Minimum Loss Design of T SQUARE the Bivariate Control Chart

• Main Authors: Chang Yi-Kun, 張義坤
• Other Authors: Chao-Yu Chou
• Format: Others
• Language: zh-TW
• Published: 1998
• Online Access: http://ndltd.ncl.edu.tw/handle/03965166858648883847

碩士 === 國立雲林科技大學 === 工業工程與管理研究所 === 86 === The economic design of control have been developed since the 1960s.The cost function of the traditional economic design of control charts is focused on minimizing the poduction cost , and consequently ignoring the society loss after the products are shipped . In this project , we add the bivariate loss function to the cost function of the traditional economic design and modify the cost function model of the loss design of the multivariate control chart, and then develop the minimum loss design of the bivariate control chart. In the minimum loss design of the bivariate control chart , there are three parameters to be determined ： the sample size (n) , the production volume between two consecutive samplings, and the control limits of the control chart. The objective function of this design is to minimize the total cost per unit , including both production cost society loss. The concept of loss function is introduced by Dr G . Taguchi . In the expression of loss function , the society loss occurs as long as the concerned quality characteristic deviates from its target. This concept is more consistent with "continuous improvement " which is emphasized by total quality management ( T Q M ) . Reducing the variability among the products and making the quality characteristic as close as its target is a basic way to decrease the quality cost.