economic tolerances and design more effective and economical control charts. The objective of this research was to apply Taguchi's quality loss function concept to tolerance design and cost based CUSUM control chart design. In this research, a modified quality loss function was developed to reflect quality loss situation where general quadratic loss curve is not appropriate. By using the quality loss function concept and the production cost, a new optimal tolerance model which minimizes total cost function was developed which apply specifically to cutting and assembly operations. This research also provided a methodology for the design of cumulative sum control (CUSUM) charts using Taguchi quality loss function concept based on the minimum cost per hour criterion. The new model differs from previous models in that the model assumes that quality loss is incurred even in the in-control period. Thjs model was compared with two other cost based CUSUM models by Wu and Goel, and Chiu. According to numerical sensitivity analysis, the new model results in longer average run length in in-control period compared to the other two models.

Share

COinS