Optimum Process Targets for Multiple Quality Characteristics Using Regression Analysis

摘要: The selection of optimum process target has become one the focused research areas to increase productivity and improve product quality. Although quality engineering literature contains a vast collection work related this issue, couple questions still remain unanswered. First, loss function approach using conventional functions, such as step-loss Taguchi been extensively used determine mainly due mathematical convenience. When historical data concerning customer associated with performance are available, well-established statistical method, regression analysis, may be more practical alternative procedure. Second, many researchers have carried out their studies based on single characteristic. From customer's viewpoint, however, products often judged than To address these questions, paper first develops multivariate empirical its then proposes an optimization scheme for most economical target.