This paper presents a method for the optimization of processes or products measured by multiple responses. All the steps of the method, which covers from problem identification to quality and costs optimization, are described. To illustrate the use of the method, an optimization study conducted on a casting process is discussed. This case study was first reported by Chang e Shivpuri (1995). This paper presents a comparison of the results obtained by these authors and those obtained through the use of the proposed method. The advantages of the proposed method are discussed and pointed out.
Quality; Optimization; Design of Experiments; Multiple Regression; Loss Function