Process parameters and screening limits settings under the non-normal process

Abstract

For the traditional statistical quality control model, one usually assumes that the quality characteristic of product is normally distributed with unknown mean and variance. However, the product characteristic is different from normal distribution in many production processes. Chen and Kao (2009) presented a bi-variate normal distribution between product and screening characteristics for determining the optimal process mean and screening limits. In this paper, the author proposes the modified Chen and Kao's (2009) model with bivariate Burr distribution. The optimal process mean, standard deviation and screening limits of product will be simultaneously determined by minimizing the expected total cost of product including the quality loss of conforming product, inspection cost, rework cost, scrap cost, and penalty cost for product / screening characteristics.